INTRODUCTION TO THE 
HISTOLOGY AND HISTOPATHOLOGY 
OF THE NERVOUS SYSTEM 
BY 
DR. PAUL SCHRODER 
PROFESSOR OF PSYCHIATRY AND NEUROLOGY IN GREIFSWALD 
AUTHORIZED TRANSLATION FROM THE SECOND 
REVISED GERMAN EDITION 
BY 
BALDWIN LUCRE, M.D. 
ASSISTANT PROFESSOR OF PATHOLOGY, SCHOOL OF MEDICINE, 
UNIVERSITY OF PENNSYLVANIA 
AND 
MORTON McCUTCHEON, M.D. 
_ INSTRUCTOR IN PATHOLOGY, SCHOOL OF MEDICINE, 
UNIVERSITY OF PENNSYLVANIA 
WITH 53 ILLUSTRATIONS 
PHILADELPHIA AND LONDON 
J. B. LIPPINCOTT COMPANY Copyright, 1923, by J. B. Lippincott Company 
PRINTED BY J. B. LIPPINCOTT COMPANY 
AT THE WASHINGTON SQUARE PRESS 
PHILADELPHIA, U. S. A AUTHOR’S PREFACE TO THE FIRST 
EDITION 
The following lectures do not claim to be a complete 
presentation of all that is known in the field of the histol- 
ogy and histopathology of the nervous system. They 
have been given for a number of semesters as lectures 
introductory to the demonstration of microscopic prepara- 
tions on the projection apparatus. They treat, for the 
most part, of general fundamental questions. Histological 
and histopathological minutiae have been presented only 
where it seemed unavoidable. No illustrations have been 
included, for these are of value only when they are numer- 
ous and good, and such are costly. References have been 
given particularly to such papers as contain good illustra- 
tions. Histological methods have been brought into the 
discussion in many places, since clear comprehension and 
interpretation of our observations are largely dependent on 
the understanding and evaluation of the methods used. 
The lectures were originally intended to be confined to 
the histopathology of the nervous system, but it was soon 
found that such discussions were not practicable without 
previous agreement concerning the normal histological 
structure. To assume any previous familiarity with the 
subject was not possible, because of the gaps in our pres- 
ent knowledge, and because of differences in interpreta- 
tion of apparently contradictory histological observations. 
Therefore the pathological part has been preceded by a 
histological part, which is more or less independent of the 
III PREFACE TO THE FIRST EDITION 
IV 
former, so that it may be used even by those who are less 
interested in pathological questions. 
In the present state of our knowledge of the histology 
and liistopathology of the nervous system, any presenta- 
tion of the subject which is not to be merely a compilation 
of the literature, must necessarily have a strong subjective 
coloring. This is exemplified in the choice of observations 
and conclusions of others, regarded as important and 
correct. No extended discussion of conflicting opinions 
has been given, for the reason that a brief coherent 
presentation of the more important questions was aimed 
at, rather than a complete presentation of the entire field. 
Any one familiar with the subject will easily recognize 
the influence of the investigations of Nissl upon the 
treatment of the material and the selection of the themes. 
Nissl’s researches form the foundation of a large part of 
the entire presentation. To his teaching and influence 
the author owes the pleasure which he has found in the 
subjects treated here. 
Paul Schroder. AUTHOR’S PREFACE TO THE SECOND 
EDITION 
The first edition has been criticised because of the 
absence of illustrations; this lias been remedied in the 
second edition. By agreement with the publisher, instead 
of expensive lithographs from drawings, half-tones, mostly 
from microphotographs, have been used. A part of the 
illustrations has been borrowed from the works of the 
great masters in the field of histology and liistopathology 
of the nerve cell and neuroglia—Nissl, Alzheimer, Bethe, 
Bielschowsky, Held. Others are reproductions from 
former publications of the author. Illustrations have 
been limited to those having particular instructive value 
since their purpose in a book such as the present one is to 
help in the understanding of the more difficult portions 
of the text. 
The general character of the work has not been altered. 
It is an introduction for those who wish to become familiar 
with the subject, and is designed as a stimulus to further 
independent investigation. It is no more a text- or refer- 
ence-book than was the first edition. 
The text has been thoroughly revised, additional mat- 
ter has been added and whole sections have been rewrit- 
ten; this applies particularly to the Second Chapter, on 
the neurofibrils; the Fourth, on the lymph paths; the 
Sixth, on the vessels; and the Eighth, on inflammation. 
Paul Schroder. 
V TRANSLATORS’ NOTE 
This book is what its name implies: an introduction to 
the histology and histopathology of the nervous system. 
It treats of general and fundamental questions, of like 
importance to the pathologist, the physiologist, the 
histologist and the neurologist. The separation of the 
text into histological and histopathological parts renders 
it of more ready value to those interested only in one or 
the other phase of the subject. 
Dr. Schroder himself kindly reviewed the translation 
within the past few months and added many new refer- 
ences as well as some new material. 
The translators have attempted to render the book 
into readable English; yet where it seemed best they have 
sacrificed smooth diction to close adherence to the text. 
In controversial subjects they have not made any com- 
ments, believing that the reader had best form his own 
opinion from the references cited. The translators have 
added an index. 
Our thanks are due to Professors A. J. Smith and 
W. H. F. Addison for kindly advice. 
University of Pennsylvania, 
June, 1923. 
Baldwin Lucre, 
Morton McCutcheon. 
VI CONTENTS 
PART I: HISTOLOGY OF THE NERVOUS SYSTEM. 
Chapter I. page 
Introduction 1 
Ganglion Cells 3 
Methods of demonstration. Nissl’s method of ganglion-cell staining, its use in 
pathology. Some types of ganglion cells in Nissl’s preparations. Significance 
of the stainable portions. The doctrine of the ganglion cell—equivalent 
picture. The non-stainable substances of the Nissl picture. The pigment. 
The nucleus. The shape of the ganglion cells. 
Chapter II. 
The Neurofibrils 22 
Apathy’s observations on fibrils. Bethe’s observations. The neurofibrils in 
lower and higher animals. The pericellular apparatus of vertebrates. The 
Golgi method. The neurone doctrine. 
Chapter III. 
The Nerve Fibres 37 
Their histological structure. Differences between peripheral and central fibres. 
Regeneration of nerve fibres. The investigations of Reic.i and others on 
peripheral nerves. 
The neuroglia. Its ectodermal origin. Weigert’s doctrine of the glia. The in- 
vestigations of Held; the glia a syncytium. The glia in the Nissl preparation. 
Chapter IV. 
The Mesodermal Tissue in the Nervous System 55 
Its amount, its behavior. The blood supply of the central nervous system. The 
structure of the vessel walls, their several layers. 
The lymph paths of the central organs. Subdural, subarachnoidal, epicerebral 
spaces. Lymph paths in the interior of brain and spinal cord; the adventitial 
and perivascular lymph spaces; confusion of these with artificial shrinkage 
spaces. 
The Histological Structure of the Nervous Tissue 68 
White and gray substance. 
VII VIII 
CONTENTS 
PART II: HISTOPATHOLOGY OF THE NERVOUS SYSTEM. 
Chapter V. page 
General Preliminary Remarks 76 
Localization and trend of investigation in histopathology. Choice of staining 
methods. The elective staining methods. 
The Pathology of Ganglion Cells 79 
Early over-optimistic hopes, and the reasons for these. Nissl’s fundamental 
investigation. General significance of ganglion cell changes. Use of Golgi 
method in pathology. Fibril stains. The most important types of cellular 
disease, according to Nissl. 
Chapter VI. 
The Pathological Changes of the Glia 91 
General laws. Glia fibrils and glia plasma. The space-filling function of the 
glia, its phagocytic function, the gliogenic granule cell. Pathological glia 
forms in the Nissl picture. Their significance for histopathology. Patho- 
logical changes in the vessel walls and the connective tissue. Arteriosclerosis 
and endarteritis. 
Chapter VII. 
Pathology of the Nerve Fibres 110 
Different behavior of the peripheral and the central nerve fibres. Disintegration 
of the myelin sheath. The secondary degenerations. The Wallerian law. 
The histopathological changes in secondary degeneration of peripheral nerves, 
of central nerve fibres. The discontinuous disintegration of the myelin 
sheath. So-called anemic foci. 
Chapter VIII. 
Some Histopathological Complexes 127 
The ectodermal type, the mesodermal type. Reparatory processes in hemor- 
rhage, softening. 
Granule Cells 131 
Origin of the granule cells in the glia, proliferating vessel wall elements, neuri- 
lemma cells. 
Some General Pathological Concepts. Inflammation 137 
The development of this concept. Small round-cel! infiltration, so-called perivas- 
cular round-cell infiltration; hemorrhagic inflammation. PART I. 
HISTOLOGY OF NERVOUS 
SYSTEM 
CHAPTER I. 
INTRODUCTION THE GANGLION CELLS. 
The anatomy of the central nervous system is a well- 
tilled field of research, but in spite of this fact our 
knowledge of structure is still inadequate as a basis for 
physiology and pathology. 
That part of our knowledge which relates to external 
form, to general structure, and to division into white and 
gray matter may in a sense be looked upon as complete. 
As regards the fibre tracts, little has been learned in the 
last few decades; indeed a whole series of fundamental 
questions cannot be answered for lack of adequate meth- 
ods, and, in particular, our knowledge of the fibre tracts 
of the cerebral hemispheres is even now rather scanty. 
Recently, chief interest has been directed to another 
branch of neuroanatomy, the finer histology. A mass of 
observations has been collected, for the most part quite 
new, although there are many contradictions as to details, 
and not all of the data can be regarded as permanent 
contributions to our knowledge. 
Hand in hand with this a similar change appeared in 
the field of pathological anatomy of the central nervous 
system. This subject was dominated in the last third of 
the nineteenth century by problems relating to localiza- 
tion and to fibre tracts, that is, by attempts to determine 
1 INTRODUCTION—THE GANGLION CELLS 
2 
exactly the seat of morbid processes and to follow the 
secondary degenerations resulting therefrom. Meanwhile, 
the study of the finer structural changes was neglected. 
This trend of research was undoubtedly due to the ad- 
vances made by the physiologists and anatomists in the 
study of localization. It followed that neurology and 
brain pathology made rapid gains, while psychiatry 
profited but little. The return to histology, so neglected 
during this time, was init iated by Karl Weigert and Franz 
Nissl. Both these men advanced our knowledge of the 
normal histology of the central nervous system to an 
unprecedented extent, Weigert concerning himself chiefly 
with the neuroglia, Nissl with the ganglion cells; while 
both showed us how to apply these new facts to liisto- 
pathology. In particular the researches of Nissl on mor- 
bid changes in the nerve cells led to great hopes for the 
histopathology of the central nervous system, and stimu- 
lated others to produce an immense amount of work. 
Nissl himself warned us that instead of centring our 
attention on the ganglion cells we should take into con- 
sideration all the elements making up the nervous system, 
without regard to prevailing opinions on their relative 
functional importance. To this study we must bring the 
principles of general pathological anatomy, correlating 
tissue alterations with clinical pictures. In this way we 
might possibly find that alterations in the supporting 
tissues (the glia and the connective tissue—vascular 
apparatus) prove more important than the changes in the 
specific nervous elements (nerve cells and nerve fibres).1 
Moreover, the researches of men like Held, Alzheimer, 
Bethe, Reich, Brodmann, Bielschowsky, and others, have 
1 Nissl, Zentralbl. f. Nervenheilkunde u. Psychiatric, 1903, p. 519. GANGLION CELLS 
3 
enriched our knowledge in the field of histology and liisto- 
patliology in numerous details, partly in entirely new and 
unexpected directions. If we wish to understand this 
progress, it must first of all be made clear what we may 
look upon as definitely established in our knowledge of 
normal structure; later on, when we come to study the 
histology of pathological changes, we will find many indi- 
cations that our knowledge does in fact rest upon a firm 
foundation of truth. 
Ganglion Cells.—Among the structural elements 
of the nervous system which we wish to discuss, at first 
separately, and then in their relation to the whole of the 
nervous tissues, the most important are, according to 
orthodox views, the ganglion cells. Whether they really 
have the supreme importance for the function of the 
organ that has long been attributed to them, has become 
somewhat questionable. 
For their staining, we possess a whole series of 
methods. The older diffusely staining substances (car- 
min, nigrosin, hfematoxylin, van Gieson’s, etc.) give for 
the most part indefinite pictures, which are especially 
difficult to interpret under pathological conditions; while 
fixation by chromic salts (Muller’s fluid), formerly so 
generally used, causes numerous artefacts. Our definite 
knowledge of the characteristics of ganglion cells we owe 
to three so-called specific methods—the Nissl, the Golgi, 
and the methods of fibril-staining. These three are 
entirely different in principle, and of all of them there are 
numerous modifications, which however lead only to 
unimportant differences in the microscopic pictures.2 
2 There are several such methods for the elective demonstration of the neuro- 
fibrils. 4 
INTRODUCTION—THE GANGLION CELLS 
These three different methods give very dissimilar pic- 
tures of the nerve cells; thus the silver impregnation of 
Golgi gives black silhouettes of the entire cell body 
(Fig. 1); the methods for the demonstration of fibrils 
(Fig. 13 and 14) show only the nucleus and fibrils, but no 
other cell substances; while Nissl’s method stains certain 
constituents of the nucleus (but in a different way), and 
certain portions of the proto- 
plasm lying between the fibrils, 
but it does not stain the fibrils 
themselves. (Figs. 2 and 4.) 
The most important method 
for pathology, and the one that 
hitherto has been most fre- 
quently and successfully used for 
the staining of ganglion cells, is 
that of Nissl. 
Nissl called it a specific method 
for the staining of nerve cells, but 
this is not to be taken literally. 
The method, to be sure, does 
not stain the ground substance 
nor the nerve fibres; hence the 
distinctness with which the cell bodies are brought out. It 
does stain certain parts of all cell nuclei, not only of the 
ganglion cells, but also of the glia, the connective tissue, 
and the vascular elements. In the cell body, only certain 
substances of the cytoplasm are intensely colored. And 
such stainable substances are normally present, in the 
nervous system, only within ganglion cells.3 The bodies, 
Fig. 1.—From the deep layers of the 
visual cortex of a young cat. Golgi 
preparation. After Cajal. (a) large 
pyramidal cell, (c) smaller pyramidal 
cells with branching axis-cylinder proc- 
esses. (g) small pyramidal cell. (d) 
axis-cylinder processes. 
3 There are also ganglion cells which possess only extremely small quantities 
of these stainable substances. GANGLION CELLS 
5 
particularly those of the large nerve cells, stain very deeply. 
But in morbidly changed parts of the nervous system, con- 
ditions are considerably different; all sorts of tissue cells, 
particularly those which are youthful or proliferating, may 
form protoplasmic substances which are stained more or 
less distinctly with Nissl ’s method. Among these are in- 
cluded plasma cells, mast cells, fibroblasts, and most of the 
progressively altered glial elements. This property elimi- 
nates Nissl’s method as a specific staining method for nerve 
cells, but is the basis for its great usefulness in pathological 
anatomy even where the recognition of structural changes 
in ganglion cells is not concerned. In particular, the method 
lends itself to the demonstration of proliferating glia, as 
well as of alterations and pathological elements in the 
connective tissue-vascular apparatus. What may be 
seen in the ganglion cells in Nissl preparations has been 
described most thoroughly and extensively by Nissl 
himself. Later authors have added little of real impor- 
tance.4 Nissl distinguishes, in the cell body, those parts 
which retain the basic aniline dye after differentiation, 
and those which do not retain it; he speaks, in a purely 
descriptive sense, of stainable and of non-stainable sub- 
stances (Figs. 2 and 4). 
The stainable parts are tinged in part deeply, in part 
less deeply. With high magnification they are seen as 
smaller or larger granules, as groups, rows, or chains of 
granules, or as larger bodies of regular or irregular shape 
(spindles, nuclear caps and branch-cones). Between the 
4 The following description of the normal morphology of the ganglion cells is 
based chiefly upon publications by Nissl, which are scattered through various journals; 
cf. especially, Neurol. Zentralblatt, Vol. XIII and XIV; Allgem. Zeitschr. fur Psychiatrie, 
Vol. LIV; Beitrdge zur Frage nach der Beziehung zwischen klinischem Verlauf und anato- 
mischem Befund, Vol. I, Nos. 1-3, Berlin 1913-15, Julius Springer. 6 
INTRODUCTION—THE GANGLION CELLS 
stainable portions there run small or broad unstained 
tracts, which, as Nissl supposed, and as was later demon- 
strated to be the case, represent the paths of the neuro- 
fibril bundles through the cell body. The stainable parts 
are continued, chiefly in the form of thin, long spindles, 
for a distance into the dendritic processes of the cell. The 
axis-cylinder process is always free from stainable sub- 
stances. The various forms, sizes, numbers, and grouping 
of the stainable parts produce a great variety of pictures, 
but it is not difficult to recognize a certain number of fre- 
quently recurring types, and to learn that, as an impor- 
tant rule, such typical structural forms are characteristic 
of certain localities in the central nervous organs. The 
position and form of stainable portions in the protoplasmic 
body are just as constant for a certain type of cell as are 
the size of the cell, its external form, its nucleus, and so on. 
Of such types, only a few of the more generally known 
ones are here given, according to Nissl’s description. 
The Large Motor Cells.—They are called motor 
cells because we know that they are somehow closely 
related to motor function. They are found in the spinal 
cord as the large cells of the anterior horns, and in their 
analogues, the motor nuclei of the medulla oblongata; 
also as the so-called Betz giant pyramidal cells in the 
motor region of the cerebral cortex. In the spinal cord 
and the medulla oblongata they are more spherical in 
form, and they send out dendrites of equal size in all 
directions. In the cerebral cortex they take on the well- 
known pyramidal form. In external form and structure 
the separate elements of the same region show striking 
similarity, but there is also considerable individual varia- 
tion. They constitute the largest nerve cells occurring THE LARGE MOTOR CELLS 
7 
in the human central organs. In general, they exhibit 
rather large stainable portions which are clearly differ- 
entiated from the unstained paths. Their striped pattern 
is typical for these cells and is most plainly seen at the 
base of the processes. The tracts which stream in from 
Spindle shaped 
stainable sub- 
stance 
Unstained paths 
Bifurcation cone 
■Unstained paths 
Bifurcation cone 
•Bifurcation cone 
Spindle shaped 
stainable sub- 
stance 
Unstained paths 
Unstained paths 
Nuclear cap 
-Basal Nissl body 
\ Mesial basal dendrite 
\ (this is not the axis- 
\ cylinder process) 
\ Points of attachment of lateral 
\ basal dendrites 
Points of attachment of lateral 
basal dendrites 
Fig. 2.—Normal large pyramidal cell from the third layer of the human frontal cortex. Nissl 
stain. After Nissl and Eanke. Magnification x 930. 
the dendrites are spirally twisted. These cells possess a 
relatively large horse-shoe shaped axone hillock, which 
does not contain stainable substances, and which passes 8 
INTRODUCTION—THE GANGLION CELLS 
into the short, broadly lance-shaped axis-cylinder process; 
in the Betz cells of the cortex it is always situated at the 
base of the cell, or at the point of exit of one of the large 
basal dendrites. The motor cells are further characterized 
by a large pale nucleus, which contains a large nucleolus, 
the nuclear membrane being usually invisible because it 
Small nerve 
-cell (not in 
!focusj 
Remnants of 
stainable sub- 
stance 
•Glia nuclei 
(Glia nuclei 
Protoplasm of the 
axis-cylinder proc-| 
ess. 
I Axis-cylinder 
-jprocesses 
Fig. 3.—Two large pyramidal cells of frontal cortex in state of “acute degeneration.” Nissl 
stain. After Nissl and Ranke. Magnification x 930. Compare with Fig. 2. 
is covered up by the adjacent stainable substances of the 
cell body. 
The large pyramidal cells of the third layer of the human 
cerebral cortex (Fig. 2) rarely reach the size of the giant 
pyramidal cells of Betz. They likewise have large pale 
nuclei, which, however, can be always recognized from 
the fact that the nuclear membrane is definitely visible. 
Adjacent to it are saucer- and shield-shaped stainable 
substances. At the base of the cell body there are similar THE SPINAL GANGLION CELLS 
9 
but larger masses without definite arrangement. The 
apical dendrite may exactly resemble that of a motor 
cell; but usually it contains only pale fine threads and 
rows of granules, and no axone hillock is present. The 
Section through part of a nerve cell 
\ Bifurcation cone 
\ Dendritic spindles 
Large spindles 
of apical 
dendrite 
Unstained. 
paths 
Large spindles 
of apical den-- 
drite 
Unstained 
paths 
Unstained 
paths 
Dendritic 
.spindles 
-Basal Nissl 
body 
Unstained 
paths 
Unstained 
paths 
Dendritic 
spindles 
I Unstained paths 
Basal Nissl body 
Glial nucleus poorly 
focussed 
Fig. 4.—Large cell from normal calcarine cortex. Nissl stain. After Nissl and Ranke. Mag- 
nification x 860. 
axone breaks through at the base between two thick 
portions of the stained substance. 
The spinal ganglion cells are spherical structures of 
very unequal size, and with only one process. Their 
stainable substance consists of larger and smaller parts, 10 
INTRODUCTION—THE GANGLION CELLS 
whicli are arranged in chains, and which concentrically 
surround the nucleus. 
In the Purkinje cells of the cerebellum the stainable 
portions are united in a net-like fashion within the cell 
Remnants of stainable substance 
Two small nerve 
cells in state of 
acute degenera- 
tion (only part 
of the cells are 
present; not 
sharply focus- 
sed) 
Much altered 
nucleolus 
Remnants of 
stainable sub- 
stance 
Remnants of 
'stainable sub- 
stance 
Remnants of stainable substance 
Two small nerve cells in state of acute degen- 
eration (only part of the cells are present; 
not sharply focussed) 
Blood vessel 
Fig. 5.—Large cell from the calcarine cortex in advanced stage of acute degeneration. Nissl 
stain. After Nissl and Itanke. Magnification x 860. Compare with Fig. 4. 
body. In addition there is a striped arrangement, the 
knots of the net-work being arranged in rows; the proc- 
esses consist chiefly of unstainable substance. 
Of such types there are many more. Nissl further de- 
scribed the cells of the large sympathetic ganglia, certain 
cells in the lenticular nucleus, the mitre cells in the olfac- NISSL’S STAINABLE SUBSTANCES 
11 
tory bulb, the large cells in the tuberculum acusticum, 
and the large pyramidal cells of the cornu ammonis. 
Interest has always centred around the characteristic 
large nerve cells; but besides these there are in the central 
nervous organs a great number of smaller, and of very 
tiny ganglion cells. The latter often possess so insignif- 
icant a cell body, or there are 
within their protoplasm only 
such scanty amounts of stain- 
able substance, that it may 
be very difficult to recognize 
them at all as ganglion cells. 
Nissl attempted to incor- 
porate this wealth of form of 
the nerve cells into a system; 
particularly he classified the 
elements with distinct or large 
cell bodies according to the 
form of their stainable portions 
(net-like, chain-like, striped, 
granular). This purely de- 
scriptive division is not generally practicable and has not 
found much support. 
The question as to the exact nature of Nissl's stainable 
substances is of more importance. It was soon realized 
that they probably have little to do with nervous func- 
tion. Nissl early recognized that the unstained path- 
ways are the really important structure of the nerve cells, 
and that the intervening stainable portions act in a sense 
only as fillers. It has been suggested that the latter 
serve chiefly nutritive purposes, or that, analogous to the 
medullary sheath around the axis cylinder of the nerve 
'basal dendrite 
Nucleolus 
Deep staining 
irregular nu- 
cleus 
— Apical dendrite 
Fia. 6.—Special type of cellular disintegra- 
tion. Nissl stain. Micro-photograph, oil 
immersion. INTRODUCTION—THE GANGLION CELLS 
12 
fibre, they isolate the individual neurofibrils and fibrillar 
bundles coursing in the unstained pathways of the cells. 
It early became manifest, that these stainable structures 
are extremely labile, and that their form and staining 
qualities are quickly altered by various factors. The 
demonstration by Nissl that such changes after certain 
injuries always occur in the same manner became the 
basis for the pathology of the nerve cell. Even in the 
domain of physiology different conditions of the stainable 
substances have been observed; thus G. Mann and others 
have shown that their quantity decreases during function; 
and this led to the belief that the stainable substances are 
used up during activity and stored during rest.5 
It has further been a matter of controversy whether 
the stainable material represents preformed substances 
existing in the living cell, or whether it is merely a pre- 
cipitate formed after death or through the action of fixing 
agents. That the stainable portions differ chemically 
from the neurofibrils, as well as from the other parts of 
the cell protoplasm which do not stain by Nissl’s method, 
cannot be doubted. However, it has been questioned 
whether they have the same arrangement and structure 
in the living cell as in the Nissl picture. For normal his- 
tology this question is of great importance, but in patho- 
logical anatomy it may be disregarded. This was clearly 
expressed by Nissl in his doctrine of “the equivalent 
(comparable) 'picture of nerve cells.” “Whether what we 
5 Kronthal (Nerven und Seele, Jena 1908; Neurol. Zentralblatt 1919) has formulated 
a theory to the effect that what we term nerve cells are really more or less large accumu- 
lations of wandering cells from the blood, which surround the fibrils at their nodal 
points and which are constantly replaced by new ones. It is said that the generally 
uniform shape of the nerve cells in different localities is produced by the constancy of 
the position of the fibrils. All pathological experiences with nerve cells contradict 
these claims of Kronthal. THE “EQUIVALENT PICTURE” 
13 
see is preformed or not is really of no moment so far as the 
application of the method to the pathology of the nerve 
cells is concerned, so long as it is accepted that a standard 
technic will invariably give us the definite and recog- 
nizable nerve cell picture, which we have described.” 
The reason for any variation in this picture must lie, 
then, only in an antemortem condition of the cell itself. 
Just what effect our reagents and technical procedures 
have upon the preformed cell structures is a question for 
Stainable substances in and about the nucleolus 
Glia cells about the nerve cell 
Nucelolus' 
Stainable sub- 
stances in and 
about the nu-- 
cleolus 
-Dendrite 
! 1 
\ Granular brown pigment, which 
> occupies almost the entire 
\ v cytoplasm 
_ Glia cells about the nerve cell 
Gjia cells about the nerve cell 
Fig. 7.—Large nerve cell with brown pigment, from the substantia nigra of the brain. Nissl 
stain. Microphotograph, oil immersion. 
anatomy to answer.,5 Experience teaches us the constancy 
of the “equivalent picture;” so that, whenever we do not 
find this picture, we are justified in believing that patho- 
logical changes have occurred in the cell during life. This 
doctrine of Nissl of the “equivalent picture” contains, if 
rightly considered, nothing really new. In fact, we make 
similar deductions every day in pathological anatomy; for 
neither at the autopsy table, nor in our collections, nor 
under the microscope, do we see tissues and organs in 
6 Neurol. Zentralblatt, XVI, pp. 105-106. INTRODUCTION—THE GANGLION CELLS 
14 
exactly the same condition as they were during life. 
Death, as well as our methods of preserving, fixing, and 
hardening, leads to alterations, but we have long been 
used to disregarding them; in determining the presence of 
pathological changes in our material, we always compare 
it with mental pictures previously gained from material 
which has been prepared according to the same technic. 
Only one assumption must be made, and that is the 
constancy of the technic; and this is of special importance 
when the evaluation of the finest histological structures is 
bellow pigment 
traversed by a 
lattice work of 
stainable sub- 
-stance, appear- 
ing somewhat 
darker in the 
photograph 
-Dendrites 
Nucleus cov- 
ered and sur- 
rounded by 
stainable sub- 
stance 
Unstained 
paths 
Dendrites 
Fig. 8.—Giant pyramidal cell of Betz, (anterior central convolution), not quite in longitudinal 
section. Nissl stain. Oil immersion. 
I 
Dendrites 
in question. Any variation in the treatment of the speci- 
mens always produces more or less deviation from the 
“equivalent picture.” If we want to compare our prep- 
arations with one another or with those of others, we 
must be certain that the microscopic technic is the same 
unless it be that we are familiar with several different 
“equivalent pictures.” 
This is the reason why Nissl emphasizes that in study- NISSL’S STAINING METHOD 
15 
ing nerve cells a certain technic of staining, as for instance 
the one advocated by him, should be followed in every 
detail. The fixation of the nerve tissue, e. g., in formalin, 
gives different pictures from those obtained after fixation 
in 96 per cent, alcohol as used in Nissl’s technic; and this 
apart from the inconstancy of formalin fixation. Even 
with identical fixation, etc., different aniline dyes give 
different “equivalent pictures.” 
Nissl recommends the cutting of 
the pieces without embedding 
them, if at all possible, since the 
embedding in celloidin or paraffin, 
and the use of absolute alcohol, 
alcohol and ether, etc., necessary 
to such methods, may produce 
various artefacts. For bringing 
out only the grosser alterations in 
the stainable substances of the 
large nerve cells it is less necessary 
to observe so strictly all technical 
details. For routine preparations, 
and in cases where examination of 
the vascular and connective tissue 
apparatus is necessary,embedding 
is to be always recommended.7 
The portions of the cell bodies which do not stain by 
Nissl methods have been mentioned several times. We 
know now that they are composed of at least two constit- 
uent elements, viz., the neurofibrils and a plasma, in which 
the fibrils are embedded. In the Nissl picture the former 
wd.S™fe„^™V,fh2 
third layer of the frontal cortex. Nissl 
stain (after drawing by Alzheimer). 
7Nissl: Enzyklopddie der mikroskopischen Technik, 1910 (second edition), Article 
“ Nervensystem. ” 16 
INTRODUCTION—THE GANGLION CELLS 
appear as unstained “paths;” but, unless very thin sec- 
tions are examined, only the coarser tracts, the pathways 
for the thicker fibril bundles, are recognizable. It may be 
demonstrated that fine threads of non-stainable substance 
permeate even the stainable portions in all directions, so 
that the latter are not to be regarded as compact masses, 
but as a spongy, porous material. 
As final constituent elements of the cytoplasm, we 
Apical dendrite 
of such cells 
Remnants of 
disintegrated 
cells 
■‘Chronically 
degenerated ” 
nerve cells 
Remnants of disintegrated cells 
Remnants of disintegrated cells 
Fig. 10.—Nerve cells from the third layer in the region of the central fissure in a state of “chronic 
degeneration.” Nissl stain. Magnification x 300. After Nissl-Rosenthal. 
have substances generally known as pigments. Of these, 
two deserve to be particularly mentioned. 
A light golden-yellow pigment may, in the adult, be 
found widely spread throughout the entire central nervous 
system. It is absent in the young child, appears in small 
quantities at about the age of six to eight years, gradually 
increases with age, and is most plentiful in advanced 
senility. Under pathological conditions it is frequently 
found in considerable quantity earlier in life (Fig. 8). 
This pigment is not of hematogenous origin; it is rather 
to be looked upon as a metabolic product of the stainable, 
and possibly, in part, of the non-stainable cell substances. 
As one can see in the Nissl picture, it is scattered through, THE NUCLEUS 
17 
and replaces the stainable portions, and it is more or less 
diffusely distributed in the protoplasmic body of the cell. 
It is blackened with osmic acid, but has probably nothing 
to do with fat (Nissl). 
A brown pigment is, even in very large quantities, a 
normal constituent of certain cells in different regions 
(locus ceruleus, substantia nigra, nucleus of the vagus). 
In the cerebral cortex it does not appear. It is developed 
very early, and then remains constant in quantity. It 
has different chemical properties from the yellow pigment; 
'Glia nucleus 
Transection of 
a capillary 
Fig. 11.—Nerve cells from the cortex with adjacent dark, rounded or irregular granules, the 
so-called incrustations of the pericellular Golgi nets. The cell change is similar to that of the 
chronic degeneration. Nissl stain. 
for instance, it is not blackened with osmic acid. It is 
composed of regular, sharply defined granules (Fig. 7). 
An important constituent of the cell is the nucleus. 
The determination of its condition is, for pathology, of 
great importance. Of its morbid alterations, not much is 
known. What we do know, we likewise owe largely to 
Xissl. In deciding how severely diseased a nerve cell is, 
and in particular whether it is still capable of returning to 
normal, the changes in the nucleus frequently give us more 
valuable information than the changes observed in the 
cytoplasm. Cells with much damaged stainable sub- 18 
INTRODUCTION—THE GANGLION CELLS 
stance, may, under certain circumstances, entirely recover, 
if only the nucleus is not badly damaged. 
The nerve cells of man have but one nucleus. Two 
nuclei are not of unusual occurrence in animals. In man 
two have been observed under pathological conditions, 
probably as expression of a developmental fault, and then 
particularly in the Purkinje cells of the cerebellum. 
Fixation in 96 per cent, alcohol, as required for Nissl’s 
Disintegrated 
cells 
Disintegrated 
cells 
Capillaries 
Capillaries 
Disintegrated 
cells 
Disintegrated 
cells 
Fig. 12.—From the third layer of the frontal cortex. Nissl stain. "Disintegration” of nerve 
cells. After Nissl and Rosenthal. Magnification x 300. In the nerve cells the nucleus is appar- 
ently enlarged on account of loss of perinuclear stainable substance and of stainability of the 
nuclear membrane; the nucleolus appears prominent. 
technic, is not very suitable for the demonstration of 
nuclear structures. For nuclear study the methods of 
fixation customary in embryology are recommended. 
Fairly good results are obtained after alcohol with Heid- 
enliain’s iron hematoxylin. 
The nuclear membrane is sharply defined in the Nissl 
picture.8 The nuclear contents are unstained in the 
8 It often shows irregular folds. THE NUCLEUS 
19 
majority of the cells; in others, a faint, washed-out sug- 
gestion of the internal structure may be recognized. In 
man, with rare exceptions, only one nucleolus is present. 
Dendrites 
Dendrites 
Cross section 
of a dendrite 
Dendrites 
Dendrites 
Dendrites 
Axis-cylinder processes 
Axis-cylinder processes 
Fig. 13.—Two medium-sized pyramidal cells from the anterior central convolution. Bethe’s 
fibril stain (after Bethe). 
This is relatively large and stains deeply. In its interior 
there is a small, round, pale dot (the so-called crystalloid), 
which rarely is visible in normal cells, but which becomes 
quite distinct if the nucleolus, because of disease changes, 20 
INTRODUCTION—THE GANGLION CELLS 
is less deeply stained. In the immediate vicinity of the 
nucleolus, there is regularly found a variable number, 
(2, 3 and more) of small, round, dark bodies, the so-called 
polar bodies.9 
\Y e must finally consider the general shape of the gan- 
glion cells. With Nissl methods, their cytoplasm is every- 
where sharply defined. The dendrites, as well as the 
axone, end not far from the cell body. Only the apical 
Dendrites 
Axis-cylinder process 
Dendrites t 
Beginning of its 
myelin sheath 
Dendrites 
Fig. 14.—Giant pyramidal cell from the anterior central convolution of man. Bielschowsky’s 
fibril stain. After Bielschowsky. 
dendrite of the pyramidal cells is, as a rule, considerably 
longer. However, the Nissl picture of a normal cell shows 
the dendrites much shorter than they are in reality; 
because only the stainable portions are visible, and these 
extend for only a short distance into the protoplasmic 
processes. But even with the use of Nissl’s method, one 
has frequently an opportunity to see the cell in its entire 
extent, namely when the normally non-stainable sub- 
stance, under pathological conditions, becomes stainable 
(e. g., in acute, as well as in chronic degenerations of the 
cell) (Figs. 3 and 9). According to Nissl,10 the lance- 
9 These are more easily recognized by the use of toluidin blue and thionin than 
with a methylene blue stain. 
10 Die Neuronenlehre and ihre Anhanger, Jena, 1903. SILVER IMPREGNATION 
21 
shaped axone also terminates sharply after a short dis- 
tance; nenrofibrils may pass from its apex over into the 
axis cylinder, but the cytoplasm of the cell body terminates 
suddenly at the lower end of the lance. 
Neurofibril preparations (see the following chapter) 
give the same picture of cell outline and of the dendrites as 
does the Nissl technic (Figs. 13 and 14). But in Golgi 
preparations the shape of the ganglion cells appears con- 
siderably different; instead of short, sharply defined proc- 
esses, we see widespread and extraordinarily richly branch- 
ing dendrites, as well as very long axis-cylinder processes 
(Fig. 1). 
This is partly due to the fact that silver impregnation 
blackens not only the cytoplasm of the cell body, but the 
fibrils as well; therefore the lanceolate protoplasmic 
axis-cylinder process and the fibrils emerging from it are 
stained alike. Besides, this method makes the cell proc- 
esses look as if they were continuous with terminal arbori- 
zations of other nerve elements, and in this way gives 
an exaggerated impression of the rich branching of the 
ganglion cells. CHAPTER II. 
NEUROFIBRILS AND THE NEURONE THEORY. 
The theory that nerve cells and nerve fibres possess 
fibrillar structure has long been held. Bethe calls Max 
Schultze the father of the fibrillar doctrine, even though 
Sclmltze had probably rather suspected the existence of 
the fibrils than had actually seen them. Communications 
on this subject date back to 1868 and 1871. Fifteen years 
later (1883) Apathy and Kupfer published simultaneously 
and independently their researches on the fibrils. The 
impulse to the thorough investigation of the neurofibrils 
in the closing years of the past, and the early years of the 
present century, was given by the extensive monograph 
of Apathy, in 1897 to the importance of this for neuro- 
pathology Bethe first drew attention, and, soon after 
him, Nissl. 
Apathy’s work was done mostly with lower animals 
(leech and earth worm). He demonstrated by specific 
staining methods the existence of certain units, the con- 
ducting primitive fibrils, which, from the anatomical view- 
point, are optically and mechanically separate elements, 
and which run uninterruptedly in the nerve fibres to their 
peripheral end. These fibrils, according to his descrip- 
tion, remain of equal thickness throughout their entire 
extent; they never branch, and they nowhere anastomose 
with one another. Wherever there is a branching of a 
nerve, a number of fibrils detach themselves from the 
original stem; but the sum of all fibrils contained in the 
1 Mitteilungen aus der zoologischen Station zu Neapel, Vol. XII. 
22 NEUROFIBRILS IN INVERTEBRATES 
23 
branches is always the same as that in the original stem. 
A greater or lesser number of fibrils, eventually only a 
single thick one, is surrounded by a sheath, so constituting 
a nerve fibre; and a number of such sheathed nerve fibres 
together form a peripheral nerve. In the earthworm and 
the leech, the fibrils vary in thickness. They are very thick 
in the motor, but, to a greater or less degree, thin in the 
sensory nerves. In the higher vertebrates, such differ- 
ences are not known. Here all axis cylinders contain 
approximately the same number of fibrils of equal size. 
Otherwise the details are the same. The axis cylinder is 
not an anatomical unit, but a complex of units (primitive 
fibrils), embedded in a common protoplasm. In the neuro- 
pil, that is, in the fibrillar mass which forms the center of 
each ganglion, at the periphery of which the ganglion cells 
are situated, the primitive fibrils in worms are split into still 
finer elements, the elementary fibrils ;2 and these latter form 
a reticulum, through free anastomosis, “the diffuse elemen- 
tary reticulum. ” But later they are collected again and 
run on, united once more as primitive fibrils. Each sepa- 
rate thread of the fine reticulum corresponds to a single 
elementary fibril. At the nodal points, usually three 
threads come together and fusion takes place, not merely 
an interweaving nor a division of one thicker branch into 
two more delicate ones. 
Another such reticulum is found, in the lower animals, 
in the body of the ganglion cells. Apathy describes this 
as follows: “Conducting primitive fibrils enter the soma- 
toplasm of the ganglion cells and here break up into 
elementary fibrils; these spread out in the cell body to form 
a conducting feltwork or reticulum, in which a regroup- 
2 Their thickness according to Apathy is 1/20 to 1/5 of a micron. 24 
NEUROFIBRILS AND THE NEURONE THEORY 
ing of the elementary fibrils takes place. The same number 
of elementary fibrils now leaves the cell body as was con- 
tained in the primitive fibrils when they entered it. The 
neurofibrils neither begin nor end in, nor do they fuse 
with, the somatoplasm of the ganglion cells; nor is there 
any connection of the neurofibrils with the cell nucleus. ” 
The arrangement and form of the fibrillar reticulum 
differs greatly in different ganglion cells; but there is a 
series of easily recognizable types. Apathy recognizes 
at least two types in the leech; in the one, all fibrils are 
about equally thick and lie in 
one zone; in the other, they are 
divided into two zones, form- 
ing an external reticular sphere 
of very delicate, and an inner 
one of coarser primitive fibrils. 
The two reticula shade into one 
another through radiating con- 
necting links. The interior one 
is collected into a single thick primitive fibril, which leaves 
the cell, and can always be traced directly into a motor 
nerve fibre. In the earth worm, however, the struc- 
ture of the reticula is in many ways different; while 
concerning the condition of the fibrils in the ganglion 
cells of vertebrates, Apathy makes only short and in- 
decisive statements. 
In the periphery of the body of the leech, according to 
Apathy, each sensory cell receives a single primitive fibril 
from the nerve. In the body of the sensory cell this 
fibril forms a delicate reticulum about the nucleus; but 
the greater part of it reunites and runs further, leaving the 
cell, and probably forming, with the fibrils emerging from 
Fig. 15.—Cell from the facial nucleus of 
the rabbit after section of nerve. After 
Bielschowsky. NEUROFIBRILS IN VERTEBRATES 
25 
other cells, another intercellular neurofibrillar reticulum. 
According to this, the fibrils would not terminate at the 
periphery. Nor, in all probability, is there any termina- 
tion of the fibrils in the muscle cells. 
If we now collect these findings of Apathy, we obtain 
the following conception of the structure of the neuro- 
fibrils in the nervous system of the worm. 
Through the peripheral nerves and the central masses 
there run everywhere distinct fibrils mostly united into 
bundles. The sensory fibres, originating from the periph- 
eral sensory cells, split up in the central ganglia, into a 
reticulum of elementary fibrils, and reassemble into 
delicate fibres. Of these fibres one part enters the motor 
cells of the ganglia, and, after forming a second reticulum 
in the cell, becomes united into a thick fibril, which passes 
into a motor nerve fibre belonging to a muscle cell. There 
is, according to this, a continuous path from the sensory 
cells of the body surface to the muscle cells of the body; it 
is probable that there are no free terminations of the fibrils 
even at the periphery, but that even there the chain is 
closed. The path in the central nervous organs runs 
always through one or more ganglion cells, not necessarily 
motor in character. 
These researches of Apathy have been supplemented 
by the physiologist Bethe, from extensive investigation 
of the fibrils in the higher animals and in man. His prep- 
arations give extraordinarily clear and distinct pictures.3 
In the dendritic processes of the ganglion cells fibrils 
run side by side, often in spiral curves, and these usually 
3 See particularly: Morphol. Arbeiten, edited by Schwalbe, Yol. VIII, 1897; 
Archiv.f. mikroskop. Anatomie, Vol. LV, 1900; further: Allgem. Anatomie u. Physiologie 
des Nervensystems, Leipzig, 1903. » 26 
NEUROFIBRILS AND THE NEURONE THEORY 
group themselves at the base of the processes into several 
bundles, and then enter the cell body; they pass through 
it isolated and without dividing, regroup themselves into 
bundles and leave it again (Fig. 13). The fibrillar picture 
becomes the more complicated the more processes the 
cell has, since as a rule, all processes exchange fibrils. 
The axis-cylinder process is different from the dendrites 
only in that its fibrils, at the end of the protoplasmic lance, 
known to us from the Nissl picture, become very closely 
bunched into an apparently solid strand, and then, after a 
short course, again separate. Many fibrils do not come in 
contact with the perinuclear portion of the cell body; they 
take an outside course, from a side branch of one dendrite 
to that of another. There remain portions of the cyto- 
plasm not occupied by fibrils, and in these spaces there 
lie the substances stainable by Nissl methods. The Nissl 
picture and the fibrillar picture, are therefore, roughly 
speaking, one the negative of the other. The course of the 
fibrils explains, as Nissl had previously supposed, the 
position of the stainable substance. 
This, according to Bethe’s preparations, is the condi- 
tion in the majority of the nerve cells of the higher 
animals. A definite and important difference from the 
lower animals consists in the absence of the intracellular 
fibrillar network. 
In certain other cells, the course of the fibrils and of 
the fibril bundles is clear only in the external portions of 
the cell body; while in the region about the nucleus, the 
individual fibrils are too closely interlaced to allow of their 
being followed. True fibrillar networks, as seen in the 
ganglion cells of the invertebrates, have been found by 
Bet he only in certain definite elements. XEUROFI BRILS IN MAN 
27 
In one other point the fibrils of the vertebrates differ 
from those of the invertebrates. In the leech, the fibrils 
enter through the axone into the ganglion cells, and after 
formation of a reticulum within the cytoplasm, they make 
their exit again through the axone. The end of a fibril is 
nowhere to be found. In the vertebrates, however, the 
fibrils terminate sharply within or at the apex of the pro- 
toplasmic processes. Only in the axis-cylinder process 
they may under favorable circumstances be traced for 
long distances. 
A number of other observers, however, state definitely 
that the overwhelming majority of the fibrils in man do 
not, as Bethe describes, traverse the protoplasm evenly 
and without any anastomosis, but that the fibrils every- 
where, within the cell body, in the dendrites, in the axone, 
and the axis cylinder form meshes and reticula; that they 
constantly separate and again fuse; and that at most only 
in certain localities definite paths in the network exist. 
On the question whether there are intracellular reticula or 
traversing fibrils in the ganglion cells of the vertebrates, a 
lively discussion has arisen. Those observers who believe 
in the reticular arrangement of the fibrils in the ganglion 
cells, base their opinion on preparations made after the 
silver-impregnation method of Cajal or some similar tech- 
nic, or on those made after the method of Donaggio. 
Both methods, Jaderholm claims, but especially the latter, 
produce artefacts through the shrinkage of the tissue and 
fusion of fibrils. If one treats consecutive sections of the 
same material according to the method of Bethe and that 
of Donaggio, there are obtained in the one case the char- 
acteristic pictures of smooth isolated fibrils, in the other, 
all kinds of reticular structures. It is therefore probable 28 
NEUROFIBRILS AND THE NEURONE THEORY 
that the contradictions may be partly traced to differences 
in technic, and to faults inherent in the various methods; 
besides this, there appear to be some authors who mean 
by a fibrillar reticulum or by a fibrillo-reticular frame- 
work totally different things from the fibrils of Apathy 
and Bethe. 
The chief significance of this question lies in its 'phys- 
iological application. If the fibrils in man are not sepa- 
rate and distinct structures, then there is less ground for 
looking upon them as a real specific constituent of the 
nervous system, as the “essential nerve substance,” as 
Apathy expressed it. Indeed the impression is gradually 
gaining ground that the neurofibrils, even in the inverte- 
brates, do not form the conducting element of the nervous 
system, but merely a framework for the nerve cells. R. 
Goldschmidt, according to Alzheimer, answers the ques- 
tion thus: “The neurofibrils represent a cellular skeleton 
arranged according to Koltzoff’s principle; their presence 
and arrangement is merely a physical necessity, which 
has nothing whatever to do with nervous function, except 
that, without it, the cell would not be supported and would 
disintegrate. The neurofibrils are not a specific functional 
structure of the nerve cell, but the expression of a law 
applying to all animal cells. ” According to Goldschmidt, 
the morphological analogy of the neurofibrils with the 
skeletal fibrils of other cells is so marked that Apathy 
erroneously regarded the skeletal fibrils of the muscle cell 
in the leech and the fibrils of the basal bodies of ciliated 
epithelium as neurofibrils. But the evidence on this point 
is conflicting. 
Disregarding these controversial points, the fact re- 
mains that what we reallv know about these fibrils in man TERMINATION OF NEUROFIBRILS 
29 
is decidedly incomplete, compared with what Apathy has 
reported in the case of the invertebrates. In the earth 
worm and the leech, distinct fibrils may be traced from the 
sensory cells of the body surface into the extra and intra- 
cellular networks of the ganglia, and back again to the 
muscle cells of the body. In the vertebrates, however, 
we have knowledge only of fibrils in the ganglion cells, and 
these fibrils mostly terminate where the cell body ends, at 
its periphery, and at the termination of the dendrites. 
Only a small portion passes beyond the limits of cell body 
at the axone hillock, and may be traced as a bundle of 
fibrils into a nerve fibre, and through this, nearly to the 
motor and sensory end-apparatus, or to the neighborhood 
of other ganglion cells, or to distantly situated gray 
masses. The controversy over the neurone doctrine was 
brought about largely by the results of investigations on 
the fibrillse. It would have been of decisive importance 
in this question to know what becomes of the intracellular 
fibrils and of the fibrils of the axis cylinders; whether, and 
in what manner, the fibrils of different cells and of axis 
cylinders are connected; and whether a continuity of 
fibrils throughout the entire nervous system can be 
demonstrated in the higher vertebrates, as Apathy has 
described it in the invertebrates. According to the basic 
principles of the neurone theory, the axis cylinders split 
up into fine branches at their ends after casting off the 
myelin sheath in case they are myelinated ;4 these branches 
(the terminal arborization) reach the surface of other 
nerve cells, and thus render possible the transmission of 
stimuli. In what manner, or whether at all, an exchange 
4 See Nissl’s critique on the general justification of this concept in: Die Neuron- 
enlehre, etc. Jena, 1903, p. 378 ff. 30 
NEUROFIBRILS AND THE NEURONE THEORY 
of neurofibrils takes place between such nerve termina- 
tions and nerve cells in the gray substance of man, is 
answered differently by different observers. Some hold 
that such a communication does not take place; others 
have seen it frequently or else only occasionally in their 
preparations. Some describe these histological relations 
as very simple, others as very complicated. 
Bethe teaches that the ganglion cells and their dendrites 
are surrounded by a terminal arborization of approaching 
nerve fibres (axis-cylinder “stockings” around the cells) 
and that these “stockings” represent a wide-spread 
arrangement in the nervous system. From them delicate 
short fibrils go to the cell and form a network, which 
closely hugs the cell’s surface; a network in which 
Bethe supposes there is a true fibrillar reticulum in 
the sense of Apathy. The fibrils now pass through the 
superficial regions of the nerve cell, and unite with the 
intracellular fibrils. These fibrillar connections between 
the endings of nerve fibres and the fibrils of cells are, 
according to Bethe, difficult to demonstrate by staining, 
and, for that reason, have been seen only occasionally 
by him. 
Much easier to demonstrate is an apparatus widely 
distributed in the body in which the delicate connecting 
fibrils and the reticula referred to above are embedded, 
the so-called pericellular Golgi nets. They constitute a 
relatively coarse anastomosing trabecular framework, 
which surrounds the nerve cells and their dendrites on all 
sides, in one or several layers.5 
6 This is regarded by Cajal as a layer still belonging to the cell body, lying beneath 
the cell membrane; Held regards it as a glial apparatus; likewise Adamkiewicz (Ztschr, 
f. d. ges. Neurol, u. Psych., Vol. LI, 1919). “NERVOUS GRAY” 
31 
Nissl shares Betlie’s view concerning the relation of the 
neurofibrils within the nerve cells to the pericellular and 
peridendritic Golgi network, but he believes that these 
relations are so simple only in certain cell types. Accord- 
ing to his conception, the axis cylinders mostly terminate 
quite a distance from the cells which they are approach- 
ing; they form, with many other terminals of nerve fibres, 
a “nervous gray,” that is a “specific nervous, non-cellular 
constituent of the gray substance," which extends far be- 
yond the region of the Golgi network, and which would 
correspond to the extracellular reticulum in the neuropil 
of the invertebrates. About the structure of this network 
we have no definite knowledge, since it has not been pos- 
sible as yet to stain it. The existence of such a “nervous 
gray” is sponsored by Nissl, because of a series of com- 
parative anatomical and embryological data. 
The investigations on the fibrils of the nervous system 
created a lively interest, due to the fact that the demon- 
stration of these fibrils was from the beginning offered as 
evidence of the incorrectness of certain ideas which were 
generally held regarding the finer structure of the nervous 
system. These conceptions had crystallized during the 
preceding few decades into the so-called neurone theory;6 
and it was this theory that was questioned on the basis 
of the views arrived at from these studies upon the fibrils. 
Even if it should be shown that the neurofibrils do not 
have the importance which was attached to them by 
Apathy, Bethe, and their followers, the investigation of 
the question has made us revise our ideas about the struc- 
ture of the functioning nervous elements; and so our 
knowledge of them has become more clearly defined. 
6 This term originated with Waldeyer, 1891. 32 
NEUROFIBRILS AND THE NEURONE THEORY 
The neurone theory is, in its nature, an anatomical 
hypothesis. Its principles may be formulated, according 
to Nissl, as follows:7 all nerve fibres are processes of nerve 
cells; the whole nervous system of the adult is (except the 
meninges and supporting structures) a vast conglomera- 
tion of neurones, that is, of structures having the dignity 
of cells, each consisting of ganglion cell, axis cylinder, and 
terminal arborization; the entire nervous system, as far as 
its specific function is concerned, is nothing but the sum of 
such neurones. A discussion of the truth of this doctrine 
must be restricted to these principles; anything beyond 
them does not directly concern the doctrine. 
What may be urged against the neurone doctrine as a 
result of studies on the fibrils is plain. Apathy’s descrip- 
tions give a well-connected account of the nervous system 
in certain low forms of invertebrates, assuming his asser- 
tion, that the fibrils are the conducting mechanism, to be 
correct. Fibrils run from the periphery of the body to the 
central organs and return to the periphery without any 
interruption at the ganglion-cell terminus. There can 
therefore be no acceptance, so far as these animals are con- 
cerned, of a nervous system made up of conglomerations 
of neurones. But in the case of the vertebrates our knowl- 
edge is too full of gaps, even with the use of present-day 
staining methods, to afford absolute proof that the 
neurone doctrine does not apply. 
But as doubts began to arise about the truth of the 
neurone doctrine, its anatomical basis was called to mind. 
It was recalled that in reality no attempt had ever been 
made to demonstrate that, as was demanded by the neu- 
7 Concerning this and for much of the following see Nissl’s discussion in his book, 
Die Neuronenlehre vnd ihre Anhanger, Jena, G. Fischer, 1903 NEURONE THEORY 
33 
rone theory, the relation of cell plus nerve fibre plus ter- 
minal arborization exists everywhere; that each nerve 
fibre arises from a cell; that cell plus fibre plus terminal 
arborization form the only constituents of the nervous 
system, and that each nerve cell has an axis-cylinder proc- 
ess (Nissl). Generalizations had been made merely on 
the basis of a few isolated observations. The motor 
tracts were pointed to repeatedly as an example, but as a 
matter of fact, Apathy’s descriptions of worms show that 
in the motor part of the fibrillar tracts, and only in it, 
there exist conditions which closely ap- 
proximate the scheme of the neurone 
doctrine. Furthermore, it has been pointed 
out that the embryological data of His, 
which have always been looked upon as 
a very important support for the neurone 
doctrine, have been much overrated in 
this connection. His has made proba- 
ble the outgrowth of axis cylinders from 
ganglion cells only for one certain type 
of cell (anterior horn cells of spinal cord,) 
and even for this type, only in certain elements found at a 
very early developmental stage. The assertion that all 
axis cylinders in the entire nervous system have similar 
origin is a generalization not supported by sufficient 
data; this fact has been forgotten under the reign of the 
neurone theory. 
A series of investigators who admit these objections 
cite, as a last support for the neurone theory, certain facts 
of clinical experience, as well as of normal and patho- 
logical fibre anatomy, which cannot be explained without 
this doctrine. This shows, however, a misapprehension of 
Fig. 16.—Motor anterior 
horn cell of man, from a 
case of acute spinal cord 
disease. Bielschowsky’s 
fibril stain. Loss of fibrils 
in cytoplasm. After Biel- 
schowsky. 34 
NEUROFIBRILS AND THE NEURONE THEORY 
the essential points at issue, since our knowledge of neuro- 
pathology and fibre anatomy has nothing to do with the 
basic principles of the neurone doctrine; such facts can be 
cited as evidence either way. The claims that are usually 
made are, that more or less circumscribed masses of mye- 
linated fibrils (fasciculi) emerge from certain regions of 
the gray cortex, the sub-cortical ganglia, and the spinal 
cord, and after a shorter or longer course enter other gray 
masses, where they lose their myelin sheath. Further, 
that after destruction of the gray matter or of the fasciculi 
at any point, secondary degenerations occur, always in 
the same direction; and that the clinical phenomena are 
always the same, no matter at what point of such a com- 
plex of gray matter and fibre systems destruction has 
taken place. No opponent of the neurone theory doubts 
the existence of separate fasciculi, nor the influence, func- 
tional, nutritive, and trophic, of certain gray masses 
upon certain fibre systems. But all this has nothing to do 
with the neurone theory. The latter claims that all 
elements of all such complexes are neurones, meaning that 
each one of their fibres originates from a certain nerve cell 
of the gray matter and ends with a terminal arborization 
at another nerve cell of another gray nucleus. This, how- 
ever, can never be proved by the aid of myelin fibre prep- 
arations or by tracing secondary degenerations. Of 
course, if we were able to destroy individual cells of a gray 
mass, or individual isolated fibres, it might perhaps be 
possible, on the ground of secondary degenerations, to 
assert something more definite concerning the relations of 
individual cells and individual fibres. But we can only 
destroy considerable portions of the gray matter, that is, 
a complex of cells plus the debated intercellular fibrillar NEURONE THEORY 
35 
networks; and even if we see that degenerated nerve fibres 
leave such a focus, or that, after section of a fasciculus, 
certain groups of nerve cells exhibit certain morbid alter- 
ations, such findings can never be used as proof that every 
one of the degenerated fibres is a direct continuation of the 
diseased nerve cells or that every one of these fibres ends 
in a nerve cell of another gray mass. 
The same objection applies to the position taken by 
those who would reconcile the views of the supporters and 
opponents of the neurone doctrine. These admit that the 
histological unity of the neurone in the adult can no longer 
be maintained, but insist that the nerve cell has a trophic 
or nutritive influence upon its fibre even in post-foetal life. 
Here again the methods of pathology are able to prove or 
disprove the influence of a gray mass upon a fibre bundle, 
but never that of the individual cell upon the individual 
fibre, and yet the latter would have to be established, 
should we wish to cite the results of experiments as either 
favoring or opposing the neurone theory. 
For these reasons it is evident that in discussion of the 
principles of the neurone doctrine all our present-day 
knowledge of fibre anatomy of secondary degenerations 
of white fasciculi must be left out of consideration. It 
throws no light on the relative merits of the neurone doc- 
trine and the theory of continuous fibrils. The neurone 
doctrine is merely an anatomical hypothesis, the proof 
of which must depend on histology; for clinical medicine 
and for the interpretation of experimental work it is 
unessential. From the viewpoint of the clinic and experi- 
mentation it is certain only that the nervous system con- 
sists of an aggregation of well-defined units composed of 
gray masses and fibre systems or fasciculi, the latter being 36 
NEUROFIBRILS AND THE NEURONE THEORY 
connected with other gray masses. More than that we do 
not know. If we wish for a term to designate such a com- 
plex, we might call it a “neural.” But it must be clearly 
understood that a “neural” is not a collection or conglom- 
eration of neurones in the sense of the neurone doctrine, 
for we do not know the real histological relation within the 
“neural” of all the cells to all the fibrils. This gives us a 
conception corresponding to the facts in every respect; 
whereas the neurone doctrine is an hypothesis which 
cannot be supported by satisfactory anatomical facts. 
It is not, then, necessary to condemn the neurone 
theory as false, rather we should retain it as a valuable 
scheme for teaching. CHAPTER III. 
NERVE FIBRES, NEUROGLIA. 
Another important constituent of the nervous system 
is represented by the nerve fibres. United by connective 
tissue into bundles, they form the peripheral nerves of the 
body. In the central nervous system they make up the 
white masses and the white tracts, and are found, too, 
everywhere penetrating into the gray matter, separately 
or in fasciculi. 
In the nerve fibres we distinguish a central continuous 
filament, the axis cylinder, and its coverings, the latter 
consisting of the myelin sheath and the so-called sheath of 
Schwann or neurilemma. The most important constit- 
uent for the function of the nerve fibre is the axis cylinder. 
The myelin sheath may be absent, as for instance in the 
non-medullated fibres of the sympathetic system; the 
sheath of Schwann may be absent, as in all medullated 
fibres of the central nervous system; and, in all probability 
both may be absent (in the so-called naked axis cylinders 
found in the end arborizations in the gray matter). The 
peripheral nerve fibres are provided with a medullary 
sheath as well as a neurilemma. 
The axis cylinder consists, as we have learned in the 
previous chapter, of a bundle of probably uninterrupted 
and distinct neurofibrils embedded in a plasma.1 Accord- 
ing to the neurone theory every axis cylinder is a direct 
continuation of a nerve cell. This may often if not regu- 
1 For the demonstration of axis-cylinder fibrils, preliminary treatment with osmic 
acid is important. With the use of other fixing methods, the fibrils shrink into an appar- 
ently solid strand (Bethe). 
37 38 
NERVE FIBRES, NEUROGLIA 
larly be demonstrated in the case of the motor-nerve 
fibres; even in the lower animals examined by Apathy, the 
thick motor fibrillse make their exit at the axonal hillock of 
a motor ganglion cell and run without interruption and 
without further reticular formation to the motor end 
apparatus at the periphery of the body. But whether 
all other axis cylinders, especially those of sensory nerves 
are, in higher animals, always direct processes from certain 
ganglion cells, is not proved. The sheaths of the myeli- 
nated nerve fibres within the central nervous organs 
exhibit some important differences from those of periph- 
eral nerves. The former do not possess the so-called 
sheath of Schwann, and their medullary sheath is an un- 
interrupted tube; while in the peripheral nerves, the me- 
dullary sheath is interrupted at regular intervals (Ranvier’s 
nodal points), and has a delicate covering known as 
Schwann’s sheath, which possesses a nucleus between 
every two nodes. 
Of the grosser physiological differences, we know 
further that the peripheral nerves exhibit excellent regen- 
erative ability, while this is lacking or poorly developed 
in the central fibres. 
If we section a peripheral nerve, the distal segment 
degenerates according to the law of Waller; this is called 
“secondary degeneration.” Axis cylinder and medullary 
sheath disintegrate, but the nuclei of the neurilemma re- 
main intact and proliferate. A portion of the proliferating 
elements eventually becomes granule cells, and these serve 
to remove products of disintegration. Another part be- 
comes arranged into cellular ribbons, and in such ribbons 
there eventually appears the new, regenerated axis 
cylinder with a medullary sheath. As to the way in which AXIS CYLINDER 
39 
this new axis cylinder forms, there are two opposing 
opinions.2 Some authors claim that the proximal part of 
the old axis cylinder, which remains intact and which 
supposedly is always connected with a ganglion cell, grows 
out and pushes forward along the path pointed out by 
the nuclei of the neurilemma; at the same time or later, it 
acquires a new sheath. Other authors believe that the 
axis cylinder arises in a “discontinuous” manner; they 
look upon the so-called neurilemma nuclei as the matrix 
of the new axis cylinder, and hold that, within cells 
arranged in ribbon-like fashion, there are formed fibrils. 
These fibrils eventually meet and finally form a continu- 
ous filament.3 For the fibres of the central nervous system 
this method of investigation cannot be used, since regener- 
ation of central fibres does not take place, after preliminary 
destruction and secondary degeneration.4 
Interesting and important communications concerning 
the structure of the peripheral nerve fibres have been made 
by Reich.5 According to his observation of normal nerves 
of adults, it may be accepted that the entire covering of 
each length of nerve fibre between two nodes of Ranvier is 
a highly complicated structure, but really a single cell. 
The nucleus of this structure is the neurilemma nucleus 
2 Cf., e.g., Edinger (Deutsche Zeitschr. f. Nervenheilkunde, Vol. LVIII, 1918), who 
takes an intermediate standpoint. 
3 Concerning the embryological development, these two views on the development 
of the axis cylinder are opposed to one another. 
4 Cf. here Koichi Miyake, (Arbeiten aus dem Obersteinerschen Institut, Vol. XIV, 
1908). Bielsehowsky (Journ.f. Psychol, u. Neurol, Vol. XIV, 1909) and others ener- 
getically espouse the regenerative ability of the central nerve fibres. However, they 
hold that these regenerations are not orderly, but without plan and irregular; so that 
reconstruction of the old path, and therefore re-establishment of function, does not 
take place. 
5 Reich, Journ. f. Psychol, u. Neurol., Vol. VIII, 1907. Similar statements are 
made by Nemiloff, Arch.f. Mikroskop Anatomie, 1908. NERVE FIBRES, NEUROGLIA 
40 
of the fibre segment concerned. This nucleus lies in a 
little mass of protoplasm which is continuous with the 
neurilemma; the latter envelopes the entire cellular struc- 
ture like a cell membrane. From this membrane, funnel- 
shaped septa go to the interior; they become reunited 
axially where the axis cylinder lies, to form a homogen- 
eous substance. It is probable that the plasma, in which 
the individual axis-cylinder fibrils are embedded, is formed 
by this substance. The spaces between the septa, which 
Nucleus of neurilemma 
Network of the cell, termi- 
nating on either side in 
the sheath ox neurokeratin 
Sheath of neuro- 
/keratin 
■M Axis cylinder 
Bands of neurokeratin extending internally to the axis 
cylinder, externally to the outer neurokeratin sheath 
Sheath of neuro- 
keratin 
Fig. 17.—Peripheral nerve fibre. Longitudinal section. Van Gieson stain. After Reich 
pass externally into the neurilemma, internally into the 
axis-cylinder plasma, are filled with myelin, a substance 
which hitherto was regarded as a second independent 
covering of the nerves, the medullary sheath (Fig. 17). 
According to these observations, the peripheral nerve 
consists, even in the adult, of a chain of cellular structures 
(neurilemma cells), which are apposed at their narrow 
sides; and only the fibrils of the axis cylinder are contin- 
uous, running through this chain of cells. 
Similar assertions had previously been made by Ran- 
vier, by Apathy (for invertebrates), and by other observers. 
Stransky 6 regarded the nucleus of the neurilemma as a 
point of proliferation of the tubular cell which extends as 
6Stransky, Journ.f. Psychol, u. Neurol., Vol. I, 1903. MYELIN SHEATH 
41 
neurilemma from one node to the other. At the node the 
cell fuses with a contiguous cell, and in so doing forms a 
sieve-like plate, through which passes the axis cylinder with 
its fibrils. According to Doinikow 7 the structure of the 
myelin sheath is as follows: the cytoplasm of the neuri- 
lemma cell consists of a dense perinuclear “court” pro- 
vided with a loose reticulum, which penetrates the medul- 
lary sheath of an internodal segment in its entire extent, 
and contains the myelin substance in its meshes. The 
extremely delicate meshes of this framework are strength- 
ened by coarser trabeculae; the latter forming for in- 
stance, the funnel-like structures which run obliquely to 
the surface of the axis cylinders. 
Within the nerves there are regularly found small 
quantities of so-called Elzliolz bodies, which stand out in 
Marchi preparations because of their very deep color. 
They have a round shape, and are of the size of small 
lymphocytes. They are formed as globules at the expense 
of the medullary sheath, and are an expression of the 
normal metabolism of the latter. Their increase is an 
indication of pathological change. In the cytoplasm of 
the neurilemma cells occur similar but smaller bodies, and 
also the protagon-like granules of Reich. 
Similarly the nerve fibres of the central organs are 
often asserted to be derived embryologically from cells 
arranged in rows. Italian authors particularly have dealt 
with this question. Among others, Fragnito has seen in 
chick embryos of 12 to 18 days, in place of the later nerve 
fibres, chains of long stretehed-out cells, each with a nu- 
7 Detailed description of the structure of peripheral nerve fibre (with literature) 
given by Doinikow (Histol. u. Histo-pathol. Arb. von Nissl u. Alzheimer, Vol. IV, 1911); 
of the central, by Jakob, ibid., Vol. V, 1912. NERVE FIBRES, NEUROGLIA 
42 
cleus. He was able to trace these nucleated spindles into 
the axis-cylinder processes of ganglion cells and even into 
the cell body. But probably the central nerve fibres differ 
from the peripheral in that, in the former, the cells which 
form the nerve fibre fuse later into a filament, and their 
nuclei then disintegrate; also in that an external cover 
(which in the peripheral nerves is called a neurilemma) 
does not remain. 
Reich supposes that a part of the glia cells may stand 
in particularly close relation to the nerve fibres in the 
central organs, and that the nuclei of these glial elements 
are in a sense analogues of the neurilemma nuclei of the 
peripheral nerves. According to Paladino and Held (see 
Jakob7) the medullary sheaths of the central nervous 
system are not only everywhere surrounded by proto- 
plasmic processes of glial cells, but probably fine filaments 
of the glia penetrate into the meshes of the medullary 
sheath and through them, extending to the axis cylinder. 
Around it they form nodes at intervals. Be this as it may, 
the writers agree in supposing that the formation of myelin 
in the central fibres is a function of the neuroglia, just as 
in the peripheral fibres it is a function of the neurilemma 
cells; and that the protoplasm of the nerve fibre is em- 
bedded immediately in the protoplasmic structures of 
the glia. 
Finally, in this series of new observations, the contri- 
bution of Held must be mentioned. He asserts that the 
neurilemma cells, hitherto regarded as of connective 
tissue nature, are derived from cells which grow out from 
the medullary tube; that they are emigrated glial cells, 
that is, elements of ectodermal origin. He compares the 
neurilemma with the membrana limitans glise of the central NEUROGLIA 
43 
nervous system. Externally is the closely applied endo- 
neurium, of connective tissue nature. This conception, 
taken with those already mentioned, would mean that in 
the peripheral nerves not only the axis cylinders but also 
all their coverings are derivatives of the outer germinal 
layer, just as is the entire central nervous system. 
Ganglion cells, nerve fibres, and the fibrillar apparatus 
which connects them constitute the specific functional 
parts of the nervous system. These parts are embedded 
in a supporting framework. Of the other organs of the 
body it is known that their supporting framework is of 
mesodermal origin. In the nervous system it is different; 
this possesses a characteristic ectodermal framework—the 
neuroglia or nerve cement. Specific nervous tissue and 
neuroglia originate from a common anlage, the ectoderm; 
and they become distinct only late in their development. 
A result of this late separation is a lasting, very close 
relation between the two. The nervous system contains 
also true mesodermal connective tissue; but, excepting 
the external coverings (dura mater, pia mater; peri- 
neurium, etc., of peripheral nerves) this is present only in 
scanty quantities and only accompanying the blood 
vascular system. 
Concerning the nature and the behavior of the neuro- 
glia, we have only very gradually obtained any definite 
knowledge. Thus it was with great difficulty that its 
ectodermal origin was established. The most important 
opponent of this conception, His, only shortly before his 
death in 1904, admitted the purely ectodermal origin of 
the glia. Until the end of the last century, “a confusion 
mocking description” (Nissl) prevailed concerning the 44 
NERVE FIBRES, NEUROGLIA 
neuroglia.8 It was not possible to differentiate between 
the nuclei of many glia cells and those of certain ganglion 
cells; many of the apparently “free" glial nuclei were re- 
garded as lymphocytes, and so on. Remnants of this con- 
fusion are even now met with everywhere in the literature. 
More precise knowledge was first gained with the help 
Fig. 18.—Normal glia cell (long-rayed astrocyte) from the white matter of the cerebrum. Golgi 
method (after Golgi, from Koellicker’s Handbuch). 
of Golgi pictures. Silver impregnation preparations show 
a large part of the glia as isolated cell units, each of which 
appears as a spider-like cell structure with longer or shorter 
processes radiating from the cell body. These processes 
are often quite long, and in such cases form a network 
(Fig. 18). Such preparations give the impression that the 
glia consists entirely of well-defined cells and their numer- 
8 For a historical review of the glia, see Nissl, Histol. u. histopathol, Arbeiten, Yol. 
I; Held, Abhandl. der math.—pkys. Klasse der Konigl. Sachs. Ges. der Wissensch. 
Yol. XXVIII, 1903; Eisath, Monatsschr. f. Psychol, u. Neurol., Yol. XX, 1906. NEUROGLIA 
45 
ous crossing and interlacing processes, and of nothing else. 
Something entirely new is presented in the teaching of 
Weigert on the glia.9 He demonstrated, by means of a 
specific staining method (now known as Weigert’s glia 
stain), that there are independent glia fibres (glia fibrils) 
which, though formed by the glia cells, are different from 
them chemically, and therefore may be electively stained 
Fig. 19.—Normal glia cell from the posterior horn of the spinal cord. Weigert’s glia stain (after 
Weigert). The glia fibrils run in a curve past the nucleus. Cell body unstained. 
(Fig. 19). He further emphasized his belief that these 
chemically different fibrils in their mature condition are 
separated from the cells, that they lie outside of the cell 
bodies, and therefore represent a true intercellular sub- 
stance. These fibrils are in part much separated from 
their parent cells, in part lie close to them; and in the 
latter case they tend to sweep past the cell in curves 
that have their convexity toward the nucleus. 
Weigert taught that the older staining methods pro- 
duced the well-known astrocyte picture (short-rayed 
astrocytes, long-rayed astrocytes) simply because they 
stained or impregnated the cell body and the glia fibres in 
9 Weigert, Beitrage zur Kenntnis der normalen menschlichen Neuroglia, 1895, 
Frankfurt a. M. 46 
NERVE FIBRES, NEUROGLIA 
the same manner. The specific glia fibrils are, according 
to Weigert’s opinion, not processes of the cell bodies, but 
lie entirely outside them and are chemically distinct. 
They are smooth, peculiarly stiff structures, which never 
divide or anastomose. 
This new theory separates the mature glial tissue into 
two constituents, well-defined glial cells and intercellular 
fibres set free from them. A comparison of the different 
parts of the central nervous system proved to Weigert 
that the cells are everywhere fairly evenly, but the fibres 
very irregularly, distributed. Beneath the ependyma of 
the ventricle there is always a thick layer of very closely 
interwoven neuroglia fibres; likewise the external layers 
of the central nervous system and all portions of the tissue 
adjoining blood vessels show a greater density of the 
neuroglia. In the interior of the gray and white sub- 
stances the number and the mass of the glia fibrils vary in 
different regions; thus for instance in the cerebral cortex 
there is always a thin but dense feltwork on the surface 
beneath the pia; from this, glia fibrils radiate into the 
depth, but they become more and more scanty, so that in 
the deeper layers glial fibrils are not to be found for 
considerable stretches, and only in the white matter a 
rich feltwork again appears.10 
Weigert’s W'Ork on the neuroglia is of fundamental 
importance. It has definitely enriched our knowdedge of 
the central nervous organs. Unfortunately, his method for 
the demonstration of the glia fibrils has a number of imper- 
fections, which Weigert, in spite of all his efforts, wras un- 
able to overcome. The method is capricious even in the 
hands of the expert, and that wdthout any apparent cause. 
10 Weigert, l. c., p. 136 ff., p. 172. WEIGERT OX NEUROGLIA 
47 
It is particularly disappointing in studies on the cerebral 
cortex, especially when no pathological increase of the 
fibres is present; furthermore it cannot be applied in 
animal work.11 
After Weigert’s publication, objections to his concep- 
tion of the structure of neuroglia were not wanting. It 
was asserted that his method does not stain all the con- 
stituents of the glia; that besides the nuclei only some 
elements, in certain developmental stages, with certain 
definite chemical properties, (the specific glia fibrils) can 
be electively demonstrated; that it scarcely shows the 
cell body at all. Therefore it is claimed that Weigert 
reached his conclusion on the separation of the glia fibrils 
from the glia cells only through a one-sided emphasis on 
the nuclei and fibrils, neglecting the cytoplasmic portion 
of the glia. 
These objections show the trend of development in 
the nuclei and neuroglia theory since Weigert. 
The most important of these later researches in the 
normal histology of the glia are those of Held.12 Held 
entirely accepts Weigert’s description of the form and the 
tinctorial reaction of the glia fibrils; but he is unwilling 
to regard these fibrils as a true intercellular substance. In 
other words, he accepts their chemical differentiation from 
the glial cell body, but he denies their physical separation. 
Held, in contrast with Weigert, tried to stain not only 
the nuclei and the fibrils, but especially to demonstrate 
the cytoplasm of the glia as completely as possible. Such 
preparations have led to a distinctly different conception 
11 Erik Muller, Archiv.f. mikroskop. Anatomie u. Entw.-Gesch., Vol. LV, 1899. 
12 Hans Held, Uber den Ban der Neuroglia, etc. Abhandl. der math., phys. Klasse 
der Konigl. Sachs. Ges. der Wissensch., Vol. XXVIII, 1903. Die Neuroglia marginalis 
etc. Monatsschr. f. Psychiatrie u. Neurol., Vol. XXVI, 1909. 48 
NERVE FIBRES, NEUROGLIA 
of the structure of the entire glial apparatus. This view 
emphasizes that the glia is a richly branching but contin- 
uous syncytial network, consisting of a protoplasmic 
substance which envelopes the functional elements of the 
central nervous system and fills all spaces between them. 
At the nodal points of this network are embedded nuclei, 
and, strengthening the whole, there are the specific 
fibrils demonstrated by AYeigert, lying, however, in the 
cytoplasm and not outside of it. 
This conception of Held, of the syncytial character of 
the glia, conforms in many respects to certain embryonal 
facts, the knowledge of which we owe particularly to His.13 
The first anlage of the central nervous system is an unin- 
terrupted plate of closely placed cells. This assumes 
tubular form and soon loses its compactness. In the 
second week its place has already been taken by the 
so-called myelospongium, which is a continuous protoplas- 
mic network. The latter is condensed externally and 
internally into a membrana limitans (externa and interna), 
in this there are embedded somewhat radially arranged 
nuclei. This myelospongium remains as a lasting constit- 
uent of the future neuroglia. To it there are added, by 
division of pre-existing nuclei, widely distributed cell 
elements, which are connected with those of the myelo- 
spongium and with each other as a syncytium. The 
inner layer of the myelospongium becomes the ependyma. 
In a still later stage the separation of these elements into 
spongioblasts and neuroblasts takes place; the former 
become glia cells, the latter, nerve cells. 
13 See his last work on the nervous system: Die Entwicklung des menschlichen 
Gehirns wahrend der ersten Monate, Leipzig, 1904. HELD ON NEUROGLIA 
49 
This syncytial character, traced embryologically by 
His, is retained permanently by the glia according to 
Held’s observation. The ramifications of the cell bodies 
form a network; the anastomoses are in many places very 
rich and dense, elsewhere they are scarce and less compact, 
or they may be recognizable only as a very delicate coating 
of the neuroglia fibrils. The fibrils of Weigert serve the 
purpose of lending greater rigidity to the reticulum; they 
are embedded within the glial protoplasma, and are not 
Glial pedicle 
Capillary 
__ Red blood corpuscles 
v Endothelial nucleus 
Fig. Large fibre-forming glia cell. Weigert’s glia stain. After a drawing by Alzheimer. 
Numerous newly formed glia fibrils along the borders of cell body and processes. 
separated from it even in the mature condition (Fig. 21). Oc- 
casionally Held has not been able to discover in his prepara- 
tions a protoplasmic coating of the glia fibrils, but he has 
likewise never been able to convince himself that such 
fibrils are free in their entire extent; he has rather always 
seen them dip somewhere into the protoplasmic glial reti- 
culum. The majority of the specific fibrils lies in the 
cytoplasm near the nuclei, but another portion lies at a 
distance in the irregular protoplasmic network. Not un- 
commonly the fibrils traverse several cells; they do not 
always belong to a single cell. 
Held distinguishes between glia cells that have no 
fibrils, those that have few, and those that have many. NERVE FIBRES, NEUROGLIA 
50 
A very frequent type of the fibre-containing cell is the one 
in which the fibrils are radially arranged. This type 
according to Held, is produced, when numerous fibrils 
sweep from the long processes of the glia cells through the 
cytoplasm in the neighborhood of the nucleus, and then 
pass out again into another process. Such elements 
appear as the well-known astrocytes shown by silver im- 
pregnation methods (Golgi). In Weigert preparations, 
as a result of the decolorization of the greater part of the 
protoplasm, we get a picture in which large numbers of 
fibrils sweep in curves past the nucleus. Other tinctorial 
methods for the demonstration of neuroglia (Eisath, 
Alzheimer, Cajal et al) also show the frequency of such 
astrocytes. Held recognizes, besides cells in which the 
fibrils are arranged longitudinally or transversely, those 
in which the majority of the fibrils pass near the periphery 
of the cell far from the nuclear portion of the protoplasm. 
Many glia cells do not show any specific fibrils at all in 
their cytoplasm. 
It was known before Weigert’s time, but it was partic- 
ularly emphasized by him, that the glia is more densely 
concentrated on the surface of the central organs than in 
most other places. Weigert recognized that the same 
condition prevails at the surface of contact between the 
nervous tissue and the blood vessels, which everywhere 
stream in from the pia; this surface of contact he called the 
“internal” surface. With this “marginal” glia Held has 
concerned himself; and, here again, contrary to Weigert, 
he has paid particular attention to the protoplasmic por- 
tion of the glia. He was able to demonstrate in his prep- 
arations that the syncytial protoplasmic network of the 
glia is everywhere condensed on such surfaces (the external MEMBRANA LIMITANS 
51 
as well as the internal) into a continuous membrane 
(membrana limitans superficialis and perivascularis). In 
this manner the ectodermal nervous tissue is everywhere 
sharply separated from the mesodermal tissue (pia, 
vessels). The membrana limitans is sometimes delicate, 
sometimes coarse; it is composed of broadened protoplas- 
mic processes, the so-called pedicles; these belong to 
several more or less distant glia cells. If we section the 
membrana on the flat, we can then verify the fact that it 
consists of individual pedicles, which are seen to form an 
irregular mosaic. The membrane itself and the pedicles 
are in many places reinforced by glia fibrils, which are 
demonstrated by Weigert’s method. 
To our knowledge of the nature and structure of the 
neuroglia these observations of Held form an important 
supplement. We learn from them that the glia represents 
a continuous reticulum which is only imperfectly and in 
some places differentiated into cells, a reticulum that in 
fact envelopes and holds together as a nerve cement the 
ganglion cells, fibrils, and nerve fibres; the rigid fibrils 
described by Weigert are merely embedded in it. 
Held regards as a component of the glia the fine 
granular or lumpy masses, so-called “nets of Bethe,” 
which envelope the nerve fibres of the white substance 
like a sheath, and the periganglionic or pericellular Golgi 
nets, already discussed in a previous chapter. We learned 
there that delicate protoplasmic filaments of the glia 
probably penetrate through the meshes of the medullary 
sheath to the axis cylinder of the nerve fibre. Held em- 
phasizes, in connection with all this, that the “reticulum” 
of the glia is largely not an ordinary but a differentiated 
type of protoplasm. 52 
NERVE FIBRES, NEUROGLIA 
These structural relations of the neuroglia in the 
central organs, together with observations made on 
pathological material, indicate that the neuroglia is not 
merely the supporting framework of the central nervous 
system, but that its protoplasm forms the pathways for 
the tissue juices as well. In these the nutrient exchange 
Fig. 21.—Four glia cells from the white substance of the spinal cord. After Held. The dark 
and dots are longitudinally and transversely sectioned glia fibrils. 
takes place between the capillaries or lymphatics and the 
functional nervous elements. 
A large portion of what we know to-day of the morbid 
changes in the glia has been gained from Nissl prepara- 
tions. To be sure, Nissl’s method stains only a portion 
of the glial structure, that is, the nuclei and the cytoplasm 
in close proximity to them; it does not stain the fibrils, 
the protoplasmic network of His and Held, or the limiting 
membranes. But the pictures obtained by Nissl methods 
are extraordinarily clear and useful; they are remarkable 
for a constancy not possessed by any other method, and 
they allow the certain recognition of even slight patho- 
logical changes. We now possess many methods (Held, NORMAL NEUROGLIA 
53 
Eisath, Alzheimer, Walter, Cajal) which present the glia 
much more completely, and which are invaluable for 
special studies; but none of them is so easy to use or has 
the constancy necessary in the study of pathological 
material, as Nissl’s stain. 
How the normal glia appears in Nissl preparations has 
been described most thoroughly by Nissl himself.14 He, 
like Weigert, distinguishes among the nuclei, those that 
are small, round, deeply staining, that is, richly chromatic, 
Glia cells from the 
i white matter 
kGlia cells from the white matter 
edium sized cell from, the cortex 
Small cells from the cortex 
Normal astrocyte, from superficial cortical layer 
Fig. 22.—Normal glia cell, alter drawings by Alzheimer. Nissl stain. 
those that are large, poorly staining, usually egg-shaped, 
and transitional forms with moderately intense or light 
staining (Fig. 22). The nuclear membrane is always 
distinct, and frequently shows a composition of closely 
placed granules. Frequently, there is only a suggestion of 
the cell body, made out rather by its optical than by its 
staining properties. The cells of the white matter of the 
brain as a rule show a somewhat larger body, which lies 
like a little pale crescent about the nucleus, or surrounds 
it as a loose network. In the white columns of the cord, 
there is frequently a suggestion of spider-like forms of the 
14 Nissl, Ilistol. u. histopathol. Arbeiten, Vol. I, 1904, and: JJber einige Beziehungen, 
etc. Archiv. f. Psych., Vol. XXXII, 1899. 54 
NERVE FIBRES, NEUROGLIA 
cell body. The cytoplasm, so far as it can be demon- 
strated by the methylene blue method, is a definitely 
web-like structure; its external delineation, in the vast 
majority of cases, is not possible, and a cell membrane 
is never present.15 
The cell body may contain pigment. Finally it is well 
to remember as an established fact that the glia cells which 
have produced fibrils tend to undergo slow and gradual 
retrogression. The nucleus becomes smaller and stains 
more deeply, the cell body likewise takes on a deeper stain, 
and finally the entire element shrinks into an irregular, 
spider-like, dark-staining body. This applies not only to 
pathological conditions, but one frequently finds such 
forms normally, for instance in the external glial layer of 
the cerebral cortex (Fig. 22 a). 
15 This is readily understood if we recall the diffuse mesh-like structure of the 
entire glia protoplasm; by Nissl’s method only the protoplasm in the immediate 
neighborhood of the nuclei is brought out, and this gradually disappears in the sur- 
rounding meshwork. CHAPTER IV. 
THE MESODERMAL TISSUE. BLOOD-VASCULAR AND LYMPH 
VASCULAR APPARATUS. GENERAL HISTOLOGICAL STRUC- 
TURE OF THE NERVOUS TISSUE. 
According to what has been said in the previous 
chapter, the neuroglia holds together the various constit- 
uents of the central nervous system in a very complete 
manner, and wherever surfaces exist forms limiting mem- 
branes. Virchow’s term “nerve cement” is a most suit- 
able one for it. 
In the central nervous organs, mesenchymal connec- 
tive tissue is not a component of the nervous tissue proper. 
But the peripheral nerves, as nerve trunks and as fibres, 
are enveloped and held together by connective tissue; 
of ectodermal origin here we have only the actual sheaths 
of the axis cylinder, the medullary sheath and the neuri- 
lemma. The glial envelope terminates sharply at the 
nerve roots as they leave the brain, medulla, and spinal 
cord; even the spinal ganglia are embedded in connective 
tissue, and each of their ganglion cells is surrounded by a 
layer of connective tissue cells. 
The central nervous system likewise contains a cer- 
tain amount of mesodermal tissue, but this is limited, aside 
from the pia and dura, exclusively to the walls of the blood 
vessels and the scanty, loose, fibrous tissue accompanying 
them. Both the larger vessels and the rich capillary 
network (Fig. 35) penetrate the substance of the central 
organs in all directions, but everywhere the ectodermal 
nervous tissue is separated from them by a glial limiting 
membrane. “The vessels are no more an integral part 
55 56 
THE MESODERMAL TISSUE 
of the central nervous system than is the pia itself” 
(Weigert). An interlacing of connective tissue and glia 
fibrils, a penetration of the connective tissue between the 
ectodermal constituents of the nervous tissue nowhere 
takes place; wherever there are vessels with their accom- 
panying connective tissue, they are surrounded on all sides 
by the smooth glial membrana limitans perivascularis. 
The arteries of the central organs all come from the 
pia. Short branches radiate into the cerebral cortex per- 
pendicularly to the surface. They split soon after their 
entrance into delicate arterioles and capillaries. A 
portion of them penetrates more deeply and supplies 
parts of the subjacent white matter. The brain stem and 
the greater part of the white matter are nourished by 
arteries which branch off perpendicularly from the circle 
of Willis. In the spinal cord the white substance obtains 
its blood supply from vessels which penetrate radially 
from the surrounding pia. Short arterial branches pass 
horizontally into the depth from the pial septum lying in 
the ventral median fissure of the spinal cord, enter the 
gray substance on both sides of the central canal, and 
split here rapidly into a number of branches which supply 
the anterior and posterior horns of the spinal cord. The 
venous vessels of the brain are partly collected in the pia, 
partly they are united in the deeper parts below the sple- 
nium of the corpus callosum into the vena magna Galeni. 
The pial veins as well as the vena Galeni empty into the 
sinus durse matris. It is important to remember the 
presence of solid, protoplasmic, connecting bridges be- 
tween neighboring capillaries. In these, new formation of 
capillary tubes takes place under pathological conditions. 
The scanty connective tissue which accompanies the BLOOD VESSELS 
57 
ingoing and outgoing vessels as adventitia is the only 
connective tissue in the central organs, and may be re- 
garded as the direct prolongation of the pia.1 
We distinguish in the vascular walls of the central 
nervous organs, as well as elsewhere in the body, three 
layers—intima, media, and adventitia.2 This is- only a 
gross external division; in reality the vascular tube is 
developed as a homogeneous protoplasmic network of the 
mesenchyme, in which muscle cells, collagen and elastic 
fibres are formed. Only the origin of the endothelial cells 
is still in dispute.3 In the present state of our knowledge 
the elastic fibres and membranes play an important role 
in histopathology; for their selective demonstration we 
possess in Weigert’s resorcin fuchsin stain an excellent 
method. The vessels show beneath the endothelium a 
special layer, the membrana elastica. In the arteries this 
is present as a compact layer of laminated structure, with 
the well-known fenestrations, whence the term membrana 
fenestrata, and exhibits a regular waviness, with projec- 
tions into the lumen of the vessel at regular intervals 
(Fig. 37). (The translators regard this appearance as an 
artefact produced by postmortem conditions and by the 
fixation and contraction of the tissue). It is less marked, 
less compact, and less wavy in the veins. (A very delicate 
elastic membrane is also present in the capillaries; 
though it stains with much greater difficulty.) Elastic 
fibres and very fine lamellae pervade the muscularis in the 
form of a relatively open network. They envelope here 
1 Held differentiates between an inner layer of the adventitia, belonging to the 
vessel, and an outer (intima piae perivascularis), which represents the continuation of 
the pia mater. 
2 Cf. Evensen, Beitrdge zur normalen Anaiomie der Hirngefdsse. Histol. u. histo- 
pathol. Arbeiten von Nissl, Vol. II, 1908. 
3 O. Ranke, Zeitschr. f. d. ges. Neurol, u. Psych., Ong-Bd., XXVII, 1914. 58 
THE MESODERMAL TISSUE 
every individual muscle fibre. Where a definite adven- 
titia is present, abundant elastic fibres also may be found 
in this layer. 
The endothelium of the vessels of the brain and cord is 
everywhere single-layered.4 It consists of large, flat, 
diamond-shaped cells, the cytoplasm of which stains 
normally with difficulty, but the borders of which may be 
made visible by gold impregnation methods. They form 
an uninterrupted inner coating of the vessels. The nuclei 
of these cells are elongated and their long axis always lies 
in that of the vessel. Their size is everywhere approxi- 
mately the same, even in vessels of the most different 
calibre. In the capillaries one finds the nuclei at regular 
intervals, alternately left and right, or above and below. 
They possess a distinct nuclear membrane, which fre- 
quently permits recognition of its composition of fine 
granules; and in their interior there are mostly two to four 
small deep-staining bodies like nucleoli. In form, the 
nuclei are much flattened; they therefore appear pale 
when viewed on the flat, and dark, when viewed on edge 
(Figs. 46 and 50; Fig. 48). The endothelial nuclei fre- 
quently contain peculiar small vacuoles, to which Nissl 
first called attention (Fig. 37). The latter appear as 
circular holes with sharp margins, as if punched out, and 
possess strongly refractile contents which do not stain 
with any of the common dyes. They are frequently found 
lying partly in the nucleus and partly in the cytoplasm. 
Apparently they are regularly extruded into the cell 
body, the endothelial nucleus then closing itself behind 
them and assuming its previous form. Their significance 
4 This is important in histopathology; the presence of more than one layer is 
pathological. LYMPHATICS 
59 
is still uncertain. They occur not only in man but in 
animals as well. Their recognition may be of importance; 
nuclei which contain them may be at once regarded as 
endothelial nuclei. 
The capillary tube consists only of a very delicate 
protoplasmic membrane (Fig. 53) which on the inside is 
lined by a delicate elastic membrane, upon which is 
placed the endothelial cell layer. Externally it passes 
into the scanty adventitia. 
The muscularis, in the majority of the arteries of the 
central organs, consists simply of one layer of circularly 
arranged muscle fibres, since, with few exceptions, only 
small vessels are concerned here. The long axis of their 
nuclei, therefore, is perpendicular to that of the endo- 
thelial nuclei. 
The adventitia consists of a loose, syncytial, protoplas- 
mic network of varying width and density. The tra- 
beculae of the network exhibit a varying number of nuclei, 
as well as collagenous and elastic fibres (Held). The 
entire adventitia may be well demonstrated by the silver 
method of Achuccaro. In the Nissl preparation one 
recognizes only the individual spindle-shaped or more 
rounded nuclei, without distinct cell bodies. 
Lymphatics.—In the brain and cord, there are no 
closed lymph channels with walls of their own. Larger 
lymphatics are only formed outside of the substance of 
the brain and cord. They emerge from the skull through 
the jugular foramen and the carotid canal and leave the 
spinal canal with the vertebral vessels. The cerebral 
lymphatics communicate with the deep cervical lymph 
nodes. There are also probably connections with the 
ear, nose, and pharynx (A. Hauptmann). Concerning the 60 
THE MESODERMAL TISSUE 
direction of the lymph current, contradictory statements 
have been made. Within the brain and cord the lymph 
circulation takes place merely in spaces around the blood 
vessels; and these spaces communicate with the great 
lymph sacs lying in the clefts of the connective tissue 
coverings of the central nervous system (dura and pia).5 
They stand also in communication with the ventricles. 
They all belong to the general lymph system of the body, 
although constituting only an accessory part of it (Bartels). 
The dura mater surrounds the brain and cord as a 
smooth walled sac. Two sickle-shaped processes (the 
falx and tentorium) penetrate into the great clefts be- 
tween the cerebral hemispheres and between the cerebrum 
and cerebellum respectively. The pia mater, however, 
which carries the nutrient vessels, is everywhere closely 
applied to the surface of the central nervous organs. It 
follows all depressions and fissures, fills them, and bridges 
them with an externally smooth investment. It has been 
customary from early times to divide the pia mater into a 
loosely meshed external layer of connective tissue trabec- 
ulae, the arachnoid, and an inner sheet of connective 
tissue (the pia in the narrower Sense6) which closely hugs 
the surface of the nervous organs; the latter contains the 
pial vessels. (For the following see Fig. 44). The space 
lying between the smooth inner surface of the dura and the 
8 Cf. Schroder, Paralyse und Entzundung, Zeitschr. f. d. ges. Neurol, u. Psych., 
Vol. LIII, 1920. 
6 Others restrict the term arachnoid to that delicate layer which is everywhere 
adjacent to the dura; and the connecting meshwork between this arachnoid and the 
inner layer of the pia (pia in the more restricted sense) is called subarachnoidal tissue. A 
sharp separation is nowhere possible. The free surface of the arachnoid, like the op- 
posite inner surface of the dura, is covered with a continuous layer of flat endothelium; 
in it there are found, at regular short intervals, small round nodules consisting of accum- 
ulations of endothelial cells; these may enlarge in all irritative conditions. LYMPHATICS 
61 
equally smooth outer surface of the arachnoid is spoken of 
as the subdural space; it is lined by a continuous layer of 
endothelium and contains small quantities of lymph. 
The loose trabeculae of the arachnoid, likewise every- 
where covered with endothelium, serves as a framework 
for another lymph space. In its entirety it forms a sac 
filled with fluid, and is particularly spacious around the 
spinal cord, protecting the cord like a water cushion from 
mechanical injuries. This is known by the term subarach- 
noidal space. 
Between the innermost layer of the pia and the exter- 
nal surface of the brain and cord, represented by the glial 
membrana limitans superficialis, lies only an imaginary 
space (His), the epicerebral or epispinal space. 
There are therefore found within the meninges the 
subdural space, between the dura and the arachnoid, and 
the subarachnoidal space, within the loose tissue of the pia- 
arachnoid. The pia and the surface of the brain and cord 
are closely apposed, and there is no space between them. 
If we now wish to obtain an idea of the situation of the 
lymph paths within the brain and spinal cord, we must 
recall the blood supply of these organs. The central 
nervous organs do not possess blood vessels in their inter- 
ior during the first developmental stages. Only when a 
later stage has been reached, vascular buds from the pia 
push in against the hitherto exclusively ectodermal tissue 
of the brain and cord. These buds everywhere thrust 
themselves against the glial membrana limitans super- 
ficialis, invaginating without ever penetrating it. Thus 
the specific ectodermal nervous tissue remains every- 
where closed and sharply separated from the vessels as 
they push forward, by the glial membrana limitans peri- 62 
THE MESODERMAL TISSUE 
vascularis; this latter is nothing more than a continuation 
of the membrana limitans superficialis and therefore of 
ectodermal nature. These relations hold for the larger 
vessels as well as for the capillaries. 
Along with these vessels, a scanty amount of connec- 
tive tissue pushes in from the pia. It plays the role of 
adventitia, and its enternal layer lies in close apposition 
to the perivascular glia membrane. 
As the result of this arrangement the subarachnoidal 
lymph spaces are continued into the interior of the brain 
and cord, as spaces which accompany all vessels, large 
and small, within their adventitial connective tissue; and 
the latter is merely a continuation of the pia. These 
spaces internally are contiguous to the vessel wall (mus- 
cularis or intima); externally they are bordered by a deli- 
cate layer of the adventitia closely applied to the glial 
membrana limitans perivascularis. In this manner a 
system of clefts, serving for the circulation of the lymph, 
encircles all vessels of the central organs down to the 
capillaries; it empties on the surface of the brain and cord 
into the great lymph sac of the pia mater.7 (For what 
follows, compare Fig. 45.) 
These lymph paths around the vessels are the adventi- 
tial or Virchow-Robin spaces. They are of greatly varying 
width, larger in the gray substance than in the white. 
Under pathological conditions, they may be distended 
with fluid or cellular elements, and then assume consider- 
able dimensions. The adventitial connective tissue which 
7 See Nissl et al in Zentralbl. f. Nervenheilk. u. Psych., 1903, p. 90; Held, Monatschr. 
J. Psych, u. Neurol., Vol. XXVI, 1909, and Abhandl. d. math., phys. Klasse der Konigl. 
Sachs. Ges. der Wissenschaften, Vol. XXVIII, 1903. Concerning the respective relations 
in retina and optic nerves, see Kriickmann, Zeitschr. f. Augenheilkunde, Vol. XXXVII, 
1917. PERIVASCULAR LYMPH SPACES 
63 
forms them is always very scanty in the smallest vessels; 
as a rule, a fine connective tissue layer rests on the vessel 
wall, but a second layer regularly is applied to the mem- 
brana limitans glise, and the space between these two 
layers is traversed by a very few trabeculae; in some places 
these are entirely absent (Held; see Fig. 45). 
Mention is frequently made in the literature of a 
second, so-called perivascular system of lymph spaces, 
which surrounds the inner or adventitial system. These 
have been termed the spaces of His by analogy with the 
epicerebral and epispinal spaces. It has been supposed 
that they represent a dilatation of the supposititious cleft 
between the usually firmly adherent adventitia and the 
limitans glise. But Held has shown that this is not true. 
Membrana limitans and external adventitial layer are 
always closely welded together; it is believed that they are 
united in such a manner as to form together the limiting 
membrane between the vascular system and the connec- 
tive tissue; there is therefore no second lymph space 
between them. What is usually described as such a 
“perivascular” space lies, as a matter of fact, external 
to the limitans glise, therefore actually within the ecto- 
dermal tissue. If artefacts are avoided by careful fixation 
of the material there is seen everywhere external to the 
glial limiting membrane another layer, consisting of 
protoplasmic and membranous portions of the glial 
syncytium whicli are arranged radially and stand per- 
pendicularly to the limiting membrane; these form a 
mesh work (Held). With ordinary methods of fixing and 
preparing tissues, especially alcohol fixation, these fila- 
ments and membranes are torn apart by shrinkage, and 
larger spaces are created; these are pure artefacts, not THE MESODERMAL TISSUE 
64 
preformed structures. In such sections one finds the vessei 
occupying a markedly eccentric position within a widely 
dilated space; closer observation shows that the vessel 
wall is encircled first by a more or less wide adventitial 
space, lined on both sides by scanty adventitial con- 
nective tissue, and traversed by a few filaments; external 
to this space lies the smooth, untorn membrana glise. 
Outside of this in turn is the large shrinkage space (Fig. 
45), traversed in places by ragged, torn, glial threads 
adherent to the glial membrane. The tear always takes 
place in an irregularly circular manner just external to the 
membrana limitans. This membrane being tough, and 
moreover being made more rigid by its glial fibrils, does 
not shrink but remains closely adherent entirely or in 
part to the blood vessel, while the soft, juicy surrounding 
nervous tissue retracts. It is these shrinkage spaces that 
are often regarded as preformed lymph spaces, are con- 
fused with the spaces of His, with which, however, they 
have nothing to do; for the spaces of His lie internal to 
the membrana limitans perivascularis, while the artificial 
shrinkage spaces are found external to this membrane. 
In the literature, the statement is often encountered 
that lymph spaces exist around the ganglion cells of the 
brain and cord, the so-called pericellular spaces. Ober- 
steiner 8 asserts that in specially well-made injection prep- 
arations one may satisfy himself that these pericellular 
spaces may be injected by way of the perivascular spaces. 
These pericellular spaces likewise are artefacts, due to 
distortion or tearing of the tissues through shrinkage. 
In fact Golgi expressed himself as opposed to regarding 
8 Obersteiner, Anleitung beivi Studiwa des Baues der nervosen Zentralorgane, 1901, 
p. 220; see also Sittig, Zeitschr.f. d. ges. Nenrol. v. Psych., Vol. VIII, 1911, PERICELLULAR LYMPH SPACES 
65 
them as pre-existing spaces; afterward Nissl, and, follow- 
ing him, Held, furnished experimental proof that they are 
only present after the use of certain fixing and hardening 
agents, particularly alcohol. It is furthermore possible to 
produce extremely large spaces, if one places fresh blocks 
of postmortem tissues for a while in water before further 
preparation; and a similar condition is seen in cases of 
antemortem cerebral or spinal edema. 
Nissl has further shown that even the term “pericel- 
lular” is inaccurate for these spaces, since they do not 
lie outside the cell surface but inside the cytoplasm. In 
the process of shrinking, the surface of the ganglion cell, 
with its fibrillar and Golgi net apparatus (see Chapter 
II) is remarkably resistant. Because of this, the surface 
of the ganglion cell is firmly united to the surrounding 
tissue. On the other hand the soft and obviously more 
watery protoplasmic body is considerably retracted during 
the process of dehydration. Hence it is within the cell 
that the tear occurs, though usually near its surface. 
Within the shrinkage space thus formed, lies the nucleus 
with the larger part of the more or less shrunken cyto- 
plasm, the external border of which is jagged or frayed. 
And the external wall of the space is covered on the inside 
by a delicate, likewise irregular fringe of cytoplasm, which 
remains attached to the Golgi nets. At the base of many 
small pyramidal cells of the cortex there is usually torn off 
with the axis cylinder a shield-shaped mass of cytoplasm; 
often it may still be recognized that the projections and 
indentations of the outer fringe fit those of the retracted 
cell body, or that still untorn filaments traverse the space 
between. In close proximity to the ganglion cells lie as a 
rule one or two glia cells (the so-called satellites). The 66 
THE MESODERMAL TISSUE 
(а) Three pale ballooned glia nuclei with a nucleolus-like body, and a pale, naked, irregular 
cytoplasm. From the neighborhood of an abscess. Nissl stain. 
(б) Large stellate glia cell, with large clear nucleus and dark cytoplasm, which has everywhere 
concave borders. From a case of amyotrophic lateral sclerosis. From the pyramidal tract. 
Nissl stain. 
(c) So-called monster glia cell; dull opaque cytoplasm with irregular contour, very dark, rod- 
shaped nucleus. From a case of diffuse sclerosis. To the left is a gliogenic granule cell. 
Nissl stain. 
(d) As in c, stained with hematoxylin, van Gieson. 
(e) Rod cell, Nissl stain. 
(/) Greatly proliferated glia cells from the cortex of a case of dementia paralytica, with a 
large nucleus and large cytoplasm. 
(ff) Glial syncytium composed of huge glia cells. 
Fig. 23.—Different forms of pathological glia cells, (a) to (e) photographs, oil immersion; 
(/) and (g) after drawings by Alzheimer. ADVENTITIAL LYMPH SPACES 
67 
nuclei of these satellites during the process of shrinking 
not infrequently get into the external portion of the 
shrinkage space, and appear as if they were lying free in 
the space. Such elements have been looked upon by some 
authors as endothelial cells or as freely circulating lym- 
phocytes, and their presence has been wrongly regarded as 
additional evidence for considering these clefts as true 
lymph spaces. 
We are apt to overlook the fact that in most prepara- 
tions we find entirely analogous spaces around the glia 
cells; these are likewise to be regarded as shrinkage, and 
not as lymph spaces. 
W e see from the above that in the brain and cord only 
the adventitial spaces of the vessels are lymph spaces; 
these are direct continuations of the great lymph sac 
which surrounds the entire central nervous system. The 
clefts around the vessels and cells, which are frequently so 
striking in our preparations, are artefacts, and not pre- 
existing spaces. Following the finest capillaries the ad- 
ventitial lymph paths everywhere penetrate deeply into 
the tissues; but the ectodermal tissue is everywhere 
sharply separated from them by glial membranse limi- 
tantes. The fluid exchange beyond these membranes takes 
place either through the delicate clefts between the indi- 
vidual ectodermal constituents or along the protoplasmic 
paths within the polymorphic glial reticulum. A series of 
observations on pathological material emphasizes the im- 
portance of the latter route. 
Whatever else has been asserted about the lymph 
paths, and even about closed lymph vessels, has not 
hitherto stood the test of critical examination. 68 
THE MESODERMAL TISSUE 
An exact knowledge of the lymph paths in the brain 
and cord, and of the significance of the adventitial spaces 
as lymph channels, is of great value in understanding a 
series of important pathological processes. Within these 
channels are collected, under pathological conditions, 
elements such as lymphocytes, leucocytes, plasma cells, 
granule cells, etc. Further, they are the places in which 
products of disintegration and of metabolism, and in 
which cells from the tissues accumulate, to be transported 
by the lymph stream. The interpretation of such find- 
ings without exact knowledge of the lymph paths has 
frequently led to erroneous conceptions. 
From the tissue elements described in the preced- 
ing chapters, the various parts of the nervous system 
are constructed. 
The histological structure of the peripheral nerves is 
simple. Parallel myelinated nerve fibres possessing a 
neurilemma are joined into round bundles by scanty 
laminated connective tissue (endoneurium) and enveloped 
by a tougher connective tissue layer which is inter- 
woven with elastic fibres (the perineurium); a smaller 
or larger number of such bundles has again a common 
connective tissue covering (epineurium), and constitues 
a peripheral nerve. 
In the spinal cord, as well as elsewhere in the central 
nervous system, we recognize two different types of 
tissue, the white and the gray matter. The gray matter 
of the spinal cord surrounds the central canal, and, in 
cross sections, has the well-known shape reminding one of 
a butterfly. The white matter in turn, surrounds the 
gray, and gives to the spinal cord the shape of a long WHITE SUBSTANCE 
69 
round rod, which shows, throughout its entire length, on 
its ventral surface, a deep cleft, the anterior median 
fissure. The white matter contains, of the specific func- 
tional nervous elements, only nerve fibres. These fibres 
are medullated but do not possess a neurilemma or nodes 
of Ranvier, as do the peripheral nerves. Their calibre 
shows considerable variation. They are of varying length; 
and, for the most part, they run in the longitudinal axis 
of the spinal cord. Only the roots which enter and leave, 
and the arcuate fibres, run transversely. Grouping into 
Capillary 
Small, deep-staining, 
round glial nuclei in the 
immediate neighbor- 
hood of the capillary 
Small, deep-staining, 
round glial nuclei in the_ 
immediate neighborhood “ 
of the capillary 
Leucocyte in 
the capillary 
Endothelial cell shown 
on the flat 
Endothelial cell 
shown half on edge 
Small, deep-staining, 
round glial nuclei in the 
immediate neighborhood 
of the capillary 
Fig. 24.—From the white matter of a case of symptomatic psychosis. Nissl preparation 
Oil immersion. 
separate bundles does not take place; the rough, incom- 
plete separation of the white matter into the so-called 
anterior, lateral, and posterior columns is due to the ante- 
rior and posterior horns of the gray matter, which project 
wing-like antero-laterally and postero-laterally, as well 
as through the root bundles. 
The medullated fibres of the white substance are em- 
bedded in the general protoplasmic glial network, which 
is stiffened by considerable numbers of Weigert’s fibrils 
running for the most part longitudinally. The glia fills 
the spaces between the parallel round nerve fibres. Where 70 
THE MESODERMAL TISSUE 
the fibres are predominantly thin, as for instance in the 
tract of Goll, we find a closely meshed lattice work of 
delicate filaments, while in places where the nerve fibres 
are thick, as in the lateral pyramidal tracts and the tract 
of Burdach, the meshes are wider. On the surface of the 
spinal cord the glia is denser and more rich in fibrils; from 
this glial border delicate and coarser glial septa make their 
way into the interior, the majority tapering into finer 
filaments as they go. Owing to its constancy there stands 
out among them a mesial septum between the two posterior 
columns, and one on each side of these. The lateral septa 
in the upper part of the spinal cord separate the tract of 
Goll from that of Burdach.9 In the radiating septa the 
larger vessels enter from the pia. The adventitia accom- 
panying the vessels nowhere fuses with the glia; the vessels 
are here, as elsewhere in the central nervous system, 
separated from the ectodermal tissue by a glial limiting 
membrane. The glial nuclei are regularly scattered 
throughout the white substance; they lie in the angles 
formed by several neighboring nerve fibres. 
The gray matter of the cord has a more complicated 
structure, being made up of a greater variety of elements. 
It contains the ganglion cells of the cord; these are not 
distributed evenly through the tissue but are arranged in 
groups, each of which contains one or several types of 
cells (motor anterior horn cells, posterior horn cells, cells 
of Clarke’s column, etc.) The gray matter contains, 
besides, the medullated nerve fibres which enter from the 
roots and white columns, and form a dense network. All 
of these fibres have the characteristics of the nerve fibres 
9 There is no such septum between the two anterior columns; here is the ventral 
median fissure, into which extends a continuation of the pia with vessels. GRAY MATTER 
71 
elsewhere in the central organs, that is, they have no 
neurilemma or nodes of Ranvier. The third constituent 
is the glia. Weigert ’s glia fibrils are arranged in a specially 
close lattice work around the central canal. They are 
present in abundance in the anterior horns, much more 
plentifully than in the white matter. The posterior horns 
show a very close network of delicate fibrils in the tract of 
Lissauer, while the substantia gelatinosa is poor in specific 
fibrils. The gray matter is considerably richer in capil- 
laries than is the white. 
In the brain and medulla oblongata the white matter 
Glia nucleus of 
normal appear- 
ance and nor- 
mal size 
Sausage-shaped 
pale glial nuclei ~ 
Sausage-shaped 
pale glial nuclei 
Transverse and 
longitudinal 
sections of tra- 
beculse of the 
proliferated 
protoplasmic 
glial syncytium 
Larger and paler 
nucleus not well 
focussed, both 
surrounded by 
protoplasmic 
trabecul® 
Transverse and 
longitudinal 
sections of 
trabecula! of 
the proliferated 
protoplasmic 
glial syncytium 
(Figs. 25 to 29 represent syncytial proliferation of glial elements. In Figs. 28 and 29, begin- 
ning granule cell formation is shown. All are photographs; oil immersion). 
Fio. 25.—Proliferated glia. Nissl preparation. (From a case of lyssa). 
shows the same general histological structure as in the 
cord; only the grouping of white and gray matter is more 
complicated. In the cerebral hemispheres, the gray 
matter constituting the cortex lies as a thick external 
covering on the white fibre masses. The basal ganglia 
and the nuclei of the interbrain, midbrain, and hind- 
brain constitute separate gray masses. 
The gray matter, just as in the spinal cord, contains 
the ganglion cells; it includes also medullated nerve fibres 
of varying number. Cells and fibres are embedded in a 
matrix which in sections shows a finely granular or some- 72 
THE MESODERMAL TISSUE 
what spongy structure. Concerning the highly compli- 
cated composition of this, only the modern selective stain- 
ing methods have brought any clear information; it is 
made up of dendrites and axis-cylinder processes of the 
embedded ganglion cells, the Golgi nets of these cells, the 
arborizations of axis cylinders from other portions of the 
gray matter, with the fine intercellular fibrillar reticulum 
which probably arises from them, and finally the diffuse 
Proliferated protoplasm of the 
glia reticulum 
Lobulated dark 
staining glia 
nuclei, resem- 
bling in shape 
leucocytes 
Lobulated dark 
staining glia 
nuclei, resem- 
bling in shape 
leucocytes 
Lobulated dark 
staining glia 
nuclei, resem- 
bling in shape 
leucocytes 
Proliferated pro- 
toplasm of the 
glia reticulum 
Lobulated dark staining glia 
nuclei, resembling in shape 
leucocytes 
Proliferated protoplasm of the glia reticulum 
Fla- 26.—Proliferated glia in the neighborhood of a vessel, from the same case as is Fig. 25. 
Nissl stain. 
protoplasmic glial reticulum, with its specific Weigert’s 
fibrils. In the lattice work of the glia, nuclei are fairly 
uniformly distributed. One regularly finds such nuclei 
in close proximity to the ganglion cells, particularly at 
their bases (so-called satellites); and more numerous than 
elsewhere about the larger vessels. This tissue, composed 
of the ectodermal constituents enumerated, is traversed 
by arteries, veins and a rich capillary network, but there 
nowhere takes place an interlacing of the mesodermal 
tissues with the ectodermal. CORTICAL LAYERS 
73 
In the cerebral cortex the various ectodermal tissue 
elements are not represented in equal quantities at the 
various levels. This leads to the formation of layers. 
The medullated nerve fibres penetrate in radiating bundles 
from the subjacent white matter, and lose, after a shorter 
or longer course, their medullary sheaths. Therefore such 
radiating fibres are to be found in greatest quantities in the 
deeper layers of the cortex, while toward the surfaces 
they become more and more scanty. About in the middle 
of the cortex there run, at right 
angles to these, several loose net- 
works of medullated nerve fibres, 
parallel with the surface. A similar 
layer of fine medullated fibres is 
found just below the external surface 
(tangential fibres). The region of 
the radiating bundles is traversed 
also by isolated fibres running ob- 
liquely. Marked variations in these 
details are found in the different 
regions of the cortex. 
The variation in the number 
of Weigert’s glial fibrils in the 
different layers of the cerebral cortex has already been 
mentioned. They are very numerous only in the most 
superficial layer of the cortex, where the tangential fibres 
are found. Toward the depth the fibrils diminish rapidly 
in numbers. On the other hand the protoplasmic glial 
network is apparently equally well developed at all 
levels. Radiating glial septa, such as are found in the 
spinal cord, are not present in the cerebral cortex. 
Fig. 27.—-Proliferated glia. Nissl 
stain. Alcoholic polio-encephalitis. 
Enlarged dark staining nuclei with- 
in dense deep staining protoplasmic 
trabeculae. 74 
THE MESODERMAL TISSUE 
The ganglion cells of the cortex also are divided into 
layers, which run parallel with the surface; these layers, 
however, are not sharply defined. Further, if we make 
perpendicular sections through the cortex, the ganglion 
cells are found to be in radially disposed rows. This 
arrangement is largely due to the course of the radiating 
medullated bundles. In general the size of the ganglion 
cells increases toward the depth; although the lower-most 
layer consists again of smaller elements. The narrow, 
most superficial layer of the cortex is free from ganglion 
cells. Except for the elements in the vessel walls, the 
nuclei here belong exclusively to the glia cells; it is this 
layer in which the tangential fibres are found. 
The number of nerve cells present in a given area of 
the cortical tissue varies greatly in different portions of 
the cortex. The ones farthest apart from one another are 
those of the motor region; they lie most closely together 
in the occipital lobes. Nissl has further shown that in 
general, in animals, the lower the animal is in the develop- 
mental scale, the more cells there are in a unit of space in 
the cortex; he attributes this variation in the numbers of 
cells to the greater or less development of that reticulum 
of very fine nerve fibrils which is termed by him the 
specific “nervous gray.” This analogue to the neuropil 
of the invertebrates is supposed to lie between the gan- 
glion cells; but it has not as yet been demonstrated with 
certainty by staining methods. 
The finer structure of the cortex in different portions 
of the cerebral surface presents further variations. For 
the experienced worker it is not difficult to determine the 
cortical area from which a given microscopic section 
comes. He makes use of such guides as the number of the CORTICAL ARCHITECTURE 
75 
cell layers, the particular structure of certain layers (for 
instance the layer of granule cells), the existence of defi- 
nite cell types (for instance the Betz giant pyramidal 
cells), the width of the entire cortex, the quantity of the 
tangential and radiating fibres,, the width of the strip of 
Gennari, etc. These local differences have been utilized 
to divide the surface of the cerebrum into anatomical 
territories; this is the object of a newly developed field of 
investigation, cortical architecture.10 Flechsig used for 
this purpose the study of the successive myelination of the 
projection and association fibres of the foetus and of the 
young child.11 Kaes 12 undertook a careful count of the 
medullated fibres of the cortex of the adult, in different 
places. The investigations of Brodmann,13 who undertook 
a cytoarchitectonic inventory of man and of the entire 
series of mammals, and mapped out the brain cortex, are 
classical and of particular value. Others have used the 
Golgi methods for similar studies.14 
10 Delimitation according to fissures and convolutions of hemispheres rarely 
corresponds even approximately with definite anatomical and physiological zones. 
11 Flechsig, Gehirn and Seele, Rektoratsrede, Leipzig, 1894 and later editions. O. 
Vogt, Journal f. Psychol, u. Neurol., 1919, Erganzungsheft. 
12 Kaes, Die Grosshirnrinde des Menschen, etc. Atlas and Text, Jena, 1907. 
13 Brodmann, Journal f. Psychol, u. Neurol, Vols. II-X, 1903-1908; Vergleichende 
Lokalisationslehre der Grosshirnrinde, Leipzig, 1909; Handbuch der Neurologie, edited 
by Lewandowsky, Vol. I. 
14 Ramon y Cajal, Studien liber die Himrinde des Menschen. German translation 
by Bresler. Leipzig, 1900-1906. PART II 
HISTOPATHOLOGY OF NERVOUS SYSTEM 
CHAPTER V. 
PRELIMINARY REMARKS ON GENERAL PATHOLOGY. PATH- 
OLOGICAL CHANGES IN THE GANGLION CELLS. 
The nervous system as a whole may be looked upon as 
a single complex organ. Damage or destruction in one 
part of this organ results in a great variety of morbid dis- 
turbances, depending on the nature and number of nerve 
connections which the seat of the initial injury has with 
other parts of the nervous system or with the periphery 
of the body. The most important question, therefore, for 
clinical and pathological studies is often the localization 
of the lesion; the great progress which has been made in 
these branches in the past few decades has been largely 
in the field of localization. Even in the pathological 
anatomy of the nervous system localization of the morbid 
changes has been largely the object of investigation. To 
the histopathology and histopathogenesis of the proc- 
esses little attention has been given for a long time. This 
neglect of histopathology was seemingly justified by the 
experience that mere localization of a lesion often permits 
conclusions to be drawn concerning its nature, inasmuch 
as certain pathological processes are exclusively, or at 
least largely, limited to definite localities within the 
nervous system. 
Staining Methods.—The principal anatomical method 
for investigation of localization is Weigert’s medullary 
sheath stain in serial sections. For the demonstration of 
products of disintegration in early stages, Marchi’s method 
76 STAINING METHODS 
77 
(blackening of fatty substances by osmic acid) has won a 
place for itself. Both methods allow histopathological 
conclusions to be drawn, but only to a limited degree. 
W eigert’s medullary sheath stain was the first of a 
series of so-called selective staining methods. Such 
methods, designed to bring out only certain definite com- 
ponents of the tissue, are both in number and in excellence 
available for the nervous system as for no other organ of 
the body. Following the medullary sheath stain came 
Nissl’s cell stain, Weigert’s glial fibril stain, and the 
series of new neurofibril and axis-cylinder methods.1 Em- 
ployment of such methods permitted a better insight into 
the complicated structure of the nervous system. 
For the study of normal histology the number of such 
selective staining methods can hardly be large enough; 
but their use in histopathology is rather limited. The 
conditions under which the histopathologist must work 
are in many respects different from those of the histologist. 
The former must above all demand from his methods 
reliability; which means, that under all conditions, 
in every specimen which he examines, he must get de- 
pendable results if he is not to waste his material. In this 
regard the pure histologist is usually much better off. 
The pathologist must demand, and this applies not only to 
the selective methods, that the preparation give constant 
equivalent pictures (Nissl), that is, that all the normal 
constituents of the tissue are represented always in the 
same manner, precise knowledge of which is already pos- 
sessed by the observer. With this knowledge at hand he 
is at any time justified in interpreting deviations from an 
equivalent picture as expressions of pathological changes 
1 The Golgi method also, with certain restrictions, belongs among the selective 
methods. 78 
REMARKS ON GENERAL PATHOLOGY 
in vivo. There must be no room for doubt as to how 
far the caprices or imperfections of technic have to be 
taken into account. Finally, it is desirable that any 
method employed should allow the recognition of as 
many different details as possible. 
There is no single method which meets all these 
requirements. Certain structural relations and certain 
pathological products can only be demonstrated by spe- 
cific staining methods. Therefore the pathologist must 
always make a choice from the great number of fixing and 
staining methods at his disposal. Even his choice of a 
fixing method is of decisive importance. For the demon- 
stration of the ganglion cells by Nissl’s method, fixation 
of the fresh material in strong alcohol is necessary. 
Formalin is required for Bielschowsky ’s neurofibril stain. 
For Weigert’s medullary sheath stain, chromating of the 
blocks is imperative (with or without previous fixation in 
formalin); preliminary treatment with alcohol is not per- 
missible, since alcohol extracts the myelin. For Weigert ’s 
glial stain and for other technics special fixing methods 
are required. 
By selective staining methods it is possible to demon- 
strate some individual constituents of the tissues (med- 
ullary sheath, glia fibres, ganglion cells, neurofibrils), 
but the histological picture so obtained is necessarily an 
incomplete one. On the other hand, the diffuse stains 
(carmin, hematoxylin—van Gieson, hematoxvlin-eosin, 
etc.) differentiate poorly the various elements from one 
another, particularly in the gray matter. It is therefore 
always recommended in histopathological studies that, be- 
sides diffusely staining dyes, selective methods beemployed. 
The best method for preliminary orientation, as well STAINING METHODS 
79 
as for the study of many details, is Nissl’s; and this is 
true, as already stated, even when changes in the ganglion 
cells are not the chief object of investigation. Nissl’s 
method is remarkable for its absolute reliability and 
constancy. It gives very clear general pictures; and while 
the nerve fibres, glial fibrils, and fine punctiform or 
spongy intercellular tissues remain unstained, yet it 
brings out better than does any other method the slightest 
changes in the glial nuclei and the surrounding cyto- 
plasm, as well as in the vessels. And in the present state 
of our knowledge these are the particular elements, 
changes in which must be studied most carefully for 
proper comprehension of morbid processes. Then, too, 
Nissl’s is almost the only method for the study of the 
cellular “architecture” (the arrangement of the cells, 
layer by layer) of the cortex and deeper parts of the brain. 
In the articles on the histopathology of the central ner- 
vous system descriptions of the morbid alterations of gan- 
glion cells have for a long time occupied much space. 
There was a time when many investigators believed, with 
Nissl, that chief emphasis should be placed on the demon- 
stration of such changes. These men cherished the hope, 
that it might be possible to found a system of patho- 
logical anatomy of the brain and cord merely on specific 
changes in the ganglion cells. The result was that many 
investigators occupied themselves exclusively with these 
cells, more or less neglecting the other tissue elements. 
Experience has taught that these hopes were ill founded, 
and that such diagnostic importance for pathological 
anatomy cannot be attributed to morbid alterations 
in the ganglion cells.2 
2 S. P. Schroder, “Franz Nissl.” Monatsschr. f. Psych. Vol. XL VI, 1919. REMARKS ON GENERAL PATHOLOGY 
80 
We understand now, in retrospect, why such excep- 
tional importance was attributed to alterations in these 
cells. The time of these views and hopes corresponded to 
the golden age of the neurone doctrine, a theory which, as 
we have seen, regarded the ganglion cells alone as the place 
of all active nervous and psychic processes. It looked upon 
the nervous system, considered from a functional stand- 
point, merely as a conglomeration of ganglion cells with 
axis-cylinder processes which connected them with each 
other and with the periphery. Another important factor 
was the publication about this time of Nissl’s staining 
method, which brought out the slightest changes in certain 
parts of the ganglion cells with unprecedented certainty 
and clarity. These two facts in conjunction had the 
effect of giving to the ganglion cell an almost dominat- 
ing position in the field of pathological neuroanatomy. 
Nissl himself, in 1890, described a change in the large 
motor cells of the anterior horns of the cord and in the 
motor nuclei of the medulla oblongata, which follows 
transection of, or tearing out of the motor nerves belonging 
to them. This well-characterized change of cell body and 
nucleus (Nissl’s so-called acute cell degeneration) appears 
constantly after injury of the axis cylinders of motor 
nerve fibres. Therefore it has been used to demonstrate 
experimentally that certain nerve fibres are connected 
with certain gray nuclei. However, these changes in the 
ganglion cells result not only from injury to the axis 
cylinder, but often follow certain generalized processes 
in the nervous system. 
A few years later Nissl3 reported the results of exten- 
sive and careful studies on changes in nerve cells of ani- 
3 Nissl, Die Hypothese der specifischen Nervenzellenfunktion, etc. Ally. Zeitsschr. 
f. Psychiatrie, Vol. LIV, 1898. STAINING METHODS 
81 
mals following “subacute maximal” intoxications. He 
investigated a variety of poisons (organic, inorganic, 
toxins), and reached the conclusion that all of these sub- 
stances produce definite cellular alterations, and that each 
of them affects the cell in a characteristic manner. His 
descriptions are masterly, and are done with a thorough- 
ness unrivaled by any other writer on the subject. 
Following this work of Nissl, contributions began to 
pour in from all sides. Changes in nerve cells were de- 
scribed in the most varied intoxications, in starvation, 
insomnia, experimental uremia, experimental anemia, 
experimental fever, freezing, from strong electric currents, 
in cerebral abscess, in injuries, asphyxia, autointoxica- 
tions, as well as in the most diverse diseases of the per- 
ipheral and central nervous organs.4 In a large proportion 
of such reports wre meet the open or hidden tendency to 
claim the described nerve cell changes as “specific” for 
the injurious factor. Many merely concern themselves 
with the large motor cells, which are the easiest to find 
and to identify. Even in the field of psychiatry search 
was made for changes in the nerve cells specific for the 
various psychoses; and here again the inclination existed 
to explain the entire pathological anatomy of the cerebral 
cortex solely in terms of changes in the nerve cells. 
Only gradually have these Utopian hopes been dis- 
carded. Nissl himself was one of the first to warn against 
exclusive attention to the nerve cells, and against exag- 
gerating the importance of their changes to histopath- 
ology. He contended that there are no specific nerve 
cell changes in any disease, and that it is not to be assumed 
offhand that certain nerve cell changes correspond to 
4 An English review of the literature (Brain) in 1899 already cites 523 references. 82 
REMARKS OX GENERAL PATHOLOGY 
definite disturbances in nervous function. A certain 
disappointment was the immediate result; but subse- 
quently, as often happens in such cases, the pendulum 
swung too far in the other direction. Long and diligent 
study of the morbid alterations of the ganglion cells has 
given us a mass of observations, a part of which at least 
will undoubtedly have permanent value. But it is now 
recognized that the changes in the nerve cells with which 
we are familiar, with the possible exception of a few 
special types, are of significance only in connection with 
the changes of all the other tissue constituents, that is, 
as part and parcel of the whole histological picture. To 
seek for specific nerve cell changes in a particular clinical 
disease picture must be looked upon today as useless. 
Whether one or another disease of the nervous system will 
prove in the final analysis to be characterized by definite 
alterations in the nerve cell equivalent picture, must be 
reserved for further investigation to determine.5 Expe- 
rience thus far gained has taught that in human pathology 
the majority of the known cellular changes may be traced 
to the general somatic disturbances which accompany the 
nervous malady (fever, anemia, malnutrition, exhaustion, 
edema, etc., including the special changes occurring in a 
long agonal period), and that they therefore cannot be 
regarded as “specific” to any disease process. 
The 'pathological changes in the nerve cells have been 
studied chiefly with the help of NissEs method. The 
older diffuse tissue stains gave only indefinite pictures of 
the ganglion cell body. Moreover, fixation of the material 
6 Such characteristic changes appear to be the cellular alterations found by 
Schaffer in familial amaurotic idiocy. In obviously closely related conditions, Spiel- 
meyer (Ilistol. u. histopathol. Arbeiten von Nissl, Vol. II, 1908) has described widely 
distributed cytoplasmic inclusions of a peculiar granular mass. STAINING METHODS 
83 
in chrome salts, formerly generally used, caused various 
artefacts; Nissl’s introduction of strong alcohol in the 
technic meant a great advance in this respect. The 
agents used for fixation of nervous tissues, such as formalin, 
bichloride and others, give, in part, equally good pictures 
of the cell body; but it must not be forgotten that the 
equivalent pictures which they produce are different from, 
and therefore cannot be compared offhand with that 
obtained after alcohol fixation. By Nissl’s method there 
are stained, besides the nucleus, only certain cytoplasmic 
substances lying between the neurofibrils. These sub- 
stances are apparently of extremely labile structure, 
rapidly and markedly changed by injuries of all kinds, 
probably even during normal function. This may be the 
reason that perfectly normal cell pictures may be said to 
be almost never obtained from human material.6 The 
injuries during the agonal period often suffice to produce 
changes; and even in decapitated human subjects the 
great majority of the cortical cells do not give the normal 
equivalent picture as we know it in healthy animals which 
have been decapitated. These experiences alone show 
that one must be cautious in evaluating nerve cell changes 
for the histopathology of certain disease processes. The 
changes described may be interpreted in different ways; 
for this reason the much used terms sclerosis, chroma- 
tolysis of nerve cells, etc. have little definite meaning. 
Only a detailed analysis of the changes can be of value, 
going hand in hand with definite knowledge as to how the 
deviations from the equivalent pictures are to be eval- 
uated in the individual case, and especially whether they 
6 Nissl in 1914 stated that in his collection of about 1000 human cases he possessed 
but a single brain which in a measure corresponded to all the requirements of a normal 
one. 84 
REMARKS ON GENERAL PATHOLOGY 
allow a conclusion as to the severity of the condition and 
possibility of recovery from it. Of such observations 
Nissl has reported a whole series. 
Of the other selective cell stains, Golgi's is of little use 
for pathological purposes. It does not give histological 
details of cellular structure, but merely shows the cell 
in its entirety as a silhouette. It is furthermore not 
dependable or constant. Care is therefore indicated in 
attempting to interpret results gained by this method in 
the field of pathological anatomy. There has been no lack 
of such attempts; thus the occurrence in Golgi prepara- 
tions of certain irregular, rosary-like thickenings of the 
dendrites has been interpreted as the anatomical expres- 
sion of the ability to retract on the part of the cell proc- 
esses (so-called theory of the plasticity of the neurones); 
and certain writers have attempted to explain physiological 
processes (sleep) as well as pathological (stupor) on this 
ground. More critical observers have demonstrated that 
such thickenings may at any time be artificially produced 
by modification in the fixation of the material. 
In diseased ganglion cells stained by Nissl’s method 
the most striking changes are in the appearance of the 
stainable substances. The normal arrangement of this 
material into lumps, or rows of granules, or heaps of 
granules, etc., to which is so largely due the characteristic 
appearance of the Nissl equivalent picture, undergoes a 
variety of changes under pathological conditions. But, 
besides this, the Nissl picture permits recognition of 
changes in the normally unstainable substances (un- 
stained paths, axis-cylinder processes), chiefly in that the 
normally unstained portions become more or less dis- 
tinctly colored. Furthermore, the method shows changes TYPES OF CELLULAR DEGENERATION 
85 
in the external form of the cells and their processes, 
changes in the size, form, staining qualities and position of 
the nucleus, changes in the structure of the nuclear con- 
tents, particularly of the nucleolus, as well as in other 
details. By far the best descriptions dealing with this 
subject come from Nissl himself; he has not only defined a 
considerable number of characteristic types of cell 
changes, but he has also traced as far as possible the 
various stages through which they pass.7 
One of these types has been termed by Nissl chronic 
cell degeneration. This may be recognized by the fact that 
the entire cell becomes smaller, that the unstainable 
parts become stainable, and that the stainable substance 
shrinks somewhat. At the same time the nucleus becomes 
smaller; it assumes a somewhat elongated form; its con- 
tents stain somewhat more intensely. In further stages 
the cell shrinks more and more, the dendrites assume a 
corkscrew form. The stain of the entire cell becomes 
denser and more uniform, and finally no structural details 
may be recognized in the interior (Figs. 9 and 10). 
Another equally well-known and recognizable type is 
termed by Nissl acute degeneration of nerve cells (Fig. 3 
which compare with normal cell, Fig. 2; Fig. 5, compare 
with Fig. 4). But this certainly does not represent the 
only way in which nerve cells can become acutely diseased. 
In contrast to the chronic type, the cell here undergoes a 
swelling and rounding of its external form. The stainable 
portion appears pale, and becomes peculiarly and uni- 
7 See for the following: Nissl, liber einige Beziehungen zwischen Nervenzellerkrank- 
vngen, etc., Archiv. f. Psych. Vol. XXXII, 1899, and Beitrage zur Frage nach der 
Beziehung zwischen klinischen Verlauf und anatomischen Befund, etc., Vol. I, No. 3, 
Berlin, 1915. (Nissl himself pointed out, later on, how ill chosen his terms “chronic,” 
“acute” etc., were, and proposed the use of a simple letter for each type of cell change.) 86 
REMARKS ON GENERAL PATHOLOGY 
formly granular, so that the entire cell appears as if it were 
finely dusted over. As the result of this, the clear defini- 
tion of the equivalent picture is lost. The nucleus becomes 
ballooned, globular, and shows on a clear background 
indications of a framework which under normal conditions 
does not stain by Nissl’s method. The nucleolus usually 
retains its size, but in it there often appear one or several 
clear spaces (“vacuoles”) which in extreme cases give 
the nucleus the likeness to a delicately reticulated sphere. 
The entire nucleus shows a tendency to move to the 
periphery of the cell body. The unstainable portions 
take on a light tone; for which reason dendrites and 
axones become visible for long distances. Finally there 
is characteristically a peculiar metachromasia with a 
violet tinge. The changes in this stage are reparable or 
else they may lead to disintegration of the cell. In this 
latter condition the pale cellular substance crumbles to 
pieces and the nucleus breaks up, pale cell shadows often 
remaining for a long time in the tissues as remnants of the 
previous cells. This acute cell degeneration regularly 
follows any interruption of the connection between nerve 
cell and muscle cell. Nissl first demonstrated this in 1890, 
describing its occurrence in the nucleus of the facial nerve 
of the rabbit after tearing out the nerve. According to 
him, the first alterations in the motor cells of the facial 
nucleus can be recognized after twenty-four hours; they 
reach their height after 18 to 30 days, then remain station- 
ary for a while. Only a small part of the cells disintegrate; 
much the larger part recover, so that after 50 to 60 days 
the cells of the diseased facial nucleus can hardly be dif- 
ferentiated from those of the normal side. The same 
acute cell degeneration is very frequently spread over the 
entire central nervous system after death from acute SEVERE CELL DEGENERATION 
87 
severe systemic diseases (toxic and infectious processes, 
violent excitement, etc.). 
More rare, blit likewise important, is Nissl’s severe 
cell degeneration. According to Nissl, the changes in the 
nucleus are characteristic for this. It becomes small, 
globular, and assumes a peculiar uniform metachromatic 
stain. The nuclear membrane becomes very prominent, 
and the nucleolus lies eccentrically. In typical cases the 
nuclear wall becomes thin in one or several places, and 
pouches out like a diverticulum. In addition there are 
frequently, but not invariably, characteristic changes in 
the cytoplasm; the protoplasm becomes transformed into 
small granular, isolated products of disintegration, each 
of which appears as a deeply staining ringlet with pale 
centre.8 Cells which present the characteristics of true 
“severe” degeneration cannot recover; they disintegrate. 
As further well-characterized types of cellular degen- 
eration Nissl describes cell wasting (Zellschwund; Fig. 
12), granular disintegration of cells, cell atrophy, foamy 
degeneration, two types of pigmentary degeneration, and 
various types of necrobiosis. There is a form of necro- 
biosis frequently found in small circumscribed foci, which 
is characterized by diminution in size of the cell, almost 
entire disappearance of the Nissl bodies, metachromasia 
of the cytoplasm, and swelling of the irregularly shaped, 
deeply staining nucleus. Sometimes, however, the cell 
changes are like those usually seen in “chronic degener- 
ation”; particularly characteristic is the occurrence of 
8Nissl points out that the presence of ringlets, without characteristic nuclear 
changes, may also occasionally be observed in the uppermost cortical layers, and may 
then probably be regarded as an artefact. He has not infrequently seen very similar 
pictures in small children. According to my experience, not only the nerve cells of 
small children but also those of animals (monkeys, dogs) show a great tendency to dis- 
integration of the cytoplasm of ganglion cells into small ring-like bodies. 88 
REMARKS ON GENERAL PATHOLOGY 
very dark staining grannies and irregularly shaped bodies 
on the surface of the cell and its dendrites (Nissl’s “incrus- 
tation” of the Golgi nets) (Figs. 0 and ll).9 Spiehneyer 
has lately termed this “chronic cellular degeneration.” 
These various forms of diseased nerve cells described 
by Nissl are types; they are, according to Nissl’s own 
statement, but rarely met with in man as pure forms. 
Indeed, the variety of the pictures is very great and dif- 
ficult to arrange in groups. Combinations of several 
types may often be found, as for instance when chroni- 
cally changed, shrunken cells possess one or other feature 
of the acute cellular degeneration, or when mixed forms 
between the chronic and severe forms of Nissl are met with. 
For deciding whether an altered cell would have been 
capable of restitution under favorable conditions, the 
nuclear changes furnish the chief criterion. Changes of 
the cytoplasm alone, even if at first glance they appear 
ever so severe, do not as a rule exclude the possibility of 
recovery. In necrobiotic processes, the nucleus is always 
severely damaged. 
The number of artefacts which may be mistaken for 
pathological changes is not very great, if Nissl’s direc- 
tions are strictly followed. One frequently finds, as has 
already been mentioned, that because of shrinkage the 
external portions of the cytoplasm become deformed and 
contracted, so that spaces are formed about the cell, which 
have been wrongly looked upon as pericellular lymph 
spaces. Another change in the ganglion cells described 
by Nissl is an artificial swelling, characterized by the 
nucleus and cytoplasm forming a uniform, lumpy, washed- 
9 Schroder, Lues cerebrospinalis, Deutsche Zeitschrift f. Nervenheilk, LIV, 1915, 
p. 131. METHODS FOR STAINING OF FIBRILS 
89 
out mass. It affects the ganglion cells in small groups, 
and as a rule causes similar changes in the intervening 
glia cells. Among the more frequent artefacts are to be 
mentioned also the fragmentation of the dendrites near 
their base, as well as the extrusion of nucleolus from nu- 
cleus and cell body, brought about by the microtome knife 
in sectioning. 
The methods for the staining of fibrils are likewise very 
useful for pathological purposes. Best suited is Biel- 
schowsky’s method; however, “at present, no fibril 
method furnishes such a constant and clear equivalent 
picture as is produced by the stainable cell substance with 
Nissl’s technic” (Bielschowsky). 10 Nor has any work 
with the fibril stain described the condition of the nerve 
cells as thoroughly and systematically under all the var- 
ious conditions as has Nissl by the use of his cell method. 
There are isolated papers by different authors based upon 
various methods. These investigations have not always 
remained free from the fundamentally false assumption 
and hope of connecting definite cell changes with definite 
functional disturbances, manifestations of disease with 
disease processes. 
The most thoroughly investigated fibrillar changes are 
those occurring in Nissl’s acide cellular degeneration, espe- 
cially following axis-cylinder section. According to Biel- 
schowsky the fibrillar framework becomes first hazy and 
indistinct; later the fibrils in the cell body form stretched- 
out bundles traceable for a long distance. With unfa- 
vorable termination of the cellular disease, signs of severe 
degeneration appear early in the fibrils. It is interesting 
to note that dendrites and axones always retain their 
10 Bielschowsky in the Handbuch der Neurologie, edited by Lewandowsky, Vol. 
I, 1910, p. 43 ff. 90 
REMARKS ON GENERAL PATHOLOGY 
normal fibrillar structure much longer than does the cell 
body (Figs. 15 and 1G). 
As further abnormal types, Bielschowsky mentions a 
spongy degeneration of fibrils, fragmentation or granular dis- 
integration, and liyalinization. The fibril changes occurring 
in senile dementia, in Alzheimer’s disease, and in amaurotic 
idiocy have been regarded as especially significant. 
An exact knowledge of the nerve cell changes, following 
Nissl especially, is indispensable for the study of patho- 
logical preparations of the nervous system. They often 
give us useful hints as to the nature of the disease, and 
as to the conditions under which death has taken place. 
We may infer from them whether the changes which 
preceded death were chronic or acute, and whether the 
degeneration would have been reparable or whether per- 
manent cellular disintegration had already set in or was 
about to do so. 
If we limit ourselves to such deductions as these we 
will recognize and evaluate the changes in the ganglion 
cells as well as is possible in the present state of our knowl- 
edge; they will moreover give us valuable guidance for the 
critical examination of pathological material; and we will 
be cherishing no hopes that are doomed to disappointment. CHAPTER VI. 
PATHOLOGICAL CHANGES IN THE NEUROGLIA, THE CON- 
NECTIVE TISSUE, AND THE VESSELS. 
The nerve cells are highly differentiated elements; 
this fact may account for their inability to proliferate. 
Multiplication of the mature ganglion cells through 
division has probably never been definitely observed in 
man.1 They do not seem to possess even the ability 
to grow through intracellular increase of protoplasm.2 
Accordingly, their reparative ability after previous injury 
is slight; replacement of disintegrated cells is probably 
not at all possible. The same thing holds true for the 
nerve fibres of the central organs; restitutio ad integrum is 
only observed when injury has been slight; and only in a 
very meagre way do they possess the ability to proliferate 
and regenerate. On the other hand, the neuroglia is able 
to proliferate actively and to multiply throughout life. 
It is always in a state of unstable equilibrium, and it is 
therefore possible to observe in it not only regressive 
changes, as in ganglion cells and central nerve fibres, but 
a variety of progressive alterations as well. It is this 
tendency to metamorphosis that makes the glia so im- 
portant in histopathology. 
Regressive changes in the glia are found either in the 
form of acute necrosis, in the immediate neighborhood of 
1 v. Orzechcnvski (Arbeiten a. d. Obersteinerschen Institut., Vol. XIII, 1906) has 
described nuclear changes in the motor anterior horn cells after amputation of an arm, 
and is inclined to regard them as attempts at nuclear division of the severely diseased 
ganglion cells; see here literature on the subject. 
2 The swelling of the cell body in the acute cellular degeneration of Nissl is prob- 
ably not due to a cytoplasmic increase but to a fluid intake. 
91 92 
PATHOLOGICAL CHANGES IN NEUROGLIA 
foci of disintegration; or else they may develop slowly and 
gradually as the result of chronic injuries, as, for instance, 
through malnutrition in arteriosclerosis. We regularly see 
such regressive changes in the glial tissue when, after 
previous proliferation, the process has become stationary. 
In general it is true that the glia is more resistant to 
injuries of all types than are nerve cells and nerve fibres. 
Under various conditions (Alzheimer’s incomplete soften- 
ing in arteriosclerosis, anemic foci, diffuse sclerosis of the 
brain, etc.) it may be observed that in circumscribed foci 
or diffusely the specific functional nervous tissue disinte- 
grates, while the glia in the same foci is not vitally dam- 
aged, but is rather stimulated to proliferate. If in the 
centre of such foci the glial tissue is also destroyed, then 
the glia in the immediate neighborhood will proliferate. 
When its ability to proliferate is not sufficient for the 
replacement of the destroyed substance, or when such 
replacement is hindered for other reasons, the glia forms 
at least a more or less thick limiting membrane (capsule). 
An important principle in histopathology is the ten- 
dency of the glia to proliferate wherever constituents of 
the specific nervous tissue disintegrate, even if other 
stimuli which produce proliferation are lacking. The 
classical example for this is the secondary degeneration of 
white tracts far distant from the primary focus of disinte- 
gration. This is a law which Weigert has particularly 
emphasized, and which, with a few exceptions, is gener- 
ally applicable.5 The condition of the glia is therefore a 
good index of disintegration of specific nervous substance. 
3 As such exceptions, Nissl cites the absence of glial scar formation following focal 
destructions in foetuses and very young children; a further exception may be demon- 
strated not uncommonly in the regions of certain focal cortical changes following vas- 
cular disease, particularly arteriosclerosis and luetic endarteritis. PROLIFERATION OF GLIA 
93 
Not rarely progressive changes in the glia may be 
easily recognized before it is possible with the methods at 
our disposal to demonstrate disease of the specific nervous 
constituents. Proliferation of the glia is often our only 
available guide to the presence of degenerative processes 
in the nervous apparatus. In this proliferation, Weigert’s 
fibrils play a minor part, an increase of these appearing 
only late in the process and not being a constant feature. 
It is rather the protoplasmic glia that is chiefly concerned; 
a marked increase of this, along with swelling and multi- 
plication of the nuclei, is the earliest and most regular 
change that is seen. If now, material showing these pro- 
gressive alterations be studied with Nissl’s stain (which 
only imperfectly brings out the glia), it is possible to 
demonstrate disorganization of even a single large nerve 
cell, or indeed of some of its processes.4 
The disintegration of nerve cells and nerve fibres is 
but one of the most frequent causes for the appearance of 
progressive changes in the glia (secondary glial prolifera- 
tion). Besides this it doubtless often happens, as has 
already been mentioned, that one and the same noxious 
factor destroys or severely damages the specific nervous 
tissue and at the same time stimulates the glia to primary 
proliferation. It is obvious that it may often be impossible 
in an individual case to tell which of these two processes 
is operative. A primary proliferation of the glia which 
of itself may secondarily bring about disintegration 
of nerve cells and nerve fibres is apparently rare; here 
belong gliomas. Many other disease processes in the 
brain and cord have been interpreted as resulting from 
conditions primarily affecting the glia, but definite proof 
of this is lacking. 
4Spielmeyer, Munch, med. Wochenschr., 1919, p. 709. 94 
PATHOLOGICAL CHANGES IN NEUROGLIA 
All the constituent elements of the glial tissue take 
part in its progressive alterations, the nuclei as well as the 
cytoplasm and the specific Weigert’s fibrils. 
An extensive literature, dealing with changes in the 
glia fibrils under pathological conditions, has accumu- 
lated since Weigert showed how these fibrils could be 
specifically stained. It has been shown that in the vast 
majority of morbid processes in the central nervous system 
Weigert’s fibrils are increased, and that, generally, their 
increase and proliferation of the protoplasmic glia very 
frequently go hand in hand. Under the influence of 
Weigert’s investigations the protoplasmic portion of the 
glia has been neglected, and principal attention in patho- 
logic-anatomical research has been more or less exclusively 
focussed upon the fibrils, particularly upon their numbers. 
Such an attitude is unjustified. Storch5 has called 
attention to an important point in relation to the multi- 
plication of fibrils, in pointing out the fact that in chronic 
disease-processes in which the structural plan of the 
nervous system remains unchanged (e. g. in secondary 
degeneration, multiple sclerosis, etc.) the newly formed 
glial fibrils, however numerous, show the same arrange- 
ment as those normally present in smaller quantities, 
while in stormy processes, as in rapid destruction, such 
regularity does not obtain. Storch therefore distinguished 
between “an isomorphic” and “a reparative” sclerosis. 
The new formation of fibrils in acute processes occurs 
chiefly, but by no means exclusively, in the region of the 
protoplasmic accumulations about the nuclei of the glial 
reticulum (glia cell bodies). In such cases the fibres tend 
6Storch: Uher die pathologisch-anatomischen Vorgange am Stiitzgerust des 
Zentralnervensystems. Virchow’s Archiv.f. pathol. Anat., Vol. CLVII, 1899. FORMS OF PROLIFERATIVE GLIA CELLS 
95 
to sweep close by the nuclei in arcs with the concavity 
directed outwards. In this way are formed the often 
described astrocyte pictures. Such “spider cells” are most 
frequent in the neighborhood of fresh foci of disintegra- 
tion. Especially in those processes which Storch termed 
reparative, one often finds multitudes of the so-called 
monster cells, with large nucleus and abundant cytoplasm, 
along the concave borders of which sweep layers or bundles 
of glial fibrils (Fig. 20). In later stages, the glial cell bodies 
undergo retrogression, and the relation of the newly formed 
specific fibres to individual nuclei is then less easily recog- 
nizable. In slow proliferation such characteristic relations 
are from the beginning less marked. 
The fibrils, however, are only one of the components 
of the neuroglia, differing in chemical composition from 
the cell protoplasm. The study of their changes is im- 
portant, but alone does not suffice for the evaluation of 
progressive changes in pathological material. The proto- 
plasm and the nuclei of the glia may have proliferated, 
without the glia fibres being definitely or at all increased. 
This applies particularly to very fresh acute processes.6 
The formation of these chemically differentiated fibrils as 
a rule occurs only late, and the proliferative processes are 
at first largely limited to an increase in the protoplasmic 
substance and to nuclear changes (Figs. 23, and 25 to 29). 
The normally scanty protoplasm about the nuclei often 
increases in such cases to considerable size, staining more 
definitely with Nissl ’s method, and showing in its interior 
fine, more deeply staining granules. The external shape 
of the cells sometimes becomes more distinctly star-like or 
6 There are also gliomata which nowhere and at no stage of development form 
typical fibrillar intercellular substance (Storch, l. c. p. 213). 96 
PATHOLOGICAL CHANGES IN NEUROGLIA 
spider-like, sometimes more ballooned and rounded. At 
the same time the delicate protoplasmic glial reticulum, 
which envelopes all nervous elements and connects the 
glia cells one with the other, swells up; its filaments and 
trabeculae become thicker, more prominent, and easily 
stainable, its meshes closer (Figs. 27 and 28). In addition, 
the nuclei regularly change their appearance; they become 
larger, often attain the diameter of the nuclei of large 
ganglion cells; their membrane becomes more prominent, 
their contents clearer, the chro- 
matin clumps together into a few 
or often into a single body, which, 
however, differs from the nucle- 
olus of the ganglion cells by its 
irregular shape (Fig. 23). Be- 
sides this growth of the proto- 
plasm and the changes in shape 
of the nuclei, there is sometimes 
a multiplication of the nuclei by 
karyokinesis, and in acute disease 
processes one may observe all stages of indirect nuclear 
division of the glia. But more often in spite of marked 
and obvious increase in the number of the glial nuclei 
karyokinetic figures are seldom found, nor are there in- 
dications of direct cell division. The mode of such an 
increase is still obscure.7 One must of course be able to 
exclude the possibility that all this is merely a conden- 
sation of glia elements resulting from disintegration of 
the functional nervous tissue. 
Fig. 28.—Proliferated glia in the neigh- 
borhood of a capillary. (From a case 
of typhus fever.) Closely placed nuclei 
in a broad irregular cytoplasm which is 
becoming partly foamy-reticular for in- 
stance at (a) (beginning granule cell 
formation). 
7 P. Grawitz assumes, for other organs, in such cases a development of new cells 
from old nuclei which had remained dormant in the tissues, and a reversion of already 
differentiated cell products (elastic fibres) to chromatin and cells (see Archiv. f. klin. 
Chir., Vol. CXI., 1919). FUNCTIONS OF GLIA 
97 
The proliferated glia arising in all these ways fills the 
spaces resulting from the disintegration of the specific 
nervous elements, these latter having no ability to re- 
generate. The protoplasmic masses which, in acute 
processes, at first, abundantly proliferate, gradually 
regress after they have produced fibrils. In the arrange- 
ment of these fibrils static influences play an important 
role. Where, as the result of the morbid processes, new 
surfaces (cysts) have been formed, there arise, as under 
normal conditions, limiting membranes, which are fre- 
quently strengthened by great 
quantities of fibrils. The glial 
scar tissue which definitely re- 
places disintegrated or destroyed 
parts usually takes up a space 
smaller than that corresponding 
to the original degenerated focus; 
shrinkage therefore takes place 
regularly, more pronounced after 
acute, less marked after chronic 
processes. 
This does not exhaust the list of the functions of the 
glia in the course of more wide-spread pathological proc- 
esses. The glia takes part, under conditions which we will 
learn later, in the removal of products of disintegration 
arising in the course of morbid processes. Wherever 
nervous elements disintegrate, as for instance in a region 
of secondary degeneration,8 the proliferating glia envelops 
the various broken-down particles and phagocytizes them; 
the larger masses are often rapidly removed piecemeal; 
Fig. 29.—Proliferated glia, from a case 
of acute anterior poliomyelitis. Nissl 
stain. Irregular deep-staining nuclei 
with pale reticular cytoplasm. 
8 When in destructive processes, the vascular tissues are involved (as in hemor- 
rhages, “softening” etc.) the removal of the detritus is chiefly undertaken by other 
elements. 98 
PATHOLOGICAL CHANGES IN NEUROGLIA 
the glial protoplasm in this manner becomes loaded with 
granular detritus and takes on a foamy reticular structure 
(Figs. 28 and 29). Gradually these enclosed granules 
disappear; and it may be assumed that one portion, so far 
as its chemical composition permits, is assimilated and 
resorbed on the spot by the glia. Another portion is 
carried away in the protoplasmic paths of the glial 
Granule cells 
Granule cells 
Granule cells 
Large round glia cell with alveolar cytoplasm 
Fig. 30.—Gliogenic granule cells. From a case of diffuse cerebral sclerosis. Nissl stain. Micro- 
photograph, oil immersion. 
reticulum, and in this manner reaches the perivascular 
limiting membranes and is pushed through these into the 
perivascular lymph spaces. From the latter it is carried 
into the general lymph stream of the body. Under certain 
conditions, however, the assimilating ability of the glia 
and the transportation of disintegrated products through 
the glia reticulum does not suffice. Then one sees that the 
protoplasmic substance around the nuclei becomes grad- 
ually loaded with enormous masses of lumps and granules 
which cause it to swell to a globular form possessing one or GLIAL GRANULE CELLS 
99 
several nuclei. The original connections with neighboring 
cells are torn in this process, and the protoplasmic masses 
become free from the general glial reticulum as rounded 
cellular elements (Figs. 30 and 31). Such cells show a 
reticular cytoplasm, in the meshes of which there are 
numerous fine and frequently also a few coarser granules. 
These glial granule cells or reticular cells then penetrate 
the limiting membranes (Fig. 32); or, if tissue continuity 
is already destroyed, they are carried away by the lymph 
stream. They are later to be found in the perivascular 
sheath, in which they partly rapidly disintegrate and give 
up their contents, or else 
are transported further.9 
The course of these 
processes is not the same 
under all conditions. In 
many instances one sees 
that coarse disintegrated 
products ingested by the glia remain for many months 
at one place, and only gradually diminish in size and 
finally disappear. In other cases, the proliferated proto- 
plasmic masses of the glia are filled up after a short 
time with fine granules, and numerous reticular cells may 
be found in the tissues and about the vessels. The latter 
condition seems to occur especially when the glia is stim- 
ulated, either through the disease process itself or through 
associated factors, to marked proliferation, as occurs for in- 
stance in the course of the simple secondary degenerations. 
It is only in pathological processes associated with 
marked destruction of nervous tissue that disintegration 
Fig. 31.—Large gliogenic granule cells. Nissl stain. 
Microphotograph, oil immersion 
9 For corresponding conditions in the retina see the work of Kriickmann, Ztschr. 
fur Augenheilk. Vol. XXXVII, 1917. PATHOLOGICAL CHANGES IN NEUROGLIA 
100 
is brought about by means of granules cells. In other 
processes, which are diffuse rather than focal, this does 
not occur; in the latter group belong for instance many 
subacute and acute, as well as a part of the chronic psy- 
choses. To the disintegrative products, their chemical 
nature and their tinctorial peculiarities in histological 
preparations Alzheimer has devoted extensive study.10 
For the study of the glial protoplasm, diffuse stains, 
such as van Gieson ’s, are of service only in the presence of 
Membrana limitans glia; perivascularis 
. - Membrana lini- 
m.l.gl.p itans glise 
perivascularis 
Emigrating granule cells 
Emigrating granule cells 
Fig. 32.—Gliogenic granule cells in the process of penetration through the perivascular glial 
membrane. After Held. Drawing. 
marked changes resulting from severe acute processes. In 
all cases Nissl's method here again gives good results. To 
be sure, it only gives an incomplete picture of the extent of 
the glial reticulum; it only permits the recognition of the 
cytoplasm where it is rather dense, or where it contains 
granular inclusions which stain deeply; but it has the 
advantage of absolute constancy and dependability, 
qualities which must not be underestimated for patho- 
logical histology. Of methods which demonstrate the 
glia more completely, those of Held, Alzheimer, Cajal, 
et al. have already been mentioned. 
10 Alzheimer, Histol. u. histopathol. Arbeiten, hcrausg. von Nissl. u. Alzheimer, Vol. 
Ill, 1910. TYPES OF DISEASED GLIA CELLS 
101 
The variety of types of the altered glial elements in the 
Nissl preparation is very great, as regards their size, 
shape, and appearance.11 Attempts to classify these 
diverse pictures and to evaluate their significance in 
pathological processes originate largely with Nissl (Figs. 
23 to 29). 
Nissl12 recognized a division into two great groups of 
Granule cells, of which the two 
lower ones are connected with 
the neighboring cells by fila- 
mentous bridges 
Lumen of the 
vessel 
Portion of the cytoplasm of a 
large glia cell with alveolar 
cytoplasm 
' Granule cells, of which the two lower 
ones are connected with the neigh- 
boring cells by filamentous bridges 
Fig. 33.—Granule cells about a vessel, from a monkey (Macacus) with focal lesions in the 
brain. Hematoxylin—Van Gieson. 
changes, the progressive and the regressive. The chief 
characteristics of the first are enlargement of the nuclei, 
pale staining of their contents, and appearance of a few 
dark-staining bodies resembling nucleoli; in addition, 
11 A good illustration of these manifold varieties is afforded in Table IX of Alz- 
heimer’s work on general paralysis of the insane in Histol. u. histopathol. A rim ten von 
Nissl., Vol. I, 1904; also in Tables XXVIII-XXXIV, ibid., Vol. Ill, 1910. 
12 Cf. Nissl, “Tiber einige Beziehungen usw.,” Archiv. f. Psych., Vol. XXXII, 
Heft 2, and Histol. u. histopathol. Arb., Vol. I, S. 455 ff. 102 
PATHOLOGICAL CHANGES IN NEUROGLIA 
swelling and pale staining of the cell body. In regressive 
alterations, on the other hand, the nuclei become smaller, 
stain more deeply, assume irregular outline, and the cell 
bodies shrink. Complicated pictures arise when pre- 
viously proliferated cells undergo subsequent regression. 
The cell bodies in the Nissl picture may in places be 
sharply defined and rounded off, or be supplied with 
processes. Very frequently they become gradually paler 
Granule tells 
Adventitial nucleus 
Cytoplasm of 
granule cells, 
the nucleus of 
which is not 
included in the 
section 
Reduced granule cells 
I Cross-section of capillary 
v Endothelial nucleus (tangential section) 
Cytoplasm of granule cells, the nucleus 
i of which is not included in the section 
Reduced granule cells in poor focus 
Fig. 34.—Granule cells about a capillary. Nissl stain. 
toward the periphery and gradually fade out in the sur- 
rounding tissue. The sharp concave borders of the bodies 
are sometimes formed by bright retractile lines; these are 
the fibrils, which cannot be stained with basic aniline dyes, 
but which may be brought out by other suitable methods. 
The increase in cytoplasm may be slight compared with 
the normal, but often is quite considerable. Not infre- 
quently several or many nuclei lie in great irregular 
masses of protoplasm (Nissl’s glial syncytium) (Fig. TYPES OF DISEASED GLIA CELLS 
103 
23 g). A frequent type is represented by elements with 
small or medium-sized, very clear nucleus, having a dis- 
tinct membrane and a single eccentrically placed nucle- 
olus-like body; the small foamy cell body may lie entirely 
to one side of the nucleus (Fig. 23 a and g). 
Among the very large types, Nissl describes particu- 
larly one which he designates as “fattened” (gemastet). 
This has a large dull-staining cell body, of round or 
sausage-shape, with cell outline fairly well defined. The 
nuclei are pleomorphic, and always lie quite eccentrically 
in the cell body (Fig. 23 c and d). These elements are 
identical with the large monster cells of the Weigert 
pictures, which form great masses of fibrils at their 
borders. They have a marked tendency to undergo regres- 
sive change, and later usually appear as round plaque-like 
bodies, the nuclei of which are much altered or have 
disappeared entirely. 
In the regressively altered neuroglia of the uppermost 
cortical layer and of the peripheral portions of the spinal 
cord, the spider-like branching protoplasm tends to take 
the same deep stain as the dark shrunken nuclei; whence 
arise characteristic spider-like bodies which are small, 
deeply stained, and ragged (Fig. 22 a). 
To another group belong elements showing remarkably 
elongated, regular or more angulated nuclei, and long, 
filamentous, frequently branched cell bodies (Fig. 23 e). 
There has been much discussion about the significance of 
such “rod cells,” which are regularly found, for instance, 
in progressive paralysis. Nissl, who first described them, 
at first regarded them as glial, later as mesodermal ele- 
ments belonging to the vessels. There is good reason to 104 
PATHOLOGICAL CHANGES IN NEUROGLIA 
believe that they may be of ectodermal as well as of meso- 
dermal origin.13 
Forms with a very irregular, lobed or split nucleus and 
scanty cytoplasm are met with in acute anterior polio- 
myelitis and other acute processes. In sections stained 
with hematoxylin and similar dyes it is often not easy to 
distinguish between these and leucocytes (Figs. 26 and 29). 
The delicate reticular formation within the syncytial 
protoplasm of this glia, as well as within the free round 
granule cells of glial origin, is also well demonstrated by 
Nissl’s method; one recognizes in them the delicate little 
filaments of the protoplasmic network including round 
spaces which are or have previously been filled with fatty 
waste products. The nuclei of these elements are small, 
and always bear stigmata of regressive alterations, such 
as irregular contour, indentations and shrivelling (Figs. 
30 to 34). 
The extraordinarily diverse forms of glia cells have 
various significance, and often enable us to draw inferences 
as to the general character of the morbid process, whether 
it is an acute or a chronic one, whether it is progressive or 
has been arrested. Large forms, like the mast cells of 
Nissl and the giant spider cells are always indications of 
an active, acute morbid process. Smaller cells, not 
showing such evident progressive alterations, indicate 
chronic irritation or gradual disintegration of the nervous 
tissue. Many small, deeply staining nuclei suggest an 
older pathological process, in the course of which glial 
fibril formation takes place. Regressive alterations in 
proliferated glial elements allow the conclusion that an 
13 U. Gerletti, “Zur Stabehenzellenfrage, ” Folia neurobiologica, Vol. Ill, 1910 and 
Histol. u. histopathol. Arbeiten von Nissl u. Alzheimer, Vol. Ill, 1910, S. 329; Spielmeyer, 
Zeitschrift f. d. ges. Neur. u. Psych., Vol. XLVII, 1919, S. 39. CONNECTIVE TISSUE AND BLOOD VESSELS 
105 
acute process has existed for some time. Reticular cell 
formation always points to severe acute injuries. The 
various inclusions (products of disintegration) have spe- 
cial significance. 
The importance of this subject in the present state of 
our knowledge is apparent. Observation of the ganglion 
cells themselves often gives us little information as to 
pathological conditions in the nervous tissue. The latter 
are often better indicated by the changes in the glial 
structure. These, too, better permit us to draw conclu- 
sions about the functional structures of the central organs, 
and are often more significant than the changes to be 
observed in the nerve cells and nerve fibres themselves. 
An important role in the histopathology of the cen- 
tral nervous organs is taken by the connective tissue and the 
blood vessels. That compact connective tissue layer 
completely enveloping the brain and cord, the pia mater, 
is the starting point of many disease processes which 
spread to the neighboring nervous tissue; or both tissues 
may simultaneously become diseased (acute and chronic 
forms of meningitis, tuberculosis, cerebro-spinal lues). 
The adventitia, with its lymphatics, is the seat of 
marked changes following necrosis and atrophies of the 
nervous tissue. In it, various cells (leucocytes, lympho- 
cytes, plasma cells, granule cells, etc.) (cf. Figs. 48 and 
50) which appear in the course of such processes, accumu- 
late in large quantities and are then transported further. 
Often the elements of the adventitia actively participate 
in these events; increase of its fibres and nuclei is there- 
fore of frequent occurrence.14 
14 Besides diffusely staining methods (hematoxylin-van Gieson’s, eosin etc.) and 
“special” methods (Held, etc.), the silver method of Achuccaro is particularly suited for 
the demonstration of the general group of connective tissues. 106 
PATHOLOGICAL CHANGES IN BLOOD-VESSELS 
In all the more extensive destructive processes of the 
nervous tissue the capillaries, including their adventitia, 
begin the work of removing detritus and of forming the 
first scar tissue. Large numbers of mesodermal granule 
cells and of young connective tissue cells arise by active 
proliferation (see 8th chapter). The increase in capil- 
laries met with in many processes takes place partly 
through budding from already functionating capillaries, 
and partly through tubulation of the mesenchymal 
vascular bridges everywhere present between neighbor- 
ing capillaries. 
Independent changes may occur in the vessel walls, 
and these frequently produce important lesions in the 
nervous system. Changes in the capillaries, the small 
arteries and veins, and particularly in the endothelial 
cells, which are well shown in the Nissl picture, and 
changes in the delicate elastica are frequent findings in 
various acute and chronic morbid processes in the brain 
and cord; systematic data concerning these are not yet at 
hand; individual observations are numerous. Well-known 
coarser changes in the vascular tubes are usually grouped 
together as arteriosclerosis and as endarteritis. 
Arteriosclerosis.—Atheromatosis of the vessels is 
very probably a collective term for various processes. In 
the cerebral vessels the best-known histological changes 
are those in the membrana elastica and in the endothe- 
lium.15 In the larger vessels the elastica loses its usual 
uniform wavy appearance, becomes stretched, irregular, 
split, and new lamellae become superimposed upon the old 
(Figs. 36, 38 and 39). In the later stages necrotic changes, 
15 Cf. Schroder, D., Zeitschr. f. Nervenheilk., Vol. LIV, 1915, S. 136 ff.; Cerletti, 
Rivista spiriment. di frenetria, Vol. XXXVII, 1911; Histol. u. histopathol. Arbeiten von 
Nissl u. Alzheimer, Vol. IV, 1910. ARTERIOSCLEROSIS 
107 
hyaline degeneration, hemorrhages, with masses of gran- 
ule cells and with new connective tissue formation, take 
place here. Proliferating muscle tissue from the media 
may also be met with. Calcareous infiltration is not 
frequent in cerebral vessels. 
The endothelial cells show exclusively regressive 
alterations; even with very 
marked splitting and increase 
of the elastic membrane, the 
number of nuclei remains 
scanty in most cases. There 
are, however,particular disease 
conditions of the vessel walls, 
usually grouped with arterio- 
sclerosis (and predominately 
found in more youthful indi- 
viduals and having a heredi- 
tary basis) in which from the 
beginning the number of cells 
is considerable and remains so. 
In the muscularis, regressive 
changes predominate in arte- 
riosclerosis. In the capillaries, 
sclerotic changes are recog- 
nized by the irregular shape of the normally regular 
vessels, by altered optical and tinctorial properties of the 
walls, by thickening of the normally very delicate elas- 
tica, and finally by regressive changes of the endothelial 
nuclei and cell bodies. 
In endarteritis obliterans (Heubner) of the larger 
vessels, proliferation of the endothelial cells is the predomi- 
nant feature, in contrast to arteriosclerosis. Such pro- 
Pia mater 
Gray matter 
of cortex. The 
capillaries in 
the cortex are 
all distended 
with blood; the 
erythrocytes are 
blackened by 
the staining 
method, so that 
the vessels 
are visible 
Central white 
matter with 
emerging radial 
fibre bundles 
Fig. 35.—The capillary network of the 
brain cortex in a section 20 micra thick. 
Weigert’s myelin sheath stain. Low mag- 
nification. 108 
PATHOLOGICAL CHANGE IN BLOOD-VESSELS 
liferation is easily recognized, because in the brain and 
cord the intima is always lined by a single layer of endo- 
thelial cells, even in the largest arteries such as the basilar 
and the internal carotid. Any increase is just as certainly 
pathological as an increase in the membrana elastica, 
which likewise normally consists of but one layer. In well- 
developed instances of endarteritis obliterans one sees 
the vascular lumen encircled by a uniformly thick, 
richly cellular layer, beneath which is the original mem- 
brana elastica bordering on the media (Fig. 42). Not 
infrequently, cellular bridges occur between the opposing 
Adventitia 
Muscularis 
Membrana 
elastica 
Adventitia' 
Muscularis 
Membrana elastica 
Fig. 36.—Cross-section through two basal arteries of a 3-month old child. Elastica stain, 
(Weigert’s resorcin-fuchsin). Low magnification. 
walls. The elastica, outside this cellular zone, is split 
into several rather irregularly arranged layers (Alz- 
heimer) (Figs. 40 and 41); or one finds, scattered over the 
entire layer, the delicate elastic fibres, which are usually 
cut longitudinally in cross section of the vessel, and which 
are bordered externally by the old membrana elastica; the 
latter remains unsplit or nearly so. The inner border of 
this layer is formed by a new elastic membrane toward the 
vascular lumen (Fig. 41). 
The arteriosclerotic changes differ from those of en- 
darteritis in the greater irregularity of the picture, in the 
many regressive phenomena of the proliferated intima, 
and in the scarcity of nuclei. ARTERIOSCLEROSIS 
109 
Nissl contrasted with Heubner’s endarteritis of the 
large and medium-sized vessels an analogous disease of 
capillaries. Again the most important change is here the 
proliferation of the endothelial cells (Fig. 43). However, 
the adventitial cells also take part in this, so that it is 
frequently impossible to differentiate between endothe- 
lial and adventitial elements. The vascular tube is lined 
with closely placed, large, richly chromatic nuclei, their 
cell bodies being quite prominent. The lumen is often 
difficult to recognize, and frequently it has the appearance 
of a series of lumina lying side by side.16 
Heubner’s, as well as Nissl’s, type of vascular disease 
occurs most frequently in lues, but not exclusively so. 
There is beside this a true gummatous disease of the 
vessel walls (Baumgarten’s syphilitic cerebral arteritis).17 
The diffuse endarteritis of the small vessels of the 
brain is closely simulated by the vascular proliferations 
which regularly occur in reparatory processes in the 
immediate neighborhood of degenerative foci (areas of 
softening, hemorrhage, etc.). Not infrequently regressive 
changes may at the same time be recognized in the mark- 
edly proliferated vessel elements. 
The infiltration of the adventitia or the surrounding 
pial tissues with lymphocytes and plasma cells is, in the 
literature, often termed periarteritis. 
16 Nissl, Beitrdge zur Frage nach den Beziehungen u,m\, Vol. I, S. 39, Berlin, 1913; 
Sioli, Archiv.fiir Psychiatrie, Vol. LXVI, 1922, p. 318. 
17 Schroder, /. c., P. 145. CHAPTER VII. 
HISTOPATHOLOGY OF THE NERVE FIBRES. 
For decades pathological anatomy of the nervous 
system meant little more than the study of the nerve 
fibres. However, this sort of investigation was not 
histological in our sense of the word; the absence of medul- 
lated fibres was used merely as a guide to the localization 
of morbid processes and of secondary degenerations result- 
ing from them. Weigert \s medullary sheath stain was 
employed for this purpose as a chemical staining reaction, 
to determine the presence or absence of medullary sheaths 
in certain places in the central or peripheral nervous 
system. The interest of those times lay not so much in 
histopathological questions as in localization and in the 
pathological physiology of certain parts of the nervous 
system based upon this. 
Under pathological conditions peripheral myelinated 
nerve fibres behave, in many respects, differently from the 
central ones. These differences in behavior are closely 
connected with differences in the histological structure, 
which we have already described. The sheaths of the 
peripheral fibres consist of a continuous regular band of 
cells of Schwann, the outer border of which is known as 
the sheath of Schwann (neurilemma), whereas their inner 
parts are filled with myelin, constituting what we know 
as the medullary sheath. In the central organs the nerve 
fibres do not have such a definite architecture; in particu- 
lar, they do not have a continuous neurilemma. We shall 
see that on the possession of this sheath depends the 
ability of the nerve fibres to undergo progressive changes, 
110 CHANGES IN MEDULLARY SHEATH 
111 
to proliferate, and, after destruction, to regenerate. 
Axis cylinder and medullary sheath always react to irrita- 
tion by undergoing regressive changes; on the contrary 
the same injurious agencies which severely damage these 
structures or even bring about their disintegration actu- 
ally stimulate the neurilemma nuclei and their protoplasm 
to proliferation. With this proliferation there is directly 
connected, circumstances permitting, the beginning of 
the new formation of fibres. Regenerative processes of 
the fibres are absent in diseases of the central white 
matter. The glia in the brain and cord it is true, acts to a 
certain extent as a substitute for the neurilemma of the 
peripheral nerves, but this substitution is obviously an 
incomplete one. The end result of the processes initiated 
by the glia after destruction of central nervous tissue is 
the formation of a scar, but never the regeneration of 
conducting nerve fibres. 
Of regressive changes in the medullary sheath the best 
known and most common is a disintegration into large 
rounded or elongated lumps. This breaking up is asso- 
ciated with chemical alterations. In contrast to normal 
myelin these coarse lumps are stained black, in the course 
of but a few days, with osmic acid (upon this depends 
Marchi’s method for the demonstration of recent secon- 
dary degeneration). They do not, however, give the fat 
reaction with sudan and scarlet R; their further fate is, 
as we have already seen, not always the same. This 
change is from the beginning irreparable; it always 
leads to the disintegration of the medullary sheath. 
Besides this, there are other, often less severe patho- 
logical conditions about which we possess as yet but little 
definite knowledge. In slight injuries small round globules 112 
HISTOPATHOLOGY OF THE NERVE FIBRES 
become extruded from the myelin, and these, after osmic 
acid treatment, take on a brown to black color (so-called 
Elzholz' bodies). These same bodies are normally present 
in small quantities. 
In the axis cylinder morbid changes are histologically 
recognizable through swelling or shrinkage, crumbling 
apart, granular degeneration of the fibrils, decrease or 
increase in the stainability of the substance of the axone. 
When the axis cylinder is severely damaged the medul- 
lary sheath regularly disintegrates conjointly with it. 
On the other hand there are disease conditions of the 
medullary sheath in which the axis cylinder remains 
intact, or shows only slight alterations. Severe changes 
involving the entire thickness of the nerve fibre at any 
point always result in disintegration of the entire distal 
portion of axis cylinder and medullary sheath; slight 
changes, which do not interrupt the continuity of the con- 
ducting fibres, may remain limited locally (so-called 
discontinuous disintegration). 
From what has been said, it is already evident that 
various morbid changes may take place in the nerve 
fibres. There are still many gaps in our knowledge as to 
details, but a mass of observations is even now at hand. 
Description of the accepted facts will be brought out 
best in a discussion of the histological processes found in 
secondary degeneration of the nerve fibres.1 This has been 
most thoroughly studied, because it may be produced at 
any time experimentally with ease and absolute certainty; 
and it also plays an important role in pathology because 
of its frequency. We designate as secondary degeneration 
those anatomical changes in peripheral nerves and white 
1 Called “secondary” in contradistinction to the “primary” changes at the focus 
of disease or injury. SECONDARY DEGENERATION 
113 
matter of the brain and spinal cord which regularly take 
place when the continuity of axis cylinders is interrupted 
at some point. Upon the investigations of Waller is 
based the well-known general law that divided nerve 
fibres undergo complete degeneration in their entire 
distal (centrifugal) segment, while the proximal stump 
still connected with the central organs remains intact.2 
No exceptions are yet known to the first part of Waller’s 
law; but there is a series of observations indicating that 
changes may be recognized also in the proximal segment. 
That in peripheral nerves the disintegration progresses 
centripetally to the nearest node of Ranvier is generally 
admitted; but frequently disintegration of further seg- 
ments has been reported. After subsequent septic infec- 
tion of the operative wound degeneration of the entire 
central segment has been observed. When motor nerves 
or roots are not cleanly sectioned but are forcibly torn, such 
a retrograde degeneration of the intraspinal or intra- 
medullary segments regularly takes place (Bregmann, 
Raimann and particularly Lugaro and von Gehuchten);3 
the same usually occurs if complicating systemic diseases 
are present (Raimann). A part of these changes in the 
central stump is probably to be explained as due to dam- 
age to the centres of origin of the injured nerves. A slight 
slow atrophy in the proximal portion of the resected 
nerve (thinning of part of the medullary sheath, increase 
of Elzholz’ bodies) probably always occurs. It is further 
known that in the absence of other injuries certain motor 
2 The question as to whether all axis cylinders arise each from a ganglion cell may 
here be disregarded. 
3 Lugaro in: Handbuch der pathol. Anatomie des Nervensystems von Flatau, Jacob- 
sohn, Minor, 1904, S. 180; Von Gehuchten, Le Nevraxe, Tome V, 1903. See first: Spatz, 
Zeitschrift fur die ges. Neurol u. Psych., Vol. LVIII, 1920, p. 327. 114 
HISTOPATHOLOGY OF THE NERVE FIBRES 
fibres of the vagus, after section, regularly undergo cen- 
tripetal disintegration, and von Gehuchten has shown that 
probably certain fibre bundles of the spinal cord and the 
medulla oblongata always degenerate in both directions, 
even after carefully performed interruption of continuity. 
In these, however, the disintegration of the central portion 
takes place somewhat more slowly than that of the per- 
ipheral. Among these bundles do not belong any of the 
usually investigated long fibre tracts. 
The histological processes regularly occurring in the 
course of secondary degeneration are independent of the 
nature of the injury which leads to interruption of the 
axis cylinder, unless it be that the injury at the same 
time change the degenerating fibres, and that in such 
manner, the process of secondary degeneration become 
complicated with a morbid process of different kind. The 
changes in simple secondary degeneration are essentially 
composed of two series: viz., processes of regressive, necro- 
biotic nature in the axis cylinder and medullary sheath, 
and progressive changes of the neurilemma or of the sur- 
rounding glia. 
In peripheral nerves 4 the beginning of disintegration 
may be histologically demonstrated after as short a time 
as 24 hours in the medullary sheath of the distal segment. 
There appear, as may be seen with simple staining 
methods,5 irregular fissures. The myelin tube disinte- 
grates into rounded, ovoid, or elliptical clumps, most of 
which include a piece of the axis cylinder; then a further 
4 Cf. the papers by von Biingner, Sehiitte, Monckeberg-Bethe, and especially: 
Stroebe (Ziegler’s Beitrdge, Vol. XIII, Zentralbl.f. AUg. Pathol., 1895); also Doinikow, 
Histol. u. histopathol. Arbeiten von Nissl u. Alzheimer, Vol. IV, 1911. 
5 More complicated technic permits the recognition of many other details (see: 
Doinikow, l. c.). PROGRESSIVE CHANGES 
115 
disintegration of the clump into ever smaller globules and 
droplets takes place. The fibrils of the axis cylinder are 
transformed into fine granules, which soon disap- 
pear entirely from the picture given by the ordinary 
technical methods. 
But along with this purely degenerative process pro- 
gressive changes set in relatively early. The nuclei of the 
neurilemma begin to proliferate on the second day. 
According to Stroebe, the height of their multiplication 
through mitosis is reached about the eighth day. Con- 
jointly with the proliferation of the nuclei there is a 
growth of the protoplasm; it pushes into all the gaps and 
clefts between the myelin clumps, surrounds them, at 
first as irregular, then as roundish or oval, well-defined, 
uni- or multi-nuclear cellular structures, in the interior 
of which the clumps of myelin become rapidly reduced to 
fine granules (von Biingner, Stroebe). In this manner 
there arise multitudes of granule cells, with round cell 
body and a delicate reticular protoplasm, in the meshes of 
which the granules are enclosed. Such elements are to 
be found from about the fourth week onwards in the 
lymph spaces of the neighboring vessels.6 Stroebe and 
Schutte mention that the genesis of these granule cells 
has formerly been often misunderstood. They have 
been looked upon as leucocytes or so-called wandering 
cells, and have been assumed to originate from the 
ruptured vessels at the primary focus of destruction of the 
nerve. It was thought that they move along the nerve 
sheaths and everywhere pliagocytize the disintegrated 
6 These processes in the region of secondary degenerations must be differentiated 
from the conditions which occur at the point of injury, that is in the region of so-called 
primary degeneration. Here the blood vessels and the connective tissue play an active 
part; the majority of the granule cells here are therefore derivatives of these structures. 116 
HISTOPATHOLOGY OF THE NERVE FIBRES 
myelin along the nerve fibres. Other authors have entirely 
overlooked the fact that the rounded masses of disinte- 
grated myelin are enclosed in cellular bodies and that they 
represent granular cells and not merely granular aggregates. 
The complete understanding of the formation of these 
granule cells from neurilemma elements was first attained 
in the studies of Reich, Nemiloff, et. al, upon the normal 
histology of the nerve sheath. According to these writers 
we must suppose that the entire process is in a sense 
nothing more than a disintegration of each interannular 
segment into a number of cellular bodies. These bodies 
each contain in their interior a portion of the medullated 
tube, which, normally present in every uninjured seg- 
ment, is here disintegrated into fragments and granules. 
Their nuclei are all derivatives of the one proliferated 
neurilemma nucleus. 
Of the cellular elements arising by growth and pro- 
liferation from the neurilemma, only a part is used as 
granule cells for the purpose of removal of the detritus. 
Another part is transformed into long spindle-shaped cells, 
which arrange themselves end to end into bands; they 
serve in the process of regeneration. We have already 
learned that there are two opposing views as to the way 
in which this regeneration takes place; according to one, 
the end of the central stump of the sectioned axis cylinder 
grows out and into the chain of young spindle-like cells; 
according to the other, these adjacent cells themselves 
produce new axis-cylinder fibrils, while the rest of their 
protoplasmic body with the nuclei becomes the elements of 
the new neurilemma, in the meshes of which myelin 
is then deposited. 
The histology of secondary degeneration of central DEGENERATION OF CENTRAL FIBRES 
117 
nerve fibres has been most frequently studied in the com- 
pact long tracts of the spinal cord, the posterior tracts and 
the lateral pyramidal tracts. The literature contains 
numerous reports of systematic investigations in animals 
as well as on human pathological material (Schieferdecker, 
Homen, Tooth, and others, especially Stroebe' and 
Jakob.8) The investigators agree that the end result is 
always the complete disintegration of the entire distal 
part of the interrupted fibres, that the glial framework in 
the degenerated field increases, and that a regeneration 
of central nerve fibres, disregarding abortive attempts, 
does not occur in higher animals or in man. Concerning 
the course of the histological processes in detail, and 
particularly whether and in what quantities granule cells 
take part in the removal of the detritus, the origin of 
these granule cells, the behavior of the glia, the role which 
the mesodermal connective tissue and the vessels play, on 
all these questions the various views differ considerably.9 
The first disintegrative phenomena in the axis-cylinder 
and medullary sheath (swelling, fragmentation, chemical 
transformation) are the same in central and peripheral 
nerve fibres. The earliest degeneration products may be 
demonstrated about the second day; the height of the dis- 
integration is reached from about the eleventh to the fif- 
teenth day, if one takes Marchi’s method as a guide. 
However, according to Stroebe and others, the further 
breaking up and transportation of the coarser fragments 
or lumps from the tissues takes place much more slowly 
7 Stroebe, Ziegler’s Beitrdge, Vol. XV, 1894. 
8 Jakob, Ilistol. u. histopathol. Arb. von. Nissl u. Alzheimer, Vol. V, 1912. 
9 In the central organs it is also necessary sharply to separate the “ primary 
degenerations,” produced through experimental injury or through disease, from the 
“secondary degenerations. ” 118 
HISTOPATHOLOGY OF THE NERVE FIBRES 
in the central than in the peripheral nerves. Abundant 
quantities of myelin droplets remain in place for months, 
and may here be demonstrated by Marchi’s method. 
According to Stroebe, the granule cells take a very small 
part in the transportation of the fragments, and the pro- 
gressive changes in the glia as well as in the mesodermal 
connective tissue are only slight. Other authors, however, 
report great proliferation of the framework and abundant 
granule cell formation. 
These contradictions are explained if one takes into 
consideration the different materials used. 
Stroebe’s statements apply to secondaiy degenerations 
which have been produced by careful and aseptic trans- 
section of the spinal cord in animals. In such material one 
can observe10 that the medullary sheath and axis-cyl- 
inder rapidly disintegrate into rounded fragments, which 
blacken relatively quickly with osmic acid, blit which do 
not stain red with sudan. From about the sixth day 
onward the glial nuclei become clearer and larger in the 
degenerated region; the cell processes become more prom- 
inent; there is a swelling of the protoplasm about the 
nuclei and in the processes, and here and there a delicate 
reticular structure may be recognized in it. The frag- 
ments and chimps, which at the beginning lie in irregular 
rows, become enveloped on all sides by the proliferating 
glial protoplasm. Then they very gradually crumble into 
fine granules, beginning at the periphery. As soon as these 
granules can be recognized as such, they are seen to lie in 
the glial protoplasm which has forced its way among 
them. During this time multiplication of the nuclei 
through division is fairly marked. Where such absorp- 
10 Kniek, Inaug.-Diss., Breslau, 1908, and Journ. f. Psychol, u. Neurol., Vol. 
XII, 1908. The subject is still more thoroughly discussed by Jakob (l. c.). DEGENERATION OF CENTRAL FIBRES 
119 
tion takes place the reticular structure of the glial proto- 
plasm becomes increasingly prominent and distinct. 
Particularly, close about the nuclei there is a considerable 
swelling of the cell bodies, and some commence to assume 
round forms; but everywhere their connection with the 
rest of the glial reticulum is definitely visible. Only 
rarely under these conditions do these bodies with nu- 
cleus and reticular protoplasm become entirely globular, 
their connections with the surrounding glia become 
severed, and in this manner free granule cells (reticular 
cells) arise. 
All these processes take place very slowly, requiring 
weeks and months. The larger fragments gradually 
become fewer, the reticular elements, connected nearly 
everywhere with one another and with the rest of the 
reticulum, become more prominent. Even after the lapse 
of a year, coarse clumps as well as fine droplets may be 
found. In the lymph spaces about the vessels, inde- 
pendent granule cells are just as rare during this time as 
in the tissue; but fine free fat droplets may be demon- 
strated within them in abundant quantities with osmic 
acid as well as with sudan. Hand in hand with these 
processes which serve for the removal of products of dis- 
integration, there is a slow, and never very marked in- 
crease in the fibrillar glial framework. 
The distinguishing features of such secondary degen- 
eration in central nerve fibres are: slow absorption of the 
originally coarse detritus by the proliferating glial retic- 
ulum, formation of “reticular structures” in the glial 
protoplasm, occasional liberation of such glial elements 
with reticular body which then appear as rounded granule 
cells; finally complete disappearance of the disintegrated 120 
HISTOPATHOLOGY OF THE NERVE FIBRES 
remnants and slight increase of the glial framework. Con- 
nective tissue and vessels do not take part. 
This histological picture however is altered if, in order 
to produce secondary degeneration, instead of making a 
clean section of the cord, pressure necrosis is produced, as 
for instance, through introduction of a glass pearl into the 
vertebral canal. In such cases we find in the region of the 
secondarily degenerating white tracts the same changes 
as after transsection, but the disintegration of the frag- 
ments into fine fat droplets takes place more rapidly, the 
reaction of the glia is more energetic, the formation of 
reticular cells in the protoplasm more abundant; above 
all, free granule cells are met with in larger numbers. The 
same stormy phenomena may be found if after simple 
section an infection from the operative wound takes place. 
These observations enable us to understand the vary- 
ing and partly contradictory findings in secondary degen- 
erations in the spinal cord of man. The human material 
on which investigations are based comes largely from cases 
of spinal cord tumors, of caries of the vertebrae, of frac- 
tures of the vertebrae, and similar conditions. In all these 
cases, pressure upon the spinal cord has continued for 
some time, and infectious febrile diseases have usually 
been superadded (cystitis, bed sores, etc.). Under such 
conditions one always finds in the region of the degenera- 
tions considerably more glial proliferation than after 
clean transsection in animals, and a much more rapid dis- 
integration of the myelin fragments into aggregates of 
fatty granules; large, often very large, numbers of free 
granule cells (reticular cells) are always to be found in the 
tissues as well as in the lymphatics about the vessels. 
Where the complicating factors (pressure, infection) are CHANGES IN PERIPHERAL NERVES 
121 
absent, as for instance in isolated descending secondary 
degeneration of the pyramidal tracts resulting from focal 
destruction of the cerebrum (Homen), the histologic 
picture in man corresponds to that obtained after simple 
transsection in animals. In both cases identical histological 
processes are concerned; the difference lies especially in the 
rapidity and extent of the reaction on the part of the glia, 
that is, it is purely one of degree. 
Secondary degeneration is not the only way in which 
nerve fibres may become altered. Stransky11 has recently 
recalled attention to a histologically well-defined morbid 
process of 'peripheral nerves, which was previously de- 
scribed by Gombault in 1880. This process, in contrast to 
secondary degeneration, leads merely to a local (discon- 
tinuous) and partial disappearance of the medullary 
sheath, to merely a slight injury of the axis cylinder, and 
only under certain conditions to definite disintegration of 
the involved fibres. It has been most frequently observed 
in peripheral neuritides of toxic origin (lead, alcohol, 
diphtheria, etc.). Gombault and Stransky have exper- 
imentally produced these conditions in animals by sub- 
acute and chronic lead poisoning. 
According to Stransky the first anatomical changes 
consist in the appearance within a part of an interannular 
segment, or in one entire or several adjacent segments, of 
numerous small round bodies (Elzholz bodies), which lie 
within the medullary substance and stain deeply with 
Marchi’s method. In normal fibres these bodies appear 
only in scanty numbers. At the same time the neuri- 
lemma protoplasm begins to swell, and the neurilemma 
11E. Stransky, “Uber diskontinuierliche Zerfallsprozesse an den peripheren 
Nervenfasem, ” Journ.f. Psychol, u. Neurol., Vol. I, 1903. 122 
HISTOPATHOLOGY OF THE NERVE FIBRES 
nuclei to multiply. Then the myelin tube disintegrates, 
always beginning at the periphery, in scattered places, 
forming groups and aggregates of smaller and larger 
bodies, which stain brown to black with osmic acid. 
They are usually embedded in large numbers in the 
spindle-shaped cytoplasm of one of the proliferated neuri- 
lemma nuclei. The remaining medullary sheath appears, 
because of this, smaller and irregularly eroded. In the 
further progress of the process the fibres in the affected 
region appear sprinkled with smaller or larger products of 
disintegration; the protoplasm is greatly proliferated; 
the nuclei much increased (eight to ten and more to one 
internodal segment of Ranvier). Later, spindle-shaped 
groups of disintegration products separate themselves 
more and more distinctly. They usually lie in the cyto- 
plasm about a nucleus, and are enveloped by a protoplas- 
mic mantle. Finally, they become independent elements, 
the contents of which are transported, at least in part, by 
the lymph stream. 
At first the axis cylinder remains intact or shows at the 
most a broadening out and but slight tinctorial alterations. 
Later on it may entirely lose its stainability in the diseased 
portion of the fibre, but at either end of this region appears 
again as a normal structure. If the process continue 
it may finally lead to definite destruction of the axis 
cylinder. Then, and only then, the well-known changes of 
secondary degeneration appear in the entire distal 
portion of the fibre. If the original injurious agent is 
removed early enough restitution of the medullary sheath 
soon begins. 
That which distinguishes this pathological process 
from secondary degeneration is its limitation to one or CHANGES IN PERIPHERAL NERVES 
123 
several fibre segments, its late extension to the axis 
cylinder, and finally the disintegration of the medullary 
sheath into fine granules and globules, always beginning 
at the outer margin, and not in toto or into coarse 
fragments or chimps. Formation of granule cells from 
proliferated neurilemma elements takes 
place in both. The mesodermal con- 
nective tissue does not participate in 
either process. 
Probably belonging to the same group 
of morbid changes is a series of slight 
alterations in the medullary sheath 
of peripheral nerves, which may occur in all sorts of 
systemic diseases. Among such diseases belong, besides 
toxic and infectious processes, disturbances of metabolism, 
arteriosclerosis, progressive paralysis, particularly when 
these lead to marasmus and cachexia. In such cases one 
finds in the nerve fibres an increase of the Elzholz bodies, 
and now and again the earliest characteristic appearance 
of discontinuous disintegration of the medullary sheath.12 
Related to this are perhaps also the slight alterations 
appearing in the proximal stumps of injured nerves; viz., 
moderate increase of Elzholz bodies, later a permanent 
thinning of the myelin mantle of the fibres.13 The 
majority of investigators, however, look upon this as a 
special “atrophic” process. 
In the central fibre tracts and in the white matter 
in general there are likewise changes which differ from 
those of secondary degeneration. Whether they are 
morbid processes which must be looked upon as true 
Protoplasmic 
tube of the 
capillary 
Endothelial 
nucleus viewed 
from the flat, 
with nuclear 
vacuole 
Fig. 37.—Capillary, long- 
itudinal section. Nissl 
stain. Oil immersion. 
12 Stransky, Arbeiten a. d. Neurol. Institut. von Obersteiner, 1907. 
13 Elzholz, Pilcz, Raimann, Stransky, Homen, u. a. 124 
HISTOPATHOLOGY OF THE NERVE FIBRES 
analogues of Stransky’s medullary sheath disintegration 
in peripheral nerves is still an open question. We find 
them chiefly in those not infrequent cases of toxic mul- 
tiple neuritis in which the central organs are affected.14 
At present there are no available histological observations 
Fig. 38. 
Fig. 39. 
Fig. 38 and 39.—A small portion of a large brain artery of a 72-year-old man, suffering from 
cerebral arteriosclerosis. Legends as in Fig. 36. 
on this matter. However, it lias already been noted by 
several authors that the cord changes in multiple neuritis 
are not always strictly limited to definite fibre tracts 
as one necessarily expects in secondary degenerations, 
and that the process is often local (“discontinuous”). 
There is not infrequently a better-known histological 
change met in the spinal cord in severe anemias.15 It is 
14 Heilbronner, Monatsschr. f. Psych, u. Neurol., 1898. 
15 Nonne, Deutsche Zeitschr.f. Nervenheilk., Vol. XIV; Boedeker und Juliusburger, 
Archiv. f. Psychiatrie, Vol. XXX; Schroder, “Die funikiilare Sklerose, ” Deutsche Med. 
Wochenschrift, 1923. CHANGES IN ANEMIA 
125 
characterized by the acute or subacute development of 
small confluent foci, which usually first appear, like many 
other pathological processes, in the posterior columns, 
but which, when more widely disseminated, may be found 
in other cord tracts. The gray matter is never involved. 
Histologically the foci may be recognized by regressive 
changes in the medullary sheath and by proliferation of 
the glia, of the same nature as is seen in places of secondary 
degenerations of central fibre tracts in man. There is a 
considerable swelling of the protoplasm of the glia net- 
work, assuming in the affected foci its well-known retic- 
ular structure. From the diffuse reticular protoplasm, 
granule (reticular) cells gradually become differentiated 
and reach the perivascular lymphatics. The medullary 
sheaths do not first disintegrate into coarse globules and 
fragments, but apparently always directly into fine 
granules. The axis cylinders are, just as in Stransky’s 
discontinuous disintegration of the peripheral nerves, 
attacked only late.16 Therefore by suitable staining 
methods many intact axis cylinders may always be demon- 
strated in the foci. These axis cylinders are frequently 
still surrounded by a thin myelin covering which may be 
demonstrated in medullary sheath preparations. How- 
ever, in all fairly advanced cases complete disintegration 
of at least a portion of the fibres takes place, this contin- 
uing into secondary degeneration, with superimposition 
of its well-known changes upon the original histological 
picture. The end result in each case is thickening of the 
glial tissue. The connective tissue and the vessels do not 
actively participate in the histopathological process. 
16 Contrary results are reported by Thumazuri: Archie. f. Psychiatrie, Vol. LIU, 
and Wohlvill, Deutsche Zeitschrift fur Nervenheilkunde, Vols. 68 and 69, 1921. 126 
HISTOPATHOLOGY OF THE NERVE FIBRES 
In the same class belongs a group of not very frequent 
extensive diseases of the white matter of the brain and 
cord, the so-called diffuse sclerosis of the brain (enceph- 
alitis periaxialis diffusa of Scliilder), and the cases of 
so-called acute or malignant multiple sclerosis. The histo- 
pathological processes in these are obviously identical 
with those described in anemia, different as the original 
injurious factors may be; and one may collectively speak 
of them as a myelinoclastic process.17 There are prob- 
ably quite a number of morbid conditions in which the 
medullary sheaths are involved first, while the axis 
cylinder for a time remains intact, and there is a reactive 
proliferation of the glia. In progressive paralysis as well, 
formation of flecks and streaks occurs, in which only the 
medullary sheaths disintegrate while the axis cylinders 
are not at first affected, and gliogenic granule cells make 
their appearance.18 
Concerning the histopathology of other focal dis- 
eases like multiple sclerosis and of tract diseases such 
as tabes, amyotrophic lateral sclerosis, etc., we know 
next to nothing. 
17 Schroder, “Encephalitis und Myelitis,” Monatsschr. f. Psych, u. Neurol., Yol. 
XLIII, 1918. 
18 0. Fischer, Allg. Zeitschr. f. Psych., Vol. LXVI, 1909, and Spielmeyer, Zeitschr. 
f. d. ges. Neurol, u. Psych., Vol. I, 1910. CHAPTER VIII. 
CONCERNING SOME HISTOPATHOLOGICAL COMPLEXES. THE 
GRANULE CELLS. THE CONCEPT OF INFLAMMATION. 
In the last chapter we extended our discussion of 
morbid changes in the various individual tissue elements 
of the nervous system to a consideration of certain histo- 
pathological complexes. The changes in the tissue ele- 
ments are numerous and various; each new technical 
method brings out new details. On the other hand the 
number of complexes is limited, and probably each tissue 
has but a few such histological complexes by which it may 
react to injuries. The course of each complex is as regular 
as the course of the changes in its constituent elements. 
But the picture may vary considerably at different times, 
as when in the beginning the mesodermal, later the ecto- 
dermal, tissue elements are chiefly involved. 
We have come to recognize from the above a series of 
pathological processes in the peripheral nerves, and in the 
white matter of the spinal cord and brain, which are quite 
similar in their nature and course. This similarity lies in 
the fact that the changes are limited to the ectodermal 
tissues,1 the mesodermal parts (vessels and connective 
tissue) not participating; that axis cylinders and medul- 
lary sheaths undergo retrogressive alterations (complete 
or partial disintegration) while the neurilemma elements 
or the glia show proliferation of their protoplasm and 
multiplication of their nuclei. We have seen that these 
processes occur in the same manner in secondary degen- 
1 The neurilemma cells, as has been pointed out in Chapter III belong to the 
ectodermal group. 
127 128 
HISTOPATHOLOGICAL COMPLEXES 
eration of the peripheral nerves as in the brain and cord. 
They occur in discontinuous disintegration of peripheral 
nerves as well as in the white matter of the cord in cases of 
severe anemia, and they are found likewise in extensive or 
diffuse lesions of the central organs. Differences can be 
observed only in the rapidity of the myelinic degeneration 
and the extent of the glial or neurilemma proliferation. 
In some cases the process has more of a subacute, in 
others more of an acute character. 
Changes of this type occur under the most varying 
conditions. They may appear either isolated and con- 
stitute by themselves the whole picture of certain histo- 
pathological processes; or they may form merely a part of 
more complicated morbid processes. We shall not go 
astray if we regard this complex of histological processes 
as the morphological expression of a regular and definite 
mode of reaction of the nervous system to definite injuries 
(<ectodermal type). The nature of the injurious agents 
may be very diverse. 
In contrast there is another type in which the vascular- 
connective tissue system plays the chief role (mesodermal 
type). This we meet wherever nervous tissue has disin- 
tegrated in toto, no matter what the size of the lesion, 
whether produced by trauma, by crushing, hemor- 
rhages, thrombotic or embolic softening, cauterization, 
freezing, etc.; provided that, besides the ectodermal 
tissue, at least a part of the vascular network has been 
injured or destroyed. Its prototype is the tissue change 
which develops in connection with hemorrhages in the 
central organs, in the gray as well as in the white matter.2 
2 See the publications of Nissl and of some of his pupils (Farrar, Histol. u. histo- 
pathol. Arbeiten von Nissl, Vol. II, 1908). MESODERMAL REACTIONS 
129 
In such cases we always find the histological picture in the 
beginning dominated by proliferative changes in the 
vascular and connective tissue system. The endothelium 
of the vessels in the surrounding zone becomes swollen 
and multiplies, and the affected capillaries proliferate. 
They form young vascular buds, which press in from the 
periphery toward the destroyed or necrotic area and grow 
into it. Another portion of the proliferating mesenchymal 
cells become fibrous elements (fibroblasts) which tempo- 
rarily replace the destroyed tissue. Finally, there arise 
from these proliferated vascular elements great numbers of 
granule cells, which become packed with detritus (Nissl). 
The glia within the focal lesion has disintegrated with 
the nerve cells and the nerve fibres and cannot actively 
participate in the process. However, the glial tissue 
outside the zone of capillary growth swells and prolifer- 
ates. This proliferation at first lags behind that of the 
connective tissue and vessels, but soon attains consider- 
able prominence. One finds diffuse, poorly defined, pro- 
toplasmic masses and the various type of progressively 
altered glia cells. Among them there are regularly met 
the gigantic “mast cells” of Nissl (Fig. 23 c and d), which 
appear after but a few days in a narrow zone near the 
lesion. We know of these cells that they form large 
quantities of glia fibrils. Within this zone the glia also 
participates, as described above, in the ingestion and 
transportation of particles of ectodermal tissue which here 
disintegrate secondarily to the actual lesion (through 
nutritional disturbances, secondary degeneration, etc.). 
Later the proliferated glia forms greater and greater 
masses of fibrils about the focus, and these gradually 
encapsulate the disorganized region. Later this mass of 130 
HISTOPATHOLOGICAL COMPLEXES 
glia gradually extends inward, and finally displaces the 
originally mesodermic scar. If the lesion is a small one, 
the connective tissue will finally be entirely replaced by 
glia fibrils; otherwise a loose connective tissue remains in 
the centre of the focus, or a cystic cavity is formed which 
is lined by a layer of mesodermic tissue, and which is 
externally bordered by a dense glial capsule. 
These processes are somewhat modified if the destruction 
of the tissue is less complete, and if the damage is a grad- 
ually progressing one, as, for instance, not infrequently 
happens in thrombotic softening or pressure necroses 
(compression). One may then observe that only a part 
of the glia within the lesion disintegrates, and that glia as 
well as connective tissue participates in the reparative 
process. Thus there arise mixed forms between the meso- 
dermaBand the ectodermal types. That the glia also 
participates in the proliferative processes in the border 
regions of completely destroyed areas has already been 
stated; this is an example of a mixture of the ectodermal 
type of reaction at the periphery of the lesion with the 
pure mesodermal type at the centre. 
The brain of the foetus and of the new-born behaves 
differently, as we know from Nissl. Both in secondary 
degeneration and in focal destruction, if the lesions are 
not too large the remaining tissue closes in, so that later no 
mesenchymal nor glial scar tissue will be present to indi- 
cate the site of the lesion. According to Spatz3 the more 
extensive lesions lead, in the fetal brain and cord, to 
softening, and are not organized as they are in adults. 
There is no formation of fibroblasts; the granule cells 
3 Spatz, Zeitschr. fur die gesamte Neurol, u. Psychiatrie, Vol. LIII, 1920 and 
Histolog. u. liisto-patholog. Arbeiten von Nissl u. Alzheimer, Erganzungsband, 1921. GRANULE CELLS 
131 
disappear with extraordinary rapidity. The end result 
is a cyst. 
The characteristic changes of acute inflammation (in 
the sense of Cohnheim) in the central and peripheral 
nervous system are, as in all other organs, of vascular 
nature. The nervous tissue itself, however, is always 
involved. The sequels of nutritional changes are necro- 
bioses of different types, and they in turn are followed, as 
are other degenerative and destructive lesions by the two 
types of tissue reactions discussed (but more especially 
by the mesodermal type). These are no more character- 
istic of inflammation than of hemorrhage, a puncture 
wound, a crushing injury, or a burn. This fact has been 
largely disregarded, whence have arisen many of the 
difficulties and misunderstandings concerning the histo- 
logical correlation of inflammation in the nervous System. 
A discussion of the histological phenomena of some more 
chronic types of retrogressive changes than those con- 
sidered above may be found in the papers of Alzheimer.4 
The important role which the granule cells take in 
acute and subacute morbid processes has been indicated. 
The histopathology of the nervous system has been 
much concerned with these.5 Gluge described (1837) 
“inflammatory globules” in areas of cerebral softening; 
their cellular nature was only later demonstrated. Since 
4 Alzheimer, Histol. u. histopathol. Arbeiten, herausg. von Nissl und Alzheimer, Vol. 
Ill, 1910, S. 522 ff. Concerning a special type of disintegration of cortical tissue, see 
also: Schroder, Ilirnrindenveranderungen bei arteriosklerot. Demenz. N aturforscherver- 
samml., Dresden, 1907. 
5 For the older literature consult: Baiimler, Inaug., Diss., Halle, 1881, for the more 
recent literature, Nissl, Histol. u. histopathol. Arbeiten, Vol. I. S. 328 ff., 1904; also: 
Merzbacher, ibid, Vol. Ill, 1910, and Held, Monatsschr. f. Psych, u. Neurol., Vol. 
XXVI. 1909 132 
HISTOPATHOLOGICAL COMPLEXES 
then there has been much dispute about their significance, 
and the most diverse assertions have been made in regard 
to their origin in the brain, cord, and peripheral nerves. 
There is no histological element in the nervous system 
which has not been regarded by one investigator or other 
as the source of the granule cells; ganglion cells, glia cells, 
cells of the vessels’ walls, fixed connective tissue cells, vari- 
ous white blood corpuscles, so-called wandering cells, and 
pus cells have been implicated (Nissl). 
Nissl has properly emphasized that all cellular ele- 
ments, even the highly differentiated ganglion cells, may 
under certain conditions ingest in their protoplasmic 
bodies certain foreign substances (granules). But he goes 
on to say that we do not always speak of these as granule 
cells, that we have rather been accustomed to reserve this 
term for histological elements of a very definite character. 
It is customary to designate as granule cells free, rela- 
tively large, rounded elements, with small, often eccen- 
trically placed nucleus, and a cytoplasm which shows a 
foamy or definitely alveolar reticular structure, containing 
in its meshes pliagocytized “granules” of heterogeneous 
source (Fig. 32). The granule cells are therefore always 
phagocytes; their business is the ingestion and transporta- 
tion of tissue detritus and of small foreign bodies.6 The 
ingested detritus is gradually broken up and as far as 
possible assimilated and liquefied. We find granule cells 
wherever a large amount of tissue undergoes disinte- 
gration or necrosis and where the remaining tissue re- 
tains sufficient vitality to remove the necrotic substances. 
The inclusions of the granule cells may be of greatly 
varying nature, and accordingly their appearance varies to 
6 Therefore also called “ Abraumzellen” (Merzbacher). GRANULE CELLS 
133 
a certain degree, according to the size and physical and 
chemical properties of these inclusions. In hemor- 
rhagic areas they appear at first choked with erythrocytes, 
later with pigments. Where nerve fibres have undergone 
disintegration these cells contain fragments of myelin 
4nd axis cylinders, and so on. Their characteristic cell 
structure is well demonstrated by Nissl’s method and by 
diffuse stains (hematoxylin-van Gieson). 
This long-disputed question concerning the origin of 
the granule cells may be settled by regarding these struc- 
tures as arising in the peripheral and central nervous 
system from at least three different sources: the glia, the 
elements of the vessel walls (involving adventitia) and the 
neurilemma. The oldest conception of granule cells 
attributed their origin to white blood corpuscles which 
had phagocytized tissue detritus. Closely associated with 
this explanation is the widely spread idea that the pre- 
sence of granule cells is as indicative of the inflammatory 
nature of a morbid process as the presence of extrav- 
asated leucocytes. We have not discussed the possibility 
that granule cells may be derived from white blood cor- 
puscles, for it must be regarded as very improbable that 
any considerable numbers of leucocytes are ever so trans- 
formed in the nervous system. This statement applies 
even to reparatory processes after hemorrhage, in which, 
in the first few days, large numbers of leucocytes are 
regularly met with; and it also applies, as we shall see, 
to suppuration. 
Nissl showed from extensive research that one source 
of granule cells lies in the connective tissue elements of the 
proliferating vascular system (mesodermal granule cells) .7 
7 Compare the illustrations in the article of Farrar in Nissl’s Histol. u. histopathol. 
Arheiten, Vol. II, Taf. I u. II. 134 
HISTOPATHOLOGICAL COMPLEXES 
This is true, however, of only a certain group of morbid 
processes, namely those which we have briefly designated 
as reparatory processes of the mesodermal type. Under 
other conditions the development of granule cells may be 
traced step by step from the glia, with the vessels and 
connective tissue taking no part whatever (gliogenic 
granule cells). Putting the two sets of observations 
together, we learn that the typical free granule cells in 
either case represent the end stage of a process of develop- 
ment that begins in fixed tissue elements, and that all 
transitional stages between these fixed elements and the 
granule cells may be found. The histological end product 
is the same in either case, whether they arise from vessel 
walls or from glia. No morphological distinction is yet 
known; and in fact Nissl recommended as the most 
suitable and appropriate name for these mesodermal 
bodies the term “reticular cells” (Gitterzelle), which 
Boedeker and Juliusburger proposed for elements that 
according to their description doubtless arise from the glia. 
As the third source of granule cells, we mentioned, in 
the seventh chapter, the neurilemma. 
We see, therefore, that granule cells in the nervous 
system may develop from different cellular elements, and 
particularly from those that possess marked proliferative 
ability. It has already been mentioned that we must regard 
both the circulation of tissue juices as well as the trans- 
portation of metabolic products from the tissues, as taking 
place within the protoplasmic paths of the glial syncy- 
tium. Under normal conditions local blockage and accu- 
mulations of such products in the shape of granules and 
droplets are not infrequent within the bodies of the glia 
cells; and even well-formed granule cells may be found. GRANULE CELLS 
135 
In morbid processes granule cells arise when through acute 
changes such considerable masses of disintegrated products 
are formed that the normal ways and means for absorption 
and transportation no longer suffice.8 They are altered 
tissue cells which have separated from their original histo- 
logical connections, and are therefore always doomed to be 
destroyed. Wherever we find them fully developed, 
regressive changes in them are recognizable (dark, pyk- 
notic, indented nuclei). When, after swelling and round- 
ing of their cell body and separation of their processes, 
they have become free and mobile, they wander, laden 
with granules, into the lymph spaces. Held has directly 
observed the penetration of granule cells close to the 
capillaries through the glial perivascular limiting mem- 
brane into the adventitial spaces (Fig. 32). Their amoe- 
boid or more passive locomotion in the tissue juices is in 
many instances facilitated by the broken continuity of 
tissue caused by destructive processes at their point of 
origin. The further fate of the granule cells is probably 
that, either at first in the tissue or later in the lymph 
stream, they burst; or that their liquefied contents are 
lost through diffusion. Their remnants may often be 
mistaken for blood lymphocytes (Held, “reduced” gran- 
ule cells). Such nuclei are not infrequently found in 
considerable quantities forming mantels about the cap- 
illaries (Fig. 34). 
Other less frequent elements must not be mistaken for 
the granule cells, in spite of their similarity. These are 
certain degeneration forms of plasma cells (known as 
Russel cells or Y-cells of Perusini) and therefore deriva- 
8 On the other hand in the embryonal brain and spinal cord fatty granule-cells play 
a considerable role in the myelinization of nerve fibres (“ Aufbauzellen, ” Merzbacher, 
c.). 136 
HISTOPATHOLOGICAL COMPLEXES 
tives of lymphocytes. We deal here not with phagocytic 
cells, which have taken up and elaborated tissue detritus, 
but with degeneration forms of plasma cells, showing a 
mullberry-shaped, or more uniformly rounded body, in 
which are included certain bodies formed intracellularly.9 
They are illustrated in Figs. 49 and 50. 
In the pathological anatomy of the nervous system 
formerly much emphasis was placed on the determination 
Fig. 40. 
Fig. 41. 
Membrana, 
elastica 
Adventitia- 
Muscularis 
Newly 
formed 
elastica 
Intima 
Adventitia 
Intima 
Membrana 
elastica 
Newly- 
formed 
elastiea 
Figs. 40 and 41.—Arteries of the base of brain from a case of brain syphilis. Endarteritis 
obliterans. Elastica stain (Weigert’s resorcin-fuchsin). Proliferation and splitting of the 
elastic membrane. 
of hyperemia and anemia in the tissue. The histological 
estimation of the amount of blood in the vessels is dif- 
ficult on account of the alterations in the blood distri- 
bution occasioned by death and by physical factors 
after death. 
The histological study of edema is even more difficult. 
The fluid contents of the tissues are so changed through 
our fixing and embedding methods that all conclusions 
based upon the examination of microscopic preparations 
9 Spielmeyei, Die Trypanosomenkrankheiten, Jena, Fischer, 1908; Perusini, Rivista 
speriment. di Frenetria, Yol. XXXVI, 1910. Schroder, Zeitschrift fiir die gesamte 
Psychiatrie u. Neurol., Vol. LXIII, 1920, p. 143. E. von Muller, Frankfurter 
Zeitschr. f. Pathologie, Vol. XXIII. INFLAMMATION 
137 
must be accepted with reservation. Reichardt 10 has 
repeatedly emphasized the point that under pathological 
conditions the brain is especially subject to great fluctua- 
tion in fluid content, and that the various resulting con- 
ditions should not be neglected in the examination of 
diseased brains. However, these conditions of swelling 
or shrinking of the tissue can only be identified by histo- 
logical examination when it is possible to recognize 
associated histopathological phenomena which are con- 
stantly associated with them. 
Under the macroscopical concept of atrophy a number 
of diverse conditions are included. All severe morbid 
processes lead, at least in the central nervous system, to a 
loss of specific nervous substance. Atrophy of some parts 
or regions is therefore the end result of all severe morbid 
processes, no matter what their nature may be. In con- 
tradistinction to such secondary atrophies, it is custo- 
mary to look upon as primary those processes which are 
assumed to have been insidious from the beginning, and 
which lead to a loss of nervous substance without stormy 
phenomena. But concerning their histopathology we still 
know very little that is certain. 
The concept of inflammation in the nervous system, 
finally, necessitates a somewhat more detailed discus- 
sion. 11 A glance through the literature and through the 
clinical text-books shows that in neurology and brain 
pathology the terms neuritis, myelitis and encephalitis 
are used very glibly and extensively, and that there are 
10 Uber die Untersuchung des gesunden und kranken Gehirns mittels der Wage, Jena, 
1906; Hirnschwellung, Zeitschr.f. Psych., Yol. LXXV, 1918. 
11 Nissl, Archiv. f Psych., Yol. XXXIII, 1900, S. 685; Schroder, Enzephalitis und 
Myelitis, Monatsschr. f. Psych, u. Neurol., Yol. XLIII, 1918; “Paralyse und Entziin- 
dung,” Zeitschr.f. d. ges. Neurol, u. Psych., Vol. LII, 1920. “tlber Entzilndung, ins- 
besondere im Nervensystem, ” Ziegler’s Beitrdge, Vol. LXXI, 1922, p. 1. 138 
HISTOrATHOLOGICAL COMPLEXES 
few of the diseases of the nervous system that are not 
generally, or at least by some of the authors, called by one 
or the other of these names. In the course of time ter- 
minology has become very loose in this respect. Anatom- 
ical descriptions accord more or less closely with these 
clinical concepts. Only in recent times have many dis- 
senting voices been raised against such usage. We find 
that a large number of investigators are content to make 
Pia mater 
Intima 
Membrana elastica 
Muscularis 
Adventitia 
Fig. 42.—An anterior spinal artery from a case of spinal syphilis. Hematoxylin-van Gieson. 
Uniform proliferation of the intimal cells. 
an anatomical diagnosis of neuritis on the strength of 
demonstrating medullary sheath disintegration with 
Marchi’s or Weigert’s method. Since this disintegration 
is also a regular finding in simple secondary degeneration, 
these observers assume that it is impossible sharply to 
differentiate between the two morbid processes, and such 
terms as traumatic “neuritis” and degenerative “neur- 
itis” have been introduced. The word myelitis is used to 
designate any, but especially an acutely developed focal 
disease of the spinal cord; and the term encephalitis is used 
equally loosely. This custom is only partly founded on INFLAMMATION 
139 
surviving though discredited pathologic-anatomical con- 
ceptions. To the old authors softening in the brain and 
cord implied inflammation; hardening (sclerosis) was 
regarded as the end result of inflammatory processes. 
And even to-day we do not invariably find an attempt to 
differentiate acute inflammatory processes from simple 
hemorrhages, softening, pressure necrosis, trauma, etc. 
After we became acquainted with “inflammation without 
softening,” the question was raised whether the greater 
part of all known morbid processes really belongs under 
inflammation. An attempt was even made to differen- 
tiate histologically between parenchymatous and inter- 
stitial inflammation of the nervous system. It has already 
been mentioned that the alleged derivation of granule 
cells from leucocytes seemed to furnish an additional argu- 
ment for regarding as inflammatory all processes in which 
granule cells occur. Serious attempts were not lacking to 
establish scientifically this conception of inflammation. 
In order to group all these histological pictures together, 
all tissue changes which result from “inadequate” stimu- 
lation were designated inflammatory.12 A long series was 
then constructed beginning with the slightest changes and 
ending with suppurative disintegration of the tissue. It 
was taught that all these changes are stages of one and the 
same process, namely inflammation, which may terminate 
sometimes at one stage, sometimes at another. 
The difficulty arises that, if we follow this conception, 
almost all morbid processes attended by progressive 
alterations of any of the tissue elements may be included 
under inflammation. Trauma would stand as the type of 
12 Friedmann: “Inflammation is the general reaction to any inadequate stimulus”; 
Virchow: “Inflammation is only different quantitatively from simple irritation.” 140 
HISTOPATHOLOGICAL COMPLEXES 
non-suppurative inflammation, a case of crushing of the 
cord would be regarded as “by its very nature” inflam- 
matory and the phenomena of secondary degeneration 
would be considered as characteristic of neuritis. But it is 
apparent from this that it would be obviously impossible 
to make a sharp distinction histologically between inflam- 
matory processes and diffuse glial proliferations or cerebral 
softening. Nor would it be possible in many cases to make 
the anatomical diagnosis of inflammation solely from the 
Endothelial cells 
Capillary tube 
Endothelial cells 
Adventitial cells 
Fig. 43.—Proliferating capillary. Nissl stain. Microphotograph, oil immersion. 
histological picture, without at the same time considering 
the etiology and the clinical course. Such conclusions are 
inevitable if one holds to the above definition; and they 
demonstrate that this conception of inflammation has no 
value for histopathology. 
The same principles apply to the elements of the 
central organs as to all other organs; we are acquainted 
with many different regressive and progressive changes 
in them, but none of them can claim to be characteristic 
or decisive for the inflammatory nature of the disease, 
unless one sweepingly regards all progressive changes 
as inflammatory. 
We may therefore regard it as an advance, that from 
different sides a tendency has developed to drop this con- 
ception of inflammation, and to recognize a series of INFLAMMATION 
141 
simple histological processes. The present tendency is to 
reserve this term for a more restricted field, and to leave it 
to the future to determine whether it be necessary or 
desirable to regroup these processes, in whole or in part, to 
form a new complex.13 
Nissl made the first step in this direction for the central 
nervous system, by sharply defining a group of tissue 
changes particularly studied by him, which we designated 
above as the mesodermal type. He showed “that wher- 
ever brain substance is destroyed, no matter in what 
way, or wherever necrotic brain substance or a foreign 
substance is surrounded by living tissue, the latter reacts 
in a very definite manner;” that this reaction is essentially 
the same in polioencephalitic hemorrhages, in the neigh- 
borhood of a sarcoma, after penetrating wounds made by 
a hot needle, after the introduction of a bit of cotton 
saturated with a streptococcic culture, or after corrosion 
of the brain substance with chromic acid. Definite and 
characteristic processes, already described, are always set 
in motion in the surrounding intact tissues. 
These changes of mesodermal type are, however, 
merely secondary accompanying phenomena; they are 
the anatomical expression of reactive processes which 
have nothing to do with the injurious agent or the disease 
which led to the disintegration or the destruction of the 
nervous tissue. If we keep this in mind, the presence of 
this type of reaction will not tempt us to regard the histo- 
logical changes in a simple hemorrhage from rupture of a 
vessel or from trauma as essentially the same as and only 
differing quantitatively from those of an abscess. 
13Nissl: “In my opinion pathological anatomy would not be fundamentally 
disturbed, if one off-handedly abolished the present-day concept of inflammation.,v 
Hist, and Histopath. Arbeiten, Vol. I, p. 466. 142 
HISTOPATHOLOGICAL COMPLEXES 
In the same manner there should be grouped those 
histological processes which are collectively known as the 
ectodermal type. We saw that they developed sometimes 
apparently without known cause,—primary; at times 
obviously as a reaction to various injuries of the tissues. 
They are often considerably intermingled with the changes 
of the mesodermal type. They have been described as the 
type of “ non-suppurative ” encephalitis. 
The above-mentioned grouping together of processes 
supposed to be inflammatory was possible only by 
focussing attention on these two types of secondary 
degenerations, by regarding them as characteristic of 
inflammation, and neglecting the other findings as imma- 
terial or as merely indicative of its stage or severity. 
This arrangement is further quite artificial in that it 
does not correspond to the experience of pathologists, 
that the several changes may merge into one another. 
Suppuration is always suppuration from the beginning, 
and is characterized as such histologically; the reactive 
processes in connection with a hemorrhage, with a non- 
inf ectious embolic softening, a crushing injury, never 
develop into an abscess, unless it be that a secondary 
infection is superadded; the so-called parenchymatous 
neuritis never develops into an exudative interstitial 
form; and the latter is never a preliminary stage of a 
“parenchymatous” neuritis. 
Inflammation is in origin purely a clinical concept, 
applied to a symptom complex of the external parts, with 
the symptoms of rubor, dolor, and tumor. For a long 
time its gross anatomical and histological basis has been 
sought. Since Cohnheim’s classical investigation the 
exudation of white and red blood corpuscles and of fluid INFLAMMATION 
143 
constituents of the blood through unruptured vessel 
walls has been regarded as the chief anatomical character- 
istic of the inflammatory process. For the nervous system 
Nissl shared Cohnheim’s viewpoint, and taught that only 
those histological processes should be termed inflamma- 
tory, in which, beside progressive and regressive changes 
in the parenchyma, exudative vascular phenomena may 
be demonstrated. 
Chief emphasis is usually placed on the demon- 
stration of extravasated white blood-corpuscles, since 
■Bone of skull 
•dura mater 
■subdural space 
'arachnoid 
•trabeculae of arachnoid in the subarachnoidal space 
'Pia mater (intima piae of Held), to it is closely ap- 
proximated the membrana limitans gliae 
Brain substance 
Fig. 44.—Schema of membranes and lymph paths of the brain. 
the demonstration of increased fluid exudate is difficult 
with the available histological methods. The presence 
or absence of erythrocytes is as a rule regarded as of 
slight significance. 
When white blood cells in the central organs emigrate 
from unruptured vessels they reach first the perivascular 
lymph spaces, and only afterwards the specific nervous 
tissues (see Chapter IV). Accordingly we find extrav- 
asated blood elements as 'perivascular accumulations or 
as a diffuse infiltration of the tissue. This brings us to two 144 
HISTOPATHOLOGICAL COMPLEXES 
important concepts for the doctrine of inflammation, 
which have been often misinterpreted, and the clear 
definition of which was established by Nissl. Accumula- 
tion of cellular elements in the lymph spaces and increase 
of nuclei in the tissues may have very different signif- 
icance, according to the nature and source of these 
elements. They are indicative of the “inflammatory” 
Perivascular 
shrinkage space 
Adventitial (A) 
nucleus 
Lumen of vessel (L) 
Adventitial 
(Virchow-Robin) 
lymph space 
Membrana 
limitans glia 
perivascularis 
, Endothelial tube 
, Connecting trabeculae of the adventitia 
External layer of adventitia 
Inner layer of the adventitia 
Endothelial nucleus 
Location of space of His (internal to rnembrana limitans glise 
• perivascularis) 
Nervous tissue 
Fig. 45.—Vessel wall and lymph spaces. Semi-schematic, after drawings by Held. 
nature of pathological processes only when they can be 
shown to have arisen from the blood stream. That which 
is in literature designated as “round-cell infiltration” is 
of very heterogeneous nature. 
Of primary importance is the different significance of 
the two chief types of white blood corpuscles, the mono- 
nuclear lymphocytes, and the polymorphonuclear leuco- 
cytes. The latter are the specific extravasated elements 
of all suppurations (Fig. 51). They show a pronounced 
tendency to penetrate rapidly through the lymphatics LEUCOCYTES 
145 
into the tissue, to form accumulations and lead to lique- 
faction necrosis (abscess formation).14 
Diapedesis of leucocytes through the injured vessel 
wall and necrobiosis (even to liquefaction) of the infil- 
trated tissue and of the infiltrating leucocytes constitute 
the findings in every abscess. Associated with these are 
the well-known reactions of the surrounding tissue, and, 
competing with them, the further progress of the lique- 
faction necrosis. 
But not only in the formation of an abscess may we 
observe the extravasation of leucocytes through the vessel 
walls. Nissl wrote: “If I destroy the cortical tissue with a 
hot needle or make a simple incision into it, or introduce a 
drop of blood, or freeze the tissue, we find in all these 
cases at first a more or less pronounced accumulation of 
multinuclear leucocytes and progressive changes in the 
parenchyma; but the leucocytes very soon disintegrate.” 
One of his pupils, Devaux, states more exactly that the 
leucocytes appear after about 12 hours, begin to degen- 
erate after 24 hours, and have disappeared after 3 days 
This means that after any single, acute, temporary injury 
leucocytes at first emigrate from the vessels into the 
tissues just as Cohnheim found in his studies on inflam- 
mation; but they disappear in a few days. They are 
transient guests and rapidly disintegrate. In suppuration 
on the contrary, the extravasation continues, fresh leuco- 
cytes keep pouring in as the old ones break up, and an 
abscess is formed with liquefaction of the tissue. How- 
ever, these phenomena are not due to the original injury, 
but to the added, persistent insult due to multiplying 
14 It is questionable whether this is due to the.activities of the leucocytes per se. 
or to those of the contemporaneous etiologic factors ox metabolic products. 146 
HISTOPATHOLOGICAL COMPLEXES 
microorganisms.15 From this it follows that the much 
discussed classical observation of Cohnheim on the 
extravasation of leucocytes is a phenomenon which lasts 
merely for a few days after the injury and rapidly disap- 
pears, except when the conditions leading to inflammation 
persist; and, above all, it is without influence on the nature 
of the consequent reactive and reparatory processes. 
In contrast to the leucocytes, which rapidly scatter 
through the tissue, we find that the lymphocytes accu- 
Fig. 46. 
Plasma cell 
Plasma cell 
Fig. 47. 
Plasma cel! 
Endothelial 
nucleus viewed 
on edge 
Plasma cell 
in poor focus 
Glia nucleus 
Plasma cell 
Plasma cell- 
Fig. 46 and 47.—Capillaries with plasma cells. Nissl stain. Oil immersion, 
mtilate, particularly in the perivascular lymph spaces, and 
that they do not simultaneously appear in the surrounding 
tissue; they form the true “perivascular infiltrations. ” 
Nissl has expressed this by stating that the lymphocytes 
respect the biological border lines between the vessels 
with their connective tissue and the ectodermal tissue, 
while the leucocytes do not. 
In many cases, especially in the more chronic morbid 
processes, we find, mixed with the lymphocytes or out- 
numbering them the plasma cells of Unna and Marschalko. 
These are mononuclear elements characterized by the 
15 For a detailed discussion see: Schroder, “Uber Entziindung,” Ziegler’s 
Beitrage, Vol. LXXI, 1922, p. 1. PLASMA CELLS 
147 
arrangement of the nuclear chromatin to resemble the 
spokes of a wheel; they possess a more or less large cell 
body containing irregularly shaped lumps of a material 
that stains deeply with basic aniline dyes, and with a 
pale perinuclear “court'’ (Figs. 46 to 48). Concerning 
the origin of the plasma cells there has been much debate,16 
Endothelial 
cells viewed * 
on edge 
Plasma cells, 
viewed on edge 
Plasma cells," 
viewed on edge 
Vascular lumen 
Endothelial 
cells viewed 
on the flat 
Endothelial cells 
viewed on the flat 
Plasma cells, viewed 
on the flat, lying on 
the vessel. 
,.*“Mast cell” 
Endothelial cells 
viewed on edge 
Plasma cells, 
viewed on the flat, 
lying on the vessel 
Fig. 48.—Longitudinal section of a vessel, infiltration of the adventitial sheath with large 
plasma cells. Nissl stain. After a drawing by Alzheimer. High magnification (oil immersion). 
but it is now generally accepted that they are derivatives, 
transitional forms, of mononuclear lymphocytes; between 
plasma cells and lymphocytes all intermediate forms occur. 
Alzheimer, Nissl, and after them many others be- 
lieved that the lymphocytes which become transformed 
into plasma cells are of hematogenous nature, that is, 
they emigrate per diapedesin through the vessel walls. 
According to the belief of Nissl and others, their presence 
16 For a historical discussion on plasma cells see: Nissl, Histol. u. histopafhol. 
Arbeiten, Vol. I. S. 247 ff.; Marchand, “ Refer at auf der Tagung Deutschen Pathol. 
Gesellschaft zu Marburg191.3. 148 
HISTOPATHOLOGICAL COMPLEXES 
is always to be regarded as proof of the inflammatory 
nature of the disease process, in the sense of Cohnheim, 
However, the histological demonstration of diapedesis of 
lymphocytes and plasma cells is difficult. In spite of 
their great numbers in many cases, hardly ever has one 
of them been observed in the act of passing though the 
vessel wall, in contrast to what is seen in the case of leu- 
cocytes. For that reason the possibility of a different 
origin of the perivascular lymphocytic and plasma cell 
infiltration cannot be dismissed, namely the lymphatics.17 
According to this view the lymphocytes do 
not come from the blood but from the lymph, 
occurring as collections and accumulations 
(occasioned by cliemotaxis or similar factors) 
within the adventitial lymph spaces of the 
vessels, which freely communicate with the 
subarachnoidal space and the general lymph 
system of the body. If this be true these 
perivascular infiltrations have not arisen by extravasation; 
and therefore do not constitute proof of the inflammatory 
nature of the lesions in which they are found. They would 
be accumulations of cells from the lymph, swimming in 
the lymph stream of the adventitial lymph spaces; their 
origin might be looked for outside of the brain and in the 
hemopoietic organs.18 
Of another cellular element which may produce the 
picture of diffuse tissue infiltration, as well as of perivas- 
cular accumulation, much has already been said; this is 
the granule cell. We have seen that granule cells develop 
Fig. 49.—Russel 
cell (degenerated 
form of a plasma 
cell having a mul- 
berry shape). 
Photograph, oil 
immersion. Nissl 
stain. 
17 Schroder, Zcitschr.f. d. ges. Psych, u. Neurol., Vol. LII, 1920. 
18 Tumors also may grow into the lymph paths of the brain, and fill the adventitial 
spaces with their cells, so that the resulting picture is like that of a perivascular in- 
filtration. GRANULE CELLS 
149 
from various fixed tissue elements, that they become free 
only when fully developed, that they then reach the 
perivascular lymph stream, and that they are in part 
locally formed from the adventitial elements of the vessels. 
After loss of their contents and disintegration of their 
cell body, the nuclei of such granule cells may very closely 
resemble lymphocytes (Held’s reduced granule cells, 
Fig. 34). The development and composition of “infil- 
trates ” thus originating is accordingly very different from 
that of true cellular extravasations: in the one case the 
Lymphocytes 
Adventitial 
lymph space 
. Russel cell in 
the tissue 
Endothelial cell 
Russel cell in 
the lymph space 
Capillary tube " 
__ Russel cell in 
"”the lymph space 
Membrana 
-limitans 
perivascularis 
Fig. 50.—Capillary, infiltration of the adventitial sheath with Russel cells. Hematoxylin-van 
Gieson. Oil immersion. 
elements immigrate from the blood stream through the 
vascular wall into the lymph spaces and into the tissue; 
in the other case, from the tissue into the lymph spaces 
(Fig. 52 and Figs. 26 to 29). It is therefore obvious that 
these two types of infiltration do not have the same sig- 
nificance in our study of the underlying morbid processes. 
It was formerly assumed that white blood corpuscles 
are everywhere present, free in the nervous tissue. It 
was therefore not difficult to suppose that tissue infiltra- 
tions could be produced through multiplication of these 
elements. Weigert and Nissl, however, demonstrated HISTOPATHOLOGICAL COMPLEXES 
150 
that structures formerly regarded as free nuclei (Ilenle) 
and thought to be identical with lymphocytes, are nuclei 
of neuroglia, the nature of which it was not possible to 
recognize with the older methods; and that normally 
lymphocytes occur not at all or very rarely free in the 
nervous tissue.19 
An increase of the nuclei in the tissue frequently 
occurs under pathological conditions, through various 
proliferative changes in the glia. Not infrequently this 
Endothelium 
~ Vascular lumen 
Adventitia 
Adventitia 
Fig. 51.—From a small recent brain abscess in the white matter of the cerebral hemispheres. 
Nissl stain. Infiltration of the vessel wall and the surrounding tissue with leucocytes. 
glial nuclear proliferation in the cortex, as well as in the 
deeper portions of the central organs, is particularly 
pronounced in the immediate neighborhood of the vessel 
walls. If, as is frequently true in such cases, the nuclei 
lie in rows quite close to the glial membrana limitans peri- 
vascularis, the picture of a true perivascular exudation 
may readily be simulated, since these nuclei possess a 
certain superficial resemblance to lymphocytes (Figs. 24, 
53, 26, and 28). 
19 Cf. Nissl, Histol. u. histopathol. Arbeiten, Vol. I. p. 440 ff.; Ranke, ibid., Vol. 
II, S. 276. RED BLOOD CELLS 
151 
It may also be mentioned that, with superficial exam- 
ination, even an increase of the elements of vessel walls 
(e. g. in endarteritis obliterans) may impress one as an 
extravasation (Fig. 43). 
In Cohnheim’s experiments on inflammation, besides 
the white corpuscles, red blood cells were found to escape 
Fig. 52.—"Small round cell infiltration” of the cerebral cortex. Nissl preparation. Low mag- 
nification. The infiltration, perivascular as well as diffuse, consists solely of granule cells and 
large glia cells. Of the latter the opaque bodies are recognizable. 
from the vessels. This observation is partly responsible 
for the conception of the so-called hemorrhagic encephalitis 
or myelitis. In human pathology hemorrhages in true 
inflammatory (exudative, diapedetic) processes are not at 
all as frequent as is usually assumed. Miliary or even 
somewhat larger hemorrhages are found for the most only in 
very acute inflammatory injuries; and even in such cases it 
is not certain whether they should be regarded as part 
and parcel of the inflammation or whether they do not 
rather represent merely secondary changes resulting from 
miliary thrombi and ruptures. 
Another group of processes, commonly designated as 
hemorrhagic encephalitis, has in all probability nothing 
whatever to do with inflammation. These are processes, 152 
HISTOPATHOLOGICAL COMPLEXES 
the recognizable morbid changes in which consist merely 
in punctiform hemorrhages, and the well-known reactions 
that regularly follow upon the hemorrhages in the sur- 
rounding tissue. 
If we now sum up the conceptions discussed in the 
preceding and in this chapter we have a series of argu- 
ments in favor of dividing up that great group of histo- 
pathological processes which even today is too often 
inclusively called inflammation. It seems better to avoid 
entirely the term parenchymatous inflammation and to 
separate from inflammation both those simple reactions 
of the supporting tissue which regularly follow degener- 
Glial nuclei 
Endothelial cell 
viewed on edge 
Capillary wall 
Endothelial cell 
viewed on edge 
Glial nuclei 
Fig. 53.—Capillary surrounded by numerous glial nuclei. Nissl stain. 
ative processes in the parenchyma, and the reactions 
consequent to destruction of tissue. If we wish to retain 
the term inflammation at all we should restrict it to those 
morbid changes in which cellular extravasations occur 
unquestionably; this occurrence at present is certain only 
in the case of suppuration. Where this is lacking it is 
well at present to be cautious about making the histo- 
logical diagnosis of inflammation. It is quite likely that 
various other histopathological changes may be met with INFLAMMATION 
153 
in inflammatory areas, such as degenerative processes in 
the “parenchyma," and regressive and progressive alter- 
ations in the ectodermal and mesodermal supporting 
tissues. However, these changes most not be looked 
upon as characteristic of the inflammatory process, but 
merely as the anatomical expression of injuries, with the 
resulting reactive tissue proliferation which is produced in 
the living organs by any injury under all conditions (pro- 
liferation of vessels, formation of fibroblasts and granule 
cells, proliferation of glia). INDEX 
A 
Abraumzellen, 132 
Abscess, 145 
Achuccaro, adventitia demonstration, 
59 
staining method, 105 
Adamkiewicz on Golgi nets, 30 (ref.) 
Alzheimer on chronic retrogressive 
changes, 131 (ref.) 
on disintegration products, 100 (ref.) 
on glial changes, 101 (ref.) 
Anemia, 136 
cord changes in, 124 
Apathy on neurofibril studies, 22 
(ref.), 32 
Arachnoid, 60 
Artefacts in ganglion cells, 88 
see lymphatics 
Arteries, 56 
endarteritis obliterans, 107 
sclerosis, 106 
syphilis, 109 
Arteriosclerosis, 106 
Astrocyte, 45, 50, 95 
Atheroma, 106 
Atrophy, 137 
Aufbauzellen, 135 
Axis cylinder, 37 
in anemias, 125 
changes regressive, 112, 114 
regeneration, 116 
degeneration discontinuous, 122 
B 
Bartels on lymphatics, 60 
Baumgarten, syphilitic arteritis, 109 
Baumler on granule cells, 131 (ref.) 
Bethe or axis-cylinder fibrils, 37 
nets of, 51 
on neurofibril studies (ref.), 30 
Betz, cells of, 6 
Bielschowsky on neurofibril changes, 
89 
on regeneration of central fibres 
(ref.), 39 
stain, 78, 89 (ref.) 
Bodies of Elzholz, 41, 112, 113, 121, 
123 
Boedeker on cord changes in anemia, 
124 (ref.) 
Brain architecture, 75 
fetal, reaction to injuries, 130 
histological structure, 73 
Brodmann on cerebral architecture, 
75 (ref.) 
C 
Cajal on architecture of cerebrum, 75 
(ref.) 
views on Golgi nets, 30 
Capillaries, 59 
changes in, 109 
Cells, of Betz, 6 
endothelial, in arteriosclerosis, 107 
glial, 46 
Nissl picture, 53 
proliferation, 93 
progressive changes, 101 
regressive changes, 101 
stainable substances, 4 
staining methods, 100 
granule, 131 
definition, 132 
in degeneration of nerve fibres, 
119 
155 156 
INDEX 
Cells, granule, demonstration of, 133 
fate of, 135 
functions, 132, 135 
gliogenic, 99, 134 
in inflammation, 148 
mesodermal type, 133 
of neurilemma, 115, 134 
origin, 133 
reduced, 135, 149 
mast, 104, 129 
monster, 95, 103 
nerve, 3 
artefacts, 88 
degeneration, acute, 80, 85, 89 
chronic, 85 
severe, 87 
disintegration, 86 
fibrillar reticulum, 23 
Golgi picture, 84 
in cerebral cortex, 74 
cord, 70 
necrobiosis, 87 
neurofibril changes, 89 
Nissl picture, 83 
nucleus, 17 
pathology of, 79 
pigments, 16 
proliferation, 91 
see lymphatics pericellular 
shape, 20 
spinal, 9 
stainable substance, 4, 11, 26, 84 
staining methods, 3, 4 
fibril, 4, 22 
Golgi’s, 4, 21 
Nissl’s, 4, 15 
unstained paths, 84 
of Russel, 135 
plasma, 136, 146 
Purkinje, 10 
pyramidal, 8 
reticular, 99, 105-, 125, 134 
rod, 103 
satellite, 65, 72 
Cells, spider, 95, 103, 104 
Cerletti, on arteriosclerosis, 106 
Compression, 130 
Cord, spinal in anemia, 124 
neuroglia of, 70 
structure of, 68 
vessels of, 70 
Cysts, 97, 130, 131 
D 
Degeneration of nerve cells, acute, 85, 
89 
chronic, 85 
severe, 87 
of nerve fibres 
discontinuous, 21 
of nerves, peripheral, 114 
primary, 115 
secondary (Wallerian), 38, 112 
of central fibres, 116 
Disintegration, discontinuous, 112 
Doinikow, on degeneration of per- 
ipheral nerves, 114 (ref.) 
on fibre structure (ref.), 41 
E 
Edema, 136 
Edinger, on fibre regeneration (ref.), 
39 
Elzholz, bodies of, 41, 112, 113, 121. 
123 
Encephalitis, 137 
hemorrhagic, 151 
non-suppurative, 142 
periaxialis, 126 
Endarteritis obliterans, 107 
Endoneurium, 68 
Endothelium, 58 
Ependyma, embryology, 48 
Epineurium, 68 
Evensen, on artery structure, 57 
(ref.) INDEX 
157 
F 
Fibres, nerve, 37 
in anemia, 124 
axis-cylinder changes, 112 
degeneration, 38 
discontinuous, 121 
primary, 115 
secondary, 112, 116 
embryology of, 41 
histopathology, no 
in neuritis, 124 
regeneration, 39, 91, in, 117 
sheaths, 38 
changes in, in 
of spinal cord, 69 
Fibrils, glial, 45, 49 
of brain, 73 
of cord, 69, 71 
proliferation, 93, 94 
reaction to injuries, 129 
see Neurofibrils 
Fischer, O., on progressive paralysis, 
126 (ref.) 
Flechsig, on architecture of cerebrum, 
75 (ref.) 
Fragnito, on embryology of nerve 
fibres, 41 
G 
Ganglion cell, see cell, nerve 
von Gehuchten, on fibre degeneration, 
113 (ref.) 
Gerletti, U., on rod cells, 104 (ref.) 
Gitterzellen, 134 
Glia, see cell, glial 
ependyma 
fibre, glial 
membrana limitans 
neuroglia 
spongioblast 
Glioma, 93 
Gluge, on “ inflammatory globules,” 
131 
Goldschmidt, on neurofibrils, 28 
Golgi, nets, 30, 51, 65 
in necrobiosis, 88 
stain, 4, 21 
for glia demonstration, 44 
for nerve cells, 84 
Granules of Reich, 41 
Grawitz, P., on glial proliferation, 96 
(ref.) 
“ Gray, nervous,” 31 
H 
Hauptmann, A., on lymphatics, 59 
Heilbronner, on fibre degeneration,. 
124 (ref.) 
Held, on adventitia, 57 
on glia derivation, 43 
on glia histology, 47 (ref.) 
on Golgi nets, 30 
on granule cells, 131, (ref.) 
on lymphatics, 63 
on neurilemma, 42 
on sheath medullary, structure of,. 
42 
Hemorrhage, 141, 151 
His, on glia embryology, 48 (ref.) 
spaces of 61, 63 
Hyperemia, 136 
I 
Infiltration, calcareous, 107 
round cell, 144 
Inflammation, 137, 152 
erythrocytes in, 143, 151 
granule cells in, 148 
hemorrhage in, 151 
leucocytes in, 143 
lymphocytes in, 144, 146 
non-suppurative, 142 
plasma cells in, 146 
suppurative, 142, 145 158 
INDEX 
J 
Jaderholm, on fibril staining, 27 
Jakob, on degeneration of central 
fibres, 117 (ref.) 
on structure of central nerve fibres, 
(ref.) 211 
K 
Kaes, on architecture of cerebrum, 75 
(ref.) 
Kronthal, theory of nerve cells, 12 
(ref.) 
Kriickmann, on glial granule cells, 99 
(ref.) 
on lymphatics, 62 
L 
Leucocytes, duration, 145 
in inflammation, 144 
Lugaro, on fibre degeneration, 113 
(ref.) 
Lymphatics, 59 
adventitial, 62 
embryology, 61 
fluid exchange, 67 
functions, 68 
intracerebral, 61 
pericellular, 64 
perivascular, 63 
subarachnoidal space, 61 
subdural space, 61 
Lymphocytes, in inflammation, 144, 
146 
in tissues, 150 
origin, 148 
M 
Mann, G., on stainable substance, 12 
(ref.) 
Marchand, on plasma cells, 147 (ref.) 
Marchi, stain, 76, in 
Mater, dura, 60 
pia, 60 
Matter, gray, of brain, 71 
of cord, 70 
white, of brain, 71 
of cord, 69 
Membrana fenestrata, 57 
limitans gliae, 48, 51, 61, 64 
in cyst formation, 97 
Merzbacher, on granule cells, 131 
(ref.) 
Metachromasia, in cell degeneration, 
86 
in necrobiosis, 87 
Miyake, on regeneration of central 
fibres, (ref.), 39 
von Muller, E., on plasma cells, 136 
(ref.) 
Myelitis, 137 
hemorrhagic, 151 
Myelospongium, 48 
N 
Necrobiosis of ganglion cells, 87 
Nemiloff, on sheaths of nerve fibres, 
(ref.), 39 
Nerves, peripheral, 68 
degeneration of, 114 
Nets of Bethe, 51 
of Golgi, 30, si, 65 
in necrobiosis, 88 
Neural, 36 
Neurilemma, 38, 39, no 
derivation of, 42 
granule cells, 115 
progressive changes, 114 
regeneration, in, 115 
Neuritis, 124, 137 
Neuroblast, 48 
Neurofibrils, 6, 15, 22 
Apathy’s conception, 22, 32 
Bethe’s conception, 25 
in degeneration, acute cellular, 89 INDEX 
159 
Neurofibrils, in earthworm and leech, 
22 
elementary, 23, 25 
function, 28 
Goldschmidt’s views, 28 
in man, 25 
primitive, 23 
reticula, 23, 26 
staining method, 4, 89 
Neuroglia, 43 
in anemia, 125 
cell—see cell, glial 
of cord, 69 
derivation of, 43 
embryology, 48 
fibril—see fibril, glial 
functions, 52 
Golgi picture, 44, 50 
granule cells, 99 
in nerve degeneration, 118 
Nissl picture, 52 
phagocytic power, 97 
progressive changes, 92 
in cells, 95 
in fibrils, 93 
proliferative power, 91 
reaction to injuries, 129 
regressive changes, 91 
relation to vessels, 55 
staining methods, 100 
Weigert method, 45, 50 
syncytial character, 47 
Neurone theory, 29, 31, 32 
Neuropil, 23, 74 
Nissl, on axone, 20 (ref.) 
on capillary changes, 109 (ref.) 
on degeneration, acute cellular, 80 
(ref.), 85, 89 
chronic cellular, 85 
severe cellular, 87 
on disintegration of cells, 86 
on equivalent picture, 12, 13 (ref.) 
on glial cell changes, 101 (ref.) 
Nissl, on granule cells, 131 (ref.) 
on inflammation, 137 (ref.), 141 
(ref.) 
on lymphatics, 62 (ref.) 
on lymphocytes, 150 (ref.) 
on mesodermal tissue reaction, 128 
(ref.) 
on necrobiosis, 87 
on nerve cells, altered, 85 (ref.) 
morphology of, 5 (ref.) 
on “ nervous gray,” 31 
on neurofibrils, 29 (ref.) 
on neurone theory, 32 (ref.) 
on pathology, general, 2 (ref.) 
on plasma cells, 147 (ref.) 
on section preparation, 15 (ref.) 
on spaces, pericellular, 65 
stainable substance, 4, 11, 26 
in ganglion cells, diseased, 84 
staining method, 4, 15, 78, 79, 83 
glia picture, 52, 53 (ref.) 
glial alterations, 101 
Nodes of Ranvier, 38 
Nonne, on cord changes in anemia, 
124 (ref.) 
O 
Obersteiner, on pericellular spaces, 64 
(ref.) 
v. Orzechowski, on regeneration of 
nerve cells, 91 (ref.) 
P 
Paladino, on structure of medullary 
sheath, 42 
Paralysis, progressive, 126 
Paths, unstained, 16 
in diseased cells, 84 
Perineurium, 68 
Perusini, on plasma cells, 136 (ref.) 
Picture, equivalent, 12, 77 
Pigments, 16 
Purkinje, cells of, 10 160 
INDEX 
R 
Ranke, O., on vessel structure, 57 
(ref.) 
Ranvier, nodes of, 38 
Reich, granules of, 41 
on sheaths of nerve fibres, (ref.), 
39 
Reichardt, on edema, 137 (ref.) 
Reticulum, diffuse elementary, 23 
intracellular, 23, 26 
S 
Satellites, 65, 72 
Schroder, on anemic changes of cord, 
124 (ref.) 
on arteriosclerosis, 106 (ref.) 
on arteritis syphilitica, 109 (ref.) 
on cerebral changes in dementia, 
137 (ref.) 
on encephalitis and myelitis, 126 
(ref.) 
on ganglion cell changes, 79 (ref.) 
on incrustation of Golgi nets, 88 
(ref.) 
on inflammation, 137 (ref.), 146 
(ref.) 
on lymphatics, 60 (ref.) 
on plasma cells, 136 (ref.), 148 
(ref.) 
Schultze, fibrillar doctrine, 22 
Schwann, sheath of, see Neurilemma 
Sclerosis, diffuse, 126 
malignant multiple, 126 
multiple, 126 
Sheath, medullary, 38, no 
in anemia, 125 
in systemic disease, 123 
degeneration, discontinuous, 121 
regeneration, in 
regressive changes, in, 114 
of Schwann, see Neurilemma 
Softening, 130 
Space, epicerebral, 61 
epispinal, 61 
of His, 61, 63 
pericellular, 64 
subarachnoidal, 61 
subdural, 61 
Spatz, on degeneration of nerve 
fibres, 113 (ref.) 
on fetal brain reaction to injury, 
130 (ref.) 
Spielmeyer, on ganglion cell changes, 
82 (ref.) 
on glial proliferation, 93 (ref.) 
on plasma cells, 136 (ref.) 
on progressive paralysis, 126 (ref.) 
on rod cells, 104 (ref.) 
Spongioblast, 48 
Stains, Achuccaro’s, 105 
Bielschowsky’s, 78 
fibril, 4 
Golgi’s, 4, 21 
for ganglion cells, 84 
for glia, 44 
Marchi’s, 76, in 
for morbid changes, 76 
for neuroglia, altered, 100 
Nissl’s, 4, 15, 78, 79, 100 
ganglion cell picture, 83 
glia picture, 52 
Weigert’s medullary sheath, 76, 78 
resorcin fuchsin, 57 
Storch, on glia fibril proliferation, 94 
(ref.), 95 (ref.) 
Stransky, on degeneration of nerve 
fibres, 121 (ref.), 123 (ref.) 
on neurilemma structure, (ref.), 40 
Stroebe, on degeneration of central 
fibres, 117 (ref.), 118 (ref.) 
o n degeneration of peripheral 
nerves, 114 (ref.) 
Substance, “ essential nervous,” 28 
stainable of Nissl, 4, n, 26 
in diseased ganglion cells, 84 
Syphilis of blood vessels, 109 INDEX 
161 
T 
Theory, neurone, 29, 31, 32 
Thrombosis, 130 
Thumazuri, cord changes in anemia, 
125 (ref.) 
Tissue, connective, changes, 105, 129, 
141 
distribution, 55 
V 
Veins, 56 
Vessels, blood, Achuccaro’s method, 
59 
adventitial changes, 105 
arteriosclerosis, 106 
capillary changes, 109 
of cord, 70 
endarteritis obliterans, 107 
reactive changes, 129 
Vessels, blood, relation to nervous 
tissue, 55 
syphilis, 109 
Weigert’s stain, 57 
lymphatic, see lymphatics 
Virchow-Robin, spaces of, 62 
W 
Waldeyer, neurone theory, 31 
Waller, law of, 113 
secondary degeneration, 38 
Weigert, glial fibres, 45 (ref.), 49 
proliferation, 93, 94 
stain for medullary sheath, 76, 78 
resorcin fuchsin, 57 
Wohlvill, cord changes in anemia, 
125 (ref.) 
Y 
Y-cells, 135