» i 1

i'W^s<mmMi>^-mi!f<ii-

^Uwiv. or
Toronto

tiBRARV

>^^^^

Aberdeen University
Studies : No. 21

Studies in Pathology

University of Aberdeen.

COMMITTEE ON PUBLICATIONS.
Convener: Professor James W. II. Trail, F.R.S., Curator of the University Library.

UNIVERSITY STUDIES.
General Editor: Peter John Anderson, M.A., LL.B., Librarian to the University.

1900. No. \. — Roll of Alumni in Arts of the University and King's College of Aberdeen, 1596-1860.

Edited by P. J. Anderson.

•I No. 1.— Records of Old Aberdeen, 1157-1891. A. M. Munro, F.S.A. Scot. Vol. I.

It No. y— Place Names of West Aberdeenshire. James Macdonald, F.S.A. Scot.

1901. No. 4. — The Family of Burnett of Leys. George Burnett, LL.D., Lyon King of Arms.
11 No. I.— Records of Invercauld, 1547-1828. Rev. J. G. Michie, M.A.

1902. No. 6. — Rectorial Addresses delivered in the Universities of Aberdeen, 1835-1900. P. J. Anderson.
II No. 7. — The Albemarle Papers, 1746-48. Professor C. S. Terry, M.A.

1903. No. %.— The House of Gordon. J. M. Bulloch, M.A. Vol.1.
II No. g.— Records of Elgin. William Cramond, LL.D. Vol.1.

1904. No. 10.— Avogadro and Dalton : The Standing in Chemistry of their Hypotheses. A. N. Meldrum, D.Sc.

II No. 11.— Records of the Sheriff Court of Aberdeenshire. David Littlejohn, LL.D. Vol. L

II No. \i.— Proceedings of the Aberdeen University Anatomical and Anthropological Society, 1902-04.
President, Professor R. W. Reid, M.D., F.R.C.S.

1905. No. ly— Report on the Alcyonaria collected by Professor Herdman at Ceylon in 1902. Professor

J. Arthur Thomson, M.A., and W. D. Henderson, B.Sc.

I. No. \«,.— Researches in Organic Chemistry. Professor F. R. Japp, M.A., LL.D., F.R.S., William
Maitland, B.Sc, Joseph Knox, B.Sc, James Wood, B.Sc

II No. i^.—MeminisseJuvat: with Appendix of Alakeia. Alexander Shewan, M.A.

•I No. \6.—The Blackhalls of that Ilk and Barra. Alexander Morison, M.D.

1906. No. \T.~Records of the Scots Colleges at Douai, Rome, Madrid, Valladolid, Ratisbon. Vol. L

P. J. Anderson.

No. \Z.—Roll of the Graduates of the University of Aberdeen, 1860-1900. Colonel William Johnston, C. B.

M No. i^.— Studies in the History and Development of the University of Aberdeen. P. J. Anderson
and others.

I. No. ^.-Studies in the History and Art of Asia Minor. Professor W. M. Ramsay, D C.L., and pupils.
I. No. i\.— Studies in Pathology. William Bulloch, M.D., and others.

c

^

Studies m Pathology a>

WRITTEN BY ALUMNI
TO CELEBRATE THE
QUATERCENTENARY
OF THE UNIVERSITY
OF ABERDEEN & THE
QUARTER-CENTENARY
OF THE CHAIR OF
PATHOLOGY THEREIN

^<l

y\

-r^

"i^

W

Edited by

William Bulloch, M.D.

Bacteriologist to the London Hospital

ABERDEEN

MCMVI

\ I

PRINTED BY

MILNE AND HUTCHISON

ABERDEEN

To
ALMA MATER^

To Thee we come again, bringing in oat hands
these symbols of our Love and Reverence. ♦

We ask one thing only— that, at these splendid
Solemnities, we may lay our humble offerings on
the Altar, and, by this, renew our vows of Devotion
and Service. ♦ . . ♦ , .

PREFACE

OHORTLY after it was determined to celebrate the
^"-^ Four Hundredth Anniversary of Aberdeen University
in 1906, the notion occurred to several of the medical graduates
in London that a fitting tribute to the occasion would be the
publication of a volume of original researches, in the domain
of Pathology, written by those who had graduated since the
foundation of the Chair of Pathology in 1882. The occasion
seemed all the more appropriate for such commemoration, seeing
that Professor Hamilton, under whose tuition the above
graduates received their primary instruction, has all but com-
pleted a quarter of a century's tenure of office. A movement
was started to sound our graduates holding appointments in
Pathology as to how far they were in sympathy with the
proposal, and the researches contained in this volume are the
responses to that appeal, and are evidence of the feeling that
prevailed among them, both for their Alma Mater and their
teacher in Pathology. All the researches in the volume were
specially written for it, and the cost of production has been
defrayed by the contributors. It remains for the Editor to
acknowledge, with deepest gratitude, the self-imposed labours
of our University Librarian, Mr. P. J. Anderson, who, to all
intents and purposes, passed the volume through the press.

W. B.

London, July, igo6.

b

i

CONTENTS.

PAGE

Professor Hamilton. By W. L. M. xiii.

The History and Progress of the Sir Erasmus Wilson Chair of
Pathology from the time of its Foundation to the present
date. By W. B xix.

The Alimentary Canal as a Source of Contagion. By Professor

Hamilton ... ... ... ... ... ... ... (i)

A Remarkable Case of Bilharziosis. By William St. Clair

Symmers, M.B., CM. (1887) (Illustrated) (39)

Malformations of the Bulbus Cordis. By Arthur Keith, M.B.,

CM. (1888); M.D., F.R.C.S (Illustrated) (55)

The Administrative Aspects of Tuberculosis. By William Leslie

Mackenzie, M.A., M.B., CM. (1888); M.D., D.P.H. ... (75)

Paroxysmal Irregularity of the Heart and Auricular Fibrillation.
By Arthur Robertson Cushny, M.A., M.B., CM. (1889);
M.D.: and C W. Edmunds (Illustrated) (95)

Researches on certain Problems of Plague Immunity. By

George Dean, M.A., M.B., CM. (1889) (in)

Experimental Study of the Immunity against Bacillus Pyocyaneus.

By William Bulloch, M.B., CM. (1890) ; M.D (157)

On Epignathus. By Alexander Low, M.A., M.B., CM.

(1896) (Illustrated) (175)

A Contribution to the Pathology of Exophthalmic Goitre. By

George Mellis Duncan, M.B., CM. (1896) (Illustrated) (199)

The Rat Theory of Plague Epidemics. By William Hunter,

M.B., CM. (1897) (Illustrated) (215)

b I

Xii. CONTENTS

PAGE

Some Experiments with Disinfectants. By Andrew Ross Laing,

M.B.,C.M.(i897); M.D, D.P.H (237)

On Eck's fistula— observations on Four Dogs, with a review of
the h'terature relating to previous work on the subject.
By John James Rickard Macleod, M.B., CM. (1898) ... (265)

On the Action of certain Bacteria in producing Cell-necrosis,
with Special Reference to those of the Bacillus Enteritidis
(G^ertner) Group. By George Ford Petrie, M.B., CM.
(1898); M.D : (287)

The Relationship between the Factors inducing Haemolysis
and those inducing the Phagocytosis of Red Blood
Corpuscles. By Robert Donald Keith, M.A., M.B., CM.
(1902); M.D (Illustrated) (301)

An Experimental Enquiry into the Relationship of Leucocytosis
to the Opsonic Content of the Blood Serum. By
James Charles Grant Ledingham, M.A., B.Sc, M.B., CM.
(1902), and William Bulloch, M.B., CM. (1890); M.D. (Illus.) (321)

Immunity in Pneumococcal Infections. By George Grant

Macdonald, M.A., M.B., CM. (1903); M.D. (Illustrated) (365)

Preliminary Note on the Bacteriology of some Diseases of Sheep.
By James Milner Adams, M.A., M.B., CM. (1905); and
Bertie Ronald Gordon Russell, M.B., CM. (1904) ... (401)

PROFESSOR HAMILTON.

OVER twenty years ago — I remember it as yesterday — a
wave of personal excitement passed over the Aberdeen
Medical School. Nothing much had happened ; it was only
that the Students of the Fourth Year (in those days the last of
the curriculum) had petitioned the University Court against
including Pathology in the Final Professional Examination.
Pathology, as a subject by itself, was then new ; Professor
Hamilton was then the new Professor. The students of
Aberdeen, no more then than now, desired to shirk work ; but
it was the end of a long course not yet re-adapted, and many
had begun it before a Pathology Chair was established. The
dispute was very keen ; it even affected old personal friendships ;
but the Court — it was in Professor Bain's first Rectorship —
went with the students, and against the new Professor. Who
raised the question, I never knew ; but it made all the world
talk of Pathology, and of its Professor.

Does any student in these luxuriant days know what went to
the making of the splendid department he now works in ?
Will he look back with me and picture the hard beginnings of
the New Prophet, the days in the wilderness, the triumphant
personal energy that, day in, day out, unfaltering, unfailing,
planned, and worked, and developed, and taught, until, like the
pointed arch of a Cathedral, the new temple of a new science
was reared, a fit sanctuary for our worship ? It is not easy to

xiv. QUATERCENTENARY STUDIES IN PATHOLOGY

go in where a man's mind dwells, and live with him again the
hopes, the disappointments, the despairs, the agonies that went
to the creation of the vast fabric of Pathology as we now know
it ; but that is what every student of Professor Hamilton's would
wish to do. The personal force that created his department
created also the mind to work in it. This little group of studies
is but a selection to mark the end of twenty-five years of
working, teaching, stimulating, investigating, developing.
There are worlds yet to conquer ; but Professor Hamilton goes
always forward. As twenty-five years ago, so now, he fights in
the front rank ; but he fights with the growing energies of all
those years to aid him, with hundreds of old pupils living in his
memory, with ever new generations of workers waiting for his
voice, with opportunity, sympathy, a whole history of success.

Do you know or do you remember what Professor Hamilton
was as a Teacher ? I see him now come in — absorbed, active,
energetic. He has undone his portfolio. He has opened his
manuscript. He has glanced round to see that all his specimens
are in their places. He looks up ; he looks down ; he puts
his fingers into his vest pockets. He looks to the window
on the right ; he begins his morning parade. Then it is a
drum-tap to attention. His fine baritone — trenchant, decisive,
agreeable — gives the topic of the day. It was his method to
select a definite idea, or part of an idea, to make it very precise
in a condensed statement, to dictate this to his class until
we all had it. Then he would analyze, illustrate, explain.
He would show us this diagram, that specimen, this section,
that section, until, when the hour was done, we had seen and
understood and went away with a vivid impression vibrating
in our memory. At the end, we crowded up to verify his
points. He kept us close to actual specimens. We handled
them, and he taught us how. We saw for ourselves. I have
had twenty-six Professors, and among them many superb

PROFESSOR HAMILTON XV.

teachers ; where so many were great, It is difficult to characterise,
for each had his individual way, each his strength ; but
Hamilton was among the most vivid, the most effective, the
most stimulating.

And in his Practical Classes it was not different. Every
leading point of his lectures and many more, he made real to
us by naked-eye specimen, microscopic section, bacteriological
culture. The pictures there seen for the first time are with
me still — clear, defined, easily placed in the memory. Is there
a better proof of good teaching ?

And in the Mortuary, the Teacher still predominated. He
made the post-mortem room a new laboratory for the school.
I knew the post-mortem room before Hamilton. How little
we learned ! I knew it after Hamilton came. How little
we missed ! He did not spare us ; neither did he spare himself.
He drummed knowledge into us ; he commanded us not only to
look, but to see ; he taught us method by making us act ;
he taught us accuracy by making us weigh, measure and record ;
he made us feel that science looks on nothing as unclean or
common.

I remember, too, when the Text-book of Pathology was
being forged. Every hour snatched from a busy day went to
the searching of monographs, the verifying of references, the
checking of theories. I am not a specialist in Pathology, and
cannot tell how the book compares with the other great text-
books ; but I should be surprised if it does not equal the best
in lucidity and wealth of factual basis.

Of Professor Hamilton's work on the brain, his particular
method of sectioning, his theory of the corpus callosum, and
how many more points, I need say nothing. They were the
furniture of our minds as students, and remain so. In the
meanwhile he has worked along other lines. His splendid
report, just issued, on Louping-ill and Braxy is one of the best

xvi. QUATERCENTENARY STUDIES IN PATHOLOGY

researches of recent Pathology. It is the result of many years
of stringently scientific labour. It takes its place as one of the
most solid achievements of the School.

But hitherto, I have been thinking of the strenuous,
commanding, irresistible, earnest teacher of Pathology. It is
all true ; but it is not all the man. He is, too, an artist in
music ; he studies pictures and cathedrals ; he sings, that in
listening to him you forget the world ; he lectures on Gothic,
that you imagine here is his real love. And now I realise how
much that voice stands for. If it was our trumpet-call on the
field, it was, too, our sweet song in the moonlight, by the camp
fires. Of all my twenty-six Professors, I hear the voices
speaking still, and I know how little I have learned that did
not come to me in those winged words. Some speak from the
dead. But some are of the living. And we are going up for
the Great Festival, where we shall hear them all again !

W. L. M.

August, igo6.

The History and Progress of the Sir Erasmus Wilson

Chair of Pathology from the time of its

foundation to the present date*

FOR several years previous to the foundation of the Chair,
it was felt that one of the main drawbacks to the
reputation of the Aberdeen curriculum was the want of
systematized teaching in Pathology. Many of the members of
the Medical Faculty, and to their praise be it recorded, several
of the oldest members, had long recognised this defect. The
perfunctory manner in which the subject had been tagged on to
the courses of Practice of Physic and Surgery, was, to say
the least of it, unsatisfactory.

Doubtless a certain amount of instruction in morbid anatomy
was to be had from the pathologist to the Royal Infirmary,
but when one calls to remembrance the disadvantageous
circumstances under which the casual demonstrations were
conducted, the wonder is that they were attended even in the
somewhat desultory manner which marked their career.

Germany and France were far in advance ot Great Britain
in this respect. They possessed, and had possessed for years.
Chairs of Pathology not only in their large centres of medical
education, but in every small University town, and were
attracting all our best young graduates bent on acquiring the
knowledge which was denied them in the schools of our

XX. QUATERCENTENARY STUDIES IN PATHOLOGY

own country. Bacteriology had commenced to blossom forth as
a new department of the subject — the year of foundation of the
Aberdeen Chair will ever be memorable as that of the discovery
of the tubercle bacillus. Physiology, moreover, had advanced
to such an extent that an experimental pathologist such as
Cohnheim or Strieker had become a possibility.

We were no worse off, however, than the Universities of
Glasgow and St. Andrews. Edinburgh at that time was the
only Scotch University, indeed the only University in the United
Kingdom, which possessed a Chair of Pathology, and even the
efficiency of this Chair had been for long stultified by its being
occupied by one who, in theory as well as in practice, was a
declared homeopath.

Opinion was thus ripe for the establishment of a Chair of
Pathology in our midst, the only thing wanting being the money
necessary for a foundation, and this was procured through the
liberafity of Sir Erasmus Wilson.

Erasmus Wilson was the son of Dr. Wilson, a native of the
parish of Gartly, in Aberdeenshire. His father was in the
King's navy, and was educated at Aberdeen University.
While, however, his son Erasmus was still in his boyhood, he
removed to London, in order to be nearer his ship, and Erasmus
accordingly owed his medical education to the southern
metropolis, having become a pupil of St. Bartholomews
Hospital in the year 1825, under Mr. Abernethy, with whom
he soon became a special favourite.

It may be conjectured, however, that, like all true Scotsmen,
Sir Erasmus had a warm side for his birthplace and scene of his
early youth. Amidst the distractions of a large practice, and
after the greater part of a life spent in London, he did not
forget Aberdeenshire, and in a princely manner established
the present Chair of Pathology by the gift of ^10,000 to
the University.

SIR ERASMUS WILSON CHAIR OF PATHOLOGY xxi.

There were few in the medical profession of his day
whose attainments presented such a varied aspect, and whose
career was marked by such unusual success. Commencing life
as an anatomist, Sir Erasmus did much, by writing and lecturing,
to advance its claims, and to make its teaching easy. The
Dissector s Manual, the Vade Mecum, and Anatomical Plates
are evidence enough of the energy he developed in the study
of the subject. In course of time, however, the founder of
our Chair of Pathology renounced his anatomical pursuits to
engage in the specialty — that of Dermatology — with which his
name, medically, is chiefly associated, and in which he gained
for himself a world-wide reputation. The works which he
published on this subject, such as his Diseases of the Skin,
his On the Management of the Skin as a means of promoting
and preserving Health, his Atlas of Portraits of Diseases of
the Skin, and many others, are well worthy to be, as they
are, classical models for all time. In the year 1869, he founded
the Chair and established the Museum of Dermatology in the
London College of Surgeons, and was appointed its first
Professor.

He was admitted a member of the Royal College of
Surgeons of England in the year 1831, of which College he
subsequently became an Honorary Fellow and ultimately
President. In the year 1844, he was elected a Fellow of the
Royal Society, and in that of 1881, the University of Aberdeen
conferred on him its honorary degree of Doctor of Laws.

One of the subjects by which Sir Erasmus Wilson was best
known, popularly, was that of Egyptian Archaeology. We need
hardly remind the reader of the works he published in this
domain, and of the gift he conferred upon the nation by the
transference of one of the Alexandrian obelisks to the banks
of the Thames. He amassed an enormous fortune, partly from
the practice of his profession, partly from judicious investment,

xxil. QUATERCENTENARY STUDIES IN PATHOLOGY

and out of this, besides his endowments in aid of medical
education, he dispensed many gifts of charity, among them
;^30,ooo to build the new wing of the Margate Royal Sea-
Bathing Infirmary.

It is something for a man to leave money in bequests for
educational and charitable purposes, but when it is gifted during
the life of the donor it means a very great deal more, and
as such the act demands correspondingly greater admiration.
There is this, however, always to be remembered, namely,
that gifts made during a lifetime for educational and charitable
purposes, like the quality of mercy, are twice blessed ; they not
only benefit humanity at large, but reflect their reward on
the donor through the pleasure afforded in watching the
progress of the object they were intended to foster. Doubtless,
it was a matter of gratification to the founder of the Chair
to see a new subject safely launched upon the curriculum of
one of our most ancient Universities.

It would be, however, a gross oversight in recording the
history of the foundation of the Chair were one to omit mention
of the part taken in the accomplishment of the deed by the late
Professor Pirrie. He and Sir Erasmus Wilson had been
close friends in their boyhood, and the intimacy and friendship
had continued into years of maturity. No more ardent
advocate for the establishment of a Chair of Pathology was
to be found than the late eminent Professor of Surgery, and
when the question of obtaining the foundation for such was
being acutely discussed in Senatus and elsewhere, the happy
thought occurred to him of making application to the friend
of his youth. In answer to a letter which he addressed to
Sir Erasmus Wilson he got back the reply that he (Sir
Erasmus) never took any serious step in life without consulting
two persons, namely, his lawyer and his wife. He saw that
what was wanted was a matter of ;^ 10,000, and having obtained

SIR ERASMUS WILSON CHAIR OF PATHOLOGY xxiii.

the approval and assent of the above, he had pleasure in
informing the Senatus that he was willing to give this sum to
found the much needed professorship. He further added that
he gave the money ** as an expression of my regard for the
institution in which my father — a native of Aberdeen — received
his medical education, and as a recognition of the honour
which the University has been pleased to confer on me by
granting me the distinguished degree of LL.D."

The first Professor appointed to the Chair was its present
occupant, Professor D. J. Hamilton. After several years
occupied in Hospital practice in Scotland and England he
gained the Sir Astley Cooper triennial prize of ;^300 for an
essay on The Diseases and Injuries of the Spinal Cord, and
thereafter resolved to devote his life to the subject of Pathology,
although at that time there was but little promise of an opening
in this Department of Medicine anywhere in Great Britain.
With this view he proceeded abroad, and had the opportunity of
studying with some of the most renowned pathologists of the
day in Germany and France. He remembers seeing the
venerable Rokitansky, the father of Pathology, sitting day by
day in the post-mortem theatre of the Vienna Allgemeines
Krankenhaus. He worked with Strieker for a winter upon
experimental inflammation, and with Schenck on embryology,
afterwards in Strassburg with von Recklinghausen, Waldeyer,
Hoppe-Seyler and others, and subsequently with Virchow,
Koch, and Pasteur.

He was thus enabled to adopt a system of teaching gleaned
from the best models of the time, and to adjust this to the
requirements of our own country.

On returning from the continent he was appointed Demon-
strator of Pathology in the University of Edinburgh under the
late Professor Sanders. This post he held for seven years
before being advanced to the Chair in Aberdeen. While

XXiv. QUATERCENTENARY STUDIES IN PATHOLOGY

Demonstrator in Edinburgh he was entrusted with the forma-
tion of a Class of Practical Morbid Anatomy and Pathological
Histology. The class was perhaps the only one of the kind at
the time in this country and, although purely voluntary, was
attended by the whole of the students enrolled in the Systematic
Class, and by many of those getting their education in the extra-
mural school. He also succeeded to the office of Pathologist to
the Edinburgh Royal Infirmary, on its becoming vacant shortly
after he became Demonstrator in the University.

On taking up his duties in Aberdeen everything had to be
organised from the commencement. Accommodation had to be
found somewhere for the new classes of Systematic and
Practical Pathology, a matter of no little difficulty in the then
restricted space occupied by Marischal College. Through the
acquiescence and kindness of the Professor of Medical Juris-
prudence, Dr. Francis Ogston, his class-room was converted
into a laboratory, and this for many years constituted practically
the only premises belonging to the Chair. Here the class of
students for years was instructed by relays in Practical Morbid
Anatomy, Histology, and Bacteriology, while the Systematic
Lectures were delivered in whatever class-room for the time
being could admit of them.

Fortunately the newly appointed Professor had brought
with him a sufficiency of material to start working upon, and
with this he had to be content for some time to come. The
Chair, unlike that in Edinburgh, did not possess any connection
with the Infirmary, and hence the teaching in the one could
not be associated with that in the other. In course of time,
however, the pathologist to the Infirmary, Dr. Rodger, resigned
office, and a strong appeal was addressed to the Managers
to have the new Professor installed in his place, an appeal
which was listened to and acted upon by them, so that the
two Departments, that of the University Chair and that of the

SIR ERASMUS WILSON CHAIR OF PATHOLOGY XXV.

post-mortem room in the Infirmary, came to be under the
direction of the one teacher. The advantage of this to the
Students attending the class has been enormous, enabling them,
as it has done, to connect the systematic teaching of the class-
room with the invaluable instruction to be obtained in the post-
mortem theatre. No Chair of Pathology can be completely
efficient without this connection, for, in default of the practical
experience to be had in the study of the effects of disease, as
manifested after death, the teaching is always liable to become
too strictly didactic, and to lapse into the domain of unproven
theory. It has been the invariable endeavour of the present
incumbent of the Chair to broaden the limits of the instruction
afforded, by attacking the subject from many sides, and thus
preventing his students from conceiving too narrow a scope of
its compass. He has striven to connect the subject with
physiology on the one hand and with clinical medicine on the
other, and has strenuously opposed any tendency to specializing
before an all-round comprehension of the possibilities of the
subject had been acquired.

After several years, however, of probationary trial, the
necessity of providing adequate accommodation, both at the
Infirmary and at Marischal College, became apparent to the
authorities concerned. A spacious post-mortem theatre was
added on to the new Infirmary, and when the extension of
Marischal College was taken seriously in hand, the building of
a pathological Department was among one of the first practical
outcomes of the movement.

With the advance of the recent discoveries in Bacteriology,
it became highly desirable that arrangements should be made
for teaching it both systematically and practically to each
student taking the course, and this naturally required the
necessary accommodation and teaching-plant. Provision was
made, accordingly, in the new buildings to meet this requirement.

XXVI. QUATERCENTENARY STUDIES IN PATHOLOGY

It is the opinion of the present occupant of the Chair, an
opinion forced upon him by long experience and voiced by him
on all available occasions, that the teaching of Pathology to be
effective must be essentially practical, that although a certain
amount of systematic instruction is absolutely necessary, yet,
that the tuition which really drives the matter home and renders
the knowledge gained practically indelible is to be acquired only
in the laboratory. It is here that the truths brought before an
audience of students in the class-room ought to be demonstrated
and fixed in their minds, for by no other means is it possible to
convince them that what has been described by the lecturer can
be seen and verified by themselves. The combination of the
two, in such wise that while the systematic description of the
subject is fresh in their memories its truths are driven in upon
their minds by actual demonstration, has always seemed to him
the essence of all good teaching.

In planning the new Department, accordingly, it was his
first care to stipulate for a teaching laboratory on such a scale
as to admit of accommodating the whole class simultaneously,
and in comfort.

Previous to the installation of the new buildings, it had been
customary to give the Systematic Course in winter and the
Practical Course in summer. This arrangement involved a
great deal of overlapping, repetition, and consequent waste of
time and energy. On application to the Senatus and University
Court, however, permission was granted to combine the two
courses of instruction in such a manner that, the systematic
description of any subject having been given in the Lecture
Room, the whole class, immediately afterwards, might be taken
into the laboratory, and there made familiar practically with the
object described. The teacher has thus been relieved from the
monotony of delivering one hundred consecutive lectures, and

SIR ERASMUS WILSON CHAIR OF PATHOLOGY xxvii.

time has proved that the new system, from the students' point
of view, is a decided advance upon the old.

The Department at the present time consists of a capacious
lecture-room, practical class-room (see illustration), a museum,
several bacteriological laboratories and apartments for special
research, a photographic room, workshops, animal accommo-
dation, and abundance of storage. It is quite possible that,
before long, we may have to add on to these, but meanwhile
they are fairly sufficient for the purposes required of them.

Mr. G. M. Duncan, M.B., CM., Senior Assistant and
Lecturer on Bacteriology, conducts a class of advanced
Bacteriology, summer and winter, for medical graduates and
those preparing for the Diploma of Public Health.

Dr. A. R. Laing, the second assistant, presides over the
part of the municipal public health office concerned with the
bacteriological examination of products from cases of contagious
disease. The municipal authorities are in league with a large
number of districts in the north-east of Scotland which are
entitled to the same privileges as the city itself in so far as they
can have brushings from diphtheria, typhoid blood, phthisical
sputum, etc., examined on subscribing a merely nominal fee to
the common fund.

One great object, all along, has been to encourage original
research, and, since acquiring the new premises, such research
work has been rendered possible. We have always a supply
of young men anxious to do original work, and it is in this
direction that it is desirable to extend the usefulness of the
Department.

Even although the Department has been in existence for
quarter of a century, it is yet far from being thoroughly equipped
for either research or teaching purposes. Makeshift appliances
have had to be utilized in the past which are not creditable
to a great University, and which have seriously trammelled

xxviii. QUATERCENTENARY STUDIES IN PATHOLOGY

the efficiency of the instruction capable of being imparted ; and
even at the present day the working plant of the Department is
anything but what it should be. It would require from ;^3000
to ;^4000 to put it on a thoroughly good working basis.

In addition to funds necessary for the complete equipment
of the various laboratories, generous aid is also required
in the endowment of research. There is no lack of young
graduates who would be willing enough to undertake re-
search studies provided we could furnish them with the means
of living for a year or two. Our students are usually not over-
burdened with independent fortunes, and, consequently, as
soon as they have graduated, most of them have to look out
for some employment which will yield them the means of
subsistence. Many of our most gifted young graduates are
thus cut off from the chance of following the higher paths
of medical science, and have to betake themselves to some
obscure assistantship or other expedient for the purpose of
earning a livelihood.

This seems a great reflection upon the efficiency of our
University, and is one which is common to all the Scotch
Universities. They have been content in the past with turning
out graduates, quite oblivious of the fact that the mere imparting
of knowledge to students is by no means the only function
of a great centre of learning, indeed, is perhaps not the most
important element in the work which should emanate from
such. A great part of this teaching might be done quite
efficiently by senior assistants or lecturers, leaving the Professor
more leisure to conduct research-work himself, and personally
supervise that of young graduates similarly engaged.

No doubt a little assistance in this direction has been
obtained from various funds, manifestly from the Carnegie
Grant to the Scotch Universities, but, with it all, the number of
endowments in connection with these is limited, and the

SIR ERASMUS WILSON CHAIR OF PATHOLOGY xxix.

restrictions set upon them are so stringent as to be almost
prohibitive of their being accepted. When one thinks of the
enormous sums of money which have been left for bursaries in
time past, and which still come pouring in on all hands, when
one considers the questionable efficiency of the subsidizing of
students' fees by the immense sum of money comprised in the
Carnegie Grant, several questions naturally assert themselves: —
Has this money been laid out to the best advantage ? is it
subserving the purpose for which it was intended ? and has it
been directed into channels which are likely to yield the highest
return in the future.

So long as Scotland, as in the 17th and i8th centuries, was
a poor country, the good purpose subscribed by leaving money
for bursaries could not be gainsaid. These bursaries in early
times did an infinity of good in encouraging learning. Deserving
merit through them met with its due reward, and was fostered
and encouraged until such time as it blossomed forth in the
cultured man of letters. But it becomes those who have the
welfare of our Universities at heart, as well as those who are
anxious about the education of the country in general, to ask
themselves whether the system has not been overdone. Scot-
land is no longer a poor country ; wealth has accumulated in it,
and prosperity has followed the steps of those of her sons who
have launched out in her various enterprises, indeed, to an
extent which, reading the annals of a period so late as the end
of the eighteenth century, could not have been anticipated by
her most sanguine of prophets.

There is evidence, accordingly, that the subsidizing of the
fees necessary for education at our Scotch Universities is in
great part uncalled for at the present day, and, moreover, there
is a suspicion abroad that it is actually doing harm by destroying
the honourable feeling of independence and self-reliance which

XXX. QUATERCENTENARY STUDIES IN PATHOLOGY

has characterised the Scottish nation in the past, and which has
been one of the mainsprings of her progress.

In face of this, think what might have been achieved with
this money had it been applied to the proper equipment of our
laboratories and the furnishing forth of our libraries. Both are
in an inefficient state, and surely common sense would dictate
that before announcing that we are prepared to turn out
competent scholars, medical men, lawyers, etc., we should be
supplied with the machinery for accomplishing this object.

We would enter an appeal in the strongest terms to
benefactors of our University in the future, if they wish to
confer a real benefit on higher education, to see to it that their
benefaction is no longer cast into the educational slough known
as the " Bursary Fund." Let it be applied to what are really
the weak points in our system, namely, the equipment of our
laboratories, the encouragement of research, and the endow-
ment of our libraries.

W. B.

THE ALIMENTARY CANAL AS A SOURCE OF

CONTAGION.

By David James Hamilton, M.B.,

Professor of Pathology in the University of Aberdeen.

"l

The Alimeniary Canal as a Source of Contagion.

The mucous membrane of the alimentary canal is pre-eminently an
absorbent surface. It is constantly bathed in liquid swarming with
bacteria, and yet only in rare instances apparently do these bacteria
pass through its walls to gain entrance to the blood- and lymph-circula-
tions. The great mass of them, many varieties being highly toxic, is
rejected by the absorbents, and so, under natural circumstances, the
blood remains free from contamination.

The cause of this cannot reside in the presence of a defensive barrier
of phagocytes in the intestinal wall, for no such barrier has ever been
demonstrated, and, moreover, it would be hard to suppose that any such
means of protection would be sufficient to deal with the enormous
numbers of bacteria of different kinds constantly passing along the
intestinal channel.

Certain bacteria, as for example that of typhoid or of tubercle, do
become absorbed, make their way into particular organs, and germinate
upon them, while the numerous other organisms lying in their vicinity
fail to be absorbed, or, if absorbed in small quantity, are destroyed
by the blood or lymph.

How this is effected, and why there should be such a selective
affinity in favour of one organism, is as yet shrouded in obscurity,
although manifestly the function, whatever it may depend upon, must
play an all important role in protecting the body under conditions of
health from bacterial invasion.

Comparative Study of Certain Contagious Diseases
of the Sheep.

The subject, it will be confessed, is one of intense interest, and
my attention has been directed to it by the study of a remarkable class

(3) BI

4 QUATERCENTENARY STUDIES IN PATHOLOGY

of contagious diseases which affect the sheep. These diseases of the
sheep seem to throw a great deal of side-light on the whole problem,
and suggest that, likely enough, many diseases of man, whose pathology
is still obscure, may be of intestinal origin, even where no such
connection has been suspected.

The diseases in question have claimed my attention for many
years past, first in a private capacity, thereafter under the auspices of
the Highland and Agricultural Society of Scotland, and, lastly, since
the year 1901, under the patronage of the Board of Agriculture and
Fisheries. In the year 1901 the Board of Agriculture appointed a
Departmental Committee to inquire into two of the diseases, namely,
those known as Braxy (Morbus subitarius ovis) and Louping-ill
(Chorea paralytica ovis)^ and our Report on these was issued a few
months ago. It gave an account of the work so far as it had gone up
to the date of publication (Report on the two Diseases of the Sheep,
known as Braxy and Louping-ill. Board of Agriculture and Fisheries,
1906).

To this Report I must refer the reader for details, and meanwhile it
is to the relationship existing between these diseases of the sheep and
many obscure diseases of man that I would desire to enlist attention.

The sheep is peculiar in respect of the many contagious diseases to
which it is liable, and it is curious that these heretofore have not claimed
more attention than has been awarded to them. Thus, there is a large
group the members of which are closely related in so far as they are
each caused by a specific organism having certain mutual affinities
and, apparently, of the same type as that of Tetanus. Several
of the group have never up till now been recognised, and those whose
characteristics have claimed attention have been investigated only
in a perfunctory manner. Previous to the work of the Board of Agri-
culture Committee, little was known of most of them which could serve
to explain their pathology and aetiology, or lead to their prevention.

So far as my own observations have demonstrated, the members of
the group are comprised in the following : — Braxy (Morbus subitarius
ovis)^ Louping-ill or Trembling (Chorea paralytica ovis), Malignant
(Edema of the sheep, Blackquarter or Quarter Evil, the disease known
as " Struck," and two diseases which, provisionally, I have named
Disease " A " and Disease " B."

(4)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION ^

Some of them, such as Braxy, appear to be pecuHar to the sheep,
while others, such as Blackquarter, are common to it and to cattle.
Louping-ill, although pre-eminently a disease of the sheep, is said to
affect other animals, such as the calf, the pig, and the goose, but only
on rare occasions.

Each of the diseases in question is caused by an anaerobic bacillus
having a great tendency to spore, and whose natural habitat is the
intestine. Doubtless they can all be inoculated upon the skin, and,
when thus transferred, occasion a " Blackquarter " slough, but it is a
mistake to suppose that even in the case of so-called " Blackquarter "
this is the usual method of invasion. The " Blackquarter " slough is
simply the manifestation of the effects of the action of the virus when
implanted superficially upon a scratch or puncture wound, and can be
called forth by any of them, while the commonest method of intro-
duction, and this holds good of the whole class, is through the stomach
and intestine.

Whether Malignant CEdema of the sheep is the same disease as
that which affects other animals and man, and which goes by the same
name, I am as yet unable to say, but quite possibly, from its occurring
so frequently in the sheep, it is altogether a different malady.

There is another disease of the same class which is peculiar to
the deer, but which as yet does not seem to have been identified or
described.

Their Periodicity. — One remarkable feature of these diseases of the
sheep and the deer is that they occur periodically, that is to say, at
stated times of the year. Certain of them, such as Braxy, Disease
" A," Disease " B," and Malignant CEdema prevail in the autumn and
winter months, while others, and more particularly Louping-ill, are
diseases of the spring ; all of them tend to vanish during the summer.
They show themselves, almost to a day, each in its season, and vanish
quite as regularly and mysteriously.

Areas Affected. — They prevail only in certain districts, and mainly
along the west coast and southern counties of Scotland and the northern
counties of England, while the east coast of the whole of Great Britain
may be said to be almost exempt from their ravages. Draw a
straight line from the north of Scotland down to the south of England,
and you practically separate the infected districts from the non-infected.

(5)

6 QUATERCENTENARY STUDIES IN PATHOLOGY

Braxy is a most destructive disease in Iceland, the Faroes, and the
west coast of Norway ; in fact it may be asserted that wherever the
waters of the Gulf Stream impinge upon the littoral of a country, there
Braxy will be found to prevail. It is quite likely that other diseases of
the group infest such countries as well as Braxy, although nothing is
known of the matter.

Pecuniary Loss. — The pecuniary loss entailed upon sheep-farming
districts afflicted by these diseases, directly and indirectly, is enormous.
It has been calculated that our loss in Great Britain must amount to
something like half-a-million yearly, and this, I fear, is really an under-
statement of the case. So dreadful is the mortality in certain areas of
Scotland that sheep farming as a profitable industry is ceasing to exist.

Chorea paralytica ovis, or Louping-ill.

It would be too large an undertaking to attempt even to outline the
features of each of the diseases I have enumerated, in a communication
of the present scope.

For the purpose I have in view, namely, that of illustrating the
importance of the alimentary canal as a source of contagion in a
number of diseases of man and the lower animals, whose pathology
heretofore has been unexplained, I shall select one, namely, so-called
" Louping-ill," that is to say, the leaping disease, a name applied to it
popularly on account of the very manifest convulsive spasms with which
the animal is affected. The late Principal Williams gave it the scientific
designation of " Chorea paralytica ovis " — a term which seems to be
singularly appropriate.

The disease, as just remarked, is one which prevails chiefly during
the spring months, although sporadic cases occur in the autumn or early
winter. The period extending from the middle of April to the middle
of June may be said to mark the limits of its occurrence in an epidemic
form, the middle of May constituting the zenith of its intensity. The
valley of the North Tyne is one of the most severely smitten areas, and
it was greatly through the interest shown in the matter by the Duke of
Northumberland that the Board of Agriculture Inquiry was under-
taken. The loss from it in the West Highlands of Scotland is tre-
mendous, while all over the southern counties of Scotland its ravages

(6)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 7

are well known. On the east of Scotland, however, from the extreme
north down to the Lothians, the disease is so rare that some of those
engaged in sheep farming hardly know what it is.

Symptomatology. — The symptoms can be divided into three distinct
stages. In the first, the animal, as in so many of these contagious
diseases of the sheep, is noticed to be somewhat dull. It may separate
from the rest of the flock, stand apart in a listless fashion with drooping
head, and be off its feed. It assumes subsequently a reeling gait as if
intoxicated, and will lean against a dyke or fence for support. A dazed
expression is often noticed in this stage as if the animal were in the
initial stage of a fever.

These symptoms may last for a period of from two to three days,
when it falls over quite unable to support itself or to regain, even
temporarily, the upright position. The limbs are now spasmodically
convulsed at intervals of perhaps a minute to a couple of minutes. The
movements are mostly of a galloping character, and so incessant that
the turf becomes worn in the area of their excursion. During the
intervals, quivering or trembling movements are perceptible, hence the
name " Trembling Disease " sometimes applied to it in the Western
Highlands. The neck is drawn back as a rule, but the muscles of
the neck are not intermittently contracted as in the case of the
limbs. The temperature in this, the second stage, may go up to
105° to 108° F., and the pulse-rate and respirations are increased in
number. The muscles of the jaws and those concerned in swallowing
are not usually involved, nor is there any squint, and the intelligence of
the animal appears to be little if at all impaired. It will nibble grass,
and it swallows milk with avidity and without impediment.

There are cases, however, which assume quite a tetanic character. In
these the muscles are in a state of rigid spasm, while, it is said, although
I have never seen an instance of this in the natural disease, the muscles
of mastication are in a like rigid condition, and in most respects the
phenomena resemble those of idiopathic tetanus. Shepherds in Louping-
ill districts will tell you of the occurrence of what they term " lockjaw "
among their sheep during the spring months, and from the examination
of the carcases of those dying with such symptoms, and for other
reasons, I have come to the conclusion that this disease is simply a
severe variety of Louping-ill.

(7)

^ QUATERCENTENARY STUDIES IN PATHOLOGY

The animal may succumb in this second or convulsive stage,
apparently from acute toxic poisoning. It passes into a semi-comatose
state, the temperature sinks, and the convulsive spasms become weaker
and weaker, preliminary to the fatal termination.

It should be mentioned that sometimes the animal suffers from
diarrhoea with the presence of blood in the rectum.

Should the disease not prove fatal within a matter of a week or less,
then there is every likelihood of its passing into the third stage
characterised by the following phenomena : — The convulsive spasms of
the limbs, so notable in the second stage, now give place to a condition
of more or less complete motor paralysis ; the limbs are outstretched
and limp, while, if the sheep be held up by the fleece, they hang down
relaxed and listless, and the animal is quite unable to use them for
purposes of support. When the feet touch the ground the fetlocks are
knuckled under in a perfectly helpless manner.

Painful sensation, so far as one can judge, does not seem to be
affected to any appreciable extent even in this stage, but the reflexes
from the limbs appear to be blunted, and in certain cases annulled.

The whole appearance of the animal closely resembles that of a
person suffering from post-diphtheritic paralysis, the nature of the
paralysis being essentially motor, and affecting the limbs by preference,
although it seems to differ from the diphtheritic form in the fact of the
palatal muscles being spared. During the course of the disease the
animal is able to swallow liquid nourishment without impediment.
The intelligence in this third or paretic stage often remains evidently
uninfluenced. The animal recognises objects about it, and will bleat
when a companion sheep is removed into a neighbouring pen. It will
eat fodder if offered to it, but is quite helpless to seek for such of its own
accord, and, as a consequence, in many instances, seems to die from
starvation as much as from any other cause.

In this highly paretic state it may live for weeks, recovery seldom if
ever taking place, even although the animal may have been fed artifically.
Some subjects of the disease are said to make a partial recovery,
probably only one limb remaining permanently crippled. I have always
been somewhat doubtful, however, of the diagnosis in such cases ; they
look to me more like the effects of spinal abscess, which is very common,
at least among lambs, during the spring months.

(8)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 9

The following case will give an idea of the most striking clinical
features during the first and second stages. The animal, a Cheviot ewe
hogg, was procured from "clean" land in the year 1902, and took the
disease accidentally on being transferred to our observation pens at
Kielder. The first symptoms showed themselves on May 14th. On
that day she was noticed to be a little dull, and was seen standing
isolated on three or four occasions during the day.

On the morning of the 15th she was found to be duller than
yesterday, and was the last sheep to rise this morning when fresh fodder
was taken into the pen. She now stood completely isolated from her
mates, and was not inclined to move with them. She turned round
occasionally, slowly, and with restricted movements. The head was
depressed, the ears failed to be pricked, and the eyelids were half closed.
When standing, the animal's hind legs were somewhat flexed as if she
were weak upon them, and when lying down all the limbs were gathered
up towards the centre of the body. The rump was drooped, the back
arched, and the front of the chest thrown forwards. She had not been
seen to feed or ruminate to-day.

After standing for a few minutes, it was also observed that the
animal became unsteady, and swayed a little from side to side, but
quickly, and, as if conscious of the abnormal condition, she moved a few
steps forwards to take up a fresh position. This, apparently, she did in
order to prevent herself from falling forw^ards, and in the intervals stood
in a drowsy, almost sleeping, attitude, the head gradually falling down-
wards with half-nodding movements, and inclined to the right side. A
little rocking of the body from before backwards was evident, and in
order to avoid pitching forwards she took a step or two. The animal,
indeed, presented very much the appearance of being intoxicated. The
whole body seemed to be in a slight quivering or shaking condition.
The rectal temperature was 106'' F., the pulse 80 per minute, and the
respirations 80 and panting in character.

The animal was visited between 11 and 12 o'clock midnight, and was
found evidently worse than during the day. She was lying against the wall
of the pen, isolated, and did not rise when disturbed, nor did she offer
any resistance when manipulated unless on two occasions, and at these
times the body exhibited quivering or trembling movements. She could
just manage to stand, and only for a short length of time ; the gait was

(9)

10 QUATERCENTENARY STUDIES IN PATHOLOGY

very unsteady ; the free movements of the limbs were restricted, as if
she was in fear of falling ; and the hind legs were occasionally snatched
up spasmodically. When last seen on this night she was lying down
quietly and looking the picture of dejection.

By the following morning (May i6th) the symptoms were more
pronounced even than on the previous night. The sheep was now
(between 8 and 9 o'clock) down on her side, and the most striking feature
of the case was the occurrence of continuous violent convulsive spasms
which persisted up till the time of death. The limbs were flexed and
extended alternately, especially the posterior ones, which lay in a direct
line behind the body. The movements of the fore limbs had a galloping
character and were so continuous and violent that the underlying ground
was worn out in a semi-circle from the friction caused by the excursion
of the hoofs. The muscles were firm and rigid, and the contraction
almost tetanic in character, although the tension was intermittent, not
continuous as in tetanus. The head was thrown violently back and
bent down on the dorsal spine ; it had also an inclination to the right
side. The muscles of the neck were very tense during this action. The
breathing was laboured, the animal gasping for breath during the height
of an attack of convulsions, and, when there was a period of relaxation,
it panted. Frequent and continuous nibbling or snapping movements
of the mouth were noticed, and tufts of grass, or the soil, were occasion-
ally clenched between the incisor teeth and the pad. There did not,
however, appear to be any trismus. A puffing or choking noise was
emitted through the nostrils, and frothy saliva began to accumulate in
increasing bulk round the lips. During expiration the frothy saliva was
projected in front of the animal for several inches to a foot.

The nostrils were widely dilated, but the breathing, although to a
limited degree nasal, took place chiefly through the mouth, and was
accompanied by guttural sounds. The muscles of the face, lips, and
ears were in a tremulous condition, and during a lull between the crises
of spasmodic attacks, the whole body was more or less tremulous. The
eyelids were widely separated, and the pupils, if anything, somewhat
contracted. The conjunctivae were of a brick-red colour from injection
of their blood-vessels, and the corneae had now a dull appearance. The
sphincter ani was contracted, and the perineal region was hot, wet, and
clammy. The thermometer was inserted into the rectum with a good

(10)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION II

deal of difficulty from resistance on the part of the animal. There was
blood in the rectum to such an extent as to soil the thermometer, and
the fingers were covered with blood after withdrawing the instrument.
The tail, for the most part, was quite limp.

The abdomen was fully distended and tympanitic, and during an hour
of more or less continuous spasmodic convulsions the distension visibly
increased.

The external temperature was evidently high, and that within the
rectum, at this time, was 108-5° F., the pulse 160, quick and short, and
the heart could be plainly felt thumping against the thoracic wail.

When the legs were held in the palms of the hands above the hocks,
they were stretched out rigidly, and tense quivering of the muscles was
experienced during the manipulation. On seizing the limbs by the
fetlocks they were snatched out of the hands violently and kicked out in
a galloping excursion. When she was raised from the ground, the hind
limbs were still extended backwards, became rigid, the hoofs were struck
into the ground, and the body was thrown forwards, the animal resting
on her knees and panting like a dog. If the support was withdrawn, the
animal pitched forwards violently one or two yards and rolled over in a
convulsive spasm. Placed upon her haunches, the spasmodic attacks
still continued, the head was drawn back and to the right side, the fore
limbs made the usual excursion in mid-air, and immediately the support
was withdrawn she fell over in a convulsion. During the hour from
ii'20 a.m. to I2'20 p.m., there were short periods of exhaustion and
total collapse.

At I2"20 p.m. the convulsive seizures ceased, with the exception of a
few rapid spasms of the fore limbs and occasional retraction of the head.
The animal gave a long deep gasp or two and succumbed in what
appeared to be a violent, almost tetanic, convulsion.

Morbid Anatomy. — On examining the carcase in this disease one of
the most notable features is the absence of lesion which might serve to
localise the peccant agent in any particular organ. Gas begins to
develop in the abdomen very soon after death, and within a few hours, in
certain instances, the wall of the abdomen may assume a greenish tint.

The abdominal cavity, as a rule, contains an excess of serous liquid,
but this is not always the case. Sometimes the liquid is thick, muddy-
looking, and, it may be, tinged with blood, while at other times it is quite

(II)

t2 QUATERCENTENARY STUDIES IN PATHOLOGY

clear and limpid, or, at the most, a delicate coagulum separates from it.
In no case have I seen peritonitis or pleurisy accompany the disease,
and hence the' conclusion seems inevitable that the organism which
causes it is not possessed of inflammatory tendencies.

A few punctiform haemorrhages may be met with along the course of
the intestine, but with this exception all the viscera may seem to be
quite healthy. Nor have I seen any evidence of meningitis or other
disease of the central nervous system.

The microscopic examination of the natural liquids and of the organs
of the body proves equally disappointing. Thus the blood is free from any
micro-organism which can be detected microscopically, and when
cultivated aerobically or anaerobically remains equally barren. The
cerebro-spinal liquid and nerve centres are devoid of any parasite which
might be taxed with a causal relationship to the disease, and, for these
reasons, the pathology of the disease for long remained to us a problem
fraught with obscurity.

During the first season (1902), in which we conducted our observa-
tions at Kielder, in Northumberland, we noticed, however, that two
kinds of the case were brought in to us. In the one there was an excess
of peritoneal liquid which was also turbid and sometimes slightly stained
with blood. In the other the peritoneal liquid was perhaps not in
excess, or, if so, it was quite clear and limpid.

On microscopic examination of the turbid liquid it was found to be
teeming with a large coarse-looking rod-organism having a great tendency
to spore, while in the case of that which was clear and limpid not a
bacillus was to be detected. The rod in question had a close resemblance
to that of Blackquarter, and, at first, we supposed that we had to do with
two diseases running side by side, namely, Blackquarter and true
Louping-ill.

On incubating the clear peritoneal liquid, however, in sealed tubes,
I found invariably that, in the space of twenty-four hours, it became
turbid, and when the tubes were opened, a whiff of gas escaped with a
small explosion. On examination of the liquid microscopically, it was
now found to be swarming with the same large sporing rod present
in the liquid which was turbid.

This threw a new light on the whole pathology of the disease.
There were evidently two forms, one in which the peritoneal liquid was

(12)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 1 3

SO full of a notable micro-organism that it could be readily detected by
microscopic examination, the other in which the organism was so
sparsely distributed in the liquid that it could not at first be detected
microscopically, but in which the same specific rod developed abundantly
on the liquid being incubated at a body temperature.

The turbid liquid was found in those animals which died within a few
days, and with acute toxic symptoms, the clear liquid in those which
lived longer and in which the disease went through all its three stages.

When the animal was slaughtered during the height of the malady
the peritoneal liquid was always of the clear variety, the turbid variety
was found only where the animal died a natural death.

Examination of the Peritoneal Liquid. — And here, perhaps, I may
be allowed to emphasise the necessity for examining the peritoneal
liquid in all diseases of a contagious nature. Casual reference is made
to its examination in the literature on such affections, but it seems
to me that the importance of examining it as a routine procedure
has not been sufficiently recognised. The morbific agent in all these
diseases of the sheep is invariably present in far greater quantity, and
freer from contamination, within the peritoneum than in any other
part of the body. The liquid can be drawn up in a pipette and sealed
off with ease, and, so long as the organism contained in it is sporing,
may be retained in an active condition for a matter of years. I have
in my possession peritoneal liquid which is three to four years old,
and which is just as potent to reproduce the particular malady when
introduced subcutaneously, as it was when removed from the original
host.

Description of the Organism. — The organism {Bacillus chorece
paralyticce, Hamilton) possesses the following characteristics : — It is a
large coarse-looking rod, sometimes elongating into a thread, or, it may
be, a chain of rods. The actual measurements, as taken directly from
the organism in the peritoneal liquid, I have found to be : — In one case,
when not sporing, 5*6 x r4yu, 7'0 x i'4/x ; and, when sporing, 4*2 x r4/x.
In another case, when not sporing 2'8/x, 4"2/>t, S'^ijl, 7fx, 9*8ya, and
ir2/x X i"05/x to r4/i. The spores in the latter case measured vo^fx to
r4/x in length. Its dimensions, therefore, like all the members of the
group, vary considerably, chiefly accounted for by the fact that involution
forms are almost always present. The ends are rounded, and it is

(13)

14 QUATERCENTENARY STUDIES IN PATHOLOGY

possessed of feeble motility. It has a considerable tendency to spore ;
the spore is located at its centre or at one end ; and occasionally,
especially after incubation in its native liquid, it assumes a drum-stick
configuration, indistinguishable from that of the Bacillus Tetani. Most
of the usual aniline dyes stain it readily, and the colour is not discharged
by Gram's process. These staining reactions hold good of the organism,
both when taken directly from the carcase and when in culture.

It is a strict anaerobe, and grows on various media but most
characteristically on alkaline glucose-beef-tea and glucose-gelatine, each
covered with olive oil. The glucose-beef-tea becomes turbid after four
to five hours' incubation at 38° C, and continues so for days while
incubation is proceeding. If removed from the incubator after, say, four
days' growth, the culture begins to settle down slowly at the bottom of
the tube in a fine precipitate of greyish colour. It does not tend to
agglomerate in a granular form, as in the case of Braxy, nor to become
attached to the sides of the tube. During the process of germination
much gas is evolved which possesses a distinctly putrefactive odour.

Examined microscopically, the culture is found to be composed of
thick stout rods, somewhat longer possibly than in the original peritoneal
liquid, slightly motile when first removed from the incubator, losing this
characteristic later on. Their actual measurements were found to be: —
4-2 X r4/x, 5*6 X i/x, 7*0 x i*4/x, and 14 x r4/x.

The ends are again rounded, but usually the growth is free from
spores. Even when the medium has been strongly alkaline to begin
with, the Braxy organism will render it acid after a few hours' growth.
The Louping-ill bacillus acts in a like fashion, but slower, and this
probably accounts for germination being more protracted in the latter
than in the former case.

Surface cultures on glucose agar grow luxuriantly ; along the central
streak formed by the inoculating wire, and from each side of this, some-
what arborescent processes extend outwards, rendering the sides of the
central streak very irregular. Such cultures are not particularly diagnostic.

The stab-culture on glucose-gelatine, however, grown at 21° C, is very
characteristic. In order to observe it, the gelatine should be covered with
olive oil, and the inoculation made with the platinum wire through this.
Quite a week will elapse before the growth reaches its best.

From the surface there passes downwards a grey-coloured streak for a

(14)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 1 5

distance of a centimetre or so, but underneath this the culture becomes
more expanded and often flattened out in a single lamella whose borders
are occupied with a series of loop-like festoons. It liquifies the gelatine
in course of time, but slowly, and when liquefaction around the culture is
complete, the organism falls down in a greyish-coloured deposit. Gas-
bells may or may not be liberated ; sometimes there is a single gas-bell
at the deepest part.

Like the other members of the group, the culture in glucose-beef-tea
bleaches dilute solutions of permanganate of potash.

The involution forms met with, as in the case of all the members
of the group, are, at first, misleading. So varied may the general
aspect of the organism be that it might be imagined that two or
even three different bacteria were under observation, a fallacy which
must be guarded against in arriving at a diagnosis of the case. The
deception is particularly in evidence when Gram's process is applied to a
culture on glucose-beef- tea. The organism invariably gives a positive
reaction, but should such a culture be retained for a week or so at the
ordinary temperature of a sitting-room, it will be found that the Gram's
reaction is irregular, certain of the members giving an intense blue
colour, while others remain only faintly discernible, or have a distinct
pink tint. The pink tint is sometimes very remarkable, and forms a
notable contrast to the intense blue of other bacilli lying in the neigh-
bourhood. I have found this anomalous and irregular staining
reaction in several members of this group of anaerobes, and do not con-
sider it as peculiar to any one in particular ; not only does the colour
diflfer, but those which have the pinkish tint appear very much more
slender than those which give the ordinary deep blue colouration, so that,
in an unguarded moment, one is apt to suppose that they represent
two distinct bacilli. On examining the culture unstained, however, or
stained with ordinary aniline dyes, the individuals will be found to be
the same organism. The anomaly is evidently due to an optical effect
consequent upon the one being intensely coloured, the other only
faintly so.

Gram's process, as the literature on the subject proves in the case of
Blackquarter, is extremely illusory when applied to these anaerobes, and
requires to be practised with the greatest care. It will sometimes happen
that the organism, taken from the peritoneal liquid directly, will give an

(15)

l6 QUATERCENTENARY STUDIES IN PATHOLOGY

undoubted reaction, while, in culture, the same organism is completely
decolorized. This I do not regard as a true reaction in any respect.
If the organism be one which gives a true reaction, this should be forth-
coming, both when it is growing in the native liquid, and when it has
been cultivated on an artificial medium.

Inoculability. — When a small quantity, say half a cubic centimetre or
less, of the peritoneal liquid containing spores is injected subcutaneously,
and more surely if a few cubic centimetres of dilute acetic acid be injected
side by side with it, the animal dies usually in from twenty to thirty-six
hours afterwards. Some sheep are more resistant than others, but the
experiment seldom fails to bring about a fatal issue, provided the animal
be a one-year-old, and the experiment made during the months of
prevalence of the disease.

I inject the liquid on the inner aspect of the thigh, as here the skin
is free from wool and is cleaner than elsewhere. In a few hours after
performing the operation the animal exhibits signs of distress. It has a
dull expression, remains apart from its mates, ceases to feed or
ruminate, and the inoculated limb is lame and has begun to swell.
By a matter of from sixteen to twenty hours, these phenomena have
become aggravated, the sheep has probably fallen over in a helpless
state, may be delirious, and finally passes into a comatose condition for
an hour or so before death.

At the autopsy the injected limb, sometimes indeed both limbs, and
the front of the abdomen, are found intensely oedematous, so much so
that occasionally the skin has ruptured and the liquid is escaping. The
muscles for some distance round the point of inoculation are infiltrated
with blood, and the areolar tissue is blown up with gas. The peritoneal
liquid, as a rule, is abundant, opaque, and blood-stained, and, on micro-
scopic examination, is seen to be literally swarming with the characteristic
rod, which is sporing on all hands. The oedematous liquid of the
inoculated limb also contains it, but not so abundantly as the liquid
of the peritoneal cavity. It is usually absent from the heart's blood.

The organism is thus one of extreme virulence for the sheep, causing
the death of the animal in a few hours after its introduction subcu-
taneously. It can also be inoculated upon the guinea-pig or rabbit,
although apparently they are naturally immune to the disease.

Portal of Entrance. — The question of how, in the natural disease,

(.6)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 1 7

the organism gains entrance to the body of the sheep and to the
peritoneal cavity, next engaged my attention. A common notion is
prevalent among sheep-farmers that the tick has something to do with
its transmission, but the theory is not supported by any well-
grounded evidence founded on reliable observation. The tick, after
hibernating upon rough dry scrub and rank grass, becomes parasitical
upon the sheep during the month of April, and hence probably the
origin of the supposed aetiological relationship. Suffice it to say,
however, that, having sifted the matter very thoroughly, our Committee
did not acquire evidence corroboratory of the allegation. Even
although the organism may be present in the body of the tick, or,
what is much more likely, merely adherent to its surface, the data
at our command failed to prove that it is ejected from the rostrum
into the skin of the sheep, and that the disease spreads from
the tick-puncture as a centre. Many of the carcases had not a single
tick upon them, nor might there be evidence of a tick-bite. We never
made out that local oedema, gas-formation, or haemorrhage developed in
the vicinity of the tick-bite, and in the case of an organism exciting
such extreme local reaction as the bacillus of Louping-ill does, this
seems strong a priori evidence against the theory of the transmission of
the disease through the agency of the tick or any other animal parasite.

Nevertheless, if sheep taken from uninfected land be put upon that
which is infected during the Louping-ill season, probably from a half to
two-thirds, or it may be the whole of them, will be dead in from three
weeks to a month afterwards.

The first case that really opened my eyes to the point of entrance of
the organism was that of a lamb admitted to the Observation Station at
Kielder during our second season, and in the year 1903.

A Cheviot lamb, perhaps about one month old, arrived at the Station
at Kielder on May 14th, 1903.

The animal could walk, but in an uncertain swaying manner. The
posterior limbs and hind part of the trunk gave way occasionally, and
the animal fell on its side. The fore limbs and head maintained their
natural posture, and even the hind quarters could in a manner be raised
by an effort on the part of the animal, but soon again gave way. When
leaning against a fence it would remain erect for a considerable time,
seemed to be perfectly conscious so far as could be judged, did not

(17) C

1 8 QUATERCENTENARY STUDIES IN PATHOLOGY

appear to suffer pain, and nibbled grass freely. When lying, the limbs
were occasionally twitched backwards and forwards.

It remained in very much the same condition until the evening of
May 24th, and was found dead on the following morning.

The examination was made probably from seven to eight hours after
death. There was a complete absence of ticks, keds, or tick-bites, and
the skin was perfectly sound and unbroken. The peritoneal cavity con-
tained about I oz. of clear liquid, from which there separated a little
flocculent albuminous precipitate. The liquid, however, did not coagulate
universally on being retained in a sealed pipette.

On microscopic examination, a (ew colourless corpuscles were found
within it containing very large granules, but nothing in the way of any
micro-organism could be detected.

The carcase, otherwise, did not show any abnormality.

Anaerobic cultures were made from the contents of the small intestine.
The tubes and their contents were heated up to 90° C. for a quarter of an
hour, cooled down, and thereafter incubated. The cultures were started
on May 25th, and by the following day were in a state of active germina-
tion, with tremendous evolution of gas. The medium employed was
glucose-beef-tea. The organism which grew seemed to be quite pure,
and consisted of a rod identical with that found in the peritoneal liquid
in other cases, relatively thick, and free from spores. It did not appear
to be motile.

The method of obtaining the material from the intestine for cultiva-
tion purposes was as follows : — The pieces of bowel, the small intestine,
after being ligatured above and below, were cut off and placed in sterile
Petri's dishes. The surface of the bowel was singed with a red-hot
spatula. An opening was thereafter made with red-hot scissors and a
platinum loop introduced right into the channel of the bowel. A little
of the intestinal contents and mucus from the wall were withdrawn and
mixed with the hot glucose-beef-tea. The wire was never allowed to
touch the wound in the intestinal wall ; the material was taken entirely
from the interior.

After twenty-six hours' growth 4 c.c. of this culture from the intestine
were inoculated subcutaneously into the left thigh of a Cheviot from
"clean" land, along with 5 c.c. of a i to 10 glacial acetic acid solution in
water. The inoculation was made on May 26th at 7*45 p.m.

(18)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 19

By 8 a.m. on the 28th, the animal was noticed to be ill. It stood
alone in the pen, pawed with its fore foot, and had, at times, an excited
look. By 12 o'clock noon of the same day it was much worse. It lay
down on its side and uttered a moaning sound as if delirious ; its
breathing was heavy. From time to time it arose, again to assume the
recumbent posture within a minute or two. When lying down, the head
was thrown straight out spasmodically and a little back.

By 4 p.m., or just about forty-four hours after receiving the injection,
the animal died.

The autopsy was performed immediately afterwards. There was a
complete absence of ticks. Both hind legs were intensely swollen from
the presence of cedematous liquid, and, curiously, the uninoculated limb
seemed to be more cedematous than that which had been the subject of
experiment. The oedema also extended over the whole of the front of
the abdomen. The liquid was serous in character, blood-stained, and
was permeated with a bacillus identical with that introduced, slightly
motile, and apparently not sporing.

The peritoneal sac was almost dry. The few drops of liquid removed
were clear, and contained a good many lymph corpuscles, and a few
bacilli.

The organs seemed to be unaltered.

A pipette was filled with the clear peritoneal liquid from the original
lamb, sealed off, and incubated at 37° C. After twenty-four hours at
this temperature, the contents of the tube had become thick and muddy.
This result was due to a splendid sporing rod having developed within it,
alike with that obtained from other animals dying of Louping-ill. The
spore was oval-shaped, was placed usually at one end of the rod, and was
of a faintly yellowish-brown colour. Many spores were also seen lying
free. Certain of the rods did not contain a spore, and these were usually
dividing.

About 2 c.c. of the incubated peritoneal liquid were mixed with about
10 c.c. of sterile distilled water, and 5 c.c. of the mixture were injected
subcutaneously into the left thigh of a Cheviot hogg, side by side with
5 c.c. of a I to 10 solution of glacial acetic acid in water.

The experiment was made on May 26th, 1903, at 5 p.m., and by the
following morning at 8 a.m. the animal was found to be very ill. It was
lying down in the pen with the legs drawn under the body, and with a

(19) C I

20 QUATERCENTENARY STUDIES IN PATHOLOGY

tendency to somnolence. It did not rise when approached, and the
breathing was rather hurried. At 5 p.m., exactly twenty-four hours after
the commencement of the experiment, the animal died.

The autopsy was made immediately after death. The inoculated le^,
as well as that on the opposite side, and the whole of the front of the
abdomen, were intensely infiltrated with slightly opaque and somewhat
blood-stained oedematous liquid. The oedema was extreme, perhaps
greater than in any case previously observed.

Examined microscopically, this liquid was found to be swarming with
a bacillus alike with that injected, dividing actively, but free from spores.
It varied in length, some of the rods being short, others fairly elongated,
and certain of them were distinctly motile.

This case, certainly, gave me an insight into the aetiology of
the disease which previously was wanting. The animal was under
our own observation when alive, the symptoms were typical of the
disease, it died a natural death after about twelve hours' illness.
Ticks and tick-bites were absent, and the carcase seemed to be
free from disease so far as could be judged by mere naked-eye
observation. The peritoneal liquid was in small quantity and was
clear ; it did not contain any organism detectable by microscopic
examination. Anaerobic cultures of the intestinal contents developed
pure growths of the Louping-ill organism, which, on being inoculated
subcutaneously on a sheep, killed it within forty-four hours, its body
permeateid with the rod inoculated. The clear peritoneal liquid, after
being incubated, became turbid, developed the Louping-ill organism, and
this, on being injected subcutaneously into another sheep, killed it,
in this case within twenty-four hours, the liquid of its tissues again
swarming with the same organism.

Here is another observation having the same bearings : —

A four-year-old Cheviot ewe was admitted living to the Station at
Kielder on May 29th, 1903. She had been ill for a considerable period,
and had been lying helpless on the farm for a week. The symptoms
were said to have been typical of Louping-ill, and by the time she
reached us a state of almost total collapse had ensued.

The animal was found dead on the following morning (May 30th),
and the examination of the carcase was made within a few hours of her
demise. The abdominal wall in front had a faintly greenish tint. About

(20)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 21

4 ozs. of liquid were found in the peritoneal sac. It was only slightly
turbid, the turbidity being in great measure due to a flocculent precipitate,
and was not blood-stained. On microscopic examination, however, most
characteristic Louping-ill rods were found in it abundantly, some sporing,
others elongating into threads.

On examining the contents of the small intestine, and more particu-
larly the secretion on the surface of the mucosa, the same bacillus as
that in the peritoneal liquid was found in tremendous abundance. It
seemed, in fact, along with a little mucus and shed epithelium, to con-
stitute the entire moist discharge lying on the mucous membrane.

The remaining viscera were somewhat congested, but otherwise not
abnormal.

On May 30th, anaerobic cultures from the mucus of the bowel were
made on glucose- beef- tea heated to 80" C. for twenty minutes, and were
subsequently incubated. They grew luxuriantly, and by June 6th the
culture had settled down at the bottom of the test-tube in the form of a
dull grey-coloured deposit Examined microscopically, the deposit was
seen to consist of a rod identical with that of Louping-ill. It was
comparatively thick, varied somewhat in length, and, at the time of
examination at least, was not motile and did not appear to be sporing.

On June 3rd, 1903, 5 c.c. of this culture, which had been allowed to
grow for forty-eight hours, were injected subcutaneously into the left
thigh of a Cheviot hogg from " clean " land, side by side with 5 c.c. of a
I to 10 solution of glacial acetic acid.

On the following day, the inoculated limb was considerably swollen ;
the animal was quite lame, and continued so up to June 6th, when it was
slaughtered by bleeding. The object in doing so was, among other
things, to ascertain whether the pure bacillus, without spores, introduced
subcutaneously increases in virulence by transference from one sheep to
another. The animal thus slaughtered had lived for three days after
inoculation, and although very ill might have recovered.

The inoculated leg was found to be considerably swollen from sub-
cutaneous oedema, and a minute and quite superficial abscess had formed
at the point of inoculation. The whole of the inner aspect of the
inoculated thigh and the front of the abdomen, together with the lower
part of the inoculated leg, were very cedematous and blood-stained.
The effused liquid contained the same bacillus as that injected, that from

(21)

22 QUATERCENTENARY STUDIES IN PATHOLOGY

the front of the abdomen possessing more of it than that from the leg.
In the former situation, also, the rods were elongating, and seemed to be
commencing to spore. There was about i oz. of peritoneal liquid, clear
and limpid.

On June 6th, 1903, at 9 p.m., i c.c. of the liquid from the front of the
abdomen was mixed with 8 c.c. sterile distilled water, and 5 c.c. of the
mixture were injected subcutaneously into the left thigh of a Cheviot
hogg from ''clean" land, side by side with 5 c.c. of a i to 10 solution of
glacial acetic acid.

June 7th, 8 a.m. — The animal was found to be lame in the inoculated
leg, and had a somewhat collapsed appearance. By 10 a.m. it was much
worse, and was leaning against the wall of the pen in a very depressed
condition. Seen again at 11*30 a.m., it occupied exactly the same
position, never having moved apparently since the last observation. At
9 p.m. the animal was prostrate and evidently dying. It was found dead
on the following morning.

The appearances of the carcase were identical with those met with
in similar experiments, only the cedematous effusion was very much
greater than usual. It prevailed not only in the inoculated limb, ran
out of it when the skin was incised, but also spread upwards all over the
left side of the abdomen. The liquid was blood-stained and turbid, and
contained the bacillus in great quantity. Some of the rods were sporing,
others dividing. The cedematous subcutaneous tissue also showed some
haemorrhages and was blood-stained.

Briefly summarized, this case is to the effect that a four-year-old ewe
was admitted to our Station af Kielder alive and suffering from Louping-
ill. The carcase did not present any visible morbid feature, but the inner
aspect of the intestinal wall, the mucous surface, was literally paved with
the Louping-ill bacillus. This, cultivated anaerobically, grew with
vigour, the growth consisting of Louping-ill rods, and when injected
subcutaneously into a fresh sheep, caused great local reaction, oedema,
and lameness of the inoculated limb. The animal was killed three days
after being inoculated, and some of the cedematous liquid from the front
of the abdomen was inoculated into a second sheep, killing it within
something like twenty-four hours, the cedematous liquid again permeated
with the specific rod.

These are samples of observations I have verified over and over again.

(22)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 23

They require little comment ; they speak for themselves. The carcase
in each case was free from evidence of disease which could be detected
with unaided vision, yet, in both, the contents of the intestine were
swarming with a specific bacillus which proved to be anaerobic, and
which when inoculated into a fresh sheep rapidly killed it with evidence
of intense local reaction. The disease germ, moreover, could be passed
from host to host, evidently gaining in virulence by its transit.

The intestine the portal of entrance in the case of all the members of the
group. — Not only in the case of Louping-ill, but in that of all the other
members of the group the intestine seems to be the portal through
which the organism gains entrance, and the fact that the peritoneal
cavity contains it, more than any other cavity or any organ in the
body, is thus readily enough explained. The peritoneal cavity is
evidently the great lymph-sac of the body. Not only do its walls
contain the lymph-vessels in connection with the intestine, but evidently
those lymph-vessels returning from the hind limbs have also a free,
although perhaps circuitous, connection with its interior. The organism
of Louping-ill fails to propagate on the blood, but grows freely enough
on the secretion of the peritoneal membrane, hence the large quantity
of bacillus usually found in the liquid.

When the organism of the disease is inoculated subcutaneously in
the sheep, death takes place so suddenly, evidently from acute toxic
poisoning, that time and opportunity are not afforded for the develop-
ment of the nervous phenomena. Where the organism, on the contrary,
is introduced into the alimentary canal, and where the animal takes the
disease but lives over, it may be, several weeks, the nervous symptoms
are well developed.

Production of immunity to the disease, — It would be apart from the
object of this communication to enter into the explanation of the
production of immunity to the disease too minutely, and the reader is
again referred to the Board of Agriculture Report (loc. cit.), already
mentioned, for a full statement of the case.

The bacillus of Louping-ill can be administered to sheep by the
mouth with impunity throughout the greater part of the year. As the
susceptible months are approached, however, namely, March, April, and
May, the danger of doing so is extreme, and a fair proportion of the
animals so treated will die with all the classical symptoms. At other

(23)

24 QUATERCENTENARY STUDIES IN PATHOLOGY

times of the year the organism may and does pass along the intestine of
the sheep without exerting any harmful manifestation. Indeed, at these
times, it exerts a most beneficial influence in rendering the animal
immune. My whole method of preventive treatment, as will be seen
from the Report, is founded upon this principle, namely, the administra-
tion of the organism by the mouth at a time of year when the sheep is
not susceptible to the disease. The organism multiplies in the intestine,
but apparently, at these times, is prevented by some means from crossing
the barrier afforded by the intestinal wall, and so does not find access to
the peritoneal cavity. Nevertheless, it undoubtedly immunizes the
the animal and protects it from an attack of the natural malady. Out of
a total of 1340 sheep treated by us according to this method during
the year 1904-5, in the very worst districts of Scotland, and where we
often shifted the animals deliberately from " clean " to " foul " pasture, we
had not a single death from Louping-ill. A culture was administered to
the animals mostly during the month of January, and not in a single
instance did we find that it had, when administered thus early in the
year, a baneful influence, and yet, nevertheless, it acted as a most
effectual protective against the natural disease.

The subject of immunization through the intestine in the case of
contagious diseases of man which are of intestinal origin has not, it seems
to me, met with that attention which its importance claims.

How the immunization in the sheep is effected, I will not at present
venture to explain. It may be that the epithelium of the intestinal
mucosa becomes resistant to the passage of the organism, and thus
prevents it gaining access to the peritoneal cavity, or it may be otherwise.
To elucidate the problem will require much patient investigation.

Strangers visiting a foreign country and residing in towns where
typhoid prevails endemically, are more likely to contract the disease
than the regular inhabitants. May the explanation of this not be that
the latter are habitually drinking typhoid, and have become immune to
the fever without actually suffering from the malady ? We know, as a
fact, that immunity to the intensely poisonous substances ricin and abrin
may be brought about by administering graduated doses by the mouth.
It seems, therefore, rational enough to suppose that in the case of several
diseases of man, especially those in which the intestine is primarily
concerned, a like immunity may be established through the alimentary
canal. (24)

THE ALIMENTARY CANaL AS A SOURCE OF CONTAGION 25

Whether the dead bacillus given in this way has a like effect I have
not as yet determined, but I think it probable that, if employed in
sufficient quantity, such may be the case. The immunizing principle is
evidently contained in the protoplasm of the bacilli, and there does not
seem any very evident reason for believing that it may not exert its
beneficial influence if administered by the mouth, and in a condition fit
to be absorbed.

During the months in which the disease is rampant, however, the
protective influence of the intestinal wall is lost in a large proportion
of cases, the organism gets over into the peritoneal sac, fructifies within
it, and kills the animal.

The spores being voided with the dejecta are taken up from the
pasture by a fresh host, and the result seems to depend very much on
the season of the year at which this happens. Should it occur during
the susceptible months the danger is extreme, while, at other periods, it
is practically nil. Nevertheless, the younger the animal the more liable
is it to the disease, and hence we may suppose that the ingestion of the
organism at an early period must have effected its immunization.

Cause of periodicity of the disease. — The insusceptibility of the sheep
to the disease at certain times of the year seems to depend directly or
indirectly on the condition of its blood. The blood of the sheep during
the spring months of the year usually constitutes an excellent medium of
culture, while at other times it, as a rule, is not only inimical to the growth
of the bacillus of Louping-ill, but is intensely bacteriolytic to it. So
that if, say, the blood of the sheep during the month of July, be mixed iji
vitro with a culture of the bacillus, the mixture covered with olive oil,
and the whole incubated at a body temperature for twenty-four hours,
probably every bacillus will be found to have vanished. During the
susceptible months, however, the organism multiplies and spores on the
blood of the sheep perhaps better than upon any other medium. It is
evidently this inhibitive action on the part of the blood during most of
the year which prevents the organism growing upon it or upon the
peritoneal liquid, for there is no reason for believing that the organism
gets into the alimentary canal with more facility during the months in
which the sheep is susceptible than at other times. The peritoneal
liquid apparently does not possess solvent powers to anything like the
same extent as the blood, and where the bactericidal action of the

(25)

26 QUATERCENTENARY STUDIES IN PATHOLOGY

blood is lessened as during the months of susceptibility, the organism is
enabled to pass the wall of the intestine, to fructify on the peritoneal
liquid, and to kill the animal acutely with all the symptoms of toxic
poisoning. The peritoneal liquid in such instances is thick and turbid,
and contains the organism in abundance.

Absence of Phagocytosis. — And here it may be mentioned that the
destruction of the bacillus by the blood occurs quite independently of
phagocytosis, for neither in the animal's blood nor in the peritoneal
liquid have 1 ever seen evidence of phagocytosis in this or any other
of these diseases. When, moreover, the organism is mixed with freshly
drawn sheep's blood and incubated, not a bacillus will be found to
have been inglobed by the phagocytes. Should there be, as there
sometimes is, a coccus or other impurity in the blood, the phagocytes
may be found to be packed with the impurity while not a single one
of the specific bacilli has been seized upon. Whether this depends
upon a powerful negative chimiotaxis existing between the specific
bacillus and the phagocyte or not is hard to say, but of the absence
of the phenomenon there cannot be a shadow of a doubt.

Cause of the Toxic Phenomena. — It may happen, however, that in
other instances the blood still retains sufficient bacteriolytic properties
to dissolve any of the organism which gets into it, although it
is not sufficiently inhibitive to prevent a certain exodus of the bacillus
from the intestine. Those bacilli which enter the blood stream,
under these circumstances, are still bacteriolysed, and apparently
the same thing happens, but to a minor extent, within the peritoneum.
The blood in such animals will be found free from bacillus and the
peritoneal liquid may be clear and limpid, and not show any of the
organism until after being incubated. These cases run a chronic course,
and in them the nervous phenomena are most marked. The difference
between the acute and the chronic case seems to depend upon the
rapidity and volume with which the bacillus gets through the intestinal
wall. In the chronic case the number is evidently small, and can be
dealt with by the solvent action of the blood, while in the acute case the
number is so great that the animal dies from rapid toxic poisoning. All
seems to depend upon the condition of the blood. If it be in its usual
state of antagonism to the growth of the bacillus, such as prevails during
the greater part of the year, apparently the bacillus cannot leave the

(26)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 27

channel of the bowel. If, on the other hand, this salutary property be
weakened but not annulled, a certain leakage as it were takes place, but
still the number of bacteria which finds entrance to the blood-stream or
peritoneal cavity is small, and they can be got rid of by the bacteriolytic
action of the blood-plasma or peritoneal liquid respectively.

Through the toxines set free from the bacteriolysed organism, in
the latter case, a condition of chronic toxic poisoning of the animal
ensues. If, again, the blood-plasma have lost its bacteriolytic,
protective, action completely, the organism not only passes the barrier
constituted by the wall of the bowel, but fructifies on the peritoneal
liquid, killing the animal with symptoms of acute toxic poisoning.

It would thus appear that the sheep is a very remarkable animal in
respect of its blood being highly protective at certain times of the year,
while at other times this protective influence is more or less completely
lost.

Whether such a relationship exists in a modified form in the human
blood with regard to certain pathogenic bacteria may be a matter of
question. No one has inquired into the subject, but, judging from
analogy, there seems little reason to suppose that such a remarkable
quality occurring in one mammal is not at least represented in another.
The fall of the leaf and the spring of the year have always been held to
be seasons of great susceptibility to certain contagious and infectious
diseases. May it not be that the natural bactericidal properties of the
blood, as in the case of the sheep, are lessened at these particular seasons ?
various pathogenic organisms being thus encouraged to grow upon the
tissues or it may be upon the surface of the various mucous membranes.

The Toxines bound up with the Bacillus, — The essentially toxic
substance of the Louping-ill bacillus seems to be bound up with the
bacillus itself, and not to be shed into the liquid of culture. When a
culture of the bacillus is made upon glucose-beef-tea, the culture
filtered, and the filtrate injected subcutaneously in large quantity and
on several occasions into the sheep, the operation does not call forth
any symptoms ; the sheep seems to remain quite unaltered in its
condition.

Seeing that the symptoms are apparently caused by the solution of
the bacilli, which find their way into the blood or lymph from the
intestine, and whose toxines are thus liberated, it occurred to me that

(27)

28 QUATERCENTENARY STUDIES IN PATHOLOGY

were a large quantity of the bacillus bacteriolysed by the blood of the
sheep in vitro, thereafter filtered, and the filtrate injected subcutaneously,
I would be enabled to get the toxine in a soluble condition and to study
its effects without the complications arising from having the bacillus
itself to deal with.

The following experiment showed that the filtrate thus obtained
proved intensely poisonous to the sheep, and induced symptoms which
were identical with those of the severest forms of tetanus : —

A 300 c.c. flask of glucose-beef-tea was inoculated anaerobically with
Louping-ill peritoneal liquid taken from an experimental sheep which
died from the disease on February 20th, 1904. The culture was allowed
to grow for forty-eight hours, and thereafter the supernatant liquid was
decanted off and the deposit of bacillus mixed with sheep's serum
containing a large percentage of blood corpuscles, as many indeed as
could be procured by shaking forcibly the bottle in which the blood was
contained. The mixture was covered with sterile olive oil, and placed
in the incubator. The blood was obtained from an Aberdeen slaughter-
house.

In the course of twenty-four hours a large number, but not the whole,
of the bacilli had been bacteriolysed. Only a few were left, and many
of these seemed to be in a state of impending dissolution. The coloured
blood corpuscles were still retained in perfect preservation.

The mixture was passed through a Berkefeld filter, the liquid which
came through being quite clear, and, so far as could be ascertained
by microscopic examination, free from any bacillus.

On July 1 6th, 1904, 4 c.c. of the clear filtrate were injected into each
thigh of a " Cross " one year old, obtained in the Aberdeenshire market.
On the following day, and up to July 28th, nothing was perceived
in the condition of the animal worthy of note. On that day it was
taken very ill, and on visiting it at 5*30 p.m. it was found to be lying
on its side in a perfectly helpless condition, the legs all extended and
uplifted, and in a state of extreme rigidity. They could be bent, but
with difficulty. The jaw was clenched, but at this period not tightly ;
the animal's intelligence seemed to be retained. The neck was stiff and
drawn backwards, the eye clear. From time to time there was spasmodic
contraction of the limbs, particularly if the animal was approached
rapidly, or if disturbed by a sudden noise, as on walking over the stall.

(28)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 29

There was not the slightest evidence of its suffering from diarrhoea, the
neighbourhood of the tail being perfectly clean. The case at this stage
resembled closely certain examples of natural Louping-ill with a tetanic
tendency.

On asking the attendant what the first indications were pointing to
the animal being ill, he said that, when making his usual visit on the
morning of the 28th, it was standing with its head towards him, looking
out between the rails of the pen. On approaching the pen, the animal
walked across it, but in a very unsteady manner, and as if unable to
maintain its balance. It did not seem to be convulsed ; the tendency, he
said, was rather to maintain the limbs in an outstretched attitude.

July 29th, 10 p.m. — The animal was much worse ; it was now in a
state of universal tetanic spasm. The limbs were all outstretched, and
the muscles hard and rigid from tetanic contraction. The jaw was so
firmly clenched that the mouth could not be opened to administer
nourishment. Slight convulsive spasms were noticed from time to time,
but the tonic state of contraction of the limbs was resumed whenever
these passed off. There was still no sign of its suffering from diarrhoea.
The respirations were 40 and the pulse 132 per minute.

July 30th. — The attendant stated that, yesterday evening, when he
visited the pen, the animal was violently convulsed from time to time,
but the rigid state was resumed immediately afterwards. The trismus
was incessant up to the last, and was so severe that the mouth could still
not be opened sufficiently to allow of the administration even of liquid
nourishment.

It was found dead next morning.

The examination of the carcase was made on the afternoon of the
day on which the animal died, and was to the following effect : — The
legs maintained the same outstretched position as during life, and were
rigid ; the jaw was firmly clenched. The stomach and intestine con-
tained a good deal of gas, so that the abdomen presented a fairly swollen
appearance and was tympanitic. The peritoneal sac, however, was free
from gas — at any rate, if present, it must have been in small quantity ; it
contained, however, ij oz. of somewhat muddy-looking serous liquid.
There was some amount of food in the first three stomachs, but the
fourth was empty, and was free from disease. The large intestine
showed a little faecal matter, but the small intestine was devoid of
contents, (29)

30 QUATERCENTENARY STUDIES IN PATHOLOGY

There was not the slightest appearance of local reaction at the points
of injection.

On the surface of the heart there were some patches of punctiform
extravasation, and the blood within the chambers was softly and only
partially coagulated.

The brain and spinal cord were free from visible evidence of disease;
there was no meningitis.

The peritoneal liquid, on being examined microscopically, appeared
to be quite free from germ impurity, the turbidity being caused by a
granular precipitate.

Remarks. — This experiment was followed by very extraordinary
results, and is outstanding as the only one which we have made in the
inoculation of Louping-ill products where the symptoms assumed this
distinctly tetanic character.

The Louping-ill bacillus was grown on 300 c.c. of glucose-beef- tea ; it
grew luxuriantly , the sediment of bacillus, which was copious, was
mixed with sheep's blood, covered with olive oil, and incubated ; the
greater part of the bacillary deposit was bacteriolysed ; the blood was
filtered and the serum injected subcutaneously into the sheep. Symptoms
did not show themselves until the twelfth day after the operation ; they
were ushered in with dulness, were followed by convulsive spasms of the
muscles, and terminated in those of ordinary tetanus, in which last con-
dition the animal died.

It might be argued, of course, that the case was simply one of
tetanus, but against this theory is to be reckoned the fact that there was
neither wound nor external injury by which the animal could have
become infected, and the inoculation punctures had vanished without
leaving any sign of local reaction.

Then again the sheep is not an animal which is liable to tetanus as
we ordinarily understand the disease. Reference has been made, how-
ever, to the tetanus-like symptoms which occasionally develop in the
course of Louping-ill, and these are sometimes accompanied by lock-jaw.
The accounts we have had from shepherds and others of such anomalous
cases, point to the animal becoming rigid while the jaw is firmly
clenched. Such cases evidently represent the disease under conditions
of the greatest severity.

The long period of incubation might be adduced as against the

(30)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 3 1

symptoms having been the effect of the introduction of the serum, but
then it is to be remembered that Louping-ill is a fairly chronic
malady, that it sometimes runs a course of four, five, or six weeks, and
that, in these cases, the nervous symptoms are best marked. We have
also noticed that where the administration of the organism by the mouth
has communicated the disease the symptoms have not shown themselves
for a fortnight to a month afterwards. The poison of Hydrophobia
requires an even longer period of incubation, and the symptoms in it,
as in Louping-ill, are largely of a nervous character.

Had the animal died from tetanus we should have had to fall back
on the disesase being of the so-called "idiopathic" variety to account for
what happened in our experiment, a disease of whose pathology nothing
is known. Indeed, these examples of " Louping-ill tetanus " open up the
whole question of what " idiopathic tetanus " in the human subject means.

Anaemia.

The study of this exceedingly interesting group of diseases in the
sheep, and of chorea paralytica in particular, naturally leads one on to
inquire whether there are any analogous diseases in the case of man.
Each member of the group is caused by an anaerobic sporing bacillus
whose natural habitat is the intestine, and which, under certain conditions
of body, passes through the intestinal wall, and gets into the peritoneal
cavity.

The anaerobic bacteria, naturally present in the human intestine,
seem to include some highly pathogenic members which only require
the proper surrounding conditions to enable them to call forth their toxic
action. And, in the first place, certain of them have an extraordinarily
powerful haemolytic action. In the case of the group of pathogenic
anaerobes parasitical on the intestine of the sheep, which we have been
considering, there are also some whose haemolytic action is intense.
Among these may be mentioned Braxy and Disease "A." The organism
in each of these diseases is so haemolytic that sometimes not a single
coloured blood-corpuscle will be found in the blood contained in the
heart. When the organism of either of them is grown anaerobically on
sheep's blood in vitro during the season at which these diseases prevail,
the haemolytic action is so profound that every blood-corpuscle may have

(31)

32 QUATERCENTENARY STUDIES IN PATHOLOGY

disappeared within twenty-four hours, the lakin^ beinjy extreme. The
colouring matter of the corpuscles is not only liberated, but their stroma
is dissolved, so that often not a vestige of a corpuscle can be recognised.

Curiously, the bacillus of chorea paralytica is not nearly so powerful
a ha^molytic as the bacillus of either of the two diseases mentioned.
The blood-corpuscles of a sheep dying of Louping-ill are preserved, and,
consequently, there is an absence of blood-staining of the carcase, while,
as just remarked, in the case of Braxy or Disease " A " there may not be
a coloured blood-corpuscle remaining, and the carcase is deeply blood-
stained.

It is a common fancy among practitioners of medicine that a large
proportion of anaemias is to be accounted for by some derangement of
the intestinal functions, and Hunter, as is well known, holds to the idea
that pernicious anaemia is caused by an organism, probably a strepto-
coccus, present in the intestine, and that this organism may be derived
from carious teeth or patches of stomatitis.

However true this may be in general principle, it seems even likelier
that the organism which hsemolyses the corpuscles is of the nature of a
bacillus, anaerobic in its habit, rather than a coccus form.

I am led to this belief by the intense haemolytic tendencies displayed
by some of the anaerobic bacilli recovered from the human intestine.
Thus, in the dejecta of persons suffering from progressive anaemia, I
have found an anaerobic bacillus which has a most pronounced haemolytic
action. In one instance, the individual, a man, was supposed to be
suffering from pernicious anaemia, so rapid and progressive was the
destruction of blood-corpuscles and so characteristic were the abnormal
forms. Recovery, however, ultimately took place.

The dejecta, from which the anaerobe was derived, were passed into a
sterile vessel and were examined within an hour or two afterwards.
From them two sets of cultures on glucose-beef-tea were started, the one
aerobic, the other anaerobic. The idea was to discover whether the
cultures so obtained had any effect in destroying the coloured corpuscles
of human blood. The dejecta were found to be swarming with a con-
fused mass of all kinds of organism, and at least one of them was
sporing.

The aerobic growth was not purified, and presented, of course, a
medley of different forms. It was desired, in the first place, to ascertain

(32)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 33

whether these, as a whole, had any destructive influence upon the blood-
corpuscles. The culture-liquid was diffusely turbid, and remained so on
removal from the incubator owing to several of the forms being motile.

The anaerobic growth, on the contrary, seemed to be pure. The
glucose-beef-tea on which it had been sown was heated up to 80° C. for
twenty minutes, so as to destroy the non-sporing organisms. The
inoculated tubes were thereafter placed in the incubator where their
contents germinated with much vigour. Within five hours the contents
were evolving gas, and by the end of twenty-four hours they were quite
turbid. So copious was the liberated gas, that the caoutchouc caps with
which the mouths of the tubes had been covered were tightly distended.
Some hours after removal from the incubator, this anaerobic culture
began to settle down in the shape of an obtuse cone, the apex of the
cone projecting above into the clear culture liquid. It was easily
disturbed, however, and the liquid again rendered universally turbid.
The aerobic culture gave an alkaline reaction, the anaerobic was power-
fully acid. The growth was allowed to proceed in each case for
forty-eight hours.

On December 26th, 1903, an admixture of each with freshly drawn
human blood was effected, in the proportion of 10 parts of blood to i of
culture. The mixture was made in a small sterile tube with the aid of
a glass rod, and thereafter aspirated into the bulb of a pipette ; the ends
of the pipette were sealed off and the tube placed in the incubator at 38°
C. for four hours.

A control experiment was made with the same glucose-beef-tea and
blood, but without organism, and treated exactly in the same fashion.

After four hours incubation the coloured blood-corpuscles in the
control had fallen down to the most dependent part of the tube, while
the surrounding medium was clear and unstained. The corpuscles were
found to have undergone very little change. They were crenated,
but yellowish-orange in colour, and in no wise exhibited signs of
disintegration.

The mixture with the aerobic culture had preserved its colour fairly
well, and on microscopic examination the coloured blood-coupuscles
were found in a state of perfect preservation ; they were not even
crenated ; the colourless, also, were little altered. As might be expected
the organisms present were various.

(33) D

34 QUATERCENTENARY STUDIES IN PATHOLOGY

The anaerobic mixture, however, presented quite a different appear-
ance ; it had only a faint yellow tint. The organism had remained in a
state of purity, a rod of fairly large size, round at the ends, and
apparently motionless. The haemoglobin of the blood-corpuscles had
vanished entirely, and, judging from the yellow colour of the medium,
had been destroyed. The stroma of the corpuscles, however, remained,
so that mere shadows- of blood-corpuscles were left, with outlines so
indistinct that it required careful focussing to render them visible ;
they were not altered in shape. The leucocytes were preserved and
seemed to be quite free from any englobed bacilli.

The experiment was performed on several occasions subsequently,
and always with the same result.

What the rod-organism was which had these haemolytic properties I
have not as yet determined, but certain it is that its destructive action on
the haemoglobin was most striking.

Whether it was foreign to the intestine may be questioned, indeed an
anaerobic rod morphologically very like it can be isolated from the
normal human dejecta, and which, apparently, has the same destructive
action upon the haemoglobin.

The anaerobic intestinal bacteria seem to be much more powerful
haemolytics than the aerobic. The majority of the anaerobes causing the
above diseases of the sheep are extremely haemolytic, and the like
capacity possessed by the bacillus of tetanus, an organism evidently
closely related to those of the various sheep diseases referred to, lends
support to the notion of haemolysis being a very common attribute of
anaerobes as a class.

The fact that in chlorosis and anaemic diseases generally, the
intestinal functions, as a rule, are in an abnormal state, evidently points
the lesson that the condition of that organ is intimately bound up with
the disordered state of the blood. The fact known to every practitioner
that iron fails to exert its beneficial influence in chlorosis unless the
intestine be got into proper working order, adds additional weight to the
supposed intestinal origin of this form of anaemia in particular, and of
progressive anaemias in general.

Granted that it is the poison derived from a blood-destroying
organism contained in the intestine which is the cause of the haemolysis,
it may, as just said, be asked whether the organism is present normally

(34)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 35

in the intestinal contents or introduced on occasion, whether, in fact, such
a blood-destroying organism, or several of such organisms, is constantly
present in the alimentary canal, and prevented from exerting its
injurious influence on the blood by the interposition of the intestinal
wall. Indeed it comes to be a question whether specific anaemic diseases
should not be regarded as functional disorders of the intestine rather
than, primarily, diseases of the blood.

We have seen that in chorea paralytica of the sheep the blood is free
from organismal impurity while the intestinal contents are permeated
with a highly contagious anaerobe, yet there is evidently some toxine
circulating with the blood which stimulates and subsequently paralyses
the motor nerve cells. The explanation I have given of the pathology
of these phenomena, and the theory is founded on fairly conclusive
experiment, is that a certain proportion of the bacteria which are the
cause of the disease, and which are resident in the intestine, get into the
blood-current and are bacteriolysed. The liberated toxines afterwards
coming in contact with the nerve cells induce the characteristic
phenomena of the disease.

May it not be that allied bacteria, of anaerobic habit and with
haemolysing tendencies, similarly, in certain morbid conditions of the
intestine, manage to pass through the barrier constituted by its wall,
become bacteriolysed in the blood-current, have their haemolysing
poisons liberated, and that such again come to act on the blood-
corpuscles, either simply depriving them of their haemoglobin as in certain
of the chlorotic varieties of anaemia, or, in addition to this, effecting a
solution of the whole corpuscle as in the pernicious varieties of the
disease ?

Tetanus.

We have seen that certain instances of chorea paralytica have quite
a tetanic character, and that apparently Tetanus can be aroused in the
sheep by injecting subcutaneously the bacteriolysed organism of the
disease.

The organism of Tetanus has a very close relationship to that of
Louping-ill, indeed to all the members of the group, and, manifestly in
their all being anaerobes, in their tendency to sporulation, and when
sporing, in the readiness with which they assume the drumstick con-

(35) D '

36 QUATERCENTENARY STUDIES IN PATHOLOGY

figuration. In the case of Tetanus, the symptoms are induced by the
absorption of a toxine secreted by the bacillus of the disease. The
organism exerts its evil influence only when introduced into a wound ; when
administered by the mouth it is apparently harmless. So universally is
the organism distributed in nature that we must be constantly swallowing
it, and yet no ill effects follow.

There is another variety of tetanus which is known as " idiopathic,"
mainly because we know nothing of its pathology, and in which there is
an absence of any wound of the surface, or other apparent point of
entrance. It has always been assumed that the organism inducing this
idiopathic variety is the same as that occurring in the traumatic form.

In view, however, of this disease, chorea paralytica of the sheep,
being so closely allied with tetanus, both with regard to the organism
producing it and in the character of the symptoms occasionally evoked,
I would be inclined to pause before admitting the truth of this allegation.
May it not be a disease caused by an organism of the same class as that
producing chorea paralytica, but intestinal in its habitat, as the organisms
instrumental in producing all the members of this class of diseases of
the sheep are? And may it not happen that it is a common inhabitant
of the intestine, but that only in certain susceptible individuals gets over
from the channel of the bowel into the blood, becomes bacteriolysed,
and so allows the toxines bound up with its protoplasm to escape, these
again acting on the nerve cells with which they have a combining
affinity ?

Tetanus of the new-born is also a form of the disease of whose
pathology we have no conception, and I would go so far as to suggest
that it may be the result of the action of such an intestinal anaerobe to
which the infant as yet has not become immunized.

Chorea, Epilepsy, and Insanity.

The same principles applied to chorea seem to me to afford a very
likely explanation of its pathology. The human disease presents many
points of resemblance to chorea paralytica of the sheep, and perhaps it
is not going beyond the mark to suppose that this disease of man like
that of the sheep is also intestinal in its point of origin, that the same
principle is at work in its production as in chorea paralytica of the sheep,

(3<5)

THE ALIMENTARY CANAL AS A SOURCE OF CONTAGION 37

namely, that, for some reason, probably a defective condition of the
blood, a particular bacterial parasite is allowed to pass over from the
intestine into the blood, becomes dissolved in it, and thus has the toxines
set free which are the cause of the nervous phenomena. In severe fatal
instances of chorea minute abscesses have been found in the brain, and I
think that quite possibly these may have been due to the organism
having got into the blood in excessive quantity, so that the plasma has
failed to dissolve them.

Then, again, may not Medullary Epilepsy be accounted for by a
collateral pathology? May not it also be intestinal in its point of origin?

Ford Robertson {Brit. Med. Journal, 1901, IL, p. 1230; Ibid: 1903,
IL, p. 1065, Edin. Med. Journal, XIX., 1906, p. 218, Review of Neurology
and Psychiatry, May and July, 1903 — Reprint), insists upon the
paramount importance of the gastro-intestinal tract in relation to
insanity, and makes out that in general paralysis of the insane the
surface of the intestine is covered with a diphtheroid bacillus which
finds its way into the tissues. He even goes the length of asserting
that locomotor ataxia may be due to a similar cause.

Cirrhosis of the Liver.

It seems likely enough that Cirrhosis of the Liver may be accounted
for on a pathology of the same kind. A particular bacterium becomes
absorbed from the intestine, it is bacteriolysed within the portal blood,
its liberated toxines are anchored upon the liver substance and stimulate
this to the production of an excess of fibrous tissue. It is a remarkable
fact bearing upon this theory that not infrequently the minute branches
of the portal vein in cirrhosis contain organismal emboli. Quite possibly
these have accumulated shortly before death, and owing to the enfeebled
powers of the blood have remained undissolved.

What the aetiology of the cirrhosis of the horse due to feeding it
upon lupines, or that of the cirrhosis of domestic animals said to be
so prevalent in New Zealand from feeding them upon ragwort {Senechio
Jacobed), * may be, I will not venture to broach for the present, but it
seems to me these two alleged causes of hepatic cirrhosis may come
to have important bearings upon the above theory.

* (See Gilruth Report, New Zealand Department of Agriculture, Division of Veterinary
Science, 1902-3, p. 228; Ibid., Bulletin, No. 9; Ibid,, Report for 1905, p. 178).

(37)

3^ QUATERCENTENARY STUDIES IN PATHOLOGY

There are many other diseases which in all likelihood are conferred
through the intermediation of the intestine. Thus Dr. W. Hunter, of
Hong-Kong (A Research into Epidemic and Epizootic Plague^ Hong-
Kong, 1904) has been led by his researches to believe that in the case
of Plague the theory of infection through the skin has been vastly
exaggerated, and that a much more likely channel of contagion is the
intestine. His observations are supported by Simpson's previous
experiments.

Lastly, may it not be the case that a great many of the exanthemata
are in reality intestinal diseases ? May it not be that the intestine is the
portal through which the respective poisons of these gain entrance to
the system ? Scarlet fever seems to me to be a case in point, for not
only is there a lesion of the fauces in this disease which might quite
well be a manifestation of a poison absorbed by the lymphoid tissue
in that neighbourhood, but we also know that one of the commonest
means of transmitting the disease is through contaminated articles
of diet.

All these considerations go to demonstrate how vastly important the
part played by the alimentary canal in the propagation of contagious
and infectious diseases may be, and of the necessity there is for a more
thorough knowledge of the whole matter. It is a field of inquiry which
has been greatly neglected, unless in the case of those diseases which are
manifestly intestinal in the fact of their causing some gross anatomical
lesion. I would enter a plea for the more thorough investigation of
those diseases whose pathology at present is unexplained, which do not
leave any such residuum, and which, heretofore, have not been suspected
to be of intestinal origin.

(38)

A REMARKABLE CASE OF BILHARZIOSIS.
By William St. Clair Symmers, M.B., CM.,

Musgrave Professor of Pathology, Queen'' s College , Belfast ;
Formerly Professor of Pathology in the Medical School, Cairo.

{i9)

A Remarkable Case of Bilkarziosis.

The case that I desire to record in this volume, is one which in the
extent and locality of its lesions is unique in several respects. The
intestinal condition is different from anything with which I am familiar,
and is the only thing of its kind that I met with during my seven years
sojourn in Egypt.

The patient was an Egyptian male of about 30 years of age, who
died in the Kasr-el-Aini Hospital, Cairo. The post-mortem examina-
tion was held three hours after death, on the 27th October, 1904.

The body was emaciated. Rigor mortis was present throughout.
There were no external appearances of further note — no jaundice.

Weight of Organs.

Heart - - - - I75 grammes.

Right Lung - - -330

Left Lung - - - 310 n

Liver - - - . 1225 n

Spleen - _ - - 200 u

Left Kidney - - - 120 m

Right Kidney - - - 120 n

Brain - _ _ . 1205 ti

The Heart. — No pericardial fluid. The organ was small and
flabby, and was dark, reddish brown in colour, being in the condition
of brown atrophy. This cardiac atrophy (subject a;t. 30) was merely
a concomitant of the general advanced emaciation of the body.

The Lungs. — No pleural effusion, no adhesions. Both organs were
small and shrunken, fairly firm and were unusually dry. The right lung
was normal to the naked eye. The left lung showed on its surface a

(41)

4 QUATERCENTENARY STUDIES IN PATHOLOGY

number of small, flat, whitish nodules of firm, fibrous consistence, varying
in diameter from three to five millimetres. These discs were not unlike
the flat fibromata so often seen on the capsule of the spleen ; they were
especially seen at the margins of the organ, where they formed a discon-
tinuous fringe resembling a row of small pearls. From analogy with
bilharzial lesions, I concluded from the naked eye examination that
these nodules were fibrous in nature, and would contain ova of bilharzia,
but the microscope revealed quite another and unexpected condition,
viz.^ the masses were tuberculous, and were composed almost entirely of
closely packed giant-cell systems, particularly at their free margin ; the
deeper portions {i.e. nearer the lung) were more fibrous in nature.

The lung itself was firm, particularly at the apex ; a few small solid,
pale yellow masses (tubercle) were apparent at wide intervals, and the
intervening parenchyma showed under the microscope a multitude of
tiny tubercles with numerous beautiful giant-cell systems, although to
the naked eye this tissue appeared to be normal. The bronchial glands
were but little enlarged, were jet black, and were not caseous. One of
them was carefully examined microscopically, and showed neither
tubercles nor deposits of ova.

There was no ascites. Numerous bilharzia worms {Schistosomum
hcematobium) were present in the portal vein.

The Liver (see Fig. i) was a typical specimen of advanced
bilharzial cirrhosis, and in every particular tallied with the description
given of this condition by the writer in the Journal of Pathology and
Bacteriology^ Vol. 9, p. 237.

It was slightly smaller than the average, weighing 1225 grammes, the
surface presented numerous small flat fibrous growths on the capsule ;
the cut section showed an enormous periportal cirrhosis, " as if a number
of white clay pipe-stems had been thrust at various angles through it,"
these white cirrhotic masses contrasting sharply with the drab colour of
the general parenchyma. Microscopically, the ova of the worm were
distributed in the same manner as was mentioned in my original
description of this disease.

The Spleen and Kidneys showed nothing of note.

The Urinary Bladder showed the slight roughness and peculiar
sandy discolouration of the mucous membrane which is so characteristic
of incipient bilharziosis of this viscus, and a scraping of these discoloured

(42)

A REMARKABLE CASE OF BILHARZIOSIS

5

areas showed, under the microscope, hundreds of the terminally spined
ova of the Schistosomum hcematobium.

The Pancreas was to all naked-eye appearance perfectly normal, but
the microscope revealed a slight amount of cirrhosis, and an unduly small
number of Langherhans' islets. However, the most interesting thing
noted was the presence of bilharzial ova in this organ, these ova were
fairly numerous and were mostly found in the cirrhotic bands, but a few
were seen lying in the glandular tissue. The ova found in the pancreas
were mostly broken and partly destroyed, but the transparent broken
shells were quite easily recognisable, especially as similar remnants of
ova are often seen in other situations, e.g., in the liver, and, moreover, it

Fig. I. — Bilharzial Cirrhosis of Liver. Antero-posterior section of right lobe. The figure
shows great increase of fibrous tissue round the portal canals.

was found that certain of the broken shells retained the characteristic
spine.

This is the only case in which I have deliberately examined the
pancreas for bilharzial ova, and doubtless the eggs will be found in other
cases, in this organ, if they are looked for. It is generally believed that
the pancreas seldom or never contains these ova — probably this opinion
dates from Kartulis' paper,* where the distinguished Alexandrian
physician reports two cases in which he failed to find the ova in the
pancreas, although they were present in various other organs. Fourteen
years afterwards Kartulis again draws attention to the absence of eggs in

* Virchow's Archiv, 1885, xcix., p. 139.

(43)

6 QUATERCENTENARY STUDIES IN PATHOLOGY

the pancreas,* and quotes Lortet and Vialleton's explanation of this
supposed fact, viz., the female lays eggs where they may find an exit
from the body of its host. Prof. Looss states that Goebel has found the
ova in the pancreas.f

I venture to intercalate the remark in this place that I have twice
found living worms in the pulmonary blood, one of these observations

Fig. 2.— Serous surface of ileum, showing fibrous masses of various sizes.

was published in the Lancet (1905, Vol. I., p. 22), the other is still
unpublished. The reason the worms have not hitherto been found in

* Virchow's Archiv, 1899, clii., p. 474.

t Mense's Handbuch der Tropenkrankheiten, 1905, Vol. I., p. 96.

(44)

A REMARKABLE CASE OF BILHARZIOSIS 7

the lungs, is simply that they have not been looked for in a sufficient
number of cases.

Fig. 2A. —Lower ileum, showing large fibrous masses on the serous surface.

The Brain was normal.

Before proceeding to describe the most pronounced lesion of our

(45)

8 QUATERCENTENARY STUDIES IN PATHOLOGY

present case, we may mention, for the sake of clearing the ground, that
numerous specimens of anchylostoma duodenale were found in the
small intestine ; the mesenteric glands were swollen, in size from a
pea to a walnut, hard and firm, and showed microscopically a marked
increase of the connective tissue, wide bands of well-formed fibres, and
also more delicate fibrils permeating the lymphoid tissue. Certain of

■p

P'^^^^^B

*^-, ■ ■^M

^^^^^^^B 9Kl^' «^

^ jJM

^^^^^^^bflL^ji

h/-'^' ''^1^1

^^M f

-jH

^^^^Bifl^^^''^ -'

^^ "-^^H

^^HHr ...

^'-^^^*^B

■V^w

HS

HHflii

Fig. 3. — Huge fibroma on serous coat of ascending colon.

these glands contained a number of bilharzial ova, but others contained
none. Here again we come upon an unusual observation — ova are
not as a rule recognised as being present in the lymphatic glands ;
Kartulis found them, however, and mentions the fact in his paper of
1885, and again in 1899.

(46)

A REMARKABLE CASE OF BILHARZIOSIS 9

The Intestine. — The prominent and distinctive lesion of the
present case is found in connection with the bowel, particularly as
regards the external or serous portion thereof.

Beginning five feet above the coecum, there were on the serous
surface of the ileum a number of small polypoid outgrowths, varying
in size — about a centimetre in length — consisting of a hard club-shaped
extremity, often flattened or discoid, and connected with the bowel by a
pedicle of loose connective tissue. These pedicles were flattened from
side to side, often forming little sheets of membrane whose linear axis
ran with the long axis of the gut. The distal portions of the pedicles
contained the small firm masses above-mentioned, masses from the size
of a split pea to a horse bean, of various shapes, discs, oryzoid, globes,
round and oval masses, like a captive balloon or airship. Many of these
fibrous bodies lacked the pedicle, and were connected to the gut by broad
bases. As we pass down the bowel towards the coecum, the pedicles
become fewer and disappear, the fibrous masses become larger, increasing
in size to that of a hazel-nut, a walnut, a pigeon's egg (see Figs. 2 and 2A).
Nearing the coecum the masses are several inches — four to six — in length,
project an inch from the bowel surface, and occupy half or more of the
circumference of the gut, looking like large sausages split longitudinally
and applied to the wall of the viscus. The masses rise abruptly from
the bowel, are situated on the side opposite the mesenteric attachment,
and are continued almost uninterruptedly down to the coecum. They
are covered by a smooth lining of peritoneum continuous with the serous
coat of the bowel — show in places secondary polypoid growths, like
those higher up the ileum, and are composed of dense white fibrous
tissue, which contains numberless ova in certain parts.

The coecum shows a similar condition of its serous coat — a mass
almost as large as a pigeon's egg. The colon is extensively affected,
showing huge masses of fibrous tissue, an inch in thickness (see Fig. 3 j
extending along the whole length of the free surface of the bowel, and
occupying from a half to three-quarters of the circumference of the
intestinal tube. The secondary polypoid masses, growing from the colon
tumours, were numerous and much larger than the corresponding iliac
tumours, although the most extensive sessile growths were on the lower
ileum. In short, as regards the external portions of the bowels, there
was in the ileo-colon region a series of fibrous growths, increasing in

(47)

10

QUATERCENTENARY STUDIES IN PATHOLOGY

size from five feet above the coecum to a foot below it, thence becoming
gradually smaller, to end in long thick flattened masses on the free
margin of the rectum. These masses were composed of firm, well
developed fibrous tissue, and contained an enormous accumulation of
bilharzia ova, these latter could be seen in mass by the naked eye, being
so numerous that they gave a peculiar light brown appearance to the
areas where they were most abundant. Thus a cross section of one of
these masses was of a glistening china white, darkened over numerous
areas, to a light brown colour (see Fig. 4).

Microscopically the fibrous tissue was, for the most part, arranged in
beautiful concentric whorls about the enclosed ova — an arrangement

Fig. 4. — Transverse section of the fibroma represented in Fig. 3. The tag of tissue on the
extreme left is a portion of the wall of the intestine, and it is seen to be continuous along the
under surface of the tumour. The darker portions are enormous accumulations of ova.

which could not be detected by the naked eye. This whorled appear-
ance was marked in the discoloured areas of the tumours, but in the
china white portions (where few ova are found) the arrangement was that
of ordinary fibrillar tissue.

The Appendix Vermiformis was particularly interesting. It was
distinctly thickened, measured 6 cm. in length and 8 m.m. in diameter,
being almost uniform in thickness along its whole length, though some-
what compressed to an oval about its middle portion. A mesentery
was present, and opposite this attachment there was a number of
delicate polypoid outgrowths from 3 to 10 m.m. in length, which had the
peculiar light brown colour characteristic of an accumulation of bilharzial

(48)

A REMARKABLE CASE OF BILHARZIOSIS

II

ova. On transverse section of the appendix the lumen was perfectly
circular, and occluded by a solid mass of soft material. External to this
was a delicate whitish ring (mucous membrane), then a wide brownish
ring of ova-infiltrated tissue (submucous tissue) and bounding the whole
was a distinct ring of white (muscle and serosa). Thus the accumulation
of ova was present between the mucosa and the muscular coat, and here
the ova were innumerable, lying in a hyaline material, this region com-
posing nearly two-thirds of the entire diameter of the appendix, the
muscular coats being pushed out and partially atrophied. The increase

Fig. 5.— Section through appendix vermiformis (x 6).

A. Muscular coat.

B. Greatly thickened submucous coat, the dark areas being masses of ova.

C. Amorphous mass in lumen of appendix.

in thickness of the appendix was due to hypertrophy of the submucous
coat, and to the innumerable ova therein embedded (see Fig. 5).

Internally, the bowel was equally interesting. The mucous mem-
brane was unduly hyperaemic throughout. Peyer's patches were normal,
but the solitary follicles were somewhat enlarged. A few superficial
ulcers were seen in the ileum, from one to three centimetres in linear
extent, and 5 cm. broad, they lay with their long axis traverse to the
axis of the bowel, were of very slight depth, clean based with somewhat
firm, rounded margins. In the ileum were a few pedunculated polypi.

(49) E

Fig. 6. — Part of descending colon, showing polypoid growths. The serous portion was greatly
^ _ thickened by fibrous overgrowth.

A REMARKABLE CASE OF BILHARZIOSIS

13

The large bowel was the seat of an extreme polyposis (see Fig. 6).
Numbers of papillomata were ranged from coecum to anus, in all 120
were counted. These outgrowths varied in size from a pea to a hazel
nut, some few being even larger. They were broadly pedunculated as a
rule, or sessile, were nodular or mulberry-like, and were smooth or
slightly roughened in places, presumably from the action of foecal matter
rubbing over them. These polypoid growths are a well-known and
frequent manifestation of intestinal bilharziosis, but I do not remember

D

Fig. 7. — Section of Papilloma (x 6).

A. Hypertrophied Lieberkuhnian crypts on side of tumour.

B. Masses of ova in centre of tumour.

C. Hypertrophied sub-mucosa of intestine,

D. Ova in submucous tissue.

to have seen them in such numbers before, nor so uniformly scattered
throughout the whole large intestine. Moreover, it is not often that
bilharzial papillomata are seen above the ileo-coecal valve. I have
examined a number of these growths (from the present case), and so far
all that I have microscopically inspected are crowded with bilharzia ova
(see Fig. 7). This is worthy of emphasis, inasmuch as the ova are very
often absent from these intestinal growths.

(51) E I

14 QUATERCENTENARY STUDIES IN PATHOLOGY

Microscopically, these tumours show the following facts : the mucous
membrane surrounding them is partially destroyed, the superficial parts
thereof are worn away, the Lieberkuhnian crypts are to some extent
atrophied, only their deeper-lying half is left, and these remnants show
considerable alteration of the individual cells composing them, viz.^ the
luminal end of the cells is indistinct, vacuolated or empty. The crypts
are separated from each other by a mass of new tissue, which, in breadth,
is about equal to that of one of the crypts, and consists of cellular tissue
composed of mononuclear round cells with but little protoplasm, and also
of short spindle cells. A few ova are seen lying in the interglandular
new tissue, but never inside a crypt itself As this partially destroyed
mucous membrane nears the tumour, it becomes greatly hypertrophied
and is lifted up by the projecting papilloma, so as to form, for a short
distance, a covering to the latter, but it soon is broken through by the
growing tumour which is thus devoid of a mucous lining. The tumour
itself arises from an over growth of the submucous tissue, this tissue is
greatly increased in thickness along the bowel, is composed of young
connective tissue interspersed by numerous groups of small round cells,
and is crowded with ova. Portions of the hypertrophied submucous
tissue, projects outwards through the mucous membrane, carrying with it
enormous numbers of ova, and showing, in the polyp itself, a great
number of congested, ectatic capillaries, and a tissue which is less highly
developed than the general hypertrophied sub-mucosa, being composed
of young fibroblasts, some delicate fibrils, and numerous round mono-
nuclear cells. These mononuclear cells are in greatest abundance over
the surface of the polyp, where they form a richly vascular cap of cellular
tissue, which often has itself a thin covering of precipitated fibrinous
material in which a few round cells are present. The papilloma contains
great numbers of ova, principally in its most central part, the ova lie free
in the young fibroblastic tissue, and occasionally one or two are seen
lying inside a congested capillary. It is, in many instances, on the
acclivities only of the papilloma that remains of Lieberkuhnian crypts
are seen, though some disorganised glandular masses are found nearer
the summit of the tumour. Such of the true mucous membrane as
exists on the surface of the papillomata {i.e., the portions spared by the
faecal attrition) is greatly hypertrophied, the glandular crypts being con-
siderably increased in length and often widened into cyst-like cavities,

(52)

A REMARKABLE CASE OF BILHARZIOSIS

IS

besides, the membrane is traversed by numerous engorged and dilated
capillaries, and thickly infiltrated with small round cells. Lying among
these abnormal Lieberkuhnian crypts there are ova, but not nearly in
such numbers as within the central parts of the tumours. I have found
no worms in these tumours (in the present case), but the worm itself has
appeared in many of the sections, lying in the sub-mucosa of the bowel,
but by no means so numerously as in some cases, and moreover those I
have, so far, seen are young males lying alone {i.e., not copulated), see
Fig. 8.

Hitherto I have avoided any special description of the ova themselves,
because I wish to emphasise the fact that when bilharzia ova are found

Fig. 8. —Portion of submucous tissue of large intestine (x loo).

A. Male worm in a capillary.

B. Ova.

embedded in abnormal growths of connective tissue (or in fibro-epithelial
tumours) they (the ova) are no longer living, but are dead, and for the
most part disorganised. To begin with, the ova are hard bodies, and
therefore cause a certain amount of tearing of the enclosing tissue when
microtomic sections are prepared. Many of the ova in intestinal lesions
are, as is well known, possessed of a lateral spine, now in sections of such
tissues as this paper deals with, by far the greater number of the ova
present no visible spine whatever, and appear completely different from
normal ova such as are found in the urine. Moreover, the ova are often
distorted, cracked, broken or shrivelled as if they had undergone some

(S3)

l6 QUATERCENTENARY STUDIES IN PATHOLOGY

solvent action, or again, empty shells, with or without lateral spines, are
seen lying closely contiguous to round bodies, which are the transversely
cut shells of ova which have been thus sectioned by the knife of the
microtome. Never is a recognisable miracidium seen within these dead
ova, but a mass of amorphous nature, deeply stainable by logwood, is all
that represents the original content of the ovum. The slightly yellowish,
chitinous (?) wall of the ovum is all that remains in many instances,
enclosing, it may be, some dark granular material presumably calcic in
nature.

To conclude, we have in the present case : —

1. A unique growth of bilharzial fibromata of the intestinal wall.

2. A marked bilharzial cirrhosis of the liver.

3. A bilharzial fibrosis of the appendix vermiformis.

4. Bilharzia ova in the pancreas.

5. Bilharzia ova in the mesenteric glands.

6. An extreme bilharzial polyposis of the large bowel.

7. A few bilharzial papillomata in the lower part of the ileum.

8. Incipient bilharziosis of the urinary bladder.

9. And, as an epiphenomenon, an acute tuberculosis of the lungs

and pleura.

(54)

MALFORMATIONS OF THE BULBUS CORDIS.

An Unrecognised Division of the Human Heart.
By Arthur Keith, M.D. (Aberdeen), F.R.C.S. (England),

Lecturer on Anatomy, London Hospital Medical College ;
Examiner in Anatomy, Royal College of Surgeons, England.

(55)

Malformations of the Bulbus Cordis.

An Unrecognised Division of the Human Heart*

In this paper I propose to set out as briefly and clearly as I may, the
evidence which has led me to the two following conclusions: — (i) that
besides the sinus venosus, the auricular and the ventricular divisions
— the three parts out of which the whole mammalian heart is believed to
be formed, there is present a fourth, namely, the bulbus cordis ; (2) that
the great majority of cases which are classified, at the present time, as
congenital stenosis of the pulmonary or of the aortic orifices are, in
reality, due to an arrest of development or malformation of the bulbus
cordis. It is with the second of these two conclusions that I now intend
to deal here, and it will be apparent that if the congenital constrictions,
which are found at the orifices of the pulmonary artery and aorta, can
be shown to be due to an arrest or vitiation of development, then the
theory, which has still a wide currency amongst medical men that these
lesions are the result of foetal endocarditis, must be finally abandoned,
and these conditions classified with hare-lip, cleft palate, hydrocephalus,
spina bifida, club-foot, atresia ani, hypospadias, and other congenital
malformations with which the cardiac lesions are so often associated.

Note. — The interpretation which I give of the malformations of the heart described in this
article is a new one, and it is not necessary, therefore, to give an exhaustive list of references.
As regards our present knowledge of this subject, Peacock's work On Malformations of the
Human Heart (2nd Edition, London, 1866), is still the most authoritative work which has been
produced in England ; Rokitansky^s Die Defecte der Scheidewdnde des Herzens (Wien, 1875),
holds a corresponding place in German medical literature, while Moussous' Maladies
Congenitales du Coeur (Paris, 1896), and Theremin's Etudes sur les Affections Congenitales dtc
Coeur (St. Petersburg, 1895), may be accepted as representative of the more recent work done
in France and Russia.

This article deals with one of the results of an effort I have made during the last two years
to systematise our knowledge of congenital malformations of the heart. I have had an oppor-
tunity of examining over 200 malformed hearts in the medical museums in London (50 of these
were obtained at the London Hospital). In the Transactions of the Pathological Society of
London, I obtained descriptions of 123 cases; from other sources in medical literature I
obtained over 200 cases more, so that, altogether, I have accounts of over 500 specimens. The
Index Catalogue, issued from the Surgeon General's office of the United States, contains 12
columns of references to records of isolated cases of congenital morbus cordis, and it was an
examination of some of these records which convinced me of the necessity of a systematisation of
our knowledge of this subject.

(57)

4 QUATERCENTENARY STUDIES IN PATHOLOGY

To understand the nature, structure and primary connections of the
bulbus cordis, one must study this chamber of the heart in fishes, of the
shark or allied tribes, in which it reaches its highest structural and
functional development. It is the last of the four pulsating chambers of
the primitive vertebrate heart. In figure i, these four chambers are
shown, the sinus venosus (i), a reservoir into which the great veins

Figure i, heart of Xiphias gladiuSy showing the four primary compartments of the heart
(from a specimen in the Museum of Royal College of Surgeons, England).

(i) Sinus venosus.

(2) Auricle laid open, showing the venous valves guarding the orifice of the sinus.

(3) Auricular canal.

(4) Ventricle.

(5) Bulbus cordis.

(6) Lower orifice of bulbus (ventricular orifice).

(7) Upper II (aortic orifice).

(8) Common ventral aorta leading to gills.

discharge their blood, the common auricle (2), the essential function of
which appears to be that of loading the ventricle ; the common ventricle
f^), and the bulbus cordis (5), out of which opens the primitive ventral
aorta leading to the bronchial circulation. It is a matter of regret that
there is so little direct experimental evidence of the exact part played by

(58)

MALFORMATIONS OF THE BULBUS CORDIS

5

the bulbus cordis in the circulation of fishes, but from an examination of
its structure, we may refer, to some extent at least, its functional
significance. A lower orifice (figure i, ^), usually provided with valves,
marks the separation of bulbus and ventricle; an upper orifice, which may
or may not be valvular, indicates the junction of bulbus and primitive aorta.
The walls are usually thick, and composed of a dense musculature ; its
cavity, fusiform in shape, with a corrugated, ridged or valvular lining of
endocardium. Its musculature is of the striped cardiac type, rhyth-
mically contractile and inhibited by the action of the vagus (Gaskell).

The bulbus cordis is commonly regarded as a sphincter mechanism
to secure competency of the aortic valves. The necessity for this
elaborate mechanism becomes more evident when it is remembered that
the entire heart of a fish is purely a branchial or respiratory pump, and
discharges its load, not into the whole arterial system, as in the higher
vertibrates, but into the limited capacity of the branchial system, which
must undergo a relatively great expansion with each heart beat — its
valves being so rendered incompetent. The valves are made competent by
virtue of the musculature which surrounds them, therefore it is possible
that the bulbus, by the degree of its tonus, may also regulate the blood
pressure within the branchial system, either by permitting a functional
regurgitation or regulating the rate of outflow from the ventricle into the
aorta. Such a theory of the function of the bulbus does not explain the
form and great development of the bulbus, nor the great thickness and
rhythmical contractions of its muscular walls. It may be provided with
valves only at its upper or at its lower orifice. In fishes, in which there
is a well developed bony operculum covering the gills, the bulbus is
small or vestigial. The movements of the operculum probably regulates
and assists the circulation of blood through the gills. Further, there is
this most important fact to bear in mind that with the evolution of a
pulmonary system and its complete separation from the systemic
circulation the bulbus cordis becomes restricted to the right side of the
heart, and can act therefore only on the pulmonary circulation. What-
ever future research may show, this much is certain that the bulbus cordis
is essentially a part of the respiratory system, and apparently is chiefly
concerned in regulating the flow of blood through the lungs. In this
connection it seems pertinent to recall the fact that physiologists have
failed to demonstrate a vaso-motor supply to the pulmonary vessels.

(59)

QUATERCENTENARY STUDIES IN PATHOLOGY

Any descriptive account of the anatomy of the heart, even by those
who write clearly and well, is particularly hard to follow, hence it is
necessary for me to use every aid to make my description intelligible.
This end will be served if, at the outset, I indicate by the use of a figure,
the part of the normal human heart which, on the evidence to be
produced, I believe to be formed from the bulbus cordis. That part is
the infundibulum of the right ventricle (see Fig. 2). The wall of the right
ventricle, when opened in the manner shown in figure 2, is seen to show

Figure 2, normal human heart with the right ventricle opened,
(i) Pulmonary artery.
(2) Aorta.

Pulmonary valves (upper limit of infundibulum).

(5) Thickenings in wall of right ventricle at the junction of the infundibulum

and body of the right ventricle.
Body of the right ventricle.
Tricuspid valve.
Left ventricle.
Right auricle.
Left auricle.

A, right septal band, B, left septal band of infundibulum. The line between A and B
indicates the position of the septal raphe of the infundibulum.

(3)
(4)

(6)
(7)
(8)
(9)
(10)

(60)

MALFORMATIONS OF THE BULBUS CORDIS 7

a distinct thickening at the points indicated by 4 and 5. These
thickenings mark the junction of the infundibulum with the bod}^ of the
right ventricle, and represent the position of the lower or ventricular
orifice of the bulbus cordis. On the septal wall of the infundibulum are
seen two muscular bands or plates, the right (A) and left (B) septal
bands, with between them a faintly marked and rather diffuse raphe.
The septal wall of the infundibulum, formed by the two muscular bands
or plates just mentioned, is only 20 to 25 mm. in length, while the lateral
or marginal wall of the infundibulum is twice that depth or more. Its
muscular walls are dense, the fibres are in the main circular in direction ;
the uppermost fibres surround the base of the pulmonary valves, and,
as is well known, render these valves competent by their tonus in
diastole. The infundibulum of the right ventricle is manifestly
different in structure, and presumably also in function, from the body of
the right ventricle, and I regard it as derived directly from the bulbus
cordis, the muscular thickening between the two parts of the ventricle
representing the ventricular or lower orifice and the pulmonary opening
the aortic or upper orifice of the bulbus cordis.

In studying the evolution of the mammalian from the primitive
four-chambered heart of the fish, one recognises that three great changes
have taken place; (i) the primitive auricle and ventricle have become
completely divided into right and left chambers ; (2) the sinus venosus
has become partly or, as in man, almost completely, submerged in the
musculature of the right auricle ; (3) the bulbus cordis has become
separated from the left ventricle and aorta, and completely incorporated
in the right ventricle as the infundibulum of that chamber. That such
was the fate of the bulbus cordis in the mammalian heart, I became
convinced four years ago from a study of the malformations and
comparative anatomy of the heart, but my evidence was incomplete
until Greil published the results of an exact and complete research into
the embryological history of the bulbus cordis, and demonstrated its fate
in the mammalian heart. (For abstract of Greil's research see Hoch-
stetter's account in Handbuch der Vergleick, u. Experim Entwickelimgs-
lehre der Wirbeltiere, parts 4, 5, 14, 15, 1903). The cavity of the bulbus
cordis is incorporated in the right ventricle by an upgrowth of the
ventricular musculature round it, the musculature of the bulbus being
thus replaced by the musculature of the ventricle, in the same w^ay as

(61)

8

QUATERCENTENARY STUDIES IN PATHOLOGY

the musculature of the auricle replaces a great part of that of the sinus
venosus. The submergence of the bulbus evidently constitutes a critical
phase in the developmental metamorphosis of the heart, and it is during
the critical phase that malformations are apt to occur and give rise to
the various lesions I shall now proceed to describe.

Figure 3, heart of a man aged 20 who died, after two months' illness, from an acute
infection with endocarditis.

(i) Pulmonary artery.

(2) Aorta.

(4) (5) Fibro-muscular contriction marking the lower infundibular or bulbar orifice.

(6) Body of right ventricle.

(7) Tricuspid valve.

(8) Left ventricle.

(9) Right auricle.

(A) Right septal band.

(B) Left septal band.

Between A and B is seen a wide infundibular raphe.
Between 4 and 5 is seen a small interventricular foramen.

The first class of cases I wish to deal with is that described as
division of the right ventricle ; they are really cases in which the

(62)

MALFORMATIONS OF THE BULBUS CORDIS 9

infundibulum and body of the right ventricle have developed to a
normal extent but they have never completely fused, a constriction
remaining between them, representing the ventricular orifice of the
bulbus. A typical specimen is shown in figure 3. It is the heart of a
young man who sought treatment at the London Hospital, and was for
two months under the care of Dr. Percy Kidd, to whom and to Dr. F. J.
Smith I am indebted for the opportunity of examining this specimen.
The infundibulum is enormously dilated as may be seen from the figure,
and is sharply separated from the body of the right ventricle by a
muscular partition, which is perforated by an orifice, 10 mm. in diameter.
The orifice is surrounded by dense fibrous tissue which resembles, in
structure, the tricuspid valve. Just below the orifice is seen a small
interventricular foramen (see figure 3, between 4, 5). The infundibular
raphe (between A and B), ends at the bulbar orifice ; the septal bands A
and B are wide and well marked. The fibrous tissue round the constric-
tion I regard as a representation of the valves and fibrous tissue at the
lower orifice of the bulbus cordis (see figure i). This condition is
certainly not common, but in the same week as I obtained the specimen
figured here, 1 received another from Dr. John Hay of Liverpool. It
was the heart of a woman who had reached middle age ; the infundibulum
was smaller and its orifice not so narrow, as in the specimen figured
here. Altogether, I have had an opportunity of examining 15 specimens
of this type, and about 50 cases are recorded. In some of them the
constriction between the infundibulum and body of the ventricle is but
slight, and the fibrous tissue of the margin is represented by a ring of
irregular endocardial elevations. The explanation which has hitherto
been offered of such a condition is that it is the result of cicatrisation
following an endocardial lesion. Evidence in support of such a hypo-
thesis is entirely lacking.

The second class of cases which I now proceed to deal with are
common, and are usually described as congenital stenosis of the
pulmonary artery ; they are, in reality, due to an arrest of development
of the bulbus cordis. In the first class of cases, just described, there was
no arrest of development of the infundibulum — the condition was due
simply to a persistance of the lower bulbar orifice. In this group, how-
ever, the condition is different (see figure 4) ; usually the bulbus is only
partly expanded ; the endocardial lining is thick and frequently cor-

(63)

10 ' QUATERCENTENARY STUDIES IN PATHOLOGY

rugated, like the bulbus of fishes ; the lower orifice may be marked, as in
the case shown, by a fibrous diaphragm, muscular in structure at its
circumference. The muscular walls are extremely thick, as if the proper
amount of muscle had been provided for a fully expanded infundibulum.
The pulmonary orifice may be constricted ; the valves may be free but
are commonly imperfectly separated. A wide interventricular orifice is
commonly present, and death usually occurs before twenty. Death is

Figure 4, heart of a girl aged 7 years. The right ventricle has been opened by a triradiate
incision.

(i) Pulmonary artery (narrow, thin- walled).

(2) Aorta (large).

(3) Infundibulum below adherent pulmonary valves.

(4) (5) Orifice of infundibulum ; below orifice is a large interventricular foramen.

(6) Body of right ventricle.

(7) Tricuspid valve.

(8) Left ventricle.

(9) Right auricle.

frequently caused by an attack of bronchitis or of pneumonia, and in the
majority of cases recent vegetations are found round the fibrous margin
of the orifice. It is the occurrence of these vegetations which has
induced so many clinicians to accept endocarditis as the cause of the
entire condition ; it is evident that no endocardial lesion could produce

(64)

MALFORMATIONS OF THE BULBUS CORDIS II

this form of infundibulum ; the right interpretation appears to be that
these congenital constrictions are specially susceptible of becoming the
site at which vegetations form.

In the third class of cases of malformation of the bulbus cordis, the
arrest of development is more complete than in the second class, but of
course stages intermediate to these two classes occur. The condition is
shown in figure 5, where the heart of a boy aged 6 yrs. with the right
ventricle laid widely open is represented. The infundibulum is small

Figure 5, heart of a boy aged 6 with the right ventricle laid widely open.

(i) Pulmonary artery.

(2) Aorta also laid open ; below its orifice is seen a wide interventricular foramen

leading into the left ventricle.

(3) Infundibulum or bulbus arrested in its expansion.

(4) (5) Lower orifice of infundibulum or bulbus. 4 represents septal band A

in section (see figures 2, 3, 4).

(6) Body of right ventricle.

(7) Tricuspid valve.

(8) Left ventricle.

(9) Right auricle.

(7 mm. in depth, 4 mm. in width), lined by opaque thickened endocardium,
with a well-marked narrow lower orifice, formed by a fibrous thickening

(65) ^

12

QUATERCENTENARY STUDIES IN PATHOLOGY

of the endocardium. The upper or pulmonary orifice is guarded by two
valves partially adherent. The orifice is so narrow that very little blood
could have entered the pulmonary artery from the right ventricle. The
aorta opens directly from the right ventricle, and the septal bands

A

B

Figure 6, heart of a girl aged two years. A represents the heart unopened ; B the right
ventricle laid open. (The specimen was given to me by Dr. J. A. Milne.)
(i) Pulmonary artery.
(2) Aorta.

Position of infundibulum or bulbus.

Lower orifice of infundibulum or bulbus.

Body of right ventricle. (Above 5 in figure B is seen the interventricular foramen.)

Tricuspid valve.

Right auricle.

Left ventricle.

Left auricle.

(66)

(3)
(4)
(5)
(6)
(7)
(8)
(9)

MALFORMATIONS OF THE BULBUS CORDIS 13

marked A and B in figures 2, 3, form here functional parts of the wall of
the right ventricle, which throws its contents into the aorta. This class
of case and the one described just before this, make up over 50 per cent.
of the cases usually classed under congenital pulmonary stenosis.

In the cases dealt with so far, the bulbus has been arrested in its
growth and expansion after the upgrowth of the ventricular musculature,
but there still remains for consideration a very considerable group of
cases, in which the arrest of development is almost complete. In figure
6 a case of this kind is represented ; in that figure two views are given
of the heart of a girl aged two years. In A, the heart is represented
unopened ; the commencement of the pulmonary artery was represented
by a thread of tissue ; in B the right ventricle is opened, and all that is
seen of the infundibulum or bulbus is a slight cleft with fibrous lining
exposed by section of the wall of the ventricle, just in front of the aortic
orifice. The septal bands A and B of the infundibulum are shown
completely fused in the muscular wall of the ventricle, and unless one
knew where to look and what to look for, the remaining trace of the
bulbus would pass unobserved. The various conditions of the infun-
dibulum of the right ventricle, which I have just described, constitute by
far the great majority (over 90 per cent.) of the cases classed as
congenital pulmonary stenosis. It will be seen that the condition is
usually one which affects much more than the orifice of the pulmonary
artery ; it involves a complete segment of the right ventricle. Further,
the condition is not one which can be accounted for by a foetal endo-
corditis ; the form and extent of the lesion can only be accounted for by
accepting the hypothesis that the infundibulum is derived from the
bulbus cordis, a fusiform chamber, with a thick, endocordial lining, which
becomes incorporated in the right ventricle during the first month of
development, and whatever agent may be at work in producing this
malformation must exert its effect at this early stage of development.

Having now indicated the part of the right ventricle, which is derived
from the bulbus cordis, and the numerous malformations to which it is
liable, it is necessary to explain the manner in which the bulbus cordis
becomes restricted to the right side of the heart. When it is remem-
bered that the aorta, as well as the pulmonary artery is derived directly
from the primitive aortic stem, it is apparent that, in the mammalian
heart, both of these vessels ought to spring from the part of the heart

(67) F I

14

QUATERCENTENARY STUDIES IN PATHOLOGY

which represents the bulbus cordis. The aortic part of the bulbus
atrophies and disappears, and in order to understand this alteration, I
propose now to describe briefly the metamorphic changes which, thanks
to the labours of His, are known to occur in the human heart during the
third and fourth weeks of development. The condition of the human
heart during the fourth week is shown in figure 7 ; A represents the
condition about the beginning of that week, and B a stage towards the
end of the same week. Both are taken from illustrations given by His.

A

Figure 7, A, heart of a human embryo, 4*5 mm. long. (His).
B, heart of an "i 5 " long- (His).

Primitive aortic stem from which pulmonary artery and aorta will be developed.
Bulbus cordis.

Upper or aortic orifice of bulbus cordis.
Lower or ventricular orifice of bulbus cordis.
Ventricle.

Auricular canal (auriculo-ventricular junction).
Left auricle.
Right auricle.

(I)
(2)
(3)
(4)
(5)
(6)
(7)
(8)

To assist in the interpretation of these hearts, I give a figure of a fish's
heart (Echinorhinus spinosus) drawn from the same point of view — in
each case the sternal aspect of the heart is represented. In their shape
the ventricle and bulbus of the fish's heart recalls the stomach ; the
auricular canal (6) corresponds to the oesophagus; the auriculo-ventricular
junction, the oesophageal orifice, the ventricle, the cardiac part of the

(68)

MALFORMATIONS OF THE BULBUS CORDIS 1 5

stomach ; the bulbus, the pyloric part of the stomach ; the upper orifice
of the bulbus, the pyloric orifice. Lesser and greater curvatures are
distinctly marked, the lesser curvature being the result of a restricted
growth along one margin of the cardiac tube, which brings the upper
orifice of the bulbus towards and against the auriculo-ventricular orifice,
in the same way as an exaggeration of the contracture of the lesser

Figure 8, heart of a fish (Echinorhinus spinosus) from a specimen in the Museum of the
Royal College of Surgeons, England.

(i) Primitive aortic stem giving off branchial aortic arches.

(2) Bulbus cordis.

(3) Upper orifice of bulbus.

(4) Lower orifice of bulbus.

(5) Ventricle.

(6) Auriculai canal.

(7) Left auricle.

(8) Right auricle.

(9) Sinus venosus.

curvature of the stomach would approximate the pyloric and oesophapeal
orifices. In the fourth week, the parts of the human heart correspond
in shape and position to those of the fish just described (see figure 7, A,
B), there being this difference that the lesser curvature is undergoing a rapid

(69)

i6

QUATERCENTENARY STUDIES IN PATHOLOGY

atrophy, with the result that the lesser curvature of the bulbus is closely
applied to the lesser curvature of the ventricle, and the upper orifice of
the bulbus is being brought directly in front of, and applied to, the
auricular canal (see figure 7, A, B). The lesser curvature becomes rapidly
absorbed.

The changes which result from the complete atrophy of the lesser
curvature will be best appreciated by an examination of two diagram-

Figure 9, (A) Diagrammatic section of the embryonic heart showing the lesser curvature
by a black line.

(B)

The same showing the result of ;

an atrophy of the lesser curvature.

(I)

Sinus venosus.

(6)

Lower orifice of bulbus.

(2)

Auricle.

(7)

Upper 1.

(3)

Ventricle (left).

(/)

Orifice of pulmonary artery.

(3')

Ventricle (right).

(7")

It aorta.

(4)

Bulbus cordis.

(8)

Auriculo-ventricular orifice.

(5)

Primitive aortic stem.

(9)

The anterior cusp of the mitral valve.

(5')

Pulmonary artery.

(10)

The interventricular septum.

(5")

Aorta

matic sections of the embryonic mammalian heart which are reproduced
in figure 9. In each section the four chambers of the primitive heart are
shown ; (i) sinus venosus ; (2) auricle ; (3) ventricle ; (4) bulbus. In A
the lesser curvature is represented by a thick black line. The
atrophy of the lesser curvature is seen to result in the following

(70)

MALFORMATIONS OF THE BULBUS CORDIS 17

modifications:— (i) The adjacent sides of the bulbus and ventricle fuse
and rapidly disappear, and thus the bulbus comes to open freely into the
right ventricle (see figure 9, A, B) ; (2) the lesser curvature is represented
by only the base of the mitral valve (see fig. 9, B, 9) ; (3) the lower or
ventricular orifice of the bulbus is carried up to the base of the anterior
cusp of the mitral valve ; (4) the upper or aortic orifice of the bulbus
becomes applied to the base of the anterior cusp of the mitral. Further,
it is to be noted that the aorta arises from the part of the bulbus which
undergoes atrophy in the disappearance of the lesser curvature of the
heart (see figure 9, B, f). It is in this manner that the part of the
bulbus corresponding to the origin of the aorta becomes stretched and
disappears, while the part of the bulbus which remains, representing
almost the entire bulbus, becomes incorporated in the right ventricle.
Thus it comes about that the part of the bulbus cordis corresponding
to the origin of the aorta disappears completely at a very early date, and
usually not a trace of it is to be seen at the aortic opening of the left
ventricle. Occasionally remnants do occur ; all the cases showing such a
condition which I have had an opportunity of examining have been
regarded as malformed by the cicatrical effect of endocarditis, but a
closer examination of the condition, and a fuller understanding of the
development of the heart, show that such an interpretation is untenable.
In figure 10 is given a representation of the heart of a man, aged 20, who
died recently in the London Hospital. It shows a condition which I
interpret as being due to the persistence of the bulbus cordis
round the orifice of the aorta from the left ventricle. From
figure 9, B, it will be seen that such a remnant should form an
endocardial ring round the aortic orifice (the area derived from
the bulbus is stippled in figure 9, B). The case is one in
which there was an acute infection of both mitral and aortic valves,
with ulceration and formation of vegetations. But the stout fibrous
ring which surrounds the orifice of the aorta, below the aortic cusps, and
is continuous with the fibrous tissue of the heart from which the aorta
arises, could not be produced by any lesion which is known to occur in
acute or chronic endocarditis. On the mitral valve, the ring is represented
by endocardial elevations which, in section, are seen to be made up of,
not inflammatory tissue, but of one closely resembling Whartonian jelly,
a tissue in which branched cells are embedded in a jelly-like ground

(71)

i8

QUATERCENTENARY STUDIES IN TATHOLOGY

substance. The fibrous ring round the rest of the orifice has the same
structure as the mitral valve — fibrous tissue with the fibres lying in
parallel bundles, with cells arranged in definite lines between the fibres.

Figure lo, heart of a man, aged 20, who died in the London Hospital from acute endo-
carditis. The left ventricle is opened and exposed.
(i) Aorta.

(2) Infundibulum of right ventricle.

(3) Anterior cusp of mitral valve.

(4) Base of left ventricle ; between 4, 4, below the origin of the aorta is seen the

fibrous endocardial ring, which I interpret as a remnant of the bulbus cordis.

(5) Left ventricle.

(6) Top of interventricular septum, below fibrous ring.

(7) Mitral cusp of aorta, showing ulceration and vegetations.

(8) Left auricle.

(72)

MALFORMATIONS OF THE BULBUS CORDIS 19

Further, I have been able to collect nine cases exactly similar to this,
but the interpretation of the cause of the condition is rendered difficult
by the fact that in every case endocarditis was present. The ostium
bulbi on the right side of the heart, when it persists, is specially liable to
become the site of an endocardial lesion ; it seems to me that this
explanation may be also employed for the left side of the heart ; a
constriction such as is caused by a persistent part of the bulbus cordis
is especially liable to become the site of endocarditis. Further, the
constriction which is seen in this and in similar cases, occurs at the
position and in the form which one would expect from a persistent
remnant of a bulbus cordis on the left side of the heart.

It would carry this paper beyond its proper compass were I to
describe cases in which the bulbus cordis has atrophied round the
pulmonary artery, and not only persisted but attained a great muscular
development round the origin of the aorta ; in this form of maldevelop-
ment is to be found the explanation of the not uncommon condition of
transposition in the origin of the aorta and pulmonary artery. Nor do
I intend to deal with that form of atrophy of the aortic part of the
bulbus which leads to complete aortic stenosis, so common in Russia
and so rare in England. I have gone far enough to show that the theory
of foetal endocarditis cannot account for the various forms of congenital
lesion which I have shown to occur in the heart, and the causation of
which has to be sought for in the first month of foetal development.
Further, these malformations cannot be accounted for except on the
hypothesis that the bulbus cordis forms an intrinsic part of the mam-
malian heart ; the bulbus is not completely absent in the mammalian
heart, as has been the current opinion up to the present. That it does
exist in the form described we have as evidence not only its persistence
as the result of maldevelopment, but also the facts of embryology and
comparative anatomy.

To render the evidence more complete, that these lesions are not the
result of foetal endocarditis, but of some agent, the nature of which we
do not yet know, I propose to give in this paragraph the results of the
examination of the heart in a group of 23 malformed human foetuses
and infants. The malformations which these foetuses and infants
showed were one or more of the following : anencephaly, hydrocephaly,
spina bifida, umbilical hernia, atresia ani, cleft palate, hare-lip and stenosis

(73)

20 QUATERCENTENARY STUDIES IN PATHOLOGY

of the oesophagus. Of these 23 cases, 14 showed a malformation of the
heart ; of the 14, 10 showed lesions of the bulbis cordis ; in some cases
it was the pulmonary part of the bulbus which was non-developed, in
others it was the aortic. The important point is that these lesions were
similar to those I have just described in older hearts. It appears from
this evidence, then, that the malformations of the bulbus are the result
of the same agent as produces the condition of anencephaly, etc., with
which lesions malformation of the bulbus is so often associated.

(74)

THE ADMINISTRATIVE ASPECTS OF
TUBERCULOSIS.

By William Leslie Mackenzie, M.A., M.D., D.P.H., M.R.C.P.E.,

F.R.S.E.,

Medical Member of the Local Government Board for Scotland.

(75)

The Administrative Aspects of Tuberculosis.

I. The Tubercle Bacillus— Tuberculine— Heredity.

How does the problem of tuberculosis present itself to the adminis-
trative mind ? To that question I shall try to give some answer.

For the modern administrator, the history of tuberculosis began
when Koch isolated his bacillus. That the disease was an infection,
communicable from man to man, is a fact as old at least as the days of
Isocrates, and older. Through the ages, the belief in its infectivity can
be traced in literary and scientific records. The Nineteenth Century
cannot claim to have discovered the fact, nor can the Twentieth Century
yet claim to have exhausted the pathology of the disorder. But it
remains true that, for the ends of administration, the whole history of
the disease before Koch may be blotted out of our books. Even with
the isolation of the bacillus, the administrative problem was weighted by
a thousand irrelevancies. The pre-Koch pathology is far from dead.
It still perverts the clinical mind. It is still repeated in the text-books.
It still crowds the lectures with antiquarian rubbish. It clouds the mind
of the student with useless knowledge. It blocks the way to frankness
of outlook and precision of practice. Curiously, it has faded most
rapidly where the lay mind has had to be convinced. For, to teach the
farmer, or the salesman, or the butcher, or the dairyman, or the mother
of children, or any of the other innumerable units that constitute an
organized society, all the delicacies contained in the ancient theories of a
wasting disease would have been a hopeless and futile task. Even the
youngest medical officer of health — fresh, enthusiastic, full of Virchow
and not ignorant of Darwin — would have been beating his head against
the rocks had he tried to rouse in the lay mind any interest answering
to his own. Until Koch, the disease was too difficult, too complex, too

{77)

4 QUATERCENTENARY STUDIES IN PATHOLOGY

little understood, to be taught to any but technically-trained people.
But when Koch came, a note of hope rang round the world. He passed
through the fire of criticism, not scathless, but carrying with him his
cardinal fact — that where his bacillus was, there also was tuberculosis.
The word tuberculosis passed from the vagueness of speculative
pathology into the circle of positive science. It was henceforth to mean
something as definite as gunpowder, or oxygen, or steam. Forthwith,
tuberculosis became a doctrine that the lay mind could grasp. It could
be taught as easily as the multiplication table, and it could be shown to
be as practical.

So far, well. But this alone, though it excited the hopes of the
world and simplified the duty of the administrator, would not have
secured the growing interest of the layman. To him a new germ may
be an interesting curiosity ; he will listen to tales about it ; he will take
pride in repeating its name. But he is nothing if not practical. If you
cannot do him some definite good, you will tire his interest and you will
provoke him to reaction. Fast on the heels of the new bacillus came
the suggestion that the disease due to it was no longer hopeless and
incurable. Then the whole world began to ask for a miracle. It seemed
for a time as if the miracle had happened and the diseased were to be
made whole. The heart sank when the signs failed. But the miracle
had indeed happened, although the revelation of it was looked for too
soon. Science on the one hand, and, on the other hand. Nature came,
once more, together, and the open-air treatment became a fact. Mean-
while, science pushed forward more and more intensively until new facts,
new methods, new habits of the organism revealed themselves, and now,
after all, the tuberculine cure of tuberculosis is no longer a dream of
possibilities, but a definitely established doctrine. The conditions are
not so simple as the natural feelings led us to imagine ; but they are not
so complex as to have baffled the patience of research. The day is here
when, not as a vague belief resting on unsolved mysteries, but as a
permissible deduction from ascertained fact, the forecast of the near
future may be — tuberculosis will be extirpated.

So far, again, well. But in the popular mind there was another
obstacle. Biology, on the authority of great names, had left us with a
crude theory of inheritance. What could it profit that we isolated the
bacillus if the personal pedigree were bad ? Did we not hear tales of

(78)

THE ADMINISTRATIVE ASPECTS OF TUBERCULOSIS 5

families swept away, member after member, each when his day came?
Were we not filled with horror at the inherited taint? Did not the
insurance companies, do they not still, base their calculations on the
belief that a phthisical inheritance ought to mean a loaded premium ?
And in some sense, they are perhaps justified. But the countenance
began to lighten and the action to grow athletic when it was, again after
long research, made clear that tuberculosis is not inherited — that it is
mainly a thing of the environment. It is, in fact, a struggle between
two organisms, a lower and a higher. The lower is the invader, the
parasite ; but the higher has now become the master. The vague hopes
of the earlier days are now planted firmly on a basis of definite
inductions. The bacillus can be isolated ; it can be killed ; it can be
traced into a thousand by-ways ; it can be stopped at a thousand points
of its path from one mouth to another ; it does not pass from generation
to generation.

To the administrator, the isolation of the bacillus made the problem
simple. To the reformer, the growing belief in the non-inheritance of
the disease has offered a new basis of action. The reformer is justified
in taking as his objective a possible society without tuberculosis. The
administrator has now to devise the methods of attaining that end.

II. Tubercular Death-roll for Scotland.

How does it stand with Scotland ? Here I have no case to prove ;
I seek only to indicate what can be done. But a single figure may be
taken as a starting point. For the present period, some six thousand
persons die in Scotland every year of pulmonary phthisis ; some four
thousand die of other forms of tuberculosis. These figures are probably
less than the truth ; for good as our methods of diagnosis are, multitudes
of cases still escape scientific scrutiny. Probably, five times ten thousand
gives some approximation to the number of tubercular persons living in
Scotland. In this, however, our data are still insufficient. But for our
present purpose one solid fact is enough — ten thousand persons die of
tubercular diseases, and a vast number will soon follow them. The
administrative problem is how to delay for the longest possible time the
death of the infected, how to reduce the number of possible infections,
how most rapidly to convert a widespread plague into an occasional
outbreak.

{79)

6 QUATERCENTENARY STUDIES IN PATHOLOGY

III. Death-rates and Deterioration of Stock.

In passing, perhaps it is worth noting that to reduce the death-rate
of a disease does not mean to confer immortality on the patient. If he
does not die of phthisis, he will die of something else ; but it is
probable that if he does not die of phthisis, he will live at least a little
longer before he dies of anything else. So elementary a fact should
not be worth stating ; but it is worth stating, for I seem to detect in the
occasional disappointment with the slow fall of death-rates an
unexpressed, perhaps unconscious belief, that they ought to go down to
nothing per cent., which is immortality. This, of course, is not com-
patible with the conditions of life known in this world ; we must be
satisfied with deferred death. The important fact is that when we save
a case from death by consumption, we are effecting a real, not an
apparent, saving. For there is no evidence that the extirpation of
consumption means the establishing of any other fatal disease. It has
been contended that by saving a larger number of tubercular people —
of course the tubercular diathesis is itself an hypothesis only — we
ultimately contribute to the deterioration of the race by a geometrical
increase of feeble constitutions. The argument applies to every infec-
tion ; but by long tradition it has, in the popular mind, a special
application to phthisis. Personally, I have not been able to discover a
sound reason for the argument. If we make it cut the other way, we
ought, logically, to assume that every cause of death whatsoever permits
the survival of a greater number of deteriorated stocks. But it is
possible to classify causes of death into selective and non-selective.
There is, however, no obvious or immediate test to enable us to
discriminate which disease is selective and which non-selective. It is
pure hypothesis to assume that phthisis is one of the selective causes,
and that every person that dies from it is better dead than living.
Neither for phthisis nor for any other infection have we any data of the
slightest value to enable us to give a conclusive solution of that
problem. In our efforts to limit the spread of tuberculosis, we may
assume, meanwhile, that every reduction in the death-rate is a social gain.
It is pitiful to learn that superstitions about Natural Selection, which is
not a god, but an abstraction, sometimes divert public funds to other
uses. It is more pitiful to learn that men with an elementary training

(80)

THE ADMINISTRATIVE ASPECTS OF TUBERCULOSIS 7

in science are capable of offering those perfunctory half-thoughts as if
they were established truths. Before we determine dogmatically who
shall die, we should exhaust our means of discovering who can live.

IV. General Sanitation versus Direct Control.

Ever since Koch's discovery, conviction has been gaining ground
that the spread of tuberculosis can be limited by administrative
methods. In many parts of Europe and America, it has been so
limited. There is a good deal of evidence for the proposition that the
isolation of phthisical cases has materially reduced the total number of
cases. It is true that, for fifty years, the death-rate from phthisis in
Britain has fallen year by year until to-day it is only about 50 per cent
of what it was. It has been assumed, without much effort at analysis,
that this steady decrease has been the result of improved "general
sanitation." Doubtless, general sanitation has contributed something, if
under "general sanitation" we include the draining of soils, the sewering
of towns, the improvement in houses, the increase in cleanliness of
habit, and most of all the steady, remorseless, systematic campaign
against infection of every form. What destroys one infection destroys
another. Incidentally, in our efforts to limit typhus, typhoid, puerperal
fever, scarlet fever, diphtheria, septicaemia, pyaemia, and many other
infections, we have been dealing, intimately and in detail, with the same
conditions as the tubercle bacillus thrives in. In killing typhus and
typhoid, we have, doubtless, without intending it, killed also tuberculosis,
but the tubercle bacillus is a slowly invading and most persistent
parasite. It gets to places that few other parasites can invade. Every-
where it finds a nidus so easily that it is naturally the last to be
expelled. Precisely because it kills slowly, it kills most. That is
probably why, when most of the other parasites steadily fall back before
isolation, disinfection and prophylactic injections, phthisis needs more
determined and subtle dealing. But with every allowance for the
insidiousness of this slow parasite, we are now justified by the evidence
in our conclusion that by direct attack, as in typhus, typhoid, scarlet
fever and the others, we shall be able to reduce the spread of the disease
by securing that the patient shall confine his infection to himself. To

(81) G

8 QUATERCENTENARY STUDIES IN PATHOLOGY

this we are now able to add direct methods of cure. When cure and
prevention can, as they ultimately will, work perfectly together,
tuberculosis will fall back to the social status of plague and cholera in
the western world.

How this end may be attained, the pioneer agencies in every land
have already shown us. The curative side has been made manifest by
the rapid growth of the Sanatorium. The preventive side has yet to
receive its full share of attention. Here I propose to indicate how, with
our present legal resources, we may, in Scotland, turn our official
battalions against Pulmonary Phthisis, which is, for practical purposes,
the only grossly infective variety of tuberculosis.

Let it be assumed that there are three, roughly separated, classes of
cases — incipient cases, intermediate cases, and advanced cases. Let it
also be assumed that each case runs an irregular history, varying from
weeks to years. Let it still further be assumed that thousands of
patients, though affected in varying degrees with the disease, are yet fit
to pursue their callings and to maintain economic independence for
years. Here we have all the factors of the administrative problem.

To meet these what have we administratively ? We have sanatoria ;
general hospitals ; infectious disease hospitals ; poorhouse hospitals ;
methods of disinfection ; educational organizations ; housing improve-
ment schemes ; improved drainage, water and ventilation ; inspection of
factories, of workshops and of work-places ; vast legal powers for the
removal of nuisances, the prevention of overcrowding, improvement of
house structure, selection of house sites, cleansing of houses, cleansing of
persons, the isolation of patients, the supervision of dairies and the
milk supply, the examination of cows, the destruction of tubercular
meat, the maintenance of paupers and their dependents, dispensaries,
and many other minor powers and organizations. To see that the
laws are effectively applied for all these objects we have the local
authorities for public health, that is the town councils in towns and the
district committees in counties ; we have the parish councils, who must
deal with phthisis as one of the largest pauper-producing agencies.
In the sequel, I shall show how these official and unofficial agencies
may be brought into perfect correlation and so secure that one vast
organised army shall march forward to the extirpation of the enemy.

(82)

THE ADMINISTRATIVE ASPECTS OF TUBERCULOSIS 9

V. What are the Organisations Available?

The available organisations may be classified into Public and
Private. Public organisations are the Local Authorities for Public
Health in Town and County ; the Parish Councils ; the Charitable
Hospitals with their outdoor departments ; the charitable Dispensaries.
The Private organisations are : — the special Homes, Hospitals and
Sanatoria, devoted exclusively, or almost exclusively, to the treament of
phthisis. The campaign against phthisis ought, as Dr. R. W. Philip has
held for many years, to correlate all the organisations Public and
Private.

In Scotland, the Local Authorities for Public Health number some
313. Of these, the majority are Town Councils; the minority are
District Committees. Of the Town Councils many are small, they have
control only of small resources, and their efficacy for any extended
Public Health movement is correspondingly small. But in a large
number of cases they are incorporated with the District Committees,
which are the Local Authorities for Public Health in the counties. On
the other hand, the death-rate from Pulmonary Phthisis is usually
highest in the towns, and the burden on the Town Councils is, therefore,
greater. But the^ phthisis death-rate also stands high in the counties,
and in some counties it is much higher than in others. In the large
cities, the phthisis death-rate is highest, and in them, fortunately, the
resources are greatest of all. Obviously, the practical policy is for the
smaller towns to co-operate with the county districts. Above all it is
necessary for both orders of Local Authority — Town Authority and
District Authority — to avoid the mistake of attempting, out of their own
resources alone, to solve a problem that more than any other requires
combination and concerted action.

The Parish Councils number over 900. Their primary business is to
administer the Poor Law. They must take charge of the indoor and
outdoor poor. They must organise indoor and outdoor relief They
must have Poor Inspectors ; they must have Poorhouses ; they must
have Medical Officers of Poorhouses ; and they must have Medical
Officers for the outdoor poor. In Scotland, there are some 69 Poor-
houses. It is practically no direct part of the Parish Council's duty to
prevent infectious disease ; their duty is to prevent and to control

(83) G I

10 QUATERCENTENARY STUDIES IN PATHOLOGY

pauperism. But among the causes of pauperism one of the chief is
Pulmonary Phthisis. And, as it is the duty of the Parish Council to
take medical charge of the sick, they must incidentally treat phthisis.
This they cannot do without proceeding, to some extent, on the modern
lines both of prevention and of treatment, and on those lines they have
largely proceeded. For several years the Local Government Board
for Scotland have, through their Superintendent of Poorhouses,
systematically year by year inquired into the methods employed
in the Poorhouses for the isolation and treatment of phthisis and
other forms of tuberculosis. The result has been that, in the
majority of the 6g Poorhouses, segregation of phthisical cases is more or
less completely carried out. In several, the open-air methods have
been organised. The cases isolated are usually chronic, advanced and
dying cases, which, as a result of the relative segregation established by
residence in the Poorhouse, have, in greater or less degree, been
removed from the general social circulation. As time goes on, the
numbers so segregated will certainly increase. Every patient segregated
even for a time is at least one focus of infection temporarily extinguished.
But again as time goes on the Public Health relations of the disease will
become more manifest, and the action of the Local Authority for Public
Health will become more imperative.

Of the other public institutions, such as hospitals and dispensaries,
many make some effort towards treatment, and a few are reserved
exclusively for phthisis. Among the pioneers in this path is Dr. R. W.
Philip, Edinburgh, who, nearly twenty years ago, established the Royal
Victoria Dispensary for Tuberculosis and the Royal Victoria Hospital at
Craigleith. The details hereafter given of the proper functions of a
municipal or county dispensary are based on the results achieved by Dr.
Philip, who has time and again pressed upon the public the necessity for
official organisation and correlation of agencies. His fundamental idea,
as given in detail below, is that the Municipal Dispensary should be, as
it were, the Court of First Instance for every case, selecting those
suitable for Sanatorium or for palliative treatment, or for complete
isolation ; investigating the homes, controlling the industries, distributing
information, and, in general, serving as a directive bureau and exchange
for the record and correlation of all preventive and curative agencies.

It is obvious that the charitable organisations shade into the private

(84)

THE ADMINISTRATIVE ASPECTS OF TUBERCULOSIS II

organisations, and are capable of being worked in association with them.
Here it is unnecessary to give further detail.

What, then, can the Local Authorities for Public Health do ? It is
on them that the chief burden of the campaign must now fall ; for they
have the powers and the duty to deal with the disease.

VI. Disinfection.

The Public Health (Scotland) Act contains many clauses that deal
with disinfection. These are all available for Pulmonary Phthisis as for
any other infective disease. Probably they were drawn up primarily for
the suppression and prevention of the common infections like scarlet
fever, diphtheria, enteric fever, typhus fever, smallpox, etc. But, when
carefully examined, they are found applicable even to a peculiarly long
and variable disease like phthisis. The Local Authority can, and, when
required by the Local Government Board, must provide disinfecting
appliances and the officers necessary to work them (46).* They may
disinfect or destroy, as necessity indicates, bedding, clothing and other
articles that have been exposed to infection. These they may remove
and destroy or disinfect and return free of charge. For all these
purposes, they may combine with other Local Authorities.

The Local Authority may also cleanse and disinfect premises when
it is certified by a Medical Officer of Health or a qualified medical
practitioner that such cleansing and disinfection would tend to prevent
the spread of infectious disease (47). They may require householders or
others concerned to carry out the disinfection, or they may through their
own officers and at the public expense themselves disinfect the premises,
etc. To those familiar with the persistence of the tubercle bacillus, it
hardly needs to be said that, to be effective, disinfection must be very
thorough-going. Recently the perfunctory methods common in the last
generation, such as fumigation with sulphur, have steadily receded
before the more effective application of disinfectants by washing, moist
brushing, or spraying. For clothing, the methods now employed are
steeping in disinfectants of adequate strength, or passing through
saturated steam under pressure. In some places, for the disinfection of

*The numbers refer to the sections of the Public Health (Scotland) Act, 1897.

(85)

12 QUATERCENTENARY STUDIES IN PATHOLOGY

rooms infected by phthisis, the favourite method is washing the walls
with freshly made solution of bleaching powder (strength i per cent.).
In others the method is spraying of walls, floors, ceilings, furniture, etc.,
with sufficiently strong solutions of formaldehyde, or carbolic acid, or
corrosive sublimate, or one of the many effective disinfectants now
available. In a room infected with the tubercle bacillus, it is absolutely
essential that all the dust should be sterilized ; otherwise the room
cannot be regarded as safe. And it is not enough that the objects in
the immediate vicinity of the patient should be dealt with ; it is equally
essential that the floors, walls, furniture, etc., should be treated with the
same scrupulous detail as if the room were to be prepared for an aseptic
surgical operation. But if this is to be done, obviously it requires
skilled and experienced officers to do it.

Further, when a patient is treated at home, the clothing and the
room should be periodically disinfected. Works or workrooms where
phthisical patients are employed should be treated in the same way. In
no case should they be left to the chance and perfunctory methods of the
unskilled and uninformed.

Are we not asking too much ? Is it necessary, in practice, to lay
such stress on infected rooms and infected workshops? The evidence
leaves us no alternative but to say that it is absolutely necessary. Case
after case has been known to result from sleeping in infected rooms and
infected bedding. Of the cumulative proof towards establishing the
practical infectivity of phthisis, no part is more striking than the clinical
and experimental evidences gathered from infected sick-rooms.

Other Sections of the Act (51 and 53) confer special powers for
dealing with infected houses, or infected houses re-let without disinfection.

Other Sections (50 and 56) enable the Local Authority to disinfect
public places and control the public movements of infectious persons.
For instance, no person suflering from any infectious disease ought
wilfully to expose himself without proper precautions against spreading
the disease in any street, public place, shop, inn, hotel, church, or any
place used in common by persons other than members of the family or
household to which the infected person belongs. And, as by Section 50,
it is an offence to expose infected matter in ashpits, etc., clearly the
Local Authority may, if they choose, prevent phthisical persons from
scattering their infected sputum in public places. The practice of

(86)

THE ADMINISTRATIVE ASPECTS OF TUBERCULOSIS 1 3

uncontrolled spitting doubtless is a factor in perpetuating the circulation
of the tubercle bacillus. Spitting ought everywhere to be controlled.
Schools, churches, halls, theatres, markets, railway stations, railway
carriages, tramcars, omnibuses, hackney carriages, etc., etc., ought to be
periodically cleansed and disinfected.

The Local Authority should also make use of their powers under
Section 58, and see that no person suffering from Pulmonary Phthisis,
or living in a house infected with that disease, shall milk any animal,
or pick fruit, or engage in any occupation connected with food, or carry
on any trade or business in such a manner as to be likely to spread the
disease. The powers of this Section are of primary importance.

VII. Isolation and Treatnnent.

Note. — The remainder of this article, except the concluding paragraphs, is taken from
a circular issued by the Local Government Board for Scotland in March, 1906.

The provisions of the Public Health Act as to removal of cases of
infectious disease to Hospital (54) and as to the provision of hospitals
and houses of reception (66) are available for dealing with cases of
phthisis as with cases of other infectious diseases. These provisions are
comprehensive and elastic, and can be adapted in practice to any type of
case — incipient cases where the danger of infection to others though, for
the time, at a minimum, may suddenly become serious ; intermediate
cases, where the patients, still able to work, may, if uncontrolled, become
dangerous ; and advanced cases, where the patients, frequently unable
to attend to themselves, may be a source of grave danger.

Hospitals for Pulmonary Phthisis may be classified as follows : —

A. Curative Hospitals (Sanatoria) for early cases.

Here the object is, by generous and selected food, by life in the open
air, and by other forms of recuperative treatment, to develop the resistance
of the body to tuberculosis, to produce a cure of the affected lungs, and
to restore the patient to normal efficiency. To achieve these objects, it
is necessary to have the patient closely supervised and carefully nursed.
The conditions of sanatorium life are in many respects the reverse of
those in the homes whence the patients come. The patients themselves
require courage and perseverance. This treatment has been more

(87)

14 QUATERCENTENARY STUDIES IN PATHOLOGY

successful than any other hitherto attempted. It is now being supple-
mented in some sanatoria by other forms of special treatment. The
progress of the patient towards increased constitutional resistance can be
tested from day to day and week to week by methods of extreme
delicacy. Continuous and stringent medical supervision is essential.

It is obvious that, to produce satisfactory results, such highly
specialised Hospitals must be in the hands of specialists experienced
and skilled in the methods of scientific diagnosis and treatment.
Otherwise, patients and their friends will be disappointed. Accordingly,
individual Local Authorities, instead of attempting themselves alone to
provide the trained medical skill and scientific laboratories necessary for
effective management, should either utilise existing sanatoria or combine
with other authorities for a properly equipped sanatorium. It should
be part of the duty of the Local Authority, or Phthisis Committee, to
discover and to keep in touch with all available sanatoria, either in the
immediate locality or at a distance.

It may be added that, in structure, sanatoria may be of the simplest
and cheapest form. Wood and iron sanatoria have been provided at
Woodilee and Gartloch Asylums at approximately ^90 per bed. Open-
air shelters can be provided at a cost of from £6 to ^8 per bed, or
even less.

B. All- Day Hospitals.

In some sanatoria, patients attend all day for treatment, and return
home at night. This method has been found very satisfactory where
the patient's home is reasonably suitable. The resources of Hospital
treatment can thus be made available without any serious increase of
ward accommodation. Extra shelters are alone necessary. The
patients are not only educated in open-air methods ; they are definitely
treated with a view to complete cure. Doubtless, as the curative
methods are further improved, day-attendance at Hospitals will grow in
favour. The patients have the advantage of being nursed and medically
supervised through the greater part of the day.

C. All-Night Hospitals.

Frequently, the patient's home is unsuitable for living in over-night.
But his work may be in the open air, and in every respect compatible

(88)

THE ADMINISTRATIVE ASPECTS OF TUBERCULOSIS 1 5

with the treatment of the disease. In such cases, arrangements might
be made for his going to work during the day and sleeping in the
Hospital at night. He would thus be kept under continuous medical
supervision, and at the same time enabled to follow his occupation without
danger to others. This arrangement could be made at a Sanatorium or
other Hospital or Colony.

D. Convalescent Colonies and Homes — Work Colonies.

It has been found that patients, after treatment in a Sanatorium, are
apt to relapse when they return home. Accordingly, it is desirable to
provide Convalescent Work Colonies or Homes for such cases. Such
Colonies or Homes, if well organised, might be made almost, if not
entirely, self-supporting. The patients would be specially selected.
They would continue the open-air treatment. They would be provided
with light labour, as a preparation for their ultimate return to full work.
Certain satisfactory experiments have been made in this direction.

E. Hospital Wards for Educative Treatment and Control.

In several localities, vacant wards of the Hospital for ordinary
infectious diseases have been used for educative treatment of phthisical
patients. The most striking example is the town of Brighton. Here,
patients are admitted for one month. They are placed under full
Sanatorium treatment for the time. They are taught to sleep with
wide-open windows or in the outdoor shelters. They are taught how to
disinfect and dispose of their sputum. They are provided with sputum
flasks and paper handkerchiefs. They have their blood examined as a
test of their condition. Meanwhile, their homes, their clothing and bed-
clothing are disinfected. Their friends are instructed in the open-air
methods and in the precautions necessary for safety. Their employers
are communicated with and advised. At the end of the month in
Hospital, they return home. They are then visited periodically by
officers of the Medical Officer's Staff. They are further assisted when
necessary. In the event of death, disinfection is thoroughly carried out.

In this way, large numbers of cases are brought under systematic
control. The cost to the Local Authority of Brighton is relatively small.
This system, which was first elaborated by Dr. Arthur Newsholme,

(89)

l6 QUATERCENTENARY STUDIES IN PATHOLOGY

Medical Officer of Health, is capable of being organised, on some scale,
at trifling expense, wherever there is a vacant Hospital Ward and an
adequate medical nursing and public health service.

F. Hospital Wards for Isolatio7i of Advanced Cases.

Many patients are too advanced in the disease to leave their beds.
Their homes may be entirely unsuitable for safe nursing. Such cases
should be isolated. For this purpose, vacant wards of an Infectious
Disease Hospital may be utilised. Here, palliative treatment is possible.
There will be adequate medical supervision and nursing. Under very
simple precautions, friends and relatives may have access to the patient.

In many localities, notably in Lanarkshire, much has been done in
this direction. The isolation of such dangerous cases is a primary duty
of the Local Authority. By Professor Koch and many other investigators,
these advanced cases are regarded as the principal sources of infection.

It will thus be seen that every variety of case may be suitably dealt
with by the Local Authority.

VIII. Dispensaries for Pulmonary Phthisis,

" In towns and other thickly populated localities, where the number
of phthisical patients is large, the Local Authority will find it advisable
to institute a Dispensary or Dispensaries. In Edinburgh, the Royal
Victoria Dispensary for Tuberculosis, organised by Dr. R. W. Philip, has
worked successfully for eighteen years, and the suggestions here made
are largely based on the experience of that Dispensary. (See paper by
Dr. R. W. Philip, Edin. Med. fournal^ January, 1906.) On the Continent,
notably in France and Germany, Dispensaries have been of immense
value in the discovery of insanitary conditions, and in the organisation
of nursing service.

The work of a Phthisis Dispensary would include the following : —

A. Medical Examination of Patients — either at the Dispensary or

in their own homes.

B. Inquiry by a medical man, or qualified nurse, into the history

of the illness, the home-conditions, the economic condition
of the family, the occupation, the suitability or unsuitability
of the accommodation for home-treatment, &c.

(90)

THE ADMINISTRATIVE ASPECTS OF TUBERCULOSIS 1 7

C. Arrangements for providing medical treatment and nursing of

patients that could be treated at home without risk of
infection.

D. Dispensing of medicines, and disinfectants.

E. Arranging for periodic washing and disinfection of clothing

when the conditions render this necessary.

F. Selection of cases suitable for any of the forms of hospital

treatment enumerated above. This is one of the most
important parts of Dispensary work.

G. Distribution of printed information among patients and their

friends.

H. Arrangements for examination of sputum and other discharges.
I. Arranging for the removal of cases to hospital.
J. General medical supervision of cases not removed to hospital,
and general guidance in all matters concerning tuberculosis
and its control.
The Municipal or District Phthisis Dispensary ought to be the
Central Bureau of information. It should keep a register of all sanatoria,
hospitals, infirmaries, work-colonies, convalescent homes, parochial
hospitals, private houses, phthisis committees, and all other organisations
that, either within the district or without, can be made available for the
inhabitants."

IX. Notification of Pulmonary Phthisis.

" For the effective application of the Public Health Act to Pulmonary
Phthisis, a system of Notification is essential. In some localities, a
system of Voluntary Notification has been organised.

" But it is open to the Local Authority, with the approval of the
Board, to add Pulmonary Phthisis to the list of diseases notifiable under
the Infectious Disease (Notification) Act, 1889. The Board will be
prepared to give favourable consideration to any application for their
approval provided they are satisfied that the Local Authority are
in a position and are ready to deal effectively with the cases notified
to them. Notification of itself has no administrative value, and unless
it is to be followed by effective measures for curative treatment of the
patients and for prevention of the spread of infection the Board will not

(91)

1 8 QUATERCENTENARY STUDIES IN PATHOLOGY

feel justified in approving of the compulsory notification of cases of the
disease."

Long ago, when laymen and very few medical men were agitating for
the notification of the ordinary infections, the same arguments were
heard : the " liberty of the subject " always " blocked the way." To-day,
the subject's liberty is secured by the very Acts that seemed certain
to destroy it. The apparent coercion of the Notification Act of 1889,
and its precursors, was only the civil instrument for unveiling to the
citizen mind a new social duty, namely, the duty to save his fellow and,
therefore, himself, from the dangers of preventable infection. In the
Statute-book there is no Act that works with greater smoothness or
greater effect. And its application to Pulmonary Phthisis is only a
matter of detail. The principle is everywhere accepted. But Notification
is only one step. It is right to take it. It will be taken. But it must
be prepared for, it must be followed, by effective action in every case.
Meanwhile, cases are waiting in thousands to be received, and, for
the moment, compulsory notification is a secondary point of tactics.
In Scotland, compulsory notification of the common infections did not
become general until 1897. It does not yet, except in one or two places,
apply to whooping-cough, or measles, or diarrhoea. But when methods
of controlling those diseases are revealed, notification will be swiftly
invoked. In Phthisis, methods of control are obvious and effective, and
notification will soon be asked for everywhere as a simple necessity
of administration.

X. Organisation and Administration.

" It has been shown above that the Public Health Act applies to
Pulmonary Phthisis ; that, for a successful campaign against the disease,
the Local Authority must employ the various institutions and agencies
suited to the various types of patients ; that sanatoria, hospitals, and
dispensaries must be worked always in concert, and that a system of
notification, voluntary or compulsory, is indispensable. It remains to
add that an organisation so highly specialised demands the services of a
special Sub-Committee of the Public Health Committee.

This Sub-Committee — the Phthisis Committee — should have charge
of the Dispensary, which, as indicated above, should be the Administrative
Centre of the whole Organisation for dealing with Phthisis. The

(92)

THE ADMINISTRATIVE ASPECTS OF TUBERCULOSIS 19

Dispensary will normally be under the direct management of the
Medical Officer of Health, and will constitute a special department of
his work.

The organisation of hospital service — sanatorium and other— would
naturally be part of the duties of a Phthisis Committee ; but the details
will vary according to the institutions available in or for the District."

XI. General Sanitation.

" It should be unnecessary to remind Local Authorities that these
direct measures towards the control of Pulmonary Phthisis must be
supplemented by indirect measures, the unremitting and systematic
removal of nuisances, prevention of overcrowding, enforcement of good
ventilation, reconstruction of insanitary houses, improvement of insanitary
areas, improvement of drainage of soil and houses, stringent supervision
of meat, of cowsheds, of dairies, cleansing of streets, proper disposal
of refuse, &c. Direct prevention should go hand in hand with general
sanitation."

XII. Conclusion.

It is often said that the law as it stands is not adapted to the
administrative control of Pulmonary Phthisis. It is alleged that new
powers are necessary ; that the present powers may bear too hardly on
individuals or particular classes ; that the peculiar economic relations of
phthisis may involve such interference as would disturb the existing
organisation of society to its foundations. It is suggested that if the
clauses of our Public Health Act were enforced for phthisis as for
infections like enteric fever or small-pox, the lives of thousands would
be made intolerable ; that, in the great industries, workmen would
boycott their infected or infective fellow-workmen, and that the hardship,
the misery, the poverty necessarily resulting would be a greater evil than
the persistence of the disease or its slow evanescence. These fears date
from a time when the knowledge of infection and its habits was rather a
superstitious belief than a reasoned doctrine. They claim to be founded
in a sensitive care for the economic welfare of the working classes ; they
are more reasonably assigned to an uninformed imagination. At the
very moment when such fears are the commonplace of discussion, tens,
hundreds, thousands of cases are clamouring at the doors, thinking little

(93)

20 QUATERCENTENARY STUDIES IN PATHOLOGY

of " social ostracism," thinking a great deal of the treatment they need
and can secure only at the hands of the public organisations. Indeed,
the fear of social hardship is rapidly vanishing before the verifiable facts
of the open-air treatment and the growing evidence of anti-toxic cures.
The powers of our Acts are now, when cases are being handled by the
thousand, found to be after all nothing but the simple necessities of the
administrative problem. The Acts are strong, but the clauses are elastic.
The highly developed Public Health Service of Scotland has, for fifteen
years in the counties and for a longer period in the large towns, been
finding its way into the most intimate recesses of social life. What ten
years ago presented itself to the lay mind as a " medical fad " is now the
commonplace of every town council and every district committee. The
demand for isolation, for disinfection, for hospital treatment, comes no
longer from the medical side ; it comes imperatively from the side of
the patient. By the systematic administrative handling of the other
infections, the public mind has been educated to accept a similar treat-
ment of phthisis. The time has now come when the executive
authorities, who have many of them been doing a little, may, with
safety and the certainty of universal assent, make a great concerted
movement to limit the spread of the disease.

No longer is it necessary to deal with private organisations alone,
with charitable societies, infirmaries, or dispensaries. It is possible, it is
necessary, that we should put in motion the statutory organisations at
present charged with the control of infectious diseases. True, it is easier
to agitate for new legal powers than to execute in detail the powers we
have. It is easier to stir opinion than to superintend spade-work. The
one requires only an interest in the subject ; the other demands
systematic action day by day. But as the case stands, the general
conviction of the medical profession has at last found a response in the
general conviction of the lay mind. The public authorities are ready.
The vastness of the work to be done has somewhat appalled those
unaccustomed to handle so great a movement. But the necessity for
handling it has become irresistible. And, after all, when the campaign
is thought out, when the organisations available are contemplated, the
problem, in passing into detail, passes easily into practice. We cannot
at one stroke undo the work of centuries, but in a single generation we
can make an effective beginning.

(94)

PAROXYSMAL IRREGULARITY OF THE HEART
AND AURICULAR FIBRILLATION.

BY

Arthur R. Cushny, M.A., M.D.,

Professor of Pharmacology, University College, London;

AND

C. W. Edmunds,

Lecturer on Pharmacology, University of Michigan.

(95)

Paroxysmal Irregularity of tte Heart and Auricular

Fibrillation.

In December, 1901, our attention was drawn by Professor Peterson,
Director of the gynoecological clinic in the University of Michigan, to a
case of marked irregularity of the heart occurring in his wards. Careful
examination of the heart was made, and a large number of sphygmo-
graphic tracings were obtained. Unfortunately, we were not acquainted
at the time with Mackenzie's methods of taking the venous and liver
pulses, and failed to take advantage of this most valuable method of
analysing cardiac irregularity. The case is of itself of considerable
interest however, and the light which seems to be thrown on it by our
experience of irregularity in animal experiments, encourages us to put it
on record.

The history of the case supplied by Dr. Peterson is as follows : —

Mrs. H. H., widow, aged 64. Admitted to hospital, December 23rd, 1901, for
ovarian fibroid ; operation December 24th. Anaesthesia (ether) for i hour and 20 mins.
Condition good at end of operation, pulse 90. Nausea persisted throughout the 25th
and 26th, the patient vomiting repeatedly, but on the 27th some improvement was
noted in this respect. The pulse on the 25th and 26th was 90-100 at the visiting
hours, but the nurse stated subsequently that she had noticed frequent irregularities.

On the 27th the pulse* was 88 at 6 a.m. ; 57 at 8 a.m. ; 53 at 10 ; 86 at noon ; 51
at 2 p.m. ; 92 at 4 ; 86 at 6 ; 52 at 8 ; 82 at 10, and 54 at midnight. At 2 p.m. the
record states " pulse very irregular." Temperature 98°-99°.

December 28th. Slight nausea and vomiting occasionally. Pulse weak and
irregular. Sphygmograms taken in the afternoon indicate the same irregularities as
occurred on the 29th. The attacks of irregularity did not last long. Strychnine 1/30
grain every four hours.

* We attach no importance to these numbers as indicating the true rate of the heart, as
it is quite impossible to determine the pulse rate in the type of irregularity shown by the
sphygmograms. The numbers given, however, indicate periods of irregularity at 8 a.m. ;
2 p.m. ; 8 p.m. and midnight ; while the intervening counts indicate a return to the normal.

(97) H

4 QUATERCENTENARY STUDIES IN PATHOLOGY

December 29th, 8'3. Pulse very irregular at 8*30, but regular at 9'i5. In the after-
noon pulse tracings were taken every hour. At 1*45 great arhythmia ; 2*45 pulse regular ;
3'45 irregular again ; 4*45 irregular ; and this persisted through the rest of the day.

From December 30th onwards nausea and vomiting stopped, and general
improvement. Periods of irregularity became less frequent, and confined almost
entirely to the night time. The patient says she wakes up with a start, and heart
becomes very irregular at once.

On January nth, 1902, the heart was very irregular during the night, but after-
wards remained regular again until, on January 18th, she was discharged from the
hospital and reached home in the evening ; very irregular during the following night.
Patient states that she first noticed the irregularity of the heart 20 years ago during an
illness, and since then it has occurred whenever she was ill, excited or exhausted.
Vomiting, to which she is occasionally subject, is always followed by heart irregularity.
The tendency to irregularity seemed to increase about ten years before the operation,
as she observed that smaller occasions gave rise to it than previously. For two years
after the operation she suffered very little from her heart, but then some over-exertion
gave rise to cardiac irregularity again, and for the last year very slight exertion or
exposure to extreme heat or cold induces an attack.

The examination of the heart by the usual clinical methods, during the stay of the
patient in hospital, gave no clue to the cause of the irregularity. No enlargement of
the cardiac dulness could be made out nor were there any murmurs present, and no
objective signs of heart failure were present apart from the irregularity.

Analysis of the Sphygmograms.

The pulse tracings were taken with Jacquet's sphygmochronograph
recording 1/5 sees. The tracings obtained during the irregular periods
were all of the same type, and may be represented by some of those
taken on December 29th. The marker above the tracing records the
time in 1/5 sees., while the intervals between successive pulses are given
below in hundredths of a second.

Fig. I. — Pulse tracing at 1*45 p.m., December 29th.

(98)

PAROXYSMAL IRREGULARITY OF THE HEART

Fig. 2. — Pulse tracing at 2*45 p.m., December 29th.

Fig. 3. — Pulse tracing at 3*45 p.m., December 29th.

jBSwlEMBffMKMEbBElWIiWMKffilMwWMlMwJKOI^

Fig. 4.— Pulse tracing at 4-45 p.m., December 29th.

(99) " I

6 QUATERCENTENARY STUDIES IN PATHOLOGY

Of the four tracings, those taken at 1*45, 3*45 and 4*45 p.m., indicate
marked and continuous irregularity, while the intervening one at 2*45
presents a quite satisfactory, regular pulse of about 88 per minute. This
rate is considerably above the average for the age of the patient, but is
not to be regarded as abnormal.* The slight variations in the intervals
between the pulsations are not greater than those observed in normal
persons. Auscultation at this time indicated no abnormality in the heart
sounds, and, in short, the heart and circulation at this time appeared
perfectly normal.

The tracings taken before and after this period of regularity offer a
marked contrast, the pulse showing great arhythmia and great variations
in the strength of the beats. The general character of the tracings shows
a number of large pulsations occurring at irregular intervals, and having
interspersed among them smaller pulsations sometimes singly, at other
times in twos or threes or in groups of larger numbers. Examination
shows that the large excursions of the lever are not so simple as they
appear at first sight. For example, in Fig. i, the descent following the
second and third rise of the lever is interrupted by oscillations which are
obviously due to a beat of the pulse in each case too weak to make a
more definite record. This is perhaps better shown in Fig. 3, in which at
a and b the large excursion of the lever is followed by a weaker contrac-
tion, while at d the descent of the lever only shows a slight oscillation,
yet there can be no question that there was here a secondary contraction,
similar in character to those at a and b, but so weak that it failed to
cause even such a beat as is recorded in ^. In the interval c, on the
other hand, there was probably no interposed systole of the ventricle.
When the stethoscope was applied over the heart there was found a
duplication of the first sound in such intervals as have been recorded at
a^ b and d, while there was no such duplication at c. At ^, Fig. 3, two
small elevations follow each other, and at /, a third one occurred which
was too feeble to do more than arrest the fall of the lever for a moment.
A little later, in Fig. 3, a succession of eleven rapid weak contractions are
seen. Counting these weak and almost unrecorded contractions, the
pulse is found to be beating in the irregular periods at 140-150 a minute.
On auscultation the heart sounds were found to be very much more

* Mackenzie. The study of the pulse, p. 53.
(100)

tAkOXVSMAL IRREGULARITV OF THE HEART ^

rapid than the pulse taken by the finger which, as has been stated, was
given as about fifty a minute during the irregular periods.

Even when these small beats are taken into account, the rhythm of
the heart is very irregular ; for example, during a succession of rapid
contractions in Fig. 3, the interval between successive pulses varies from
0*25 sec. to 0'4i sec. There exists no such "regular irregularity" as has
been described under the terms Pulsus bigeminus or trigeminus. In
these one normal pulsation is followed regularly by one or two move-
ments ot smaller dimensions, but in our tracings no such sequence is to
be made out.

Even more striking than the irregularity in the pulse rhythm in this
case is the marked variation in the strength of the beats. This variation
in strength is of course a very common phenomenon in irregularity of
the heart. In the numerous tracings published of late years by Wencke-
bach, Mackenzie, Cushny and others, this variation in strength is
universally noted. But in most of these the strength of the beat varies
directly with the length of the preceding interval ; the ventricle accumu-
lates more energy during a long interval, and, at the same time, receives
more blood from the auricle, and these two factors together cause a very
powerful contraction and a very large output of blood. In this case,
however, the size of the pulse bears no relation to the length of the pre-
ceding interval. It is true that when there has been a great fall in the
pressure from the absence of or the inefficiency of the preceding beat, the
next contraction is often very large, but this is obviously not the
determining factor in most instances, for a strong contraction is often
preceded by a shorter interval than a smaller one, and, in other cases,
two intervals of equal length are followed by pulsations of very different
height. The variations in the size of the pulse must therefore be due not
to different degrees of recuperation in the contracting power of the
ventricle, but rather to variations in the amount of blood which it
receives during its diastole from the auricle. The amount of blood
entering the ventricle may be altered by a variety of causes. One
condition which occasionally causes pulses of different sizes is where the
ventricle and auricle are beating at different rates. For example, if the
auricle beats three times to the ventricle's twice, one of the auricular
beats discharges no blood into the ventricle, because the auricle contracts
against the contracted ventricle. The next pulse is correspondingly

(lOl)

S QtJAtEROENTfiNAkV StUDtES ik PAtHOLOGV

small. This condition, however, leads to a rhythmical variation in the
strength of the pulse, as has been shown by one of us in the case of
certain poisons which lead to a ventricular rhythm independent of and
different from that of the auricle.* No such rhythmical variation in the
size of the pulse is to be made out in this case, so that the condition is
obviously different from that occurring in animals under these poisons.
An irregular and intermittent beat of the auricle might, however, cause
the occasional discharge of blood into the ventricle during the diastole,
and thus lead to a large pulse wave from the next ventricular contrac-
tion ; the ventricular rhythm must then be much more rapid than
that of the auricle, and, of course, must be developed indepen-
dently. Another factor which may determine the variations in the
strength of the pulse in this case is imperfect diastole of the ventricle.
The blood can flow from the auricle only when the intra-auricular
pressure is higher than the intra-ventricular, and, if the ventricle remains
half-contracted during the diastole, the inflow must be small and the
next pulse correspondingly low. If this be the correct explanation of
the variations in the size of the pulse, these are, of course, due to the
ventricle alone. The auricle might be beating at the same rate as the
ventricle, but unable to empty itself against the half-contracted ventricle,
or it might be entirely inactive, the blood simply entering the ventricle
from the auricle through the pressure exerted on it by the right heart.

Which of these two factors determines the variations in the size of
the pulse we are unable to determine from the evidence of the tracings.
If the venous pulse had been registered simultaneously with the arterial,
much light would have been thrown on the point, and it would probably
have been possible to decide whether slow, intermittent, irregular con-
tractions of the auricle are the explanation, or variations in the degree of
relaxation of the ventricle. In the absence of such evidence, the latter
appears the preferable. It is a feature familiar to everyone who has
watched an irregular heart in the living animal, while it is extremely
rare to find the auricle beating so much more slowly than the ventricle
as would be indicated by the number of full pulses.

The regular beat of the ventricle which is registered by the sphygmo-
graph in normal persons follows an impulse which originates in the
" rhythmic area " around the great veins of the heart, and is propagated

* Journal of Physiology^ xxv., p. 49.

(102)

PAROXYSMAL IRREGULARITY OF THE HEART 9

to the ventricle by way of the auricle. Two factors are thus involved
directly ; the discharge of the impulse from the rhythmic area must be
regular, and its path through the auricle unobstructed ; and the ventricle
must always be in such a condition that it can respond to the stimulus,
while its irritability is too low to permit of its initiating a contraction
spontaneously. Irregularity of the heart arises either from irregularity
in the discharge of impulses or from an abnormal condition of the
ventricle which leads to spontaneous contraction before the descending
impulse reaches it ; in rarer cases the ventricle is unable to respond to
the descending impulse and a beat lapses. Careful examination of the
pulse often determines whether the irregularity arises from ventricular
disability or from the irregular action of the originating area.* For in
intermittent pulse due to ventricular lesion the intermission is almost
exactly twice the interval between two normal beats, while, if the dis-
charging area be at fault, the intermission is considerably shorter. The
fact that the intermission due to ventricular abnormality must be a
multiple of the normal length of a pulse wave, follows directly from the
consideration that the impulses descending through the auricle are
undisturbed in rhythm, and the intermission can end only on one of
these reaching the ventricle and originating the first normal beat after
the irregularity. That the intermission is shorter when it is due to an
irregular discharge from the rhythm-giving area was first deduced from
animal experiments, and has been found to conform with clinical experi-
ence in too many cases to permit of further doubt.

In our case of irregularity there is no reason to suppose that the
ventricle failed to respond to the impulses it received. The length of the
apparent intermissions or of the periods of short quick pulses is not a
multiple of the single beats. For example, in Fig. i, the first single beat
is 0*44 sec. in duration, and if the next intermission had been due to the
failure of the ventricle to contract on the next regular stimulus its length
would have been 88 instead of 73. Similar reasoning in each case shows
that the irregularity is due, not to variations in the ventricular receptivity
of stimuli, but to an irregular discharge of impulses or to an irregular
transmission through the auricle or auriculo-ventricular fibres.

* Cushny. Journal of Exper. Med, , iv. , p. 327.
Mackenzie, loc. cit.^ p. 297.
Wenckebach. Ztschr. f. klin. Med., 1899, xxxvi., p. i8i.

(i03)

lO QUATERCENTENARY STUDIES IN PATHOLOGY

The question arises whether these irregular impulses originate from
the same point as those in the normal regular pulse, or whether the
irregularity of the discharge is due to the fact that some other point takes
up the function, and, being only imperfectly adapted to it, issues only
irregular impulses. The latter view has been suggested by Mackenzie,*
who holds that in certain forms of continued irregularity the ventricle, or
the auriculo-ventricular fibres, assume an automatic or endogenous
rhythm, while the auricle may assume the ventricular rhythm, or may
remain quiescent or "paralysed." Mackenzie supports this view by a
careful analysis of his tracings, and, in the particular cases observed, the
interpretation may be correct, although an alternative explanation might
possibly suffice without the necessity of adopting the inversion of the
cardiac rhythm postulated by him. But in animal experiments, in which
the ventricle has been induced to take up an endogenous rhythm by
means of drugs, no such irregularities of the heart rhythm are observed,-]-
and, in numerous unpublished experiments in which rapid induced
shocks were applied to the ventricle, one of us was able to invert the
cardiac rhythm, the ventricle giving the rhythm to the auricle, without
rendering the rhythm of either irregular after the first few seconds. An
endogenous ventricular rhythm may therefore be perfectly regular in the
mammalian heart, and it cannot be assumed per se to be the explanation
of irregularity in the tracings under discussion.

In our tracings, then, there is nothing to suggest that the ventricle
originated the rhythm anywhere ; although we cannot exclude the
possibility of the whole phenomenon being due to endogenous ventri-
cular influence, it seems rather to have merely responded to the impulses
which it received at irregular intervals from the auricle. Not infrequently,
the impulses followed each other so rapidly that the ventricle could not
dilate sufficiently in the interval to permit of the entrance of blood from
the auricle. In other cases blood failed to enter in quantity, although
there seems to have been sufficient interval, and here the ventricular
contraction caused a very small pulse wave.

The condition of the auricle and rhythmical area in the irregular
periods appears to have been one of extreme activity if, as we believe,
the rhythm in this case was given to the ventricle by some other part of

* Study of the pulse ; BrzL Med. Journal, March 5th, 1904.
t Cushny. Jourtial of Physiology^ xxv. , p. 49.

(104)

PAROXYSMAL IRREGULARITY OF THE HEART II

the heart. This impulse-emitting area was obviously in a state of great
irritability, for the impulses amounted to 150 per minute, and at the
same time the rhythm was extremely irregular.

The sudden onset of irregularity and its temporary character suggest
that while there may be cardiac conditions favouring irregularity, the
actual change from the regular beat must be due to affection of other
organs. And the fact that it accompanied the nausea and vomiting, and
tended to recur when no exertion had been made, and especially in the
night time, suggest that the immediate cause of the irregularity was
some abnormality of the central nervous system acting on a damaged
heart. We are aware that the tendency at present is to regard such
conditions as of peripheral origin, and to exempt the central nervous
system from responsibility in such phenomena. At the same time, the
history of the case inevitably suggests this view, and we hope in the
following pages to show indisputably that the central nervous system is
capable of giving rise to very marked abnormalities of the heart action.

Animal Experiments.

In the course of the long series of experiments and demonstrations
on the dog's heart carried out in the pharmacological laboratory of the
University of Michigan, during the last ten years, it has happened
occasionally that on opening the chest the heart was found to be beating
very rapidly and irregularly. Doubtless other workers in this field may
have had similar unfortunate experiences, but we are unaware of any
recorded cases. Our dogs were anaesthetised with morphine 0'2-0'3 G.
hypodermically, and chloretone administered by the stomach tube. The
operation consisted in performing tracheotomy, prolonging the median
incision to the lower end of the sternum, sawing through the sternum
along its whole length and hooking the two sides of the chest apart, thus
exposing the pericardium, which was opened. The myocardiograph
was then applied to the ventricle and auricle, and tracings taken on a
kymograph with smoked paper. The anaesthesia was invariably deep
enough to prevent any manifestation of pain, and the eyelid reflex was
absent before the incision was made. Sometimes, however, spontaneous
respiratory movements returned while the sternum was being cut through,
or, if these were present before, they became quickened and deepened.

(105)

12 QtJAtERCENTENAkV STUDIES IN PATHOLOGY

And it was soon noted that when this change occurred, the tendency to
cardiac irregularity was much greater than in those experiments in which
more profound anaesthesia had been induced. Attempts were made to
record the changes in the pulse during the operation, but in the experi-
ments in which this was done, the irregularity was not developed. In
several cases, however, it was noted that the heart rhythm was normal
before and during the first part of the operation, but, that when respiratory
movements were induced by cutting through the sternum, the pulse,
which had previously been of the usual slow vagus type seen in the dog,
suddenly became accelerated and irregular. On examination of the
heart in those cases, before the apparatus was applied to it, the ventricles
were found in rapid, irregular movement ; the relaxation was often very
imperfect between three or four successive contractions, and then more
complete for one or two beats. The impression was given that the
ventricles were responding to a very rapid series of impulses which
prevented their diastole, and that they could only relax when their
irritability was reduced by fatigue, and then the more complete diastole
followed. The auricles were widely dilated, and no systole occurred in
them ; they were not wholly paralysed and inactive however, for, on
close inspection, the fibres proved to be in a state of continual inco-
ordinated contraction, each part of the auricle undergoing continuous
fibrillary contraction independent of all the other parts. The heart was,
in fact, in the condition known to physiologists as auricular delirium or
fibrillary contraction in the auricle.

Auricular delirium is not infrequently produced by poisonous doses
of bodies of the digitalis series,* and is very readily induced by rapid
induction shocks passed through the auricular walls in animals. Several
papers -|- have appeared recently dealing with the subject from the
laboratories of Berne and Brussels, and we also have devoted some
attention to it. In this condition, the ventricle beats much more rapidly
than normally, and the beats are irregular in rhythm, and vary in the
degree of systole and diastole. Stimulation of the vagus, in our experi-
ence, which coincides with that of Kronecker and Spallitta, has no effect

* Cushny, Journal of Exp. Med.^ ii., p. 233.

t Kronecker and Spallitta. Arch. Internat. de Physiologies ii., p. 223.
Philips. Ibid., ii., p. 271.
Fredericq. Ibid.^ ii., p. 281.

(106)

I'AROXVSMAL tRRfeGtJLARltY O^ TttE HEART 1^

whatever on the movement of the auricles, while the ventricle is rendered
slower and more regular or may be arrested completely. Philips believes
that the fibrillation of the auricle is lessened by vagus stimulation. (It
may be mentioned that the tremor in the tracing of the auricle during
delirium often appears less marked under vagus stimulation, but this is
due to the fact that the ventricle is slower and communicates less
movement to the auricle ; the fibrillation of the auricle causes no
movement of the lever in itself) The more regular contraction of the
ventricle under vagus stimulation is obviously due to the fact that the
impulses from the auricle pass less readily through the block ; when they
are altogether excluded by section of the auriculo-ventricular band, the
ventricle beats regularly (Fredericq.) The pulse was recorded by Philips
as well as the movements of the ventricle and auricle, and during the
auricular fibrillation it is accelerated and very irregular, as we have also
observed in our experiments. See Figs. 5, 6.

Fig. 5.— Tracings (Hurthle's tonometer) from the carotid of a dog. The lower tracing is
the normal ; the upper one was taken during auricular fibrillation from rapid electrical stimula-
tion of the auricle.

Stimulation of the dog's auricle with rapid induced shocks causes
fibrillary contractions, which pass off when the current is shut off, or very
soon afterwards. On repetition several times the interval between the
cessation of stimulation, and the return of the normal auricular contrac-
tions, becomes longer, until finally it may continue for an hour or more
after the electrodes have been removed. Such cardiac depressants as
chloroform have no remedial effect, even when pushed until the ventricle
almost ceases under them. The only way in which we have been able to

(107)

14 QUATERCENTENARY STUDIES IN PATHOLOGY

restore the auricle to regular contraction in persistent fibrillation — and
this is not successful except occasionally — is by the application of cold
directly to the heart. Large quantities of chilled salt solution poured on
the exposed heart sometimes restores it to regular contraction.

The auricular delirium, which we have described as occurring in dogs
during the operation, was obviously the result of sensory stimuli carried
from the chest wall to the central nervous system. This might affect the
heart either by inducing changes in the blood pressure (shock) or by nervous

Fig. 6. — Tracings (Hurthle's tonometer) from the carotid of a dog. From ^ to .5 in the
lowest tracing the auricle was stimulated with a very weak tetanising current, and imperfect
fibrillation resulted at intervals. From B onwards, the stimulation ceased. In the middle
tracing a stronger series of shocks was applied to the auricle at C, and it immediately passed into
fibrillation. The highest tracing was taken during prolonged fibrillation from auricular stimulation.

impulses passing along the cardiac nerves to the heart. In the descrip-
tions of shock given in surgical textbooks, the heart is often said to be
irregular, but personal enquiry among our surgical friends failed to elicit
any information in regard to the character of the irregularities, and we
were led to the conclusion that cardiac irregularity in shock is extremely

(108)

PAROXYSMAL IRREGULARITY OF THE HEART 1 5

rare or only seen in moribund cases. We are therefore disposed to
attribute the condition observed in our experiments to reflex influences
on the heart. These impulses must reach the heart either by the vagus or
by the sympathetic fibres. And, as a matter of fact, we have observed
one instance in which vagus stimulation was followed by delirium. In
this dog, the heart was peculiarly irritable and much accelerated. The
myocardiograph had been attached to the right ventricle and auricle, and
a tracing was being taken. The effects of electrical stimulation of one
vagus in the neck were being demonstrated, and the usual response
followed — arrest of the auricle and ventricle in diastole. But when the
current was shut off, the heart, instead of returning to its normal rate,
passed into fibrillary contractions in both auricle and ventricle. This is
the only instance in which we have seen the unpoisoned heart affected in
this way from the cessation of inhibition. In several experiments in
which the irritability of the heart was exaggerated by large doses of
members of the digitalis series or of barium, section of the vagus led
immediately to fibrillary contractions in the auricle and ventricle.

As has been stated, we have obtained no tracings of the pulse in
dogs in which the heart passed into auricular delirium during the
operations. The phenomenon is a rare one in our experience, although
we have observed it sufificiently often to have no doubt as to its
occurrence, and as to its association with the operation. In several
experiments the. pulse was recorded during the operation by means of
Von Frey's or Hiirthle's tonometer, attached to the carotid artery, and
the anaesthesia was maintained at a point at which, while pain was absent,
the reflexes were retained as far as possible. In none of these, however,
was the auricular delirium developed, although the pulse was much
accelerated ; occasionally intermissions of the pulse occurred of the
auricular type, indicating perhaps a tendency to fibrillation. The
acceleration was very slight, or entirely absent in experiments in which
atropine had been previously injected in quantities sufficient to paralyse
the cardiac vagus. The acceleration in dogs without atropine was thus
due for the most part to lessened vagus tone, from the centre being
inhibited by impulses from the seat of operation. It seems likely that the
auricular delirium observed in rare cases in dogs during the operation is
similarly to be explained by the disappearance of inhibition ; the auricles
must in these cases, however, be in some quite unusual condition of
irritability. (109)

l6 QUATERCENTENARY STUDIES IN PATHOLOGY

Of course we cannot claim to have shown definitely any connection
between this type of irregularity in the dog's heart and that in the case
described. At the same time there exist similarities between them, and
the sudden onset of the irregularity in each suggests a common cause ;
in the dog the lesion acts through the central nervous system, and the
history of the patient suggests that here also the irregularity was of
central origin. We had hoped to have the opportunity of continuing
the investigation in other cases of irregular heart, but as circumstances
preclude our working further together, we have decided to put this
imperfect investigation on record, in the hope that others may be
interested in the suggestion made, and may be able to prove or disprove
its correctness.

Summary.

A case of paroxysmal arhythmia, with marked acceleration of the
heart, is described, and it is shown that the irregularity is due to irregular
discharge of impulses and not to defects in the contraction of the
ventricle, which appears to respond to the impulses received. A similar
form of irregularity in the dog's heart is described, and it is shown that
this is due to the ventricles receiving irregular stimuli from the auricle,
which is in a state of fibrillation. This form of arhythmia is shown to
occur in the dog occasionally from peripheral irritation, and it is probable
that this gives rise to reflex inhibition of the vagus centre, which, acting
on an abnormal heart, causes auricular fibrillation. The suggestion is
made that in the patient described, and in other cases of paroxysmal
arhythmia, the condition is due to auricular fibrillation, possibly from
inhibition of the vagus centre.

(no)

RESEARCHES ON CERTAIN PROBLEMS OF
PLAGUE IMMUNITY.

By George Dean, M.A., CM., M.B.,

Bacteriologist-in-Charge, Serum Department^ Lister Institute of Preventive Medicine ;
Formerly University Assistant and Additional Examiner in Pathology, Aberdeen University.

(Ill)

Researctes on Certain Problems of Plague Immunity.

The object of the present communication is to record the results of
certain researches on Plague Immunity which have been carried out in
the Plague Department of the Lister Institute.

The chief problems dealt with in this contribution may be arranged
as follows : —

1. Does the B. pestis produce a toxin capable of demonstration and

suitable for experimental purposes ?

2. If such a toxin is obtainable, does its injection into the animal

body result in the production of an antitoxin, and in how far
does the ordinary anti-plague serum, which undoubtedly
confers protection on such animals as rats and mice, owe its
immunising property to an antitoxic action ?

3. By what mechanism does an anti-plague serum exercise its anti-

infectious effect ?

It is unnecessary here to enter into all the details of the methods of
preparing anti-plague serum, but for the sake of clearness in discussing
certain of the questions connected with the above problems it seems
desirable to give a brief outline of the subject.

As will be readily understood the most rigid precautions are necessary
to obviate the risks of infection during the process of immunisation.

The ordinary procedure for obtaining an anti-microbial serum is
employed. The animals first undergo a course of intramuscular, or
intravenous, injections of killed cultures of the bacilli, till a certain degree
of immunity is acquired. In the case of the plague bacillus this pre-
liminary stage in the immunisation occupies a considerable period,
extending from four to six months, or even longer, the slow absorption of
the oedematous and indurated swellings which result from the sub-
cutaneous and intramuscular inoculations being chiefly responsible for

(113) I

4 QUATERCENTENARY STUDIES IN PATHOLOGY

the prolongation of the process. On the other hand, considerable risks
of thrombosis, causing sudden death, are attached to the intravenous
injection of killed cultures. This danger appears to be much greater
with killed than with living cultures, and is probably due to the setting
free by the heating to which the culture is subjected of a ferment or of
nucleo-protein.

The further stage in the process of immunisation consists in giving
the animal a series of intravenous injections of living virulent bacilli.
The general experience has been that the injection of even a small
quantity of living culture greatly increases the protective action of the
serum.

Roux and Dujardin-Beaumetz at the Pasteur Institute Paris, and
Calmette at the Lille Institute, prefer intravenous injections throughout,
and use bacterial emulsions, prepared from agar cultures, for both the
early and later stages, whereas Tavel Krumbein and Glucksmann at
Berne use old broth cultures, killed by heating to 65-70° C. (Hafifkine's
Prophylactic Fluid), injected subcutaneously or intramuscularly in the
early stages, and young broth cultures injected intravenously in the later
stages of immunisation.

When " anti-plague serum " is referred to in this paper, a serum
prepared by one or other of these methods is indicated. In certain of
the following experiments the serum used was from animals which had
been treated with heated or unheated filtrates only.

Brief Review of the Literature on Plague Toxin.

Yersin, Calmette and Borrel (1895) stated that the filtrates from broth
cultures of the B. pestis have no toxic action on experimental animals,
but that emulsions from agar of the bodies of the bacilli, which have been
killed by heating to 58° C, cause the death of guinea-pigs and rabbits
when injected subcutaneously or intraperitoneally.

Roux is stated by Metchnikoff (1897) to have obtained very
powerful plague toxins by employing an organism, the virulence of
which had been elevated by animal passage in collodion sacs. The
filtrate from a bouillon culture, precipitated by ammonium sulphate, gave
a toxin of which the lethal dose for a mouse was one-fourth of a milli-
gramme, for a rabbit four centigrammes. The guinea-pig is the most

("4)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 5

resistant of the usual laboratory animals. The toxin is destroyed by a
temperature of 70° C.

Lustig and Galeotti (1897) obtained toxic substances by extraction
with a one per cent, solution of caustic potash, and subsequent slight
over-neutralisation with a one-half per cent, acetic acid solution. The
resulting precipitate is highly toxic, and is employed as a vaccine.

The German Plague Commission (1899) points out that certain
animal experiments indicate that the toxic action is not an important
feature of the pathogenic effects produced by the plague bacillus. They
refer to the condition observed in monkeys, in which a great part of the
body is affected with oedema, and where the exudate swarms with bacilli,
and yet in these animals strikingly few symptoms of toxaemia are
displayed.

Notwithstanding this view of the Commission, it cannot be denied
that toxaemia is a marked feature in many cases of human plague. In
a considerable proportion of all cases the lesions remain localised in the
affected lymphatic glands, no bacilli being present in the blood-stream,
and yet these cases die with the symptoms of profound toxaemia. It is
certain that the pathogenic effects of the plague bacillus are the result of
toxic action.

The English Plague Commission (1898-99), in speaking of the subject
of toxin, states, " We, for our part, have tested the effect which is exerted
by heat upon the filtrate from a plague culture. It results from these
experiments that the soluble toxins which are, under certain conditions,
developed in the fluid nutrient medium in which the plague bacilli are
grown, are apparently not affected by exposure to a temperature of even
100° C."

The Austrian Plague Commission (Albrecht and Ghon, 1900),
obtained results in the main similar to Markl's, to be described directly.
Markl's work was done in part with the Commission. They obtained
toxic filtrates from broth cultures, the toxicity of which increased with
the age of the culture up to a certain period, two to three months, after
which no increase took place. The filtrates from cultures even only five
days old were toxic, and the toxins differed only quantitatively from
those obtained from older cultures.

The bacillary bodies, from cultures on agar, which had been killed by
heating to a temperature of 55°-6o° C, produced toxic effects similar to
those observed with the filtrates. (115) I J

6 QUATERCENTENARY STUDIES IN PATHOLOGY

The toxic filtrates were, in their experience, fairly stable, but 20
minutes in water at a temperature of 100° C. completely destroyed the
toxicity for mice.

By means of injections of the filtrates they were able to obtain
evidence of active immunity against toxin, " Giftfestigkeit," and the
animals which had acquired this showed also some increase of resistance
to bacterial infection.

Markl (1898 and 1901) concluded, from a large number of experi-
ments, that the plague bacillus, when grown in bouillon, under aerobic
conditions, is capable of producing soluble toxins. The bacillus is
grown in bouillon, in flat flasks, to allow a free supply of oxygen, since,
under anaerobic conditions, no toxin formation takes place. Though
toxin may be demonstrated in a very short period, even within 24 hours,
two months at room temperature are required for the maximum toxicity
of the cultures to be attained. The optimum temperature for toxin
production is room temperature ; 37° C, while admitting of toxin
formation, is less favourable, a circumstance due partly to the injurious
influence on the bacillus of the higher temperature, partly to its
deleterious action on the already formed toxin. This susceptibility of
the toxin to the action of heat is a marked feature ; a temperature of
70° C, for 1 5 minutes completely destroys the toxicity of the filtrate for
mice, whereas it still remains toxic for rats, rabbits and guinea-pigs.
From this observation Markl argues that the filtrates contain two different,
but related, toxins. The relationship is best shown by the fact that
animals which have been injected with heated filtrates furnish an antitoxic
serum which is capable of protecting mice against the unheated toxin.

Markl believes that, by injecting goats and horses, with toxin, an
antitoxin may be obtained, and he suggests the advisability of employing
an anti-plague serum, obtained by a combined method of immunisation
with bacilli and with toxins.

He regards the toxin as a product of genuine bacillary secretion, and
not merely as the result of the degeneration and death of the bacilli,
with subsequent soaking-out of intracellular poisons. He bases his
opinion on a number of observations : — First, the toxin is demonstrable
at a very early period in the history of the culture, when the number of
dead and degenerated bacilli is too small to account for the toxin
present ; second, the toxin production bears a certain relationship to the

(116)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 7

virulence and habit of the culture. To obtain powerful toxins, the
bacillus must have been recently isolated from the animal body. After
a period of cultivation on artificial media, and especially if high tempera-
tures have been employed, it loses its toxigenic property, which may,
however, be recovered in part at least by fresh passage through the
animal body.

Kolle (1903), in proceeding to study the production and nature of
plague toxin, selected for his experiments such cultures as, from their
behaviour in the animal body, might be expected to prove toxigenic on
artificial media. He observed that, after a large number of passages
through the rat, plague cultures showed a tendency to lose their property
of multiplication in the blood, tissues, and organs of the animal body.
The death of the animals resulted almost entirely from the toxins
produced at the site of inoculation and in the neighbouring buboes. In
other words, the cultures, before their frequent passage, tended to produce
a plague septicaemia, after their passage a plague toxaemia. The
toxaemic symptoms came on at an early period, when the multiplication
of the bacilli in the rats was occurring solely at the site of inoculation.
Kolle employed an alkaline bouillon.

Whatever the nature of the flask employed for the cultivation of the
bacillus provision was made for an abundant supply of oxygen by having
only a moderately thin layer of medium at the bottom of the vessel.
Occasionally a little blood from the rat or rabbit was added to the
bouillon, in which case the culture was observed to be very abundant.
The cultures were grown some at 37° C, others at 35° C, others at 22°
C, and still others in the ice-safe at from 6° - 10° C.

A number of methods were employed to obtain the toxin, apart
from the bacilli. By the addition to the cultures of phenol and toluol,
and subsequent centrifugalisation, a sterile supernatant fluid was obtained
which was tested for toxicity. The Chamberland filter was also
employed, the filtrate being tested for sterility, as in certain cases the
plague bacilli have been found to pass the filter.

Kolle employed for his experiments rats of 180-200 grammes weight.
In his experience the guinea-pig and rabbit are too resistant, while the
mouse gives irregular results or proves too susceptible. Intraperitoneal
injection was in every case employed.

In young cultures^ of from 3-4 days, no toxin could be demon-

(117)

5 QUATERCENTENARY STUDIES IN PATHOLOGY

strated in the filtrates. On the other hand, if the bacillary bodies were
present, the cultures had a markedly toxic action. From this it must
be deduced that the toxin is bound to, or more probably contained
within, the bacillary bodies. Such toxins have an incubation period of

6 - 8 hours, death usually occurring within 24 hours. The symptoms
are those of collapse and giddiness, and point to the action of a brain
and heart poison. The common effects of a bacterial disease are mani-
fested in the spleen, bone-marrow and liver, viz., fatty degeneration and
coagulative necrosis, The peritoneum is injected and may present small
haemorrhages, which may also occur in other organs. Kolle regards this
as the primary specific plague toxin, since in the human subject similar
effects are not infrequent. In animals which survive the acute toxic
effects marasmus may occur.

In older cultures of from 7-10 days, where the growth has taken place
at 37° C, an increase in the toxicity of the filtrates, and in the character
of the resulting intoxication, can be determined, whereas the filtrates
from cultures at lower temperature produce effects similar to those
obtained from cultures i - 2 days old at 37° C. The older the culture, up
to a certain point, the higher is the toxicity. The cultures at 30° - 37°
C. reach their maximum toxicity in 3 - 4 weeks, whereas those at lower
temperatures require 8-10 weeks.

The toxic action of the filtrates from the older cultures is characterised
by a shortening of the incubation period. Symptoms of intoxication
occur in 10-15 minutes, and in the case of the larger doses death may
occur in I - 2 hours. The most prominent of the symptoms produced by
these filtrates from old cultures are convulsions, from which the animals
may recover. These effects are produced by non-specific substances of
the nature of cadaverin and putrescin, which have also been demonstrated
in old cultures of B. pyocyaneus, B. typhosus and B. cholerae. Associ-
ated with these substances, however, there are present in the old cultures
specific toxins which may or may not be identical with those occurring
in young cultures.

Kolle holds that the specific toxic substances originate not from a
secretory process on the part of the bacillus, but are produced by the
death of the bacilli with subsequent soaking-out of the intrabacillary
toxin.

Such intra-cellular toxins never give rise to the formation of anti-

(118)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 9

toxins. Kolle tested various anti-plague sera for antitoxic properties
with negative results. He found that normal horse serum possessed the
property in some degree, and that the slight toxin-neutralising properties
of sera obtained by immunisation did not exceed the limits found in
normal serum.

He attempted to immunise horses by intravenous injection of the
primary toxin from young cultures, in others he employed the toxins
from old cultures, but in neither case could any antitoxic property be
demonstrated in the serum.

The serum and toxin were mixed, and injected peritoneally. Where
4 lethal doses of toxin were employed, 8 rats so treated all died. When
2 and 3 lethal doses of toxin were employed, a certain number of rats
recovered, even when normal serum was used. Thus, out of i6 rats
injected with a mixture of plague serum and toxin, 9 died ; out of 8
injected with plague toxin and normal serum, only 5 died.

Kolle therefore concluded that the toxins found in plague cultures
were intracellular toxins, and that consequently, in accordance with a
general biological law, their injection into the animal body does not
stimulate to antitoxin production. The question of the tenability of
this so-called " general biological law " will be discussed later.

The three following investigations are grouped together because, in
all, the toxic or immunising substance, is derived from the animal body.

Terni and Bandi (1900), proposed to employ, as an anti-plague
vaccine, the peritoneal exudate obtained from guinea-pigs and rabbits
which had been infected intraperitoneally, and killed during the death
agony. The fluid, after being tested to ascertain that it was free from
organisms other than the B. pestis, was incubated for 12 hours, and then
submitted to a temperature of 50° - 52° C. for two hours. The fluid
was brought to a suitable volume by means of a weakly alkaline normal
salt solution, containing 0*5 % phenol. The authors claimed that good
immunity was obtained, even in 4 to 5 days, and that the injection did not
aggravate an already existing infection, but rather tended to prolong
life. The blood of an individual who had received an injection of 1-5 to
2'5 c.c. showed, within 8 to 10 hours, a hindering influence on the
development of the B. pestis.

Hueppe, F., and Kikuchi, Y. (1905), starting from the work of Bail,
Kikuchi and Weil, propose to employ, as an immunising agent, the

(119)

10 QUATERCENTENARY STUDIES IN PATHOLOGY

peritoneal exudates obtained from guinea-pigs infected in the peritoneum
with B. pestis. By means of two prophylactic injections of this fluid
or "aggressin," they succeeded in conferring a high degree of immunity
on mice, guinea-pigs, and rabbits. The injection of the substance was
followed by a period of hypersensibility. The theory of the method and
its relation to those of other workers will be discussed at a later period.

Klein (1905), while investigating the vitality of the B. pestis in
certain organs, spleen and lung, and in the buboes of animals dead of
plague, found that though all the bacilli had been destroyed by the
process of drying over sulphuric acid, an emulsion of the organs, when
injected into mice, was capable of causing the death of these animals
within 20 hours.

All the phenomena of acute plague are exhibited, without, however,
the B. pestis being present in the tissues. When the amount of emulsion
injected was insufficient to cause death, the animals, after recovery from
the local swelling and constitutional disturbance which resulted from the
injection, were found, on being tested at a later period, to be refractory
to infection, even when a virulent plague bacillus was injected.

Guinea-pigs, inoculated cutaneously with a bacillus of moderate
virulence, develop a sub-acute form of plague, death occurring from the
fourth to the ninth day. The lesions found are necrotic buboes, necrotic
nodules in the spleen and liver, and particularly necrotic nodules and
patches in the lungs. In sections of such organs the central parts of the
necrotic nodules and patches are found to be crowded with bacilli,
whereas the peripheral parts, consisting of debris, contain few, if any,
bacilli. From this it may be inferred that the necrotic changes are due,
not to bacilli themselves, but to their toxin.

Klein proposes to prepare a prophylactic substance from such
material. The bubo, enlarged spleen, necrotic lung and liver of guinea-
pigs which have died from sub-acute plague, are cut out, minced
aseptically, and then spread out in thin layers on sterile glass dishes, and
dried over sulphuric acid at 46° - 47° C. There is thus obtained " a
material which not only can be very easily and rapidly prepared, but
which is of a uniform and reliable efficacy, and in every way indeed
superior to any of the other prophylactics." Klein calculates that a
single large guinea-pig would yield 800- 1000 human doses of the new
prophylactic.

(120)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY II

It is unnecessary here to go more minutely into the results obtained
on animals, or the advantages claimed for the use of this material.
From our present standpoint, however, the toxity of the substance is of
interest.

(i) It kills a large percentage of mice within 20-24 hours in doses
of I - 5 milligrammes.

(2) Prepared from acute cases, it kills 15 -25% of half-grown white
rats in doses of 5-8 milligrammes ; prepared from sub-acute cases,
10-12 milligrammes are required to produce a fatal effect.

(3) 20 milligrammes fail to kill a guinea-pig of 200 - 300 grammes
weight.

Klein holds that the effects produced by the substance are due not
merely to the bacilli contained in the material, but depend largely on the
presence in the necrotic area of a " tissue toxin."

Besredka (1905 and 1906) has proposed a new method for the
preparation of typhoid and plague endotoxins.

The growth from an agar culture of B. pestis is emulsified, heated
to 60° C. for one hour, and then dried in vacuo. The lethal dose for the
mouse of this dried substance, injected subcutaneously, is five decimilli-
grammes. To obtain the plague endotoxin in the soluble state, the
dried bacilli are put in normal salt solution and horse serum in fixed
quantities. The mixture is placed in the ice-safe overnight, and is
centrifugalised on the following day. The clear supernatant fluid
contains the endotoxin, and the deposit of a sticky consistence is formed
of atoxic plague bacilli.

The soluble endotoxin is highly toxic for the mouse, e.g., 0*02 centi-
gramme of dried bacilli, to which i c.c. of physiological salt solution
and 4 c.c. of horse serum have been added, corresponds to about 20
lethal doses of endotoxin.

The plague endotoxin keeps for months in the ice-safe. It is
thermostable; heating to 55° C. for one hour leaves it intact; it is
slightly attenuated by heating for one hour to 65° C, but it is not
destroyed by heating to 57° C. for 15, or even 24 hours.

Its essential characteristic, according to Besredka, is that its action is
neutralised by anti-plague serum.

Besredka also investigated the endotoxin of the typhoid bacillus, and

(121)

12 QUATERCENTENARY STUDIES IN PATHOLOGY

found that 32 lethal doses of the endotoxin are neutralised by 0'2
grammes of the serum of an immunised horse.

Similarly, he found that the serum of horses which had been injected
intravenously, either with plague bacilli or with the soluble endotoxin,
was capable of neutralising the plague endotoxin.

The Writer's Experiments on Plague Toxin.

The writer's experiments on the subject of Plague Toxin were
directed chiefly to the study of the influence of certain factors on the
toxicity of cultures of diflerent races of the plague bacillus.

Technique used for obtaining the Toxin. — In certain cases the cultures
were well shaken with toluol, the bacilli allowed to sediment, and the
bacterial free supernatant fluid examined; but for the most part the
filtrates of cultures were investigated. The filtration was carried out by
means of a " tandem " filter, i.e.^ the fluid was first passed through a
Berkefeld, and afterwards through a Pasteur-Chamberland filter. The
more porous Berkefeld filter removes the coarser particles, and this
prevents the blocking of the Pasteur-Chamberland candle which,
employed in this way, always gave a sterile filtrate. The loss of toxin is
perhaps less by this method than if the Pasteur-Chamberland candle
alone were employed, because in the latter case the blocking of the pores
by the larger particles tends to make a very close filter. In every ca.se
it is desirable to test the sterility of the filtrates and centrifugalates.

Temperature. — The temperature most frequently employed for the
growth of the cultures was what is generally accepted as the optimum
temperature for the growth of the B. pestis, viz., about 30" C.

The Medium, — The medium employed was the ordinary bouillon
such as is used in the preparation of diphtheria toxin. In most cases it
was made neutral or very slightly alkaline to litmus, but, with the object
of ascertaining whether toxin formation would be aided by the use of an
alkaline broth, in certain cases the broth was prepared as for diphtheria
toxin, 7 c.c. per litre of normal caustic soda being added after neutralisa-
tion with litmus as the indicator. The results obtained with the filtrates
from two flasks of the same broth, in the one case neutral, in the other
case alkaline, are shown in Tables 1. and II. In this case, the neutral
broth appeared to be the more favourable medium for toxin production.

(122)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 13

TABLE L— Toxin 15.

Filtrate from a 2 month's growth on Neutral Broth.
Test on Mice of 20-25 grammes weight.
Subcutaneous injection.

Mouse.

Volume of Toxin.

Result.

I.
2.
3.

5-

•05 C.c.
"I II
•3 "
•5 "

I-Q II

Lived.

Died 48 hours.
Died 24 hours.
Died 24 hours.
Died 24 hours.

TABLE IL— Toxin 16.

Filtrate from 2 month's growth of the same bacillus and on the same broth as
Toxin 15. In this case the broth was made alkaline, 7 c.c. of normal caustic
soda per litre being added after the neutral point to litmus was reached.

Test on Mice of 20-25 grammes weight.

Subcutaneous injection.

Mouse.

Volume of Toxin.

Result.

I.

•05 C.C.

Lived.

2.

•I M

Lived.

3-

•3 "

Lived.

4-

5.

•5 "
I'O II

Died 48 hours.
Died 24 hours.

It was observed that the addition to the broth of i : 4 to i : 8 of
normal horse serum heated to 65° C, gave a very abundant growth, in
some cases with pellicle formation, but so far as this was investigated no
increase in toxin formation could be demonstrated.

The cultures were grown for the most part in large flat bottles each
containing i to ij litres, but Erlenmeyer's flasks, with a thin layer of
medium in the flask, were also employed. Previous experience had

(123)

H

QUATERCENTENARY STUDIES IN PATHOLOGY

shown that under these conditions the supply of oxygen is sufficient to
satisfy an organism so exigeant in this direction as B. diphtheriae.

Appearance and Age of the Culture. — The cultures vary greatly in the
appearances they present. In certain cases the growth begins as minute
flocculi in the bouillon, the surrounding fluid remaining clear. Other
cultures show an almost uniform turbidity. To certain of the flasks
cocoa butter or butter was added, and in these good stalactite formation
was observed. No connection between the type of growth and the
toxicity could be determined. The culture, as it increases in age,
becomes very alkaline, and in two months there is usually a large deposit
containing bacterial debris and masses of crystals of triple phosphates.

One change is of considerable significance in relation to toxin
formation. A platinum loopful from a culture a few days old, planted
on an agar plate, gives an abundant growth, whereas, from a flask two
months old, little or no growth may be obtained unless a larger quantity
of material be employed for inoculation. In the two months' old culture,
almost all the bacilli present have lost the usual morphological characters
of the plague bacillus, the deposit in the flask consisting almost entirely
of bizarre involution forms or mere detritus.

The Period of Maximum Toxicity. — Some preliminary tests were made
on the toxicity of the filtrates from cultures lO days and two months old
in bouillon of the same brew. Such a test shown in Tables III. and IV.
indicates that at lo days, if toxin is present, its concentration is much
less than at two months.

TABLE III.— Toxin 7.

Filtrate from 10 day's growth on Neutral Broth.
Test on Mice 15-20 grammes weight
Subcutaneous injection.

Mouse.

Vol

ume of Toxin.

Result.

I.

•05 c.c.

Lived.

2.

•I n

Lived.

3-

•5 ••

Lived.

4-

I'O y\

Lived.

(124)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 1 5

TABLE IV.— Toxin 13.

Filtrate from 2 month's growth on same Neutral Broth as No. 7.
Test on Mice 15-20 grammes weight.
Subcutaneous injection.

Mouse.

Volume of Toxin.

Result.

I.

2.

3-
4-

5.

•05 c.c.
•I II
*3 "

•5 "

I'O M

Lived.

Died 48 hours.
Died 48 hours.
Died 24 hours.
Died 24 hours.

It thus appears that the maximum toxicity of the culture is syn-
chronous with the period when most of the organisms have undergone
degenerative changes or have died ; it seems probable, therefore, that the
toxic substances have been set free as the result of a process of autolysis.
Reference will be made to this in discussing the question of toxins.
Most of the filtrates tested were from cultures 8 to 10 weeks old.

Great variations were observed in the toxicity of the filtrates from
different brews of bouillon made by the same methods and with the
same proportions of the components. Similar inexplicable variations are
well known in the preparation of diphtheria and tetanus toxin.

It would be impossible here to give the protocols of all the tests
carried out in ascertaining the toxicity of different filtrates, but those
recorded are typical of the results obtained in the case of the more toxic
filtrates. A number examined were found to be much less toxic than
Toxin No. 10. As this toxin was employed for a test toxin in testing
the sera, the protocols of tests on mice and rats are given (Tables V.,
VI., VII. and VIII). It will be noted that there is a certain amount of
irregularity in the results on mice, which makes the determination of
the minimal lethal dose a matter of some difficulty.

(•25)

i6

QUATERCENTENARY STUDIES IN PATHOLOGY

TABLE v.— Toxin io.

Filtrate from culture on Neutral Bouillon lo weeks old.

Tested on Mice of 20-25 grammes weight.

Subcutaneous injection.

First Test.

Mouse.

Volume of Toxin.

Result.

I.

•01 c.c.

Lived.

2.

•02 II

Ill— recovered.

3-

•03 M

Ill— recovered.

4-

•04 1.

Died 8 days.

5-

•05 M

Ill— recovered.

6.

•I M

Died 24 hours.

7.

•2 II

Died 24 hours.

8.

•5 "

Died 24 hours.

9-

•5 "

Died 24 hours.

TABLE VI.— Toxin 10.

14 days after previous Test on Mice of 15 grammes weight.

Subcutaneous injection.

Second Test.

Mouse.

Volume of Toxin.

Result.

T.

•05 C.C.

Died 24 hours.

2.

•I II

Died 24 hours.

3-

•2 II

Died 24 hours.

4-

5.
6.

•3 "
•4 II

•5 "

Died 48 hours.
Died 24 hours.
Died 24 hours.

7-

•6 II

Died 24 hours.

(126)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 1 7

TABLE VII.— Toxin 10.

14 days after previous Test on Mice of 15 grammes weight.
Intraperitoneal injection. — -

Second Test (contimied).

Mouse.

Volume of Toxin.

Result.

I.
2.
3.

•02 C.C.

•I II

Died 24 hours.
Died 24 hours.
Died 24 hours.

TABLE VIII.— Toxin 10.

Testing Toxicity on Rats.
Subcutaneous injection.

Rat.

Weight in
grammes.

Volume of Toxin.

Result.

I.

125

•I C.C.

Died 3 days.

2.

125

•5 "

Died 24 hours.

3.

150

I'O II

Died 24 hours.

4-

155

2*0 II

Died 48 hours.

It is interesting to note that rats and mice offer much less resistance
to the action of plague toxin than do rabbits and guinea-pigs. That the
latter animals have the more powerful defence is shown by the experi-
mental results given in Tables IX. and X.

(127)

i8

QUATERCENTENARY STUDIES IN PATHOLOGY

TABLE IX.— Toxin io.

Testing Toxicity on Guinea-pigs.
Subcutaneous injection.

Guinea-
pig-

Weight in
grammes.

Volume of Toxin.

Result.

I.

240

•I c.c.

Unaffected— lived.

2.

240

•5 •'

3-

260

I'O ti

4-

250

2'0 M

5-

250

3-0 M

6.

260

4-0 M

TABLE X.— Toxin
Testing Toxicity on Rabbits.
Subcutaneous injection.

Rabbit.

Weight in
grammes.

Volume of Toxin.

Result.

I.

550

•05 c.c.

Unaffected — lived.

2.

500

•I M

3.

520

•5 "

4.

526

I*0 11

5-

950

2*0 It

6.

760

3-0 1.

7-

800

4-0 II

All guinea-pigs and rabbits alive at the end of one month.

In regard to the stability of the toxin, the writer's experience was in
agreement with Markl's. The toxins were found to be rather unstable,
especially at ordinary hot weather temperatures as, for example, when
Toxin 10 was tested two months after the test recorded in Tables V., VI.,
VII. and VIII., it was found to have so far deteriorated that the minimal

(128)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 19

lethal dose had risen from the O'l c.c. of the Table to 0*8 c.c. Higher
temperatures, as might be expected, cause similar deterioration in a much
shorter time. The effect of exposing the toxin to a temperature of
70" C. is shown below : —

TABLE XL
Showing influence of heat to 70° C. on Toxin No. 10.

Mouse.

Weight in
grammes.

Volume of Toxin.

Result.

I.
2.
3.

20
20

I c.c.

I II
I It

Dead in 24 hours.
Dead in 6 hours.
Dead in 8 hours.

Toxin heated to 70°
C. for 30 minutes.

4-
5-
6.

15

20

I C.C.
I C.C.

I c.c.

Lived.
Lived.
Lived.

Where irregularities occur, such as are frequent in the testing of
plague toxin, it is extremely difficult to trace the influence of the
virulence of the race of bacillus on the toxicity. It was found impossible
to establish any definite relation between virulence and toxicity, but a
strong impression was formed that as a culture diminished in virulence
the filtrate obtained from it became less toxic.

Incubation Period. — The incubation period lasts from 6 to 24 hours
according to amount of toxin injected. In certain cases, where very old
cultures are used, convulsions occur almost immediately after the injection,
but these are probably due to the presence of cadaverin, as has been
suggested by Kolle. These frequently pass off in a short time to be
succeeded by the symptoms of toxic action.

Effects and Symptoms resulting from the Injection of Toxin. — If the
injection is subcutaneous, within 24 hours a soft oedematous swelling,
often very extensive, is observed in the vicinity of the site of inoculation.

(129) K

20 QUATERCENTENARY STUDIES IN PATHOLOGY

In acute cases, after the period of incubation, the animal hunches
itself up in a corner of the cage, refuses to eat, and, when disturbed, moves
in an inco-ordinate manner, as if suffering from alcoholic intoxication.
Convulsions are frequent just before death, which appears to be due to
collapse.

In chronic cases the early symptoms are less acute, the animal
gradually loses flesh till a profound state of marasmus may result. In
some of the rats suffering from chronic toxaemia following the injection
of toxic filtrates, we have observed a form of paralysis resembling
diphtheria paralysis in the guinea-pig. The hind limbs are first affected,
then the fore limbs, and ultimately the respiratory muscles may be
involved. These animals are often greatly emaciated. Convulsions
frequently occur immediately before death.

Post-mortem Changes observed in Animals dying of Toxcemia. — In the
animals which die of acute toxaemia, say within 24 hours after sub-
cutaneous injection of a toxin, the most marked feature is oedema round
the site of inoculation. This may be very extensive, affecting the whole
abdominal wall, and small haemorrhages are sometimes present. The
neighbouring lymphatic glands may be enlarged. Haemorrhages are
sometimes present in the viscera, especially where the injection has been
intraperitoneal, in which case the peritoneum and the mucous membrane
of the alimentary canal are injected or haemorrhagic.

In the animals which survive several days or weeks, wasting of the
musculature is a marked feature. Fatty degeneration of the liver is a
frequent lesion, and necrotic areas are present in a few of the cases.
Where death occurs at a late period the spleen may be enlarged.

The Question of a Plague Antitoxin.

With the view of ascertaining whether the injection of toxic filtrates
would stimulate the animal body to antitoxin production, two horses
were specially employed for immunisation with the filtrates of cultures.
In the case of one, the horse " R " filtrates from bouillon cultures,
which had been heated to 70" C. for one hour, were alone used. This
animal, in the three months preceding the removal of the blood, the
serum from which was tested, received injections amounting to 1080
c.c. of filtrates.

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 21

The second animal, after a course of injection of heated filtrates, was
further treated with injections of filtrates of unheated cultures. In the
three months preceding the bleeding from which the serum tested was
obtained, nearly two litres of unheated toxin were injected in doses of
from 60 to 300 c.c. Among other toxins, nearly a litre of Toxin No. 10,
the tests of which have been recorded, was employed in the immunisation.
As already mentioned, the other horses whose sera were tested were
immunised by the ordinary methods. The case of the horse " Pr."
requires some comment. This animal had been immunised by the Berne
method, and had received large injections up to 300 c.c. at one dose of
concentrated Haffkine's Prophylactic Fluid. This was followed, as is
usual, by a number of intravenous injections of living broth culture. On
reviewing the treatment which this animal had received before the
removal of the sample of serum tested, it was noted that within the six
weeks of treatment preceding the interval which always intervenes
between the last injection and the venesection, the animal received the
following injections : — One of 40 c.c. and another of 90 c.c. of broth
cultures four days old, one of 100 c.c. two days old, and one of 40 c.c.
six days old. In regard to this, it is to be noted that these injections
were given rather more frequently than usual, and that 180 c.c. of the
material was 4 to 6 days old when at least traces of toxin might be
expected to be present.

It is well known that out of a number of horses immunised against
diphtheria and tetanus toxins, only a certain number of these respond
freely and quickly to the stimulus of toxin injections, and are capable
of producing antitoxins of the highest potency. It is probable that this
variability among the individuals of the same species, in the response to
stimulation by toxin, may extend to other forms of immunisation. It is
quite conceivable that the dosage mentioned may have in this case been
a sufficient stimulus to excite the mechanism necessary for antitoxin
production in a highly responsive organism.

To ascertain whether the injection of heated filtrates had given rise
to antitoxin production, the following preliminary test was carried out
with the serum of horse " R," and controlled with the serum of normal
horse "Z," the test toxin employed being Toxin No. 10. The toxin
and serum were mixed and kept one hour at 35° C. It will be seen in
Table XII. that the evidence is in favour of antitoxin production.

(131) K I

22

QUATERCENTENARY STUDIES IN PATHOLOGY

TABLE XIL

TEST OF SERUM OF HORSE " R."

Test Toxin No. lo. Toxin and Serum mixed and placed i hour at 35° C.

Rat.

Weight in
grammes.

Volume of Toxin.

Volume of Serum.

Result.

I.

100

2 c.c.

I C.C.

Died in 8 days.

2.

100

2 M

I M

Lived.

3-

no

2 M

I M

Lived.

Control : — Serum of Normal Horse

"Z."

4-

100

2 c.c.

I C.C.

Dead in 24 hours.

5-

100

2 If

I M

Dead in 24 hours.

6.

no

2 It

I It

Dead in 24 hours.

('32)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 23

A test was carried out on one day with the serum of a number of
horses. It will be seen from the Table XIII. that the only serum which
showed marked antitoxic properties was the serum of horse " M," which
had been immunised with toxins only.

TABLE XIII.

SERUM OF SEVERAL HORSES VARIOUSLY IMMUNISED.

Test Toxin No. 10 ; tests all carried out on one day.
Toxin and Serum mixed and kept i hour at 35° C

Rat No.

Weight in
grammes.

Vol. of Toxin.

Vol. of Serum.

Result.

Horse

"M."

I.

95

2 c.C.

I'O C.C.

Lived.

2.

100

II

O'l II

Lived.

3-

100

11

0'02 n

Lived.

Horse

"R.D."

4.

100

2 C.C.

I'O C.C.

Lived.

5-

90

.,

O'l 11

Died 48 hours.

6.

90

"

0"02 II

Died 24 hours.

Horse

"M.B."

7-

90

2 C.C.

I'O C.C.

Lived.

8.

90

„

O'l II

Died 48 hours.

9-

90

11

0"02 M

Died 48 hours,

Horse

" P.N."

10.

100

2 C.C.

I'O C.C.

Died 48 hours.

II.

95

It

o-i .1

Died 48 hours.

12.

95

II

0'02 II

Died 24 hours

(133)

u

QtJATERCfeNTENARV STUDIES IN J'ATHOLOGV

The test on the serum of " M. " was repeated and controlled with the
normal horse " N. H.," see Table XIV. The same Test Toxin No. lo
was used, and the toxin and serum were mixed and kept two hours at
35° C.

TABLE XIV.

TEST ON SERUM OF " M." REPEATED.

Serum of Normal Horse " N. H."

Rat No.

Weight in
grammes.

lOO
90
95

Volume of Serum.

I c.c.

o'5
o*i

Volume of Toxin.

Serum of Horse " M."

2 c.c.

2 II
2 II

Result.

Died 24 hours.
Died 48 hours.
Died 48 hours.

4-

100

I c.c.

2 C.C.

Lived.

5-

100

01

2 II

Lived.

6.

95

0"02

2 n

Lived.

7-

90

O'OI

2 II

Died 24 hours.

(•34)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 25

Reference has been made to the serum of the Horse " Pr." The
results of a number of tests relating to the serum of this horse are shown
in the accompanying Tables :— XV., XVI., XVII. and XVIII.

TABLE XV.
TESTING SERUM OF "Pr." (6. IV).

Mice of 15-20 grammes weight.

Toxin (Subcutaneously at least = 4 lethal doses; intraperitoneally
lethal doses).

Serum and Toxin mixed and placed i hour in the incubator at 37° C.

Mixture injected subcutaneously.

= 10

Mouse.

Volume of Toxin.

Volume of Serum.

Result.

I.

•2 c.c.

•I C.C.

Lived.

2.

•2 II

*l II

Lived.

3.

•2 II

•2 II

Lived.

4.

•2 M

•2 11

Lived.

Mixture injected intraperitoneally.

5. -2 M -I n

6. *2 II 'I n

7. '2 II '2 II

8. '2 II '2 II

Controls : — Same quantities of Toxin ; serum of normal Horse " F."
Subcutaneous.

Lived.

Lived.

Died in 48 hours.

Lived.

9-
10.

•2 c.c.

•2 II

•2 c.c.
•2 II

Died 24 hours.
Died 16 hours.

II.
12.

Intraperitoneal.

•2 II
•2 II

•2 II

•2 II

Died 24 hours.
Died in 10 hours.

(135)

26

QUATERCENTENARY STUDIES IN PATHOLOGY

TABLE XVL

TESTING SERUM OF **Pr." (27, IV.) ON RATS.

Test Dose about 10 M. L. D.

Toxin and Serum mixed and kept 2 hours at 37° C, then injected sub-
cutaneously.

Mouse.

Weight in
grammes.

120
120
115
115
115
no

105

100
Controls

Volume of Toxin.

I'O C.C.

Volume of Serum.

•01 C.C.

Died 24 hours.

Lived.

Lived.

Lived.

Lived.

Lived.

Lived.

Died in 27 days.

Done with Serum of Normal Horse " F."

•02
•05

*!

•3

•5
10
I'O

Result.

9-

120

I C.C.

2 C.C.

Died 3 days.

10.

105

tr

2 II

Died 3 days.

II.

no

II

I II

Died 24 hours.

12.

120

M

I II

Very ill ; recovered.*

* The rats frequently bite the site of inoculation, and in certain cases, as the result of
this, leakage may take place.

(136)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 2/

TABLE XVIL

TESTING SERUM OF " Pr." ON MICE.

The Serum injected subcutaneously at one side of the body, and 24 hours
later the Toxin at the other side.

Test Toxin No. 10 about 4 to 6 M. L. D.

Only small mice available.

Mouse.

Weight in
grammes.

Volume of Serum.

Volume of Toxin.

Result.

I.

8

•6 C.C.

•3 C.C.

Lived.

2.

10

•6 M

•3 "

Lived.

3-

10

•6 ..

•3 "

Lived.

4.

10

•6 M

•4 .1

Lived.

5-

10

•6 n

•4 .,

Lived.

6.

10

•6 ..

•4 M

Lived.

Controls: — With Serum of "B." (old Diphtheria Horse).

7-

8

•6 C.C.

•3 C.C.

Very ill ; recovered.

8.

10

•6 ..

•3 "

Died 4 days.

9.

10

•6 M

•3 "

Died 7 days.

10.

10

•6 „

•4 "

Died 48 hours.

II.

10

•6 n

•4 ,1

Died 48 hours.

12.

15

•6 ,.

•4 M

Very ill; died 12 days.

(137)

2g

QUATERCENTENARV studies tN PATHOLOGY

TABLE XVIII.
TESTING SERUM OF "Pr." ON RATS (27, IV.).

Toxin No. 10.

Serum injected on one side of body; Toxin (at least 10 M. L. D.) 24 hours
later on the other side.

Rat.

Weight in
grammes.

Both subcutaneous at opposite sides of
abdomen.

Serum.

Toxin, 24 hours after
Serum.

Result.

95

95

100

105
115
120

I c.c.

1-5 c.c.

Controls : — With Normal Serum of Horse

Z."

7-

95

I c.c.

1-5 c.c.

8.

100

9-

100

10.

no

II.

115

12.

120

No symptoms.
No symptoms.
No symptoms.
No symptoms.
No symptoms.
No symptoms.

Died 24 hours.
Died 48 hours.
Died 24 hours.
Died 48 hours.
Ill; recovered.
Died 48 hours.

(138)

R£SEARCHiES ON CERTAIN PROBLEMS OF PLAGtJE IMMUNITY 29

These tests appear to show : —

1. That the serum is capable of neutralising the toxin when mixed

in vitro^ as tested by subsequent subcutaneous or intraperi-
toneal injection into rats and mice.

2. That the serum injected into rats or mice, 24 hours before the

toxin, is capable of conferring protection on these animals.

3. In the case of rats, 002 c.c. of serum, when mixed with the toxin,

neutralises the toxic action of about 10 M. L. D., and i c.c.
of serum protects against at least 10 M. L. D. injected at
another part of the body 24 hours after the serum.

Kolle says that the serum of the normal horse has considerable
antitoxic properties, but, as already stated, he was unable to obtain an
increase beyond what he regarded as within the limits of normal
variation.

If the serum of an animal normally contains antitoxin, we might
anticipate on the basis of Ehrlich's theory that it would be possible by
stimulating the organism to obtain an increase in the production of the
antitoxin. In our experiments, for purposes of control, the sera of four
normal horses were employed, but, in addition, the serum of five horses
immunised by the ordinary method of obtaining anti-plague serum
showed little or no increase in antitoxic power, and these may, therefore,
be classed among the controls.

The evidence from these experiments seemed to be in favour of the
possibility of obtaining a serum with antitoxic properties in excess of
those found in normal serum.

Relation of Neutralisation of Toxin to Precipitin Reaction. — Certain
experiments were carried out with the view of ascertaining whether the
occurrence of the precipitin reaction of Kraus might not play some part
in the plague toxin-antitoxin reaction.

The serum of certain of the horses gave a well marked'fprecipitation
with the Test Toxin No. 10. The serum of the horse " R D." gave 3 to 4
times the precipitum given by the serum of the horse " M." In the
previous tests we have seen that whereas i c.c. of the serum of " R D."
was required to neutralise 10 M. L. D. of Toxin 10, 0*02 c.c. of the serum
of " M." produced the same effect. There is, therefore, no direct pro-
portionality between the two properties of the serum.

iM9)

30 QUATERCENTENARY STUDIES IN PATHOLOGY

The Protective Action of Anti- Plague Serum. — Though much
sceptical criticism has been directed against anti-plague serum, it has
been admitted by everyone who has seriously approached the subject
that, however disappointing the results of its use on the human subject
may be, it manifests definite protective properties when tested on
animals. A typical test showing the effect on rats is shown in the
annexed Table XIX.

It will be noted that death occurred in the case of the animals which
received the largest dose of serum. The writer has observed similar
results in the case of anti-streptococcic serum. Tavel, Krumbein, and
Glucksmann also record similar results, and attribute them to the well-
known phenomena of "diversion of the complement" easily demon-
strated in test tube experiments on bacteriolysis.

A common experience is to find that animals treated with serum
survive till about the tenth day, and then succumb to the disease. This
is probably due to the fact that the immune serum has by that time
been metabolised and execreted, and that some of the bacilli which have
not been destroyed are able to re-infect the animal.

We have already seen that the antitoxic properties of the serum of
the horse " MB." are not of a high order, and probably do not exceed
the limits found in normal serum.

A number of tests were carried out to determine whether the serum
possessed bacteriolytic properties. The methods employed and the
results obtained are dealt with in the following section.

Testing Plague Serum for Bacteriolytic Properties.

A number of methods were employed to ascertain whether
bacteriolysis could be demonstrated in the case of anti-plague serum.
Neisser's (1904) technique was found to give the most consistent and
accurate results.

Small test tubes of about 10 X I'S cm. are employed. Suitable
dilutions of the immune serum are prepared in normal salt solutions, say,
I : ID, I : 100, I : looo, etc., and the quantity of serum to be tested is
pipetted into each test tube of a series, and to this 0*5 c.c. of fresh normal
serum is added to supply complement. Three drops of bouillon are
introduced into the mixture, and the volume in each tube is brought up
to 2 c.c. A measured quantity of a suitable dilution of the culture is

(140)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY

31

m

S

w

en

^

0

X

tlH

w

0

H-1

>,

pq

D

<

P^

U-t

ti3

C/3

0

'i^

t-H

H

fy)

W

H

P^

2 . o

C/3 h

O «-! !=^

> in

CO

as

■^ "TJ n3 X) "^

<D dJ <U CJ (U

> > > > >

*hJ '►Iq '^Iq 'hIJ 'k3

+

+ +

C O

10

O M CO 10 o

o o

*^ b

LO u-> 10 O O O

M t^ o^ >-• CI M

N M M M M M

O LO

a o

M ro '^^t- 10 ^ r^ 00

O O

ON O

00000

0000

'^'^00

M M CO r(- 10

(HI)

32

QUATERCENTENARY STUDIES IN PATHOLOGY

then introduced into each tube. The tubes are placed in the incubator
at 37° C, and are examined at stated intervals, say in 3, 16, 24-48 hours,
the same volume from each tube being pipetted on to an agar tube or
Petri dish. The absence or diminution in the number of colonies on
these tubes or plates after their incubation is, with certain reservations,
regarded as the indication of bacteriolysis having taken place. The
condition of the small test tubes, after their incubation for 24-48 hours,
affords valuable information, as will be exemplified in the following test :
Test of the Serum of Horse " RD." — The arrangement in thia experi-
ment was as above. The bacterial dilution was prepared by emulsifying
a 2 mgm. ose from a 24 hours culture on agar of a highly virulent plague
bacillus in 10 c.c. of normal salt solution ('85%). The bacilli were well
distributed by shaking, and a 2 mgm. loopful of this dilution was introduced
into each of the tubes containing the immune serum, fresh horse serum
and broth. The agar tube planted at once from the control tube, which
contained salt solution and bouillon but no serum, gave a fair number of
colonies, which became confluent over the water of condensation. A
measured quantity from each tube was planted on agar, after 3 and after
16 hours. The details and result of the experiment are shown in the
Table XX.

TABLE XX.

Volume of Anti-
plague Serum,
Horse "R."

Volume of comple-
mental normal
horse serum (i6
hours old).

Condition in agar
tubes, after 24
hours at 37° C.

Condition in tube containing
immune serum and complement
after 48 hours.

VO C.C.

0-5 c.c.

0

Growth in agglutinated
colonies through the fluid.

•3

1

0
0

II It
II II

•03

1

0

It II

"OI II

r

0

M fl

•003

1

Colonies ; few.

.1 II

•001 M

1

Colonies; few.

II II

•0003 II

1

Colonies ; many.

Homogeneous growth.

•0001 11

1

Colonies ; many.

II It

0

1

Uniform growth.

II ti

(142)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 33

The experiment shows, therefore, that while there was an apparent
inhibition of growth, probably largely due to the agglutinating power of
the serum which was i : 300 with a virulent culture, in no single tube had
complete destruction of the bacilli taken place, since growth occurred in
every one of the tubes containing the serum mixture.

A number of similar experiments, some of them carried out with the
immune horse serum so fresh that the addition of complement was
unnecessary, gave perfectly consistent results, e.g., the fresh serum of
horse " M B." was tested as above with a bacillus of low virulence, in full
strength and in dilutions of i-io, and i-ioo, and i-iooo, and growth
took place in all the tubes.

No matter what method was adopted, it was found impossible to
obtain evidence of the serum possessing any bacteriolytic property, even
when bacilli of low virulence were used.

Discussion as to the Nature of the Mechanism by which Anti-
Plague Serum Exercises its Anti-Infectious Effect.

In considering the properties of anti-plague serum we have seen that
an antitoxic action can play little part in the protective influence exerted
by this serum. It is conceivable that the plague bacillus produces and
throws off a large number of toxic particles or receptors under the
stimulus of the antagonistic influences which it encounters when intro-
duced into the animal body. If this be the case, it seems reasonable to
assume that the bacilli, even when grown in artificial media, possess,
though in smaller numbers, those toxic particles which are liable to be
set free, especially on the death or during the disintegration of the
microbes.

In dealing with the literature of plague toxin reference has been made
to several methods of immunisation which are based on the common
assumption that certain toxic or immunising substances are produced in
a concentrated form in the animal body from which they are recoverable.
Thus Terni and Bandi, Hueppe and Kikuchi used peritoneal exudates,
and Klein the necrotic areas resulting from plague infection in different
organs, as their immunising agent.

(143)

34 QUATERCENTENARY STUDIES IN PATHOLOGY

The first work in this direction was done by Denys and Van de Veldc
(1895) who showed that a substance " leucocidine," highly toxic for
leucocytes, could be obtained in the pleural exudates resulting from the
injection of staphylococci into the pleural cavity of the rabbit, and that
the serum of rabbits immunised against staphylococci contained a
substance " anti-leucocidine," capable of neutralising the toxic action of
leucocidine. The leucocidine could be demonstrated to be present in
cultures though not so abundantly as in the exudates.

The whole subject of the toxic and immunising action of the body
fluids of infected animals has been re-opened by the investigations of Bail
and his co-workers. This work embraces a theory of the method of
attack on the animal body by certain groups of bacilli, and of the nature
of the defence by means of which immunity is attained. Bail's views
may be briefly stated as follows : — Certain organisms obtain a footing in
the animal body by means of substances which they throw off, and to
which Bail gives the name of " aggressins." These are demonstrable in
the bacterial-free body-fluids, exudates, etc., of the infected animal.

The aggressive property of such fluids can be shown in the following
manner. A non-lethal dose of the bacillus becomes lethal when mixed
with its specific aggressin. The presence of aggressin, along with a
lethal dose, causes the occurrence of a disease of a severer and more
acute type, and one in which the pathological lesions are of a more
pronounced character.

It is, in the case of the strictly parasitic bacilli, such as anthrax,
fowl-cholera and plague, where a single virulent organism is probably
sufficient to cause infection, that the aggressin action is most marked.
The living bacillus can, without losing its vitality, secrete or throw off*
particles which prevent its destruction by phagocytes, and enable it,
at the same time, to multiply and produce its pathogenic effects.

The immunity obtained by the injection into the animal body of
natural aggressins is attributed by Bail to the production of anti-
aggressins.

Wassermann and Citron (1905) hold, that aggressins can be obtained
outside the animal body by shaking bacteria in serum or in distilled
water. The fluids obtained by these means, after removal of the living
germs, are capable of aiding the infective process. Their view is, that
the aggressins, whether obtained from the animal body or by this method,

(144)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 35

act by combining with the lysin of the serum, and thus interfere with the
bacteriolytic process.

Such artificial aggressins must be regarded as closely related to
certain bacterial toxins. One need only recall here, that the method
employed by Wassermann and Citron is practically identical with
Besredka's method for obtaining plague toxin.

The powerful immunising action of the natural aggressins, compared
with the weak immunising action of these artificial aggressins, is used by
Bail as an argument in favour of the essential difference in the nature of
the two groups of bodies.

It is impossible here to enter into a full discussion of the relation
of the aggressins to the " free receptors" of Neisser and Shiga (1903), and
to the substance, demonstrated by Pfeiffer and Fried berger (1905), in
serum, after its treatment with emulsions of typhoid and cholera
organisms.

The intimate relation which these subjects bear to plague immunity is
emphasised by certain work carried out in the Laboratory here, by
Captain Douglas, I. M.S. (1906), recorded in a paper read at the
Pathological Society of London, in which he showed that, contrary
to the view taken by the German and English Plague Commissions,
the supernatant fluid of Haffkine's Prophylactic Fluid has powerful
immunising properties, as had been previously maintained by Haffkine.
Where the prophylactic fluid was prepared by emulsifying the bacillus
from agar in normal salt solution, the supernatant fluid, in this case also,
was demonstrated to possess an immunising property. The writer
has shown that the injection of the supernatant fluid in the horse gives
rise to the production of a small amount of antitoxin. It is difficult
to reconcile with these results of Douglas the statement made by Bail
and Weil (1906), that attempts to immunise with watery extracts against
the true parasites prove absolutely futile, since the B. pestis has been
classified by them in this group.

Neisser and Shiga in the case of the dysentery bacillus of Shiga and
the typhoid bacillus, obtained analogous results which they attributed to
the presence of free receptors in the filtrates from the autolysed bacilli.
In that case bacteriolytic sera were obtained ; in the case of plague we
have shown that no matter whether the two fluids employed by Douglas
and the writer or the living bacillus are used in the treatment of the

(145) L

36 QUATERCENTENARY STUDIES IN PATHOLOGY

animals, the serum obtained does not owe its anti-microbial properties to
bacteriolysis. The explanation offered by Wassermann and Citron,
of the action of the watery extracts in hindering the anti-bacterial
action of a serum, viz., that the suspended bacillary particles
intervene in and prevent the bacteriolytic process, cannot, in this
case, be regarded as altogether appropriate, unless the preparation of the
bacilli for phagocytosis by the serum is regarded as a stage in an
incomplete bacteriolysis. Neither can the immunity in the case
mentioned be reasonably regarded as the result of the production of arlti-
aggressins, even if we assume that the natural and artificial aggressins
are identical. Bail states that, at least in the case of typhoid and cholera,
heating to 60° C, severely damages the aggressin, whereas in this case a
temperature of 70° C. was employed. The conditions, therefore, for anti-
aggressin production were not fulfilled in these experiments. It is,
however, conceivable that an aggressin modified by heating might still
bie capable of stimulating to the formation of an anti-aggressin, just as a
modified toxin or toxoid is capable of stimulating to antitoxin pro-
duction.

The questions which are of great interest in relation to immunity
generally, and especially to plague immunity, viz., whether the natural
and artificial aggressins are the same or different bodies, their relation to
toxins, and whether there exists an aggressin immunity sui generis, must
be regarded as subjects still sub judice.

It has been shown in the preceding part of this paper that the
protection conferred by anti-plague serum is little, if at all, dependent on
antitoxic action. Nor does the humoral view of immunity which has
been applied by many writers, among whom may be mentioned Kolle
and Martini (1902), give a satisfactory explanation of its action since, as
has been shown, no complete bacteriolysis in vitro can be demonstrated.

We must no\y, therefore, consider the part played by the serum in
relation to phagocytosis, the importance of which as a factor in immunity
is now universally admitted. Metchnikoff (1901), in reviewing the
subject of the anti-infective or protective sera, says that the action may
be direct on the micro-organism, and that it may be microbicidal,
properly so-called, agglutinative, or fixative. In addition to these
properties, and acting along with them, he upholds the view that these
sera have a stimulating effect on the leucocytes.

(146)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 37

The subject with especial relation to plague has been studied among
others by Denys and Tartakowsky (1900), who attributed the anti-
microbial action to the influence exerted by the serum on the process of
phagocytosis.

Bordet and Gengou (1901) demonstrated the presence of substance
sensibilisatrice in anti-plague serum by their method which is based on
the observation that only sensitised bacilli are capable of removing alexin
or complement from a fresh serum. The haemolysis of sensitised red
blood corpuscles is used as the test for free complement.

Employing this method, they showed that plague bacilli, which
had been in contact with heated immune serum, removed the complement
from fresh serum, and must, therefore, have been sensitised. These
authors point out that certain micro-organisms, such as the B. pestis,
are capable in this way of absorbing complement without undergoing any
morphological changes, though the possibility of a physiological change
having been effected is not excluded.

Markl (1903) made a very thorough investigation into the mechanism
of the action of anti-plague serum. He found that all the phenomena
could not be explained on the assumption that the serum had a
bacteriolytic action, since bacteriolysis takes place only in the case
of bacilli of low virulence. When he submitted plague bacilli in vitro to
the action of an immune serum which had been reactivated by fresh
normal serum, and made microscopical preparations at intervals of time,
he found that the bacilli showed no marked changes such as would
indicate that they had suffered any harm. On the other hand, if, in place
of the normal serum, he used peritoneal lymph containing leucocytes, he
found the leucocytes packed with bacilli which lost their normal staining
reaction, and, at the end of 12 hours, in the case of bacilli of moderate
virulence, the bacilli were represented only by dust-like detritus. Markl
holds that, in the case where peritoneal lymph, rich in leucocytes, is used
to reactivate the immune serum, not only is there great destruction of the
bacilli within the leucocytes, but there occurs extra-cellular destruction of
the bacilli which is inversely proportional to the virulence of the bacillus.
When a highly virulent plague culture is injected intraperitoneally
the multiplication of the bacilli can be followed. There is very little
exudate, in which only a few leucocytes are present, and there is a rapid
increase in the number of bacilli followed by death in 24 hours.

(147) L I

38 QUATERCENTENARY STUDIES IN PATHOLOGY

When anti-plague serum has been injected into the animal 24 hours
before the bacillus, there is observed, during the first hour, an increase in
mononuclear leucocytes, around which the bacilli are clumped, some
of them being englobed by the mononuclear leucocytes. After the first
hour the polynuclear leucocytes multiply rapidly and englobe the bacilli,
so that in three hours no free bacilli, and in six hours no bacilli
either intra or extra-cellular, are observable. The exudate now consists
entirely of polynuclear leucocytes. The animal in this case recovers.

When the number of bacilli is so great that the resistance conferred
by the serum is insufficient to protect the animal, the bacilli which
remain undevoured by the phagocytes continue to multiply till the death
of the animal results.

The bacilli lying free and those within the mononuclear leucocytes
retain their normal morphology and staining reactions, whereas, those
within the polynuclear leucocytes tend to stain badly and to assume
coccal-like forms.

In regard to the nature of the action of the serum in relation to
phagocytosis, Markl says : — " To determine whether the injection of the
immune serum has an influence on the cells of the body, or whether the
bacilli are directly influenced, the following experiment was performed.
A suspension of plague bacilli in normal salt solution was added to
inactivated immune serum, and the mixture was placed three hours in
the incubator. After centrifugalisation the clear fluid was pipetted from
the sediment, which, after being several times washed with salt solution,
was injected intraperitoneally into rats. These bacilli thus acted on by
the immune serum behaved in the animal body exactly like bacilli which
had been injected along with free immune serum. They were englobed
by the phagocytes.

The phagocytosis occurred so rapidly and was so extensive that no
doubt remains that the immune serum has a direct influence on the
bacilli, entering into union with them and conferring on them a positive
chemiotaxis." *

He concludes that in the case where bacilli of low virulence are
injected extracellular bacteriolysis plays an important part, whereas,

* A discussion of the relation of Markl's work to the views of Douglas and Wright is
avoided in this paper as of too controversial a nature.

(148)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 39

where highly virulent bacilli are injected, phagocytosis is the only agency
at work in their destruction.

The writer's experience and results are largely in agreement with
Markl's, but, as already mentioned, in dealing with the subject of
bacteriolysis in vitro, evidence of complete destruction of the bacilli,
even of low virulence, could not be obtained. The results observed by
Markl when the bacilli were injected into the peritoneum can also be
obtained in the subcutaneous tissue. The writer has found that when a
bacillus of high virulence is injected into the subcutaneous tissue of the
abdomen, or where infection is caused by shaving the skin and rubbing
the culture on the shaved area, i.e., by the cutaneous method, an enormous
multiplication of the bacilli takes place within the lymph spaces, which
become distended with lymph in the vicinity of the point of inoculation.
Microscopical preparations from this region show a complete, or almost
complete, absence of leucocytes. On the other hand, if the animal has
been immunised by an injection of anti-plague serum, given say 24 hours
before the injection of the bacilli, instead of the diffuse indefinite swelling,
or absence of any swelling, visible to the naked eye found in the
unprepared rat, a well marked generally sharply circumscribed swelling
is observed, and microscopical examination shows in it the presence of an
enormous number of polymorphonuclear leucocytes, many of which are
packed with bacilli. Animals in which this condition is observed either
recover or die at a much later period than the controls.

Summary.

Toxins can be demonstrated in the filtrates of bouillon cultures of
the B. pestis. These toxic substances are probably set free by the
disintegration of the bacillus through autolysis, but, since Albrecht and
Ghon and Markl were able to demonstrate their presence in young
cultures, the possibility is not excluded that they are toxins in the
narrower sense, i.e., secretory products of the bacillus. The toxin is
much more toxic for the rat and mouse than for the rabbit and guinea-
pig. The toxicity for mice is destroyed by heating to 70° C. for 30
minutes.

Under certain conditions, the injection of toxin stimulates the animal
body to the formation of antitoxin, and the experiments recorded here

(149)

46 QUATERCEt^TENARY STUDIES IN PATHOLOGY

appear to show that this antitoxin is capable of neutralising the toxin,
either when mixed with it or when the serum is injected twenty-four
hours before the toxin, and when the two substances are injected at
different parts of the body.

It seems doubtful whether it will be possible to maintain a hard
and fast line of demarcation between the toxins proper and the endo-
toxins. The injections of emulsions of bacilli from agar, as well as of the
toxic filtrates from broth cultures, may, in certain cases, stimulate the
animal organism to the production of an antitoxic serum as was shown
in the Laboratory here, by Todd (1904), in the case of the B. dysenteriae
of Shiga. The view that it is a general biological law, as stated by Kolle,
that the injection of endotoxins into the animal body does not stimulate
to antitoxin production, cannot be unreservedly accepted, but must
be subjected to further investigation.

Whether the toxin and antitoxin in the case of plague differs or not
essentially from those obtained in the case of diphtheria and tetanus,
it must be admitted that it is impossible to obtain them in anything like
the same degree of concentration.

The neutralisation of toxin by the antitoxin in vitro is not due to the
occurrence of a precipitin reaction. The serum of a horse giving a much
greater precipitum with a particular toxin may possess a much lower
antitoxic value than the serum of a horse which gives a much smaller
precipitum with the same toxin. The fact that the neutralisation also
takes place in vivo is against the view that a precipitin reaction plays a
part in the phenomenon, for Rostoski, Michaelis and Oppenheimer
(1902), showed that this reaction never takes place in the blood of
the living animal.

So far as has been demonstrated experimentally, antitoxic and
bacteriolytic actions play little part in the protection afforded by the
anti-plague serum. The action of the serum can probably be best
interpreted as follows : — The serum acts, not by any influence on the
cells of the animal body, but directly on the bacilli, which, after being
submitted to its action, no longer repel but rather attract the leucocytes,
by which the microbes are then englobed and digested. This was first
shown in the case of the B. pestis by Markl.

The change in the microbe has been expressed by stating that the
immune serum converts the negative chemiotaxis of the bacillus into

(ISO)

RESEARCHES ON CERTAIN PROBLEMS OF fLAGtJE IMMUNITY 4!

a positive chemiotaxis. The substance producing the change has been
variously designated as fixateur, substance sensibiHsatrice, opsonin and
incitor.

It is still an open question whether we are here dealing with a stage
in an incomplete bacteriolysis in which the amboceptor acts as the
intermediary between the bacillus and the leucocyte. The view is one
which can be maintained with much reason. Certain organisms, such as
the Cholera vibrio, when acted on by an immune serum containing
amboceptor and complement, undergo bacteriolysis, which can be seen
under the microscope, or estimated by the absence of growth. Other
organisms, such as the B. pestis, are acted on by an immune serum,
but in this case the solution or death of the bacillus does not take place
outside the cells. That the organism has, however, suffered change
is shown by the altered behaviour of the leucocytes to which we have
just referred.

Whether the bacillus undergoes an extra-cellular bacteriolysis, or
is digested after being phagocytosed, it must not be overlooked, that
in both cases it is highly probable that the ultimate protection of the
animal is effected by the phagocytes, which, in the former case, take
up the toxic particles set free by bacteriolysis, in the latter, take up and
digest the bacillus and retain the toxic particles liberated during the
process.

When the extra-cellular or intra-cellular bacteriolysis is carried on
with too great energy so that the leucocytes are not strong or numerous
enough to deal with the toxic products, the death of the animal results.

The question as to whether anti-plague serum owes any of its
protective power to the " stimulines " of Metchnikoff and the " anti-
aggressins" of Bail must be decided by further investigations.

Conclusions.

1. That toxins can be obtained from broth cultures of the plague
bacillus.

2. That these toxins are probably endotoxins, as the toxicity of the
filtrate increases with the age of the culture, and becomes considerable
only after many of the bacilli have been autolysed.

(151)

4^ QUATERCENTENARY STUDIES IN PATHOLOGY

3. That the injection of large doses of these toxic filtrates is capable
of stimulating the animal body to the formation of an antitoxin.

4. That the neutralisation of the toxin by antitoxin is not merely due
to a precipitin reaction for {a) it occurs in vivo where no precipitin
reaction takes place, and {b) a serum with little or no antitoxic properties
may be capable of producing a marked precipitin reaction and vice versa.

5. That neither antitoxic nor bacteriolytic actions play much part in
the protection afforded by an anti-plague serum.

6. That the anti-microbial action of an anti-plague serum is, as was
first shown by Mark!, almost, if not entirely, due to a direct action of the
serum on the microbe, by means of which its negative chemiotaxis is
converted into a positive chemiotaxis, so that it no longer repels but
rather attracts the phagocyte. The substance in the serum producing
this action is thermostable, and has been variously named fixateur,
substance sensibilisatrice and incitor.

(152)

RESEARCHES ON CERTAIN PROBLEMS OF PLAGUE IMMUNITY 43

LITERATURE.

Albrecht & Ghon (1900), " Ueber die Beulenpest in Bombay im Jahre 1897,"
Gesammtbericht des von der kaiserlichen Akademie der Wissenschaft in
Wien, zum Studium der Beulenpest nach Indien entsendeten Commission.
Th. IL €., p. 780.

Bail, O. (1905)' " Untersuchungen iiber die Aggressivitat des Cholera vibrio."
Archiv f. Hyg., Vol. 53, p. 302.

Bail, O. (1905)5 " ^Jer akute Tod von Meerschweinchen an Tuberkulose." Wien.
klin. Woch., XVIIL, p. 211.

Bail, O., and Kikuchi, Y. (1905), "Bacterizide Reagensglasversuche mit

Choleravibrionen." Archiv f. Hyg., Vol. 53, p. 275.
Bail, O., and Weil, E. (1906), " Unterschiede zwischen aggressiven Exsudaten

und Bakterienextrakten." Centralbl. f. Bakt. L, Orig. Bd. XXXX., p. 371.
"Bericht iiber die Thatigkeit der zur Erforschung der Pest im Jahre 1897, nach

Indien entsandten Kommission," erstattet von Dr. Gaffky, Dr. Sticker,

Dr. Pfeiffer, u. Dr. Dieudonne." "Arbeiten aus dem kaiserlichen

Gesundheitsamte," 1899. ^d. XVI., p. 3.

Besredka (1905), "Etudes sur le Bacille Typhique et le Bacille de la Peste."

Annales de ITnstitut Pasteur, XIX., p. 477.
Besredka (1906), " De I'Anti-Endotoxine Typhique et des Anti-Endotoxines en

General." Annales de ITnstitut Pasteur, XX., p. 149.
Bordet (1895 J, "Les Leucocytes et les Propri^tes Actives du Serum chez les

Vaccines." Annales dTnstitut Pasteur, T. IX., p. 462.
Bordet, J., and Gengou, O. (1901), "Sur I'Existence de Substances Sensibili-

satrices dans la plupart des Serums Antimicrobiens." Annales de I'Institut

Pasteur, Vol. XV., p. 289.
Denys, J., and De Velde, H. Van (1895), "Sur la Production d'une Anti-

Leucocidine chez les Lapins Vaccines contre le Staphylocoque Pyogene."

La Cellule, Vol. XL, p. 359.
Denys and Tartakowsky (1900), "Procede d'Inoculation augmentant I'Action

de Serum Antipesteux dans une proportion considerable." Bull, de

I'Academie Royale de Belgique, T. XIV., p. 419.

(153)

44 " QUATERCENTENARY StltOlES IN PATIIOLOGV

Douglas, S. R. (1906), "On Plague Immunity." Path. Society of London,

1906.
HuEPPE, F., and Kikuchi, Y. (1905), "Ueber eine neue sichere und gefahrlose

Immunisierung gegen die Pest." Centralbl. f. Bakt. I. Orig. XXXIX., p. 610.

Klein (1905), "Preliminary Report of the Local Government Board on a new
Plague Prophylactic."

ICoLLE (1903), "Studien iiber das Pestgift." Festschrift zum sechzigsten
Geburtstage, von Robert Koch, p. 351, 1903.

KoLLE, W., Hetsch, H., and Otto, R. (1904), "Weitere Untersuchungen iiber
Pest, im Besonderen iiber Pest-Immunitat." Zeitschrift f. Hyg. Bd.
XLVIIL, p. 368.

KoLLE and Martini (1902), " Ueber Pest." Deutsche med. Woch., 1902,
pp. I, 29, 45, 60.

KoLLE, W„ and Otto, R. (1902), " Vergleichende Wertpriifungen von Pestserum
verschiedener Herkunft." Zeitschrift f. Hyg., Bd. XL., p. 595.

Markl, G, (1898), "Beitrag zur Kenntnis der Pesttoxine." Centrabl. f. Bakt.,
Bd. XXIV., pp. 641-728.

Markl, G. (1901), "Weitere Untersuchungen iiber die Pesttoxine." Zeitschrift f.
Hygiene., Bd. XXXVIL, p. 401.

Markl, G. (1905), "Zur Kenntnis des Mechanismus der kiinstlichen Immunitat
gegen Pest." Zeitschrift f. Hyg., Bd. XLIL, p. 244.

Metchnikoff (1897), "Sur le Peste Bubonique." Annales de I'lnstitut
Pasteur, Vol. XL, p. 737.

Oppenheimer and Michaelis (1902), " Mittheilungen iiber Eiweiss-pracipitine."
Archiv. f. Anat. u. Physiologic (Physiol. Abt.). Supplement, p. 436.

Neisser, M. (1904), "Die Methodik des bactericiden Reagensglasversuches."
Gesammelte Arbeiten zur Immunitatsforschung von Ehrlich, Berlin, 1904,

p. 493-
Neisser and Shiga (1903), "Ueber freie Receptoren von Typhus- and

Dysenterie-bazillen und iiber das Dysenterietoxin." Deutsche med. Woch.,

No. 4, p. 61.
Pfeiffer, R., and - Friedberger, E. (1905) "Ueber antibakteriolytische

(antagonistische) Substanzen normaler Sera." Deutsche med. Woch. No. i,

pp. 6-8.
RosTOSKi (1902), "Zur Kenntniss der Praecipitine." Verhandlungen der

Physiol. Med. Gesellschaft zu Wurzburg, N.F., Bd. XXXV., Nr. 2.

(154)

kfiSEARCHfiS ON CERTAtN PROBLEMS Ot PLAGUE IMMUNITY 45

Tavel, Krumbein and Glucksmann (1902), " Ueber Pestschutzmaassregeln."

Zeitschrift f. Hyg., Bd. XL., p. 239.
Terni, C, and Bandi, I. (1900), "Nouvelle Methode de Preparation du Vaccin

Antipesteaux." Revue d'Hygiene, XXIL, pp. 62-74, 1900. (Ref.

Centralbl. f. Bakt. I., Bd. XXVIL, p. 629).
Todd, Chas. (1904), "On a Dysentery Toxin and Antitoxin." Journal of

Hygiene, Vol. 4, No. 4, p. 480.
Wassermann and Citron (1905), " Zur Frage der Bildung von bakteriellen

Angriffsstoffen im lebenden Organismus." Deutsche med. Woch., p.

IIOI.

Weil, E. (1905), " Untersuchungen liber Infektion und Immunitat bei Hiihner-
cholera." Archiv. f. Hyg., LIL, p. 412.

(155)

EXPERIMENTAL STUDY OF THE IMMUNITY
AGAINST BACILLUS PYOCYANEUS.

By William Bulloch, M.D.,

Bacteriologist to the London Hospital^ and Lecturer on General Pathology^
London Hospital Medical College^ London^ E.

(157)

\:iinijh

KiVu.

UJJI3AII

4:iAu..)A

i.M .r'DOU.iuM i:^iALiJiV/ vH

Expenmenkl Study of the Immunity against Bacillus

Fyocyaneus.

Introduction.

Since G^ssard (^) in 1882 isolated pure cultures of bacillus pyocyamus
for the first time, this remarkable microbe has been the subject of a great
amount of investigation. A special interest in its pathogenic properties
was aroused by the publication in 1889 of Charrin's monograph on "la
maladie pyocyanique." Charrin (2) noted that pigeons, guinea-pigs, rabbits
and frogs were susceptible, and in particular that the rabbit exhibited
certain characteristic symptoms. In this animal he witnessed several
types of the pyocyanic disease, viz : —

\. Superacute type, fatal in 24 hours.
2. Acute type, fatal in i -2 days.

;\ 3. Subacute type, fatal in 1-3 weeks.

4. Chronic type, fatal in several months.

In the chronic cases, after an incubation period lasting from 15 days
to two months, Charrin saw the onset of paralysis affecting one or both
hind limbs, and occasionally spreading to the anterior extremities. The
paralysis was of a spastic type, and death was preceded by a condition of
marasmus. Among the anatomical changes, Charrin observed enteritis,
nephritis, and in one case amyloid infiltration.

By growing bacillus pyocyaneus in bouillon for a week, and then
filtering the culture through porcelain, Charrin found that, in large doses
(60-80 c.c), the rabbit developed symptoms of disease.

In 1896 appeared the important research of Wassermann (3), who
demonstrated that the guinea-pig is the most sensitive animal to infec-
tion by the pyocyanic bacillus. The intraperitoneal inoculation of ^^ of
a platinum loop of a culture was sufficient to cause death in 24 hours.

(159)

4 QUATERCENTENARY STUDIES IN PATHOLOGY

Wassermann also made special inquiry into the existence of a
pyocyanic toxin, which he obtained by growing cultures of the bacillus
for about six weeks, at the end of which time he destroyed the vitality
of the culture by means of toluol. Such toluolised cultures were highly
toxic, killing guinea-pigs rapidly in a dose of '5 c.c. Even with young
toluolised cultures, he obtained evidence of toxic effect, and, as the bodies
of pyocyanic bacilli killed with chloroform were much less toxic, he
concluded that in the broth there must be a specific pyocyanic toxin.
The toxin was not entirely robbed of its toxic properties by heating
to 115^ C.

In his immunisation experiments Wassermann found great differences,
according as the animals were injected with the pyocyanic toxin or with
the bacilli themselves. In the former case the serum protected against
both the toxin and the culture, whereas the serum of animals,
immunised with bacilli, protected only against the bacilli, but was
unable to neutralise the lethal effect of the toxin. In most cases the
immunised animals (guinea-pigs) could withstand many times the lethal
dose of the living cultures, but only small multiples of the lethal toxin
dose. As had been shown by Pfeiffer and Kolle in the case of cholera
and typhoid, it was not possible to raise the immunity to an unlimited
degree.

The nature of the immunity induced was also studied by Wasser-
mann, and although he was unable to demonstrate any bactericidal
powers in the immune serum in vitro, he found that in vivo the pyocyanic
bacillus is destroyed in the peritoneal fluid, the destruction being
accomplished without the agency of leucocytes. He considered that the
process was similar to that which obtains in the case of typhoid and
cholera. To the sera of animals immunised with increasing doses of
pyocyanic toxin he ascribed true antitoxic properties.

Working along the same lines as Wassermann, Gh^orghiewsky (4)
obtained divergent results, in so far that he found the pyocyanic bacillus
to be taken up by the phagocytes in the peritoneal cavity, and destroyed
in their interior.

When a non-lethal dose of bacillus pyocyaneus was introduced into
the peritoneal cavity of a normal guinea-pig, there was a definite
leucocytosis at the end of two to three hours, and within six hours all the
injected bacilli were engulfed by the leucocytes, in the interior of which

(160)

STUDY OF THE IMMUNITY AGAINST BACILLUS PYOCYANEUS 5

they were destroyed. When a similar experiment was performed on a
well -vaccinated guinea-pig, the intra-cellular destruction of the bacilli
was quicker and more perfect. Gheorghiewsky thus attributes the
immunity against bacillus pyocyaneus to phagocytosis, and not to a
destruction of the bacilli by the serum. He was unable to demonstrate
any bactericidal action in the serum either in vitro or in vivo.
Incidentally Gheorghiewsky also noted a peculiarity of the immune
serum which had been observed by Charrin, viz., that it prevents the
bacillus from forming pyocyanin.

P. Muller(S) was also unable to demonstrate any bactericidal powers
in pyocyanic immune serum in vitro in aerobic conditions of the experi-
ment. In anaerobic conditions, however, he considered that the
bactericidal action was very marked, and that it could be inhibited by
heating the immune serum to 55° C. The subsequent addition of a
normal serum to the inactive heated serum restored the bactericidal
power.

Margarethe Breymann (6) made a special investigation into the
question of the toxicity of filtrates of pyocyanic cultures. The organism
grown in bouillon for periods varying from 7-50 days, was passed
through the Chamberland filter, the filtrate being inoculated into mice,
guinea-pigs and rabbits. She found that very large doses (100 c.c.) could
occasion death of rabbits, with great emaciation, quantities less than
this being practically without effect. Mice and guinea-pigs withstood
respectively doses of i c.c. and 40 c.c. of the filtrate. On concentrating the
filtrates over sulphuric acid, they were found to be non-toxic, even in
quantities representing 100 c.c. of the original filtrate. The cause of
this loss of toxicity was probably due to the disappearance of volatile
bodies — the toxic action of which had been already maintained by
Charrin (7).

Maclntyre (S) grew bacillus pyocyaneus in large quantities on agar,
and after 14 days he precipitated with alcohol, collected the precipitate on
a hard filter, and extracted with alcohol and ether. He found that 5 mgr.
of the germ substance extracted in this way killed guinea-pigs, after
intra-peritoneal injection, in 12 hours. The amount injected was about
I -50,000th of the body weight of the animal. Injected subcutaneously,
one part of cell substance to 10,000 parts of body weight produced no
permanent ill effects.

(161) M

6 QUATERCENTENARY STUDIES IN PATHOLOGY

Heating to I20°C. in the autoclave did not rob the toxic substance
of nnuch of its toxicity. Maclntyre failed to produce immunity to intra-
peritoneal injection by a previous series of subcutaneous inoculations.

In his most recent publication on the subject of pyocyanic immunity,
Wassermann (9) still maintains the distinction between the antitoxic and
bactericidal properties of the immune serum ; although he admits that
the antitoxin does not, as in the case of other antitoxins, follow the law
of multiples. He explains this by assuming that the pyocyanic
" toxin," in addition to the true exotoxin, contains endotoxin which
has dissolved out of the bodies of the bacilli. The bactericidal action
he still holds to be similar to that which obtains in the case of cholera,
although the extra-cellular destruction of the B. pyocyaneus is slow
compared with that of the vibrio cholerae.

Experimental.

Four races of bacillus pyocyaneus were used in the following experi-
ments, and they were isolated from the throat, an ulcer of the leg, from
the blood, and from the intestine respectively, of human beings. In all
cases the cultures were highly pathogenic for guinea-pigs, and less so for
rabbits.

When freshly isolated, one loopful of an agar culture injected into the
peritoneum of a guinea-pig killed the animal acutely within 12 hours, the
autopsy revealing intense peritonitis with great desquamation of endo-
thelium, and accumulation of leucocytes in the thick fibrinous exudate.
Minute haemorrhages were numerous on the mesentery. The liver and
kidney, on microscopic section, showed evidences of severe cloudy
swelling. Cultivations from peritoneal exudate, liver, and heart blood,
showed that the bacillus had become generalised over the body, the
condition being really a pyocyanic septicaemia.

Cultures from the heart blood were inoculated into large flasks
containing 3-6 litres of i % peptonised bouillon, and grown for periods
varying from two days to three months. Within twenty-four hours a fine
bacterial film appears on the surface, and within three or four days it has
become converted into a thick membrane. The flasks were then
thoroughly shaken, so as to dislodge the membrane to the bottom. As
soon as a new membrane appeared, the process was repeated. When
the flasks were kept untouched in the incubator for three to four weeks,

(162)

STUDY OF THE IMMUNITY AGAINST BACILLUS PYOCYANEUS 7

the general colour of the medium was a yellowish-brown. On shaking
violently, this was instantly converted into a bright grass green colour
from the action of the oxygen of the air, and the formation of pyocyanin
and fluorescent green. In the course of time the consistence of the
medium alters, so that, instead of the limpid clear broth, a thick stringy
mucinous-like material occupies the whole flask.

Filtration Experiments.

Cultures of the bacillus in broth were taken at various times and
passed through the Berkefeld filter. With young cultures up to 7 days
this was easy, but when the mucinous cultures had to be dealt with,
filtration was found to be extremely difficult, even the most open-pored
Berkefeld filters becoming rapidly clogged up.

Filtrate from 2 day old broth culture.

Animal.

Quantity inoculated.

Result.

Guinea-pig.

5 c.c. intraperitoneal.

No effect.

M

10 c.c. II

No effect.

II

20 c.c. M

Slight loss of weight.

II

40 c.c. M

Loss of weight — recovery.

Rabbit.

5 c.c. intravenous.

No effect.

II

10 c.c. intraperitoneal.

No effect.

M

20 c.c. II

Slight loss of weight.

"

50 c.c.

Loss of weight — recovery.

Filtrate from 7 day old culture.

Animal.

Quantity inoculated.

Result.

Guinea-pig.

Rabbit.
II

10 C.C. intraperitoneal.

30 c.c. II
20 c.c. II
40 c.c. II

No effect.

Slight loss of weight.

No effect.

No effect.

('63)

M I

QUATERCENTENARY STUDIES IN PATHOLOGY

Filtrate from 30 day old culture.

Animal.

Quantity inoculated.

Result.

Gumea-pig.
Rabbit.

II

20 C.C. intraperitoneal.

20 C.C. II
40 C.C. II

No effect.
No effect.
Loss of weight — recovery.

Filtrate from 60 day old culture.

Animal.

Quantity inoculated.

Result.

Guinea-pig.
Rabbit.

30 C.C. intraperitoneal.

40 C.C. II

Loss of weight — recovery.
Loss of weight — recovery.

From these experiments, it is manifest that even larp^e doses of
filtrates of cultures of various ages fail to produce lethal effects. It
was considered that increase in virulence of the cultures might lead to
increased toxin production, and to this end the bacillus was passed
through several series of guinea-pigs and rabbits, cultures being ultimately
made from the last members of the series. Filtrates of these cultures
were found to be incapable of producing lethal effects in doses up to
40 C.C for guinea-pigs and 50 c.c. for rabbits. Wassermann has suggested,
that although the toxin is largely a secretory product of the bacillus
pyocyaneus, it may be kept back in the mucinous material, and in
this way be prevented from passing the filter. This, however, does
not apply to young cultures up to the first seven days, and here, as
the experiments show, a soluble toxine, if such exist, is present only
in very minute quantities.

Action of dead cultures of B. Pyocyaneus.

Dead cultures prepared in various ways were used in a number
of experiments. Following Wassermann, the cultures were grown on
bouillon or on agar. The bouillon cultures, after 2, 7, 21, 31 and 60 days,

(164)

STUDY OF THE IMMUNITY AGAINST BACILLUS PYOCYANEUS 9

were covered with a thick layer of toluol, and were then repeatedly
shaken until complete sterility was obtained.

Agar cultures were grown en masse in Roux's bottles, the bacterial
growth being subsequently washed off in a small quantity of normal
saline solution. The thick emulsion obtained in this way was then
removed by a pipette to a sterile stoppered bottle, and killed by toluol
or chloroform.

Cultures in broth or on agar were also killed by heat at 65° C. and at
100° C, the toxicity being afterwards tested.

Effect of Toluolised Bouillon Cultures.

Animal.

Quantity inoculated.

Result.

Guinea-pig.

I C.C.

2 day culture, intraperitoneal.

Dead 12 hours.

n

o"5 c.c.

i< II

Dead 24 hours.

fi

0'2 c.c.

II II

Very ill — recovered.

Guinea-pig.

I c.c.

7 day culture, intraperitoneal.

Dead 12 hours.

It

o'5 c.c.

M II

Dead 24 hours.

II

0'2 c.c.

II II

Ill— recovered.

Guinea-pig.

I c.c.

21 day culture, intraperitoneal.

Dead 8 hours.

II

0-5 c.c.

II It

Dead 12 hours.

II

0-2 c.c.

II It

Dead 3 days.

M

o'l c.c.

If It

Ill— recovered.

tt

0*05 c.c.

II II

Ill — recovered.

Guinea-pig.

I c.c.

31 day culture, intraperitoneal.

Dead 8 hours.

II

0-5 c.c.

II II

Dead 24 hours.

ft

0'2 c.c.

II II

Dead 4 days.

ti

O'l c.c.

II ti

Very ill — recovered.

Guinea-pig.

I c.c.

60 day culture, intraperitoneal.

Dead 24 hours.

11

0-5 c.c.

II II

Dead 6 days.

II

0-2 c.c.

II If

Ill — recovered.

From these results it is manifest that Wassermann's so called
"toxin" can produce, in doses of '5 - I'O c.c. acute lethal effects. Post-
mortem, the main changes are witnessed in the peritoneum, which
presents evidences of acute inflammation, and in the liver and kidney,
which are inflamed, and show acute cloudy swelling.

(165)

16

quatercentenarv studies in pathology

Effect of Broth Cultures killed at dj^ C.

Animal.

Quantity inoculated.

Result.

Guinea-pig.

I C.C.

7 day culture, intraperitoneal.

Dead 12 hours.

II

o'5 C.C.

7

Ill — recovered.

It

I C.C.

2 1 day culture, intraperitoneal.

Dead 12 hours.

M

05 C.C.

II II

Dead 2 days.

It

I C.C.

31 day culture, intraperitoneal.

Dead 24 hours.

II

0-5 C.C.

M II

Ill — recovered.

M

I C.C.

60 day culture, intraperitoneal.

Dead in 8 days.

II

0-5 C.C.

II II

Loss of weight.

Effect of Bouillon Cultures exposed at 100° C for | hour

Animal.

Quantity inoculated.

Result.

Guinea-pig.

2 C.C.

7 day culture, intraperitoneal.

Ill — recovered.

I C.C.

II II

Ill— recovered.

2 C.C.

21 day culture, intraperitoneal.

Dead 4 days.

I C.C.

II :i

Ill — recovered.

2 C.C.

31 day culture, intraperitoneal.

Ill — recovered.

I C.C.

II It

No effect.

Effect of Emulsions from 24. hour agar cultures killed by Toluol.

(A 24 hour culture, which covered the whole surface of a Roux's bottle, was
emulsionised in 20 c.c. saline solution — the emulsion being then sterilised with
toluol.)

Animal.

Quantity injected.

Result.

Guinea-pig.

II

I C.C. emulsion.
o'5 C.C. II

Dead 24 hours.
No symptoms.

Effect of 2^ hour agar cultures killed by chloroform.

A 24 hour culture which covered the whole surface of a Roux's bottle
was exposed to chloroform vapour, the bottle being well closed by
an indiarubber bung.

Before the action of the chloroform, o*i of a loopful of the living

(166)

STUDY OF THE IMMUNITY AGAINST BACILLUS PYOCYANEUS 11

culture killed a guinea-pig in 12 hours. 10 loopfuls of the chloroformed
culture produced no symptoms in another guinea-pig.

I c.c. of a very thick emulsion of a chloroformed culture produced
death in 4 days.

From these experiments it is manifest that toluolised bouillon cultures
are lethal in doses of -5 - i c.c. The bacillary bodies themselves, killed by
toluol or chloroform, are much less toxic. The toxicity of toluolised
broth cultures may be largely due to poisons which have slowly diffused
out of the bacillary bodies. It is also possible that it is due to secondary
products produced by the fermentation of the medium (bouillon) in
which the bacillus has been growing. The absence of toxic action
in young filtrates is in favour of the latter view, and against the opinion
of Wassermann, that the so-called " pyocyanic toxin " is a true exotoxin.

Effects of Autolysed Cultures.

Autolysed cultures were made according to the methods of Conradi (1°)
and Neisser (^i). Following the first of these methods, bacillus pyocyaneus
was grown in Roux's bottles at 37° C. for 20 hours. At the end of this
time the culture mass was washed into a saline solution and allowed to
digest in this for 24 hours, after which time the fluid was passed through
the Berkefeld filter.

In the case of autolysates made by Neisser's method, large agar
cultures were suspended in saline solution, heated to 60° C. for i hour,
and then placed in the incubator for 2 days to digest. At the end of
this time the fluids were passed through the filter.

Animal.

Amount inoculated.

Result.

Rabbit.

5 c.c.

autolysate

(Conradi) intraperitoneal.

No effect.

10 c.c.

II

II II

No effect.

15 c.c.

11

II n

No effect.

10 c.c.

•,

•1 intravenous.

No effect.

10 c.c.

II

(Neisser) intravenous.

No effect.

15 c.c.

II

II intraperitoneal.

No effect.

20 c.c.

II

II

Death in 8 days.

Considering the large quantity of bacilli employed in the preparation,
these autolysates appear to be practically non-toxic in doses up to 15 c.c.
A lethal effect was first seen in doses of 20 c.c.

(167)

12

QUATERCENTENARY STUDIES IN PATHOLOGY

Virulence of B, Pyocyaneus obtained direct from man. — Conservation
and Exaltation of Virulence.

The virulence of the pyocyanic bacillus obtained from man varies.
In the case of the races obtained from the leg ulcer and from the intestine,
the virulence of the first subcultures were such that i loopful (2 mgr.)
injected intraperitoneal caused an acute lethal pyocyanic sepsis in the
guinea-pig within 12 hours. With cultures from the throat and from the
blood, the virulence was such that '25 loopful killed guinea-pigs acutely.
The virulence was well maintained, even for months, in cultures grown
on agar stabs at 37° C. for 24 hours, and subsequently at 20° C.

From each of the four races used in the experiments, passages were
made through both rabbits and guinea-pigs to determine whether and to
what extent exaltation is possible. After the death of the animal the
bacillus was recovered from the heart blood, grown on agar for 24 hours,
and inoculated into the peritoneum of the next animal in series.

No.

Dose.

Result.

Rabbit

I.

2 loops culture, intraperitoneal.

Death 12 hours.

2.

I II II

Death 3 days.

3-

I loop culture from Rabbit

I,

intraperitoneal.

Death 12 hours.

4-

•5

I,

M

Death 7 days.

5-

I loop culture from Rabbit

2,

intraperitoneal.

Death 12 hours.

6.

'5 " "

2,

II

Death 12 hours.

7.

•4 U M

2,

II

Death 2 days.

8.

•4 loop culture from Rabbit

5.

intraperitoneal.

Death 2 days.

9-

•3

5.

II

Death 5 days.

10.

•3 loop culture from Rabbit

8,

intraperitoneal.

Death 3 days.

II.

•25 M II

8,

II

Recovery.

12.

•3 loop culture from Rabbit

10,

intraperitoneal.

Death in 2 days.

13-

•25

10,

M

Death in 5 days.

14.

•25 loop culture from Rabbit

12,

intraperitoneal.

Death in 8 days.

15-

•20 II II

12,

II

Recovery.

(168)

STUDY OF THE IMMUNITY AGAINST BACILLUS PYOCYANEUS 1 3

In neither of the four series in rabbits did less than -3 loopful kill
acutely irrespective of the number of passages.

Guinea-pigs (intraperitoneal inoculation).

No.

Dose.

Result.

I.

I loop culture.

Dead 8 hours.

2.

•5

Dead 12 hours.

3.

•3

Dead 24 hours.

4.

'I "

Dead 4 days.

5-

•3 loop from guinea-pig 3

Dead 8 hours.

6.

•2 ir 1

3

Dead 12 hours.

7-

"I II 1

3

Dead 24 hours.

8.

•I If 1

7

Dead 12 hours.

9-

■05 II 1

7

Dead 4 days.

10.

•I II 1

8

Dead 12 hours.

II.

•05 II ' 1

8

Dead 24 hours.

12.

•03 If 1

f 8

Ill — recovered.

13-

•05 II 1

t II

Dead 24 hours.

14.

*04 II 1

1 II

Dead 8 days.

15-

•03 II 1

1 II

Ill — recovered.

16.

•04 II 1

1 14

Ill — recovered.

17.

•04 II 1

1 14

Ill— recovered.

Although the virulence of the cultures was considerably greater than
has been reported by most authors, it was found impossible to exalt the
virulence beyond certain limits. For the rabbit 0*3 of a loopful and
for the guinea-pigs 0*05 of a loopful were the smallest doses which
had lethal effects in spite of many passages. The pyocyanic bacillus
is therefore very different in this respect from microbes like pneumoccus
and streptococcus, and resembles bacteria like typhoid bacillus.

Immunity against Bacillus Pyocyaneus.

A large number of experiments were carried out on rabbits for

(169)

t4 QUATERCENTENARY STUDIES IN PATHOLOGY

the production of immunity against bacillus pyocyaneus. The methods
employed in different animals comprised : —

1. Immunisation with increasing doses of pyocyanic filtrates.

2. ti II toluolised broth cultures.

3. »i M heated cultures.

4. ti M heated cultures followed by

living cultures.

5. II II autolysed cultures (Neisser's

and Conradi's methods).

The general result of these experiments, several of which were
performed in each series, has shown that the immunity acquired is never
of a very high degree. In this respect my results differ from those
of Wassermann. Even in animals which have been apparently solidly
immunised over a considerable period with toluolised cultures, death
ensued acutely when about 3 - 4 lethal doses of the living culture were
injected into the peritoneal cavity. A few of the immunisation
experiments will sufifice to make this apparent.

Experiment i.

1904.
Nov. 8. — Rabbit's serum tested — no agglutination effect on B. Pyocyaneus, in

dilution of i m 2.
Nov. 8. — Rabbit inoculated with 8 c.c. autolysed filtrate (Conradi).
Nov. II. — Serum agglutinated bacilli in 5 min., in dilutions from i in 4 — i : 100.
Nov. 14. — Agglutination complete i in 800.
Nov. 21. — 15 c.c. autolysed filtrate inoculated.
Nov. 27. — Agglutination complete i in 1700.
Dec. 29.— 20 c.c. autolysed filtrate inoculated.
Dec. 30. — Animal seriously affected.

1905.
Jan. 4. — Agglutination i in 2000.

Jan. 14. — Animal showed paresis of hind limbs and emaciation.
Jan. 20. — Animal died ; complete paraplegia, marasmus, (maladie pyocyanique
of Charrin). Post mortem — Parenchymatous degeneration of
kidney.

(170)

14.—

It

26.—

II

6.—

It

14.—

tt

24.—

II

2.

II

14.—

M

24.—

II

STUDY OF THE IMMUNITY AGAINST BACILLUS tYOCYANEUS I^

Experiment 2.
1905.

Jan. 9. — Rabbit received 10 c.c. autolysed filtrate.

10 c.c. autolysed filtrate.

20 C.C. autolysed filtrate.

Feb. 6. — It Jg- c.c. toluolised broth culture.

\ c.c. toluolised broth culture.

\ c.c, toluolised broth culture.

Mar. 2.— II \ c.c. broth culture ; killed at 60° C.

I c.c. broth culture ; killed at 60° C.

\ loop virulent culture; rabbit died in 14 days with

paralysis.

24. — Control 1. \ loop living culture; dead in 2 days.

tt 2. \ loop living culture ; dead in 5 days.

II 3. Y(7 loop ; ill ; recovered.

Experiment j.

April 4. — Rabbit received 5 c.c. autolysed filtrate.

II. — n 1 c.c. broth culture; killed at 60° C.

I c.c. broth culture ; killed at 60° C.
\ c.c. toluolised broth culture.
I c.c. toluolised broth culture.
Y^Q loop living culture.
\ loop living culture.
J loop living culture.
5. — Control rabbit, J loop living culture— dead 2 days.
5. — n \ loop living culture ; ill — recovered.

June 14. — Rabbit received 2 loops living culture — died in 4 days.

The bactericidal power of the serum of the immunised rabbit was
tested by Wright's method on June 12th, with the following result: —

22. —

May

I. —

II. —

22. —

29.—

June

5-—

Immune Serum.

Culture Dilution.

Result.

I +

in 10

Growth

I +

in 100

I 4-

in 1000

I +

in 10,000

I +

in 100,000

I +

in 1,000,000
(171)

l6 QUATERCENTENARY STUDIES IN PATHOLOGY

The serum, mixed in equal parts with a i - i,cxx),ooo dilution of the
living culture failed to destroy the growth. The number of bacilli in the
I - 1,000,000 dilution, as estimated by Wright's method, showed —

10 mm3 of I : 1,000,000 culture dilution, i colony.

10 mm3 of I : 1,000,000 m o i?

10 mm3 of I : 100,000 n 13 n "i Average

10 mm3 of 1:100,000 »i 17 It / 15.

Thus the serum had exerted no bactericidal effect in vitro, inasmuch as
not one single pyocyanic bacillus had been destroyed. As the serum was
examined direct from the body of the rabbit, it would appear to be
unlikely that in vivo the bacilli are destroyed by bactericidal action, as
has been maintained by Wassermann.

The results of all the immunisation experiments have failed to show
anything like the degree of immunity obtained by Wassermann.

As was mentioned above, Gheorghiewsky came to the conclusion that
pyocyanic immunity owed its power to the destruction of the bacilli by
the phagocytes. Wright and Douglas have, however, shown that the
blood fluids play a cardinal role in the production of phagocytosis, and
on testing the sera of rabbits immunised against the bacillus pyocyaneus
it was actually found that the opsonic power of the serum may be con-
siderably greater than that of normal animals. Such immune serum as
we have seen possesses no bactericidal power in sensu strictori.

A few experiments were also made with pyocyanine isolated from
cultures. The cultures, rich in this pigment, were shaken up with
chloroform, the pyocyanine passing into solution. By allowing the
chloroform to evaporate, the well-known blue crystals were obtained.
As was found by Ledderhose, Charrin and Legros, pyocyanine prepared
in this way presented no toxic properties. I have not been able to
confirm the statement of Charrin and Gheorghiewsky that immune
serum inhibits pyocyanine formation.

Conclusions.

1. Guinea-pigs are highly sensitive to the action of bacillus
pyocyaneus.

2. The filtrates obtained from living cultures of virulent pyocyanic
bacilli are not toxic to guinea-pigs and rabbits except in enormous doses.

(172)

STUDY OF THE IMMUNITY AGAINST BACILLUS PYOCYANEUS 1 7

3. Bouillon cultures killed by toluol are toxic to guinea-pigs and
rabbits.

4. The toxicity of such cultures is largely referable to endotoxins
which have diffused out of the bodies of the bacilli. Some of the toxic
action may also be due to products developed from the medium in which
the bacillus is growing.

5. The degree of immunity induced by living or dead cultures is not
high, the animals seldom being able to withstand more than a few lethal
doses of the living culture.

6. The blood serum of animals immunised with living or dead
cultures possesses no bactericidal properties when tested in vitro upon
cultures of the bacillus.

7. The serum of animals immunised with dead or living cultures
shows an increased opsonic effect in comparison with normal serum.

8. The recovery from pyocyanic infection in rabbits is probably due
to phagocytosis, preceded by the opsonic action of the serum.

(173)

1 8 QUATERCENTENARY STUDIES IN PATHOLOGY

LITERATURE ON THE SUBJECT.

1. G^ssARD. " De la pyocyanine et de son microbe." These de Paris, 1882.

2. Charrin. " La maladie pyocyanique." Paris, 1889.

3. Wassermann. " Experimentelle Untersuchungen liber einige theoretische

Punkte der Immunitatslehre." Zeitschr. fiir Hygiene, 1896, XXIL,
p. 263.

4. Ghi^orghiewsky. "Du mecanisme de 1' immunite vis-a-vis du Bacille

Pyocyanique." Annales de V Inst. Pasteur, 1899, XIII., p. 298.

5. Muller, p. "Zur Lehre von den bactericiden und agglutinierenden

Eigenschaft des Pyocyaneus-immunserum." Centralbl. f. Bakter., 1900,
XXVIII., p. 577.

6. Breymann, M. "Ueber Stoffwechselprodukte des Bacillus Pyocyaneus."

Centralbl. f. Bakter., 1902, XXXI., p. 481.

7. Charrin. " Multiplicite des secretions d'un meme microbe pathogene."

La Semaine med., 1897.

8. MacIntyre. "The intracellular Toxin of Bacillus Pyocyaneus." Journal of

the Amer. Med. Assoc, 1904, XLIL, p. 1072.

9. Wassermann. "Immuntat bei B. Pyocyaneus." Kolle- Wassermann,

Handbuch der pathogenen Mikro-organismen, 1904, IV., p. 1212.

10. Conradi. "Ueber losliche, durch aseptische Autolyse erhaltene Giftstoffe

von Ruhr und Typhusbazillen." Deutsche med. Wochenschr, 1903, p. 26.

11. Neisser and Shiga. "Ueber freie Receptoren von Typhus-und Dysenterie-

bazillen, und iiber das Dysenterietoxin." Deutsche med. Wochenschr,
1903, p. 61.

(174)

ON EPIGNATHUS.

By Alexander Low, M.A., M.B.,

Senior Assistant to the Professor of Anatomy and Lecturer on Embryology,
University of Aberdeen.

(»75)

On Epignatlius.

The condition known as epignathus is comparatively rare and of much
interest.

The study of its aetiology is a subject involving important biological
and pathological problems. Windle (26) in 1899 published a paper in
which he reviewed in detail the literature of the subject, and discussed
the various and diverse opinions regarding the nature of this class of
monstrosity.

Since then, on the one hand, the work of Marchand (i^), Bonnet (J),
Wilms (24), and E. Schwalbe, has altered our opinions as to the tera-
tomata, while, on the other hand, the experimental embryological work
of Driesch (1°), Wilson (^S), Spemann (22), Loeb (16), O. Schultze (21),
Bataillon (3), has broadened our view by the artificial production of
malformations, more especially of double formations.

Recently, E. Schwalbe (20) has expressed the opinion that the
epignathi, and those formations related to them, merit detailed work.
Looking on epignathus as a connecting link between the true double
formations and the teratomata, by the study of these we hope to obtain
hints as to the origin and nature of epignathus.

Although records of cases of epignathus are to be found scattered
throughout literature, still a large proportion of the cases are very
imperfectly described. Many of the specimens have been more or less
decomposed, and unfit for detailed microscopical examination ; or the
specimen has only been examined after having been preserved as a
museum specimen for a length of time, and in such cases the history has
been incomplete or altogether wanting.

I have been fortunate in receiving a very perfect specimen of epignathus,
which occurred in the practice of Dr. Alexander G. Gall,* of Aberdeen.

* I wish to express my indebtedness to Dr. Gall for bringing me the specimen in perfect
preservation, and also for furnishing me with notes of the maternal history.

(177) N

4 QUATERCENTENARY STUDIES IN PATHOLOGY

In the present paper I describe this specimen, then give a summary of
the structure and relations of the epignathi, and finally review some of
the more recent theories as to the origin of epignathus.

I.— DESCRIPTION OF A CASE OF EPIGNATHUS.

The maternal history of the case is as follows : — The mother is
twenty-six years of age, has been married for six years, and has born
three children — two males and one female — in every way normal and
healthy. The duration of the present pregnancy, as calculated from the
end of the last menstrual period, was only thirty-four weeks ; but, as the
foetus is full time, conception must have taken place previous to the last
menstruation. Towards the end of the first month of pregnancy there
is a very definite history of the mother having received a severe nervous
shock.

No special peculiarity was noted about the foetal membranes, liquor
amnii or umbilical cord. The part of the umbilical cord still attached to
the foetus is normal, and shows one vein and two arteries.

The foetus is of the male sex, and well developed, the head only
showing a malformation. The hair of the scalp is thick, dark, and as
much as i8 mm. in length, the eyelids are open, and the free margins of
the nails of the fingers project beyond the tips of the digits. Both
testicles lie in the scrotum. The foetus measures in length 488 mm.
from vertex to sole of foot, 320 mm. from vertex to tip of coccyx, and,
with the tumour mass, weighs 3*2 kilograms. The cranial length is
120 mm., breadth 95 mm., and circumference 327 mm. There is a centre
of ossification in the lower end of the femur.

The foetus has all the appearance of having been born at full time.

From the mouth of the foetus there projects a tumour mass almost
as large as the foetal head. As seen from the front, the tumour is
roughly pyramidal in shape, its base lying against and obscuring the face
of the foetus, while its rounded blunt apex projects downwards in front
of the thorax. (Fig. i.) The surface of the tumour is roughly divisible
into an upper irregular nodulated part, covered with natural-looking
skin, and a smaller part below and to the left, covered by a smooth
glistening membrane, varying in colour from greyish to dark red.

(178)

ON EPIGNATHUS e

(Fig 3.) Between those two parts is a furrow, and along this line of
junction there is a fringe of darkish hair as much as 18 mm. in length.

The only recognisable foetal parts are a single well-formed nostril
and a mouth deformed, but showing lips and gums. The nostril opens
upwards, and is situated between the rudimentary mouth above and the
fringe of hair below. There is thus distinct evidence of a much
deformed foetal head and face, hanging by means of a stalk from the
mouth of the normal foetus, the abnormal head being really inverted.

Fig. I. — Epignathus viewed from the front and left.

This semblance becomes apparent on examining (Fig. i) where
above and to the right is seen the mouth cavity with lips and gums, and
immediately under this the opening of the single nostril, and still lower
down the surface covered by smooth glistening membrane, similar to
that covering rudimentary brain substance in a case of anencephaly.

(179) N I

6 QUATERCENTENARY STUDIES IN PATHOLOGY

Viewed from the right side, the base of the tumour is closely applied
to the face of the foetus, from the tip of its nose to the chin, for a
distance of some 80 mm. By pushing the mass towards the left, it is
seen to be attached by a stout stalk issuing from the mouth of the foetus,
and this stalk of attachment is not adherent to cheeks, lips or gums.

Except for the widely opened mouth the foetal head is not deformed.

Fig. 2. — Epignathus viewed from the right side.

The greater part of the right side of the tumour is covered with
natural looking skin with fine lanugo hairs. Near the centre of this
surface there is an oval depressed area, 50 mm. by 38 mm. in size, the
appearance of which suggests an irregular mouth cavity. This cavity is
surrounded by fleshy flaps like lips, covered by skin externally and
mucous membrane internally ; while, again, inside these are firm con-

(180) '

ON EPIGNATHUS y

centric gums. These gums enclose an irregular cavity filled up with
eight stalked bodies covered by mucous membrane, and from the size of
a pea to that of a pigeon's &^'g.

The largest of these bodies has the shape and appearance of a foetal
heart, while another much elongated process is quite similar to an
abortive piece of umbilical cord.

Fig. 3. — Epignathus viewed from the left side.

The left surface of the tumour mass is still more irregular and
lobulated (Fig. 3). Above and behind there is one large and rounded
lobule about the size of a tangerine orange, while below and in front are
grouped six smaller nodules. This lobulated surface in its upper half is
covered by natural looking skin with fine lanugo hairs, while the lower
half is covered by a smooth membrane of a dark red colour with no hairs.

(181)

8 QUATERCENTENARY STUDIES IN PATHOLOGY

The whole specimen was now injected with a four per cent, formalde-
hyde solution, and placed in a solution of the same strength for two days.
Thereafter the specimen was frozen and a sagittal section made, just to the
left of the middle line, through both foetus and tumour.

The mesial section of the foetus, except for the head condition,
presents normal appearances (Fig. 4). The body of the hyoid bone
shows a centre of ossification, and is at the level of the middle of the
body of the third cervical vertebra. The cricoid cartilage lies at the
level of the disc between the fourth and fifth cervical vertebrae, while the
upper border of the manubrium is opposite the first dorsal vertebra.
The trachea bifurcates on the disc between the third and fourth dorsal
vertebrae. Thus the various structures do not lie at a lower level than
usual. The thyroid and thymus glands are well developed. There is no
abnormality of the heart. Dark green meconium distends the large
intestine, microscopically, this shows the usual constituents of meconium
including delicate lanugo hairs. The bladder is distended, reaching to
the umbilicus.

Coming to the tumour mass itself we can obtain a general idea of its
nature by noting the appearances presented by the sagittal section
depicted in Fig. 4. The section is made just to the left of the middle
line and the right face is seen viewed from the left. The lower part of
the section is of a dark red colour, soft and vascular, while the upper
part, lighter in colour, varies in appearance and consistency. In the
centre is an irregular piece of bone in which two dental sacs are seen cut
across. Above and in front of this bone is fibrous tissue containing a
large lobule of fat, and covered by normal skin. Passing up into the
stalk of attachment of the tumour is a large serous cavity containing a
closed knuckle of bowel. The lower part of the section is largely made
up of rudimentary nervous tissue, in which there is a large cyst with a
thin smooth wall. Just above this cyst there is a degenerate and
collapsed eyeball lined by dark pigment.

Detailed Structure of Tunnour.

Each half of the tumour was now carefully and completely dissected.
Many pieces of the various tissues were removed for microscopical
examination. These pieces were embedded in celloidin, and sections cut

(182)

ON EPIGNATHUS

Fig- 4-— iVIedian sagittal section of head of foetus with Epignathus.

lO QUATERCENTKNARY STUDIES IN PATHOLOGY

and stained by different methods so as to establish the structure and the

identity of the different tissues and organs.

The following tissues and organs are present in the tumour : —

Skin and appendages. — A large part of the surface of the tumour is

covered by skin with fine hairs. Microscopic sections of this show

a well-developed epidermis with numerous hair follicles and both

sebaceous and sweat glands.

The lower part of the tumour is covered by a thin dark red membrane,

which consists of a stratified epithelium resting on a loose fibrous tissue.

Immediately under the epithelium are very numerous blood capillaries,

distended with blood cells, and giving the dark red appearance to this

part of the tumour.

Fig. 5- — Abortive heart attached to deformed hard palate of Epignathus. (Natural size.)

Bo7ie aiid teeth. — The bone mostly forms irregular pieces radiating
from the centre of the mass. To the right, one piece of bone can be
identified as two fused superior maxillary bones. The palatal and
alveolar processes of these bones can be distinctly recognised, and
embedded in the alveolar processes are eight well-developed dental sacs
containing teeth.

Month. — In relation with these alveolar processes is an irregular
mouth cavity seen on the right aspect of the tumour (Fig. 2). This

(184)

ON EPIGNATHUS

II

cavity is surrounded by lips, which are covered on their outer aspect
by normal skin, and on the inner by a stratified mucous membrane.
No striped muscle fibre is seen in the lips.

There is an irregular hard palate covered by mucous membrane,
and attached to it, and filling up the cavity of the abortive mouth, are
nine stalked bodies, from the size of a bean to that of a pigeon's egg.

Abortive heart. — One of these bodies is much larger than the others,
and has the appearance of a foetal heart (Fig. 5). This measures 22 mm.
in length, 12 mm. transversely, and 7 mm. from before back, and is
attached by a stalk about 6 mm. in diameter. The anterior wall is quite

Fig. 6.— Abortive umbilical cord, with two large blood-vessels, attached to deformed hard
palate of Epignathus. (Natural size.)

thin, and a slit made in it opens into a flattened cavity. The cavity
is triangular in shape, has a slightly ridged appearance, and leads up into
a vessel with a lumen about 2 mm. in diameter. An incision made
through the posterior wall opens into another ventricle-like cavity. The
posterior wall is from 2 to 3 mm. thick. The interior of this cavity
is more ridged and irregular, and from it there also passes a thin walled
channel. There is a thin septum between the two cavities, but it is
not quite complete at its base.

(185)

12 QUATERCENTENARY STUDIES IN PATHOLOGY

The walls of these ventricles are made up of a loose net-work of cells
with a lining of endothelial cells. No cardiac muscle fibres are seen.

Abortive timbilical cord. — Near the abortive heart, and also attached to
the deformed palate of the parasite, hangs an elongated structure which
has all the appearance of a piece of umbilical cord (Fig. 6). It measures
1 8 mm. in length, 3 mm. in diameter, and has a somewhat bulbous
free end. It has the gelatinous appearance of umbilical cord, and small
blood vessels are seen running along its whole length. Microscopically,
it is made up of typical retiform cells, and is covered by a layer of
flattened epithelium one cell deep. In the cross sections are seen four
small blood vessels. The attached end of the cord passes into a mass of
vascular tissue, in which break up branches from two large blood vessels
(Fig. 6).

A single nostril. — Below the irregular mouth is the left half of a nose
with one nostril (Fig i). The opening of the nostril measures 3 by 2 mm.,
and leads into a passage 15 mm. long, bounded on the one side by
a process from the deformed superior maxilla, and on the other side by
an irregular piece of cartilage. The passage is lined by a thin, very
vascular mucous membrane.

A degenerate eyeball — Immediately below and deep to the nostril is a
degenerate eyeball. The eyeball is firmly embedded in fibrous tissue, is
somewhat crumpled, contains a clear fluid, and measures 14 mm. by 7 mm.
There is neither cornea nor lens. The outer coat of the eyeball is largely
cartilaginous, and inside this is a thin, vascular coat, lined by a continuous
layer of typical, hexagonal, nucleated retinal pigment cells. There are
no traces of the other layers of retina.

Intestine. — In the upper part of the tumour are several separate serous
cavities containing irregular closed knuckles of intestine. These knuckles
are distended with a clear gelatinous material, the largest knuckle having
a diameter of 17 mm. A series of pieces from the walls of the different
knuckles of bowel were examined microscopically. Each piece shows
the four coats of normal bowel, viz., serous, muscular, submucous and
mucous. The muscular coat consists of the usual two layers of fibres.
Both layers are well developed, more especially the outer, which is formed
of longitudinal fibres uniformly disposed. The submucous coat contains
numerous small blood vessels. The mucous membrane is lined with

C186)

ON EPIGNATHUS 1 3

small cubical epithelial cells, resting directly on a well marked muscularis
mucosae. No intestinal glands are seen in any of the sections.

Salivary Gland. — Embedded in the tissue, near the deformed mouth,
is an irregular glandular structure about half-an-inch in diameter. This
microscopically shows the normal structure of a salivary gland. In the
sections numerous alveoli and ducts are seen. Both serous and mucous
alveoli are present. The mucous alveoli are not so numerous, but are
readily recognisable from thdr clear distended mucous cells.

Thyroid Gland. — Tissue, containing thyroid gland substance, was
found in an indefinite fibrous mass under the skin of the upper part of
the tumour. This glandular structure has a tough fibrous capsule i mm.
thick, and inside this a yellowish brown material, with here and there a
cystic condition. Microscopically, vesicles of thyroid gland are to be
made out. The vesicles are mostly elongated, and some are much
distended, so that their lining cells are flattened. The smaller vesicles
have walls, consisting of a single layer of cubical epithelium-cells, with
deeply-stained nuclei.

Liver Cells. — In the left half of the tumour there is a solid nodule as
large as a walnut. This has a smooth surface, and has the colour and
consistency of liver. Microscopic sections give no indication of the
lobular structure of liver, but show large cells, with the nucleated
granular appearance of hepatic cells. Ramifying among these cells are
numerous blood capillaries.

Nervous Tissue. — The lower part of the tumour is made up of an
irregular soft mass, covered by the smooth membrane already described.
On cutting into this it is seen to consist of a light yellowish material, of
very friable consistency. Several pieces of this were embedded, and
sections, stained by different methods, examined microscopically. All
the sections agree in showing a rudimentary or embryonic nervous tissue.
In the sections is seen a nuclear tissue with no cell boundaries, the
nuclei lying in a syncytial protoplasm. The nuclei stain very readily,
and many are dividing by mitosis. The tissue is similar to embryonic
nervous tissue in an early stage of development.

Although many different parts of the tumour were examined micro-
scopically, no nerve fibres, either medullated or non-medullated, were
observed. On the right side, some fibres from the great palatine nerve
were traced on to the stalk of the tumour, but only for a short distance.

(187)

14 QUATERCENTENARY STUDIES IN PATHOLOGY

Blood' Vessels. — In all parts the tumour is very vascular. In the
course of dissection, several large blood-vessels were traced out, and, in
the microscopic sections, numerous blood-vessels and capillaries are to
be seen.

Attachment of the Tumour.

The mouth cavity of the fcetus is distended by the stalk of attach-
ment of the tumour. The lips, although compressed, arc quite free, the
upper lip being pushed upwards, while the lower lip is turned inwards
over the alveolar margin of the lower jaw. The lower jaw is pushed
downward and backward, so as to be in a plane much posterior to that of
the upper alveolar margin. The mouth measures 48 mm. from angle to
angle, and 40 mm. from above down.

The stalk of the tumour is attached mostly to the under aspect of
the hard palate, but its posterior part is continued up along the posterior
border of the nasal septum to the under aspect of the basi-sphenoid. At
its attachment the stalk measures 24 mm. by 22 mm. To permit of the
posterior part of the stalk passing up to the sphenoid, the soft palate is
completely split into two halves. Each part of the soft palate is com-
plete, but is compressed backwards against the lateral wall of the
naso-pharynx, where each terminates in a uvula-like process. For its
size and fairly extensive attachment, the stalk causes comparatively
slight disturbance of surrounding parts. The cartilaginous nasal septum
is complete and well developed. The left nasal fossa is normal and has
well developed turbinals ; the right also is little altered, but in it lies a
tongue-shaped body which is attached to the stalk of the tumour, and
passes forwards round the posterior border of the hard palate and
extends to the anterior nares protruding as a small nodule from the
nostril. This tongue-shaped body is covered with fine hairs and
sebaceous material, and compresses somewhat the turbinals of the right
nasal fossa.

The floor of each nasal fossa, as seen from above, is intact, and
covered by mucous membrane. On examining the hard palate, from the
under aspect, its anterior half is complete, but behind it is prolonged
down on the sides and front of the stalk, as thin, bony laminae. Thus
when the stalk is removed the hard palate is deficient behind over an
area measuring 18 mm. from before back and 21 mm. from side to side.

(188)

ON EPIGNATHUS jc

The stalk of attachment consists of a core of fibrous tissue on which
are prolonged delicate laminae of bone from the hard palate, and also
from the vomer which is represented by two thin separate bony laminae
on each side of the nasal septum. In the fibrous tissue is embedded a
piece of bowel. On its outer aspect the stalk is covered by mucous
membrane. Between the mucous membrane and the laminae of bone the
main blood vessels for the tumour pass down. On the left side two
blood vessels of fair size pass down from the posterior palatine foramen
to the tumour, where they form a profuse vascular network. Along with
these blood vessels, some bundles of the great palatine nerve can be
traced on to the stalk for a short distance.

Another large artery— as large as the radial artery in the adult-
issues from the left spheno-palatine foramen, crosses over the roof of the
left nasal fossa, gains the posterior border of the nasal septum, and
descends in the thickened tissue behind this into the right half of the
tumour.

Under the mucous membrane of the posterior part of the stalk a
large thin-walled vein runs up under the fibrous periosteum of the basi-
sphenoid, and passing through a canal in the cartilage, between the
basi-sphenoid and basi-occipital, communicates with the left cavernous
sinus.

Summary of the attachment and structure of the Tumour.

The tumour is attached to the posterior half of the hard palate and
partly to the basi-sphenoid. It is of interest to note that the nasal
septum is complete. The relations of the attachment are such as to
suggest that the tumour in its origin might have had an early connection
with the mesial nasal process, and had been carried back along with the
mesial nasal process through between the developing palatal processes.

The blood supply of the tumour is mainly derived from a much
enlarged left spheno-palatine artery, and also from a branch of the left
posterior palatine artery. The blood is returned by two veins, one large
vein opening into the left cavernous sinus, and another smaller vein,
a tributary of the left posterior palantine vein.

Structurally, the tumour is one of the higher forms of epignathus,
containing recognisable foetal organs. A peculiarity of the specimen is,

(189)

l6 QUATERCENTENARY STUDIES IN PATHOLOGY

that it shows secondary stalked structures attached to the deformed
palate of the parasite. The question arises as to whether these are to be
reckoned a secondary epignathus.

Representatives of all three embryonic germ layers are present, as is
shown in the following table of structures identified : —

Ectodermal. Mesodermal. Endodermal.

Skin. Connective tissue. Intestine.

Hairs. Fat. Thyroid gland.

Sebaceous glands. Cartilage. Liver cells.

Sweat glands. Bone.

Embryonic nervous tissue. Smooth muscle.

Retinal pigment cells. Blood vessels.
Teeth.
Salivary gland.

II.— ON THE STRUCTURE AND RELATIONS OF THE

EPIGNATHI.

According to Ahlfeld (i), Hoffman, in 1688, was the first to describe
a foetus with a tumour hanging from its mouth. To this condition
Geoffroy Saint-Hilaire (^3) gave the name " epignathus," defining it
as a faulty developed second head which is attached to the palate of the
principal head.

Ahlfeld (2) says, that, by " epignathus," one understands an acardiacus
amorphus which stands in connection with the mouth cavity, mostly with
the hard palate of its twin brother. According to his view, all swellings
from the mouth of the foetus are derived from a second foetus, even
although there are no evident foetal parts.

More recently, Windle (^^j gives the following definition. " An
epignathus, then, is a congenital tumour, teratoid in its nature, or
possessing recognisable parts of a foetus, which is attached within
the cavity of the mouth or upper part of the pharynx." As he points out,
the definition must include all forms, from the simple hairy polypus of the
pharynx to the tumour with fully developed portions of a foetus.

Wittenberg (^7) gives a somewhat wider meaning to " epignathus," and
employs it for all teratomata, for which we have to accept, that the first

(190)

ON EPIGNATHUS 1 7

place of origin has been in the stomodaeum. Thus would be included
amongst the epignathi certain teratomata which lie outside the dura
mater in the middle fossa of the skull, the germ of which may have been
in connection with the buccal evagination of the pituitary body.

E. Schwalbe (20) arranges the epignathi in four chief groups, at the
same time admitting that there are transitional forms between the several
groups. These groups are as follows : —

Group I. — A second foetus is attached, by its umbilical cord, to the
palate or near the palate of a foetus. This second foetus may be more or
less well developed.

Group II. — From the mouth of a foetus there hang parts of the body
of a second foetus, which parts can readily be recognised as fully formed
portions of a foetus, e.g., lower extremities, external sexual organs, etc.

Group III. — Out of the mouth of a foetus there projects a formless
mass in which no parts similar to foetal organs are to be recognised.
Microscopic examination gives the structure of a teratome.

Group IV. — A larger or smaller tumour is found attached to the
palate or in the mouth cavity of the foetus. Microscopic examination
shows that the tumour is composed of several tissues, and is of the type
of the mixed tumours.

Examples of Group I. are very rare. Ahlfeld describes and illustrates
a case recorded by Baart de la Faille. In this case there was a typical
epignathus attached to the base of the skull of a practically normal foetus.
In addition to this, there were two acardici acephali — pelves with lower
extremities, with fork-like umbilical cords attached to the palate of the
first well-formed foetus.

Examples of Group II. are more common. The case now described
belongs to this group as it shows quite recognisable foetal parts — a nostril,
mouth. Wasserthal describes an epignathus with a lower extremity
attached. Otto (^9) describes the case of a male foetus with an epignathus
showing male external genital organs. Kreutzmann (^^) records the case
of a female foetus with an epignathus showing female external genital
organs.

In examples of Group III. there are no recognisable organs, although
the tumour is usually more or less covered by epidermis, and, like a
teratome, may contain fibrous tissue, cartilage, bone, muscle, gut

(190

1 8 QUATERCENTENARY STUDIES IN PATHOLOGY

epithelium, rudimentary brain substance, etc. To this belongs
Schwalbe's case I., which had the structure of a typical teratome, and
contained derivities from all three germ layers.

Group IV. includes the most simple forms, namely, the hair-clad
pharyngeal pol3'pi. These usually consist of some connective tissue and
fat, with perhaps a piece of bone, and are covered by cutis and epidermis,
with hairs and glands. Thus Arnold describes the case of a girl of 13
years, from whose soft palate a hairy polypus was removed. The
polypus was mainly composed of fat, and was covered with skin, showing
hairs and sebaceous glands, but no sweat glands.

As regards the site of implanation, we find that in the larger
proportion of recorded cases of epignathus, the attachment is either to
the palate or to the superior maxillary bone. This is seen on examining
the list of cases published by Windle. In rather over one-fourth of the
cases the tumour has a connection with the base of the skull. In a few
cases the attachment has been to the soft palate or some part of the
lateral wall of the pharynx.

Of those attached to the base of the skull, the majority have no
connection with the interior of the cranium. However, in a small
number of cases the stalk of attachment has some relation to the
interior of the cranium. In Windle's case there was a tubular stalk
leading through a patent pituitary foramen into the interior of the skull.
In a case recorded by Hill (^'^), a branch of the internal carotid artery
passed through a canal to the tumour. In our own case, a large vein
from the tumour passed through the base of the skull to open into the
cavernous sinus. In two cases recorded by Baart de la Faille, the stalk
of the tumour passed distinctly through the pituitary foramen to obtain
attachment to the sella turcica. In Wittenberg's case the pharyngeal
swelling is continued through an opening in the sphenoid bone directly
into the cavity of the skull. Another variety is where the tumour on
the outside is connected, by a stalk through the sphenoid, with a tumour
inside the skull, as in a case recorded by Breslau and Rindfleisch (8).
Miiller {^^) describes an epignathus attached in the mouth of a foetus,
while quite separate from it lay a tumour in the right middle cranial
fossa. Finally, a case is recorded by Beck (5), where a teratome was
confined entirely to the pituitary fossa. Wittenberg argues that all such
cases are to be included among the epignathi.

(192)

ON EPIGNATHUS lO

Thus in their attachment the epignathi point to the first settlement of
their germ having taken place in the stomodaeum.

III.— ON THE ORIGIN OF THE EPIGNATHI.

In looking for the cause of the disturbance of normal development
which produces an epignathus, there is little suggestive in the aetiology.
Out of the twenty-two cases noted, the mother was a primipara in three,
and a multipara in nineteen. There is no evidence of any association
with previous twins or malformed foetuses, or of any hereditary tendency.
In several of the cases, as in that now recorded, there is a history of the
mother having received a severe nervous shock early in pregnancy.
Hydramnios is a frequently associated condition, but this may be more
an effect than a cause.

In the discussion of theories which relate to the origin of the
epignathi, we find that two views have been advanced as to its
development.

According to one view, epignathus is a single malformation — that is,
arises from the body of the fcetus.

The generally accepted view, however, is that epignathus is to be
looked on as a double malformation — the two components being
asymmetrical, the one (autosite) being practically normal, and the other
(parasite) incomplete and attached to the first.

Thus Ahlfeld holds the view that an epignathus is derived from a
second fcetus, even although there are no evident foetal parts — that is, he
regards it as a double malformation. His theory is that there are two
embryonic areas, one in front of the other, that one embryo lags in
development and comes to lie under the forebrain vesicle of the other
and eventually in the stomodaeum, its allantois becoming fixed to the
mouth and hence acardiacus. He postulates two unequal embryonic
areas with their heads towards each other.

Arnold, in discussing the question of the origin of these tumours,
divides them into two groups. " Autochthonous " teratomata, such as
the hairy pharyngeal polypi, he traces back to the abnormal development
of a " dislocation " of embryonic material in the mouth or pharyngeal
cavities — that is, they are derived from one embryo. The " heteroch-

(193) O

20 QUATERCENTENARY STUDIES IN PATHOLOGY

thonous" tcratomata, which have recognisable parts of a second foetus,
he admits have origin in a second embryo.

Marchand {^'^) looks on the higher foetus like forms of epignathus as
asymmetrical double malformations. He draws attention to the fact
that epignathus is a parasite in intra-amniotic relation with the autosite.
Hence he says there must have been an unequal division of a single
embryonal anlage.

He thinks the origin of epignathus is most likely from the fertilisation
and development of a polar body.

Bonnet (7) upheld this hypothesis, but went further, suggesting the
possibility of the belated division of one or more blastomeres, in an early
stage of segmentation.

Wilms (^^) also supported the same view with regard to teratomata
of the ovary and testis, and held that certain details of their structure
pointed to an origin from blastomeres.

Windle, in speaking regarding the mass of germ plasm out of which
the major forms of epignathus are developed, suggests the theory that " it
is not the equivalent of a germ plasm of full embryo-producing power,
but that it possesses varying and more limited potentialities of develop-
ment."

Beard (4) applies his doctrine of multiple germ-cells to explain the
origin of tumours. He considers that the early blastomeres of a
metazoon form "an asexual foundation or larva, the phorozoon, upon
which the germ-cells, and with these an embryo, take their origin."
Normally, one of these germ-cells undergoes development, but two or
more may develop. If two primary germ-cells "develop either together
or at different times, but with abnormalities on the part of one, there
may result a more or less rudimentary embryo, an embryoma of Wilms,
a tumour."

E. Schwalbe accepts the view that the higher foetus forms owe
their origin to a blastomere, which in an early stage was disturbed from
the development of the autosite. He allows the possibility of an origin
out of germ material of a later development stage. He points out that
the prospective potency of blastomeres always becomes more circum-
scribed with progressive development. Through an unequal division of
the blastomere material, or through a disturbance of a blastomere at a

(194)

ON EPIGNATHUS 21

time when it still has the potency to form a whole embryo or part of
such, a parasitic malformation can originate.

If at first the development of such a blastomere "sists" it becomes
" displaced," and may lie in various regions of the developing embryo.

With progressive development there is a decrease of the potencies of
the several blastomeres. Hence Schwalbe argues that the more
complicated the structure of an epignathus, the earlier must be placed
the teratological process to which it owed its origin.

Accepting the view that epignathus is an asymmetrical double
malformation, we have a certain amount of experimental work bearing
on the cause of the disturbance of normal development giving rise to
such.

Thus there is evidence to show that, by the total separation of the
first two blastomeres of a single ovum, duplicate twins can result.
Incomplete separation of the first two blastomeres of a single ovum may
give rise to a symmetrical double malformation. All varieties of these
symmetrical double malformations result according to the extent of
separation and relative position of the blastomeres.

In the production of an asymmetrical double malformation like
epignathus it may be that, as held by Wilder (^3), there has been a
complete separation of the first two blastomeres of a single ovum, and
then a fusion of the later blastomeres owing to close proximity.

Possibly, however, such asymmetrical malformations are due to an
incomplete separation of a blastomere or blastomeres in a later stage of
development — perhaps at a stage when such blastomere or blastomeres
have not the full potentiality of a complete embryo.

O. Schultze (21), in working with the eggs of amphibians, found that
among over-ripe eggs there was a larger proportion of double formations.
Over-ripe eggs tend to break up into non-nucleated pieces, and perhaps
these may become fertilised and produce a double formation. Boveri
has shown that a blastomere, either with or without a nucleus, can be
fertilised and develop.

Polyspermia has been accepted by some as an origin of double
formations ; the researches of O. Hertwig (''^), Driesch, Fol and others
negative this. Besides, it has been shown that physiological polyspermia
takes place in many animals.

The most probable view is that double formations originate after the

(195) o I

22 QUATERCENTENARY STUDIES IN PATHOLOGY

egg has been normally fertilised. Thus in lower animals, by artificial
means, such as shaking, compression, increase of temperature, altering
the action of gravity or changing the osmotic pressure, the early blasto-
meres may be separated and each give rise to an embryo. Loeb and
Bataillon, by changing the degree of concentration of the surrounding
medium, brought about separation of blastomeres and double malforma-
tions. Hence it has been suggested that the production of double
malformations is favoured by pathological changes in the mucous
membrane of the Fallopian tube and uterus, causing alteration in its
secretions.

(196)

ON EPIGNATHUS 23

BIBLIOGRAPHY.

In his paper "On the condition known as Epignathus," published in the
"Journal of Anatomy and Physiology for 1899," Windle gives a detailed list,
with references to all cases of epignathus published to that date. Since then
further cases of epignathus have been described in the following papers : —

Marchand. "Demonstration eines Epignathus." Centralbl. f. Gynakol.

1902, Jhrg. 26, p. 529.
Martin. "Epignathus." Miinch. med. Wochenschr., 1900, p. 236.
ScHATZ. " Die Gefassverbindungen der Platzentakreislaufe eineiiger Zwillinge,

U.S.W.," Arch. f. Gynakol., 1887, Bd. 29, p. 419.
ScHWALBE, E. "Der Epignathus und seine Genese." Beitrage zur Path. Anat.

u. allgem. Path., 1904, Bd. 36, p. 242.
Stanze. " Ein Fall von Epignathus und polypenformigen Fibrolipoma

myxomatodes congenitale der rechten Nasenhdhle." Inaug. Diss. Miinchen,

1902.
Wittenberg. " Ueber den Epignathus und seine Genese." Berlin, 1905.

GENERAL REFERENCES.

1. Ahlfeld, Fr., '75. "Beitrage zur Lehre von den Zwillingen. I. der

Epignathus." Archiv. fiir Gynakologie., 7 Bd., p. 210.

2. Ahlfeld, Fr., '80-82. " Missbildungen des Menschen." Leipzig.

3. Bataillon, E., '01. " La pression osmotique et les grandes problemes de

biologie."

4. Beard, J., '03. "The Embryology of Tumours." Anat. Anz., 23 Bd.

No. 18-19, p. 486.

5. Beck, H., '83. " Ueber ein Teratom der Hypophysis Cerebri." Zeitschrift

fiir Heilkunde, 4 Bd.

6. Bonnet, R., '99. " Giebt es bei Wirbeltieren Parthenogenesis ? " Ergebnisse

der Anatom. und Entwicklung., 9 Bd.

7. Bonnet, R., '01. "Zur Atiologie der Embryome." Monatschrift fiir

Geburtshiilfe und Gynakologie., 13 Bd.

(19;)

24 QUATERCENTENARY STUDIES IN PATHOLOGY

8. Breslau u. Rindfleisch., '64. " Geburtsgeschichte und Untersuchung

eines Falles von Foetus in fcetu." Virchow's Archiv., 30 Bd., p. 406.

9. Driesch, H., '93. " Entwickelungsmechanische Studien."

10. Driesch, H., '00. " Studien iiber das Regulationsvermogen der Organismen."

Leipzig.

1 1. FoRSTER, A., '05. " Kritische Besprechung der Ansichten iiber die Entstehung

von Doppelbildungen." Wiirzburg.

12. Hertwig, O., '02. "Lehrbuch der Entwicklungsgeschichte des Menschen

und der Wirbeltiere." Jena.

13. St. Hilaire, G., '37. "Histoire gen^rale et particuliere des Anomalies de

I'Organisation." Bruxelles.

14. Hill, A., '85. " Dissection of a Double Monster (Epignathus)." Journal of

Anatomy and Physiology, Vol. 19, p. 190.

15. Kreutzmann, L., '95. "Ueber einen Fall von Epignathus mit Missbildung

des Herzens." Inaug.-Diss. Marburg.

16. LoEB, J., '94. "Ueber eine einfache Methode, zwei oder mehr zusammenge-

wachsene Embryonen aus einem Ei hervorzubringen." Pfliigers Archiv.
f. d. ges. Physiologic, LV., p. 525.

17. Marchand, '97. " Missbildungen." Eulenburgs Real Enzyklopadie der

gesammten Heilkunde., 15 Bd.

18. MiJLLER, E., '81. "Beitrag zur Kasuistik der menschlichen Missgeburten. I.

Ein Epignathus." Archiv. fiir Gynakologie., 17 Bd., p. 298.

19. Otto, R., '78. " Ueber einen Epignathus." Archiv. fiir Gynakologie, 13 Bd.

20. ScHWALBE, E., '04. " Der Epignathus und seine Genese." Ziegler's

Beitrage zur pathologischen Anatomic und zur allgemeinen Pathologie,
36 Bd., p. 242.

21. ScHULTZE, O., '99. "Zur Frage von der Entwickelung der Doppel-

bildungen." Centralblatt fiir allg. Path, und path. Anat., 10 Bd., p. 393.

22. Spemann, H., '03. " Entwickelungsphysiologische Studien am Triton-Ei."

Arch. f. Ent.-Mech. d. Org., 16 Bd.

23. Wilder, H. H., '04. "DupUcate Twins and Double Monsters." American

Journal of Anatomy, Vol. 3, p. 387.

24. Wilms, M., '99-02. "Die Mischgeschwiilste." Leipzig.

25. Wilson, E. B., '02. " The Cell in Development and Inheritance."

26. WiNDLE, B. C. A., '99. " On the condition known as Epignathus." The

Journal of Anatomy and Physiology, Vol. 33, p. 277.

27. Wittenberg, W., '05. " Ueber den Epignathus und seine Genese." Berlin.

(198)

A CONTRIBUTION TO THE PATHOLOGY OF
EXOPHTHALMIC GOITRE.

By George Mellis Duncan, M.B., CM.,

First Assistant i and Lecturer on Bacteriology , Pathological Department ^
University of Aberdeen.

(•99)

A Contribution to the Pathology of Exophthalmic Goitre,

The pathology of exophthalmic goitre, despite the extensive
researches which have been carried out upon the subject, still remains
somewhat obscure, although the bulk of the evidence would appear to
point to a primary thyroidal origin.

The theories which have been advanced from time to time with
regard to the pathology of this disease may be roughly classified under
the three following headings: — (i) The thyroid theory. (2) The
central nervous theory. (3) The cervical sympathetic theory.

It is very difficult to explain the various symptoms of Graves'
disease upon a purely nervous origin, although many of them can be
reproduced experimentally by lesions inflicted upon the great nerve
centres. Filehne and Bienfait (Bull. Acad. roy. de med. de Belg.,
Bruxelles, 1890) were able to imitate certain of the symptoms of the
disease by lesions made experimentally upon the restiform body.

Mendel (Deutsche med. Wochenschrift, Feb., 1892) has recorded a
case of Graves' disease in which atrophy of the restiform body was
found after death.

Nor can lesion of the sympathetic system, whether irritative or
paralytic, explain all the various phenomena. Schiff, who destroyed the
sympathetic nerve fibres accompanying the blood vessels of one lobe of
the thyroid gland in the dog, found that both lobes remained identical
in structure.

The post mortem appearances found in the nervous system have been
extremely inconsistent and irregular, and have resembled in many cases
those produced by toxic influences. Clinically, however, there would
appear to be a very intimate connection with the central nervous system :
the frequent development of emotional disturbances throughout the
course of the malady has from the first been recognised. Sudden
nervous influences, such as shock, moreover, have a very marked effect

(201)

4 QUATERCENTENARY STUDIES IN PATHOLOGY

in aggravating the symptoms, and indeed are often associated with their
commencement.

On the other hand, evidence continues to accumulate in favour of the
primary implication of the thyroid gland. Mobius (Die Basedow'sche
Krankheit, Wien, 1896), Gauthier (Lyon med., 1895, Ixxx., 5-12), Renaiit
(Courrier med., Paris, 1895, xlv.), Greenfield (British Med. Journ., 1893,
Dec. 9), and Kocher (Mitteil. aus den Grenz. der Med. und Chirurg.,
1902, Bd. ix., p. i) all support this view.

Certain pathological alterations are invariably found in the gland,
whether the enlargement be evident or slight. The number of glandular
elements becomes vastly increased, and the secreting epithelium changes
its character. The whole appearance of the gland suggests " a prolifera-
tion for the performance of increased function " (Greenfield). The older
theory that the goitre was probably due to dilatation of the vessels has
not received support from the numerous examinations made of its
structure. Of course, with this increase in its tissue, and probable
increase in its function, the gland becomes more vascular, still extreme
vascularity is rarely an outstanding feature, and hence the enlargement
of the gland can in no sense be regarded as due primarily to this cause.
Edmunds (" Pathology and Diseases of the Thyroid Gland ") has pointed
out that when a portion of the thyroid gland is removed, the remaining
part undergoes compensatory hypertrophy. The vesicles enlarge, and
tend to become oval or branched ; the lining membrane becomes
convoluted ; the secreting cells, instead of maintaining their cubical
shape, become columnar. These changes are almost identical with those
found in the goitre of Graves' disease, and would certainly appear to point
to an increased activity of the organ. With the hyperplasia of glandular
tissue there is usually some change in the microscopic characters of the
colloid in the vesicles ; it appears to become more mucinoid. According
to Greenfield, fibrous overgrowth may ultimately ensue, which, in
advanced cases, may spread from the septa into the lobules, and
produce e'xtensive atrophy.

The hyperplasia of glandular tissue and the evidence of increased
activity of function suggest the probability that the disease is due to
excessive formation of thyroid secretion, and much clinical evidence can
be brought forward in favour of this view. The striking contrast between
the symptoms of Graves' disease, and those of myxoedema, would lead

(202)

CONTRIBUTION TO PATHOLOGY OF EXOPHTHALMIC GOITRE 5

one to infer a corresponding difference in the pathology of the two
affections, for while the condition of the gland in Graves' disease is most
likely one of hypertrophy and increased functional activity, that pre-
vailing in myxoedema is undoubtedly one of atrophy, with corresponding
loss of function. The beneficial results following immediately upon
operative procedures for the removal of portions of the hyperactive
thyroid tissue also support this view, namely, that the disease is probably
due to over-secretion.

Further, the condition of thyroid^m, produced by over-dosing with
thyroid extract, resembles Graves' disease in many of its symptoms.

As just remarked, the secretion of the exophthalmic thyroid is in all
probability not normal colloid. The microscopic appearances of the
material in the vesicles are certainly not those of normal secretion.
Granted that there is an increased functional activity of the glandular
tissue, as the character of the lining epithelium of the vesicles and the
analogy with the changes seen in compensatory hypertrophy would seem
to indicate, then the scantiness of the colloid contents of the alveoli most
Hkely would point to a rapid and extensive absorption of this altered
secretion, and its absorption might account for the varied nervous and
other symptoms of the disease. The pathological changes discovered in
the central nervous system and sympathetic might also be due to this
cause ; they appear to resemble the lesions known to be the result of
toxic poisoning.

The following account of the condition of parts found in the
thyroid gland and elsewhere, is founded upon the examination of three
instances of the disease. In Case i, permission was fortunately obtained
for a complete post mortem examination. The portions of gland
in the two other cases were excised during life ; that described under
Case 2 is particularly interesting, as illustrating the changes which had
taken place nearly two years after operation.

Case 1.

This case was the only one which afforded opportunity for a complete
post mortem examination. The patient was a young girl, aged 16, who
had suffered for about two years from all the typical clinical features of
exophthalmic goitre. She died very suddenly after over-exertion,

(203)

6 QUATERCENTENARY STUDIES IN PATHOLOGY

apparently from cardiac syncope : for a short time previously she had
exhibited signs of aortic disease. At the post mortem examination,
which was performed some twenty-four hours after death, the following
appearances were noted : —

The body was somewhat emaciated. The limbs were rigid. The
dependent parts were moderately livid. The pupils were equal and of
medium size. The exophthalmos, which had been so evident during life,
had almost vanished.

£ Itithts

Fig. I.— The Thymus Glanc

Heart. — The cavity of the left ventricle was dilated, and its wall was
slightly hypertrophied. The aortic valve was incompetent to the water
test. The edges of the two aortic cusps were puckered and calcareous.
The edge of the mitral valve presented a somewhat thickened appear-
ance, and was at one point also calcareous. The left auricle was slightly
dilated. (204)

CONTRIBUTION TO PATHOLOGY OF EXOPHTHALMIC GOITRE 7

Lungs. — The pleural surfaces were free from adhesions. With the
exception of some slight hypostatic congestion at the bases, the lungs
were of normal appearance.

The Spleen weighed 223 grammes. It was distinctly larger than
usual. There was a well-marked notch in its anterior border. The pulp
was firm and fleshy ; its cut surface was of a deep red colour, dotted
over with very prominent grey Malpighian bodies.

The liver, kidneys, gastro-intestinal tract, and internal genital organs
were of normal appearance.

^ Tnekt

Fig. 2.— The Thyroid Gland.

The Brain, with its membranes, presented nothing abnormal to
naked-eye examination.

The orbital vessels were not dilated, but the orbital cavities contained
somewhat more fat than in health. The cervical sympathetic ganglia
were also of normal appearance.

The Thymus Gland (Fig. i) was enormously enlarged. It extended
downwards in the form of a tongue-shaped mass, adherent to the

(205)

8

QUATERCENTENARY STUDIES IN PATHOLOGY

anterior surface of the pericardium. The upper portion of the gland
consisted of two finger-like processes, the tips of which were just
observable above the suprasternal notch. The left innominate vein was
firmly adherent to the posterior surface of the mass. The entire gland
measured 4J inches in length ; it varied in breadth from i J to 2 J inches.
The weight was 46 grammes. The cut surface was faintly granular, and
had a greyish-pink colour.

Fig. 3.— Section of Thyroid Gland (Case i).
Shews the small size of the alveoli and the scantiness of colloid. Zeiss Ohj. A, Oc. 4.

The Thyroid Gland (Fig. 2) was also greatly enlarged, the enlarge-
ment being confined to the lateral lobes, and practically symmetrical,
while the isthmus had remained unaffected, and was about the size of a
crowquill. The tip of the lobe on each side lay about an inch below the
upper border of the thyroid cartilage ; the base on each side was nearly
an inch below the lower border of the cricoid cartilage. The right lobe
measured 2f inches long by if broad. It weighed 39-5 grammes. The
left lobe measured 2 J by if inches, and weighed 29 grammes. The cut

(206)

CONTRIBUTION TO PATHOLOGY OF EXOPHTHALMIC GOITRE 9

surface of the gland was of a dark pink colour, and presented a granular
fleshy appearance. There was no evidence of any increase in size of
the parathyroids.

Microscopic Examination.

Thyroid Gland. — There was an undoubted increase in the number of
alveoli, which were also, as a rule, much smaller than those of the normal
gland. Their size, however, varied greatly, and in shape they tended to

Fig. 4. — Section of Thyroid Gland (Case i).

Shews the irregularity in shape of the larger alveoli and the papillary projections of the wall.

Zeiss Obj. A, Oc. 4.

be oval or irregular. In the case of the larger vesicles, curious papillary
projections of the lining membrane were frequently seen. (Figs. 3 and
4.) The alveolar epithelium had lost its typical cubical shape, and had
assumed a high columnar form (Fig. 5). The cells were stained sharply
with the usual staining solutions ; their nuclei were rounded or oval, and
stained deeply with logwood. In one or two instances the nuclei shewed

(207)

lO QUATKRCRNTENARY STUDIES IN PATHOLOGY

evidence of karyokincsis, and the cells were evidently in a state of great
activity. Colloid was, comparatively speaking, scanty ; many of the
small alveoli indeed shewed no signs of colloid in the interior. The
colloid in the large alveoli had shrunk away from the epithelial lining,
and shewed round its margin a vacuolated appearance. It frequently
contained desquamated epithelial cells, and in some instances leucocytes.
In sections stained with eosin and haematoxylene it usually shewed a
somewhat patchy staining, which differed very markedly from the homo-

Fig. 5. —Section of Thyroid Gland (Case i).
Shews columnar shape of the lining epithelium. Zeiss Obj. D, Oc. 4.

geneous staining of the colloid of normal thyroid. Small deposits of
colloid were often seen between the alveoli, lying in what appeared to be
lymphatic spaces. The blood vessels, although fairly numerous, were
not dilated. The connective tissue between the alveoli was somewhat
scanty. Scattered sparsely through the specimen were small deposits of
lymphoid cells (Fig. 6). These deposits were of irregular shape, and
consisted of a small central mass closely packed with cells, which radiated

(208)

CONTRIBUTION TO PATHOLOGY OF EXOPHTHALMIC GOITRE II

out in short rows into the tissue between the neighbouring alveoh*.
These lymphoid deposits were not nearly so evident in this as in the two
other specimens.

The Thymus Gland. — The condition of this gland appeared to be
that of a true hypertrophy. Examined microscopically, the tissue looked
like normal thymus with this exception, that the differentiation between
the cortex and medulla was not quite so sharply defined. The lymphoid
cells, the concentric bodies of Hassall, and the connective tissue frame-

Fig. 6. — Section of Thyroid Gland (Case i).
Shews a portion of a small lymphoid deposit. Zeiss Obj. D, Oc. A.

work were present in natural proportions. There was a complete absence
of the eosinophile cells recorded by certain observers in this disease.

The Central Nervous System. — Sections were made at different levels
of the upper cervical cord, the medulla, and the pons, and were stained
by various methods. The parts seemed to be quite healthy.

The Cervical Sympathetic System. — Sections were made of the
cervical sympathetic ganglia of both sides, but all presented a normal
appearance.

(209) P

12 QUATERCENTENARY STUDIES IN PATHOLOGY

The Spleen.— T\\Q Malpi^hian bodies were much larger than they
vshould be, and this increase in size appeared to be due to an augmenta-
tion in the number of lymphoid cells surrounding the arterial branches,
while the usual structure of the Malpighian body was in other respects
retained, and the spleen pulp generally did not shew any change.
Possibly this increase in the size of the Malpighian bodies may be
similar in its characters to the lymphoid hypertrophy, which occurs in the
thymus and other lymph-structures in this disease.

Case 2.

Clinical History. — The patient was a female, aged 19, who was first
seen at the Aberdeen Royal Infirmary in February, 1904. She presented
at that time a typical picture of exophthalmic goitre, the exophthalmos,
rapid pulse, muscular tremor, and nervous excitability being particularly
well marked.

Upon the 26th of February the right lobe of the thyroid was removed
by Mr. H. M. W. Gray. In order to avoid injury of the recurrent
laryngeal nerve, the posterior border of the lobe was shaved off. During
the operation, bleeding was somewhat troublesome, and large vessels,
both arteries and veins, could be seen running over the surface of the
gland below the capsule. The patient made a good recovery from the
operation, and in April, 1904, it was noted that her general condition was
much improved ; the exophthalmos had almost disappeared, the
remaining lobe of the thyroid had much diminished in size, and the
general excitability was very much less. The muscular tremors and
rapid pulse still remained fairly well marked.

In January, 1906, she was readmitted to hospital. The general
nervous excitability was extreme, the exophthalmos had reappeared, and
muscular tremor was very distinct. The rapid pulse had again become
pronounced.

Upon the 2nd of February part of the left lobe of the thyroid was
removed. The gland appeared to be somewhat vascular, and bleeding
was again rather troublesome.

Three weeks after this second operation it was noted that the general
condition of the patient was very greatly improved. The muscular
tremors had gone, the pulse was nearly normal, and the exophthalmos
had again disappeared.

(210)

CONTRIBUTION TO PATHOLOGY OF EXOPHTHALMIC GOITRE

Examination of Specimens.

The right lobe of the thyroid, which was removed at the first
operation, weighed 29 grammes. It measured 2| by ij inches, and
varied in thickness from f to i inch. The entire lobe was enlarged
symmetrically, and faintly lobulated. Its cut surface had a reddish-pink
fleshy appearance ; it did not seem to be very vascular, although several
moderately large vessels were seen upon the capsular surface. The
parathyroids had not been removed ; they probably remained attached
to the posterior border of the lobe.

Microscopic Examination. — The general appearances were very similar
to those recorded in Case i ; there was the same tendency for the alveoli
to become irregularly branched, and for the lining epithelium again to
assume the columnar type. The sections, however, had a much more
patchy appearance ; areas of small, almost rounded, alveoli, alternating
with areas of large and extremely irregular vesicles, into which they
gradually merged. The lining epithelium of the spaces was high
columnar in type, this character being especially distinct in the small
alveoli. The small vesicles shewed in many instances an irregularly
stellate mass of colloid ; more frequently they were empty, or contained
a small amount of granular material which stained of the same tint as
colloid. The large alveoli contained much colloid matter which had
shrunk away from the alveolar wall ; it stained rather indefinitely, was
usually vacuolated, and sometimes contained desquamated cells. The
vessels were numerous and full of blood. Here and there throughout
the section were small irregular deposits of lymphoid cells, in character
similar to those seen in the first case. These small lymph-deposits
invariably contained a few scattered plasma cells, and a small number of
these plasma cells was also found in various parts of the section, generally
in relation with one of the blood vessels.

The portion of the left lobe removed at the second operation con-
sisted of two fragments. The larger of these measured 43 by 30 by 11
millimetres, and weighed 8J grammes. It was flat and roughly oval in
shape, and its surface was slightly lobulated. The smaller fragment
measured 23 by 14 by 8 millimetres, and weighed 1*9 gramme. It was
about the size and shape of a small bean. The cut surface of the

(211) p \

14 QUATERCENTENARY STUDIES IN PATHOLOGY

fragments was of a pink or yellowish-pink colour, and presented a some-
what mottled appearance.

Upon microscopic examination the sections presented the same
patchy appearance as was seen at the examination of the specimen
removed at the first operation, but the areas of large alveoli now pre-
dominated. In many instances these large alveoli were of quite regular
shape, and were filled with colloid, which stained homogeneously, and
was of normal appearance. Their epithelium had a distinct tendency to
revert to the normal cubical type, while the epithelium of the smaller
alveoli still retained its high columnar arrangement. These small
alveoli, however, undoubtedly contained more colloid than previously,
although it still remained granular or vacuolated. Probably the most
striking microscopic feature was the large amount of lymphoid tissue.
Every section shewed numbers of rounded or oval lymphocyte deposits,
many of them of considerable size. Occasionally one of the gland
alveoli appeared to have become enveloped by the larger lymph depots,
and to have undergone atrophy in consequence. Plasma cells were
particularly numerous. Each of the lymph deposits contained several of
them ; indeed, long rows and columns of these cells could be traced
between the smaller alveoli, and in parts formed quite extensive deposits.
They appeared to follow the course of the capillaries and small blood
vessels, and to accumulate around their walls.

Case 3.

M. W., a female school teacher, aged 41 years, had suffered from
symptoms of Graves' disease for three years. Upon examination she
shewed well marked thyroid tumour, distinct exophthalmos, rapid pulse,
loss of flesh, and very evident muscular tremors. She complained
greatly of this last symptom, and of nervous excitability and palpitation.
The right lobe of the thyroid was distinctly more enlarged than the left.

On July 4th, 1903, Mr. Gray removed the right lobe of the thyroid
by an operation exactly similar to that adopted in the last case.
Haemorrhage was comparatively slight, and easily controlled. She stood
the operation well, and made a rapid recovery. Very great improvement
in her condition resulted. Tremor almost disappeared, the exophthalmos
rapidly lessened, the pulse rate fell, and the enlargement of the left lobe

(212)

CONTRIBUTION TO PATHOLOGY OF EXOPHTHALMIC GOITRE 1 5

of the thyroid was greatly decreased. When last seen in April, 1904,
this improvement continued to be maintained.

Examination of the Specimen.

The portion removed consisted of the right lobe of the thyroid, with
the exception of a thin strip along its posterior border. The lobe was
enlarged symmetrically, and weighed 35 grammes, and measured 2 J
inches by i J. At its thickest portion it measured i inch. Its cut surface
had the same pink fleshy appearance seen in the previous cases.
Parathyroids could not be found.

Microscopic Examination. — The size of the alveoli varied greatly,
and the large vesicles were, in shape, extremely irregular and branching,
while papillary projections from their walls were of frequent occurrence.
Colloid was more plentiful than in the former cases, but it presented the
same vacuolated characters, and in many instances was granular. In
every section there were a few scattered lymphoid deposits. These
generally contained a small number of plasma-cells, but there were no
plasma-cell-collections in the connective tissue or round the blood vessels.
The connective tissue between the alveoli was more abundant than in the
other cases. Blood vessels were fairly numerous, although the tissue
could not be said to be highly vascular.

General Summary.

The microscopic appearance of the thyroid was essentially the same
in all three cases, although it differed slightly in certain minor details.
There was in each case the same glandular hyperplasia, and the same
tendency to irregularity in the shape of the glandular alveoli, many of
which shewed papillary projections from their walls. The columnar
shape of the lining epithelium was always very marked, and the
tendency towards the columnar form could be distinctly traced even in
those alveoli which approximated very closely to the normal type in size,
shape, and in the character of their colloid contents. The amount of
colloid present, as compared with the normal, was in all cases diminished ;
in the smallest alveoli, in fact, colloid was usually absent. The characters
of the colloid, as seen microscopically, were altered in all cases. The

(213)

l6 QtJATEkCEt^TENARV STUdIES l^ PAtHOLOGV

small alveoli frequently contained only a small quantity of granular
material, which stained somewhat deeply with eosin, and bore no
resemblance to ordinary colloid ; in other instances they contained a
small amount of granular and vacuolated colloid. The colloid of the
large alveoli was invariably vacuolated, and stained in a patchy manner ;
leucocytes and desquamated cells were frequently seen within it. The
sections always shewed numerous lymph-deposits, in character alike with
those described under Case i. None of the thyroids exhibited any
tendency to interstitial fibrotic change, and there was no evidence in
support of the theory that the blood vessels played any part in the
formation of the goitre.

The general appearances in all three instances were strongly
suggestive of increased activity of the gland ; the colloid secretion being,
moreover, of abnormal character. They were in accord in most respects
with the published descriptions of the changes of the gland in the
disease.

Case I would seem to shew that, in this instance, at least, the disease
was of thyroidal origin. There was a complete absence of any of the
changes which have been described as occurring in the central nervous
and sympathetic systems.

With regard to Case 2, the microscopic appearances of the second
portion of the gland removed seemed to indicate some attempt upon
the part of the thyroid tissue to revert to its normal condition as a result
of the first operation. Certain of the alveoli were undoubtedly larger
and of more regular shape than those seen on section of the first
specimen from this case. The colloid contained in these was practically
of normal character, and the lining epithelium approached very closely
in some of them to the normal cubical type.

In conclusion, I must express my indebtedness to Dr. George
Edmond for the opportunity he afforded me of making the post mortem
examination of Case I ; and to Mr. H. M. W. Gray for permission to use
the clinical histories of Cases 2 and 3.

(214)

THE RAT THEORY OF PLAGUE EPIDEMICS.
By William Hunter, M.B., CM.,

Government Bacteriologist^ Hong-Kong.

(215)

Tke Rat Theory of Flague Epidemics.

In all the plague treatises which have been published during the
past year or two, one finds a considerable amount of attention devoted
to the question of the occurrence of plague infection in rats, and
the dangers attached to the presence of such infected vermin in the
direct vicinity of man. The dictum of Koch and many epidemiologists,
that plague is primarily a disease of the rat, secondarily a disease of
man, and that epidemic plague is entirely dependent on the presence of
widespread plague infection in rats, has become widely recognised and
accepted by many plague experts. If we look into the evidence in
favour of such a conclusion, however, little information of a definite
nature can be obtained. Up to the present time, no research has been
accomplished which would justify the conclusion that plague-infected
rats are the chief sources from which the virus is communicated to man.
It must be admitted that there is much evidence in favour of such a
method of transference, but, notwithstanding the constant influx of
additional evidence, there yet remains to be shown the direct connection
between epizootic and epidemic plague.

It is only within the past few years that the occurrence of disease in
animals has become fully recognised as an important disturbing factor
in the successful dispensation of modern sanitation. With the advent of
more perfect knowledge in regard to the ways of infection, attention has
become directed to the occurrence of diseases in animals, diseases which,
when compared with those found in man, appear to be similar. The
question has, therefore, arisen — a new one indeed — as to the part played by
animals in the dissemination of certain diseases amongst the human race.

At the present day, one cannot take up a book dealing with plague,
without being struck with the prominent part devoted to rats as
disseminators of infection. In fact, on the perusal of many such manu-
scripts, one is drawn to the conclusion that rats, and nothing but rats, are

(217)

4 QUATERCENTENARY STUDIES IN PATHOLOGY

responsible for the existence of epidemic plague. Going into the
question more thoroughly, however, it is difficult to gauge exactly the
premises upon which such a conclusion has been drawn. We have to
consider whether rats are the only disseminators of plague — in fact, do
they disseminate plague epidemics at all ? — or, granting that they are re-
sponsible for the occurrence of such — are they solely responsible, or simply
partners in the wholesale serving up of the infection to the human species ?

Plague in rats is by no means a recent discovery. Great
mortality in rats was noted by the ancients during outbreaks of plague.
Classical, mediaeval, and modern literature furnishes us with many
interesting details in regard to the susceptibility of various animals to
plague infection. The general lay opinion in regard to the matter was,
that during plague epidemics an increased mortality was observed
amongst animals in the infected district, and that such a death-rate was
noticeable especially amongst rats.

It would be difficult to determine who propounded the rat theory of
plague epidemics. The prominence given to epizootic plague, during
recent years, has arisen out of better acquaintance with the bacteriological
and epidemiological aspects of the disease.

In the year looo, Avicenna noted the presence of a high rat mortality
during plague epidemics. He says : — " Et de eis, quae significant illud,
est, ut videas mures, et animalia, quae habitant sub terra, fugere ad
superficiem terrae, et pari sedar,* id est commoveri hinc inde sicut
animalia ebria."

Again Nicephorus Gregoras, in 1348, and Orraeus, in 1771, in their
Treatises on Medicine, look suspiciously upon rats as having some
relation to outbreaks of epidemic plague.

Simpson, in his recent treatise on plague, and in his Report on
Plague in China and Hong-Kong, has given us an excellent resume of
the beliefs held by the Chinese in regard to the prevalence of the disease
in animals such as rats. It is evident that the opinion is widely diffused
amongst the Chinese that there exists some relation betw^een the epizootic
and the epidemic. In their writings, the Chinese make frequent mention

* Pari sedar — an Arabic expression, the interpretation of which has given rise to some
difference of opinion. In Avicenna arabum medicorum principis. Liber quartus canonis,
ex Gerardi Cremonensis versione, et Andrea; Alpagi Bellunensis castigatione (Venetiis, mdxcv.),
torn. XL, tract. IV., cap. 3, p. 69, pari sedar is translated not as above but as "exire
manifeste." — Ed.

(218)

THE RAT TttEORV OF t>LAGUE EPIDEMICS ^

that during, or immediately preceding, plague epidemics, severe outbreaks
of disease occur amongst their cattle, their fowls, and their house vermin.

Reports such as these are extremely interesting to investigators of
the present day, who are endeavouring as far as possible to bring rats into
close relationship with the occurrence of plague amongst human beings.

Within the past decade, much evidence of a more weighty character
has been brought forward in support of the rat theory of plague. The
experience gained by the outbreaks of plague in the seaports of Oporto,
Alexandria, Sydney, Kobe, London, etc., goes far to establish some
connection between rat and human plague. In these ports, as well as in
others, the disease clung fast to the harbour. The neighbouring cities
and villages were not, or only slightly, affected by the disease.

Again, in the grossly overcrowded and filthy cities of the Orient, the
introduction of human plague has scarcely ever been followed by the
establishment of an actual endemic plague focus. In such cities or
ports, in which plague broke out epidemically, these epidemics would
not appear to depend on the arrival of plague-infected human beings,
but upon the introduction of plague-infected rats.

In different plague epidemics, grain stores, in which rats preferably
lodge, have frequently become the central points of the outbreak. For
example, in Bombay, plague broke out first among the Banniahs
(German Plague Commission), and in Oporto the food stores appeared
to be the centres from which the epidemic spread. (Kossel and Frosch.)

Of great importance are rats on board ship. Plague-infected rats
may be present on board ships, without the occurrence of human plague.
Further, plague rats, or the bodies of plague rats, on board ship, are
probably of greater importance than a plague-infected man. The
possibility of the infection of merchandise through plague-infected ship
rats is of great moment, and cases of such have been reported during
the past few years. According to Kossel and Nocht, merchant ships
have frequently arrived in Hamburg from plague-infected ports. No
cases of plague in man have occurred on board, but, on discharging the
cargo, numerous dead rats were found, many of which were plague-
infected. Similar instances have been found recently in Bristol, London
and Hong-Kong. In such cases, therefore, it is obvious that all
precautions must be exercised to prevent contact of the healthy rats on
shore with those on board ship or with the infected merchandise.

(219)

6 QUATEkCENTENAKY STUDIES IN PATHOLOGY

The reports of English writers, during the past few years, admit the
importance of the rat in disseminating plague infection, yet the majority
of these investigations show that considerable difficulty is experienced
in perceiving how the infection can be conveyed from the rat to man.
Snow's observations note the incidence of plague in rats, but do not
suggest any connection between the epizootic and the epidemic. The
Indian Plague Commission does not lay any weight on plague-infected
rats as agents in the dissemination of plague epidemics. Bruce Low's
papers on Bubonic Plague, dated July, 1902, lead one to believe that
man and the rat are reciprocally infected. No evidence is forthcoming,
however, in regard to the question from a priori grounds. To explain
the transference of the disease from the rat to man, Hankin concluded
that plague in man stood in relation to the accessibility to rats, and that
probably some intermediary insect was necessary to communicate the
infection from rat to man. About the same time, Simond stated that
the epidemicity of plague was due to migrations of plague rats, and not
to human intercourse. Further, the experience of Ashburton Thompson
in Sydney, during the recent outbreaks of plague, is that plague rats
were the sole source from which the infection was communicated to man.
Still, as this writer has remarked, considerable systematic research is yet
required in order to establish definite proof that man and the rat, in the
usual circumstances of propinquity, are reciprocally infective. In Sydney,
every effort was made to settle the question. Much supplementary
evidence was obtained, and the conclusion drawn that infected rats play
an important,, if not the most important, role in the dissemination of the
disease.

At present the general trend of opinion is against the causes which
have hitherto been advanced to explain epidemic plague. It miist be
difficult for a certain class of professional men to get away from the
time-worn causes of epidemic disease. Indeed, in all text-books on
plague, one still finds the old dogmas of infection most carefully
reproduced, namely, the solution of the problem by a recognition of: —

{a) The communication of the infection from the sick direct.

{U) Indirect means.

{c) Place Infection, etc.
Such a table of causes would amply explain the epidemicity of plague,
but the adherents of such a classification of etiology would appear to

(220)

THE RAT THEORY OF PLAGUE EPIDEMICS 7

take no cognisance of the advances made by bacteriological research.
As may be gathered from the most recent works on such a subject, the
ways and means afforded to the B. pestis to produce infection, the course
of plague infection, and the exact problems underlying the spread of the
pest, must undergo remodelling upon the lines indicated by the results of
modern epidemiology.

At the present time the general trend of opinion is against the
hitherto advanced causes of epidemic plague, namely, direct and indirect
mode of communication of the disease, and place infection. My own
experience of plague epidemics leads me to conclude that, apart from
cases of primary pneumonic plague, the dangers of one person infecting
another are over-estimated, and that place infection, apart from the
presence of plague-infected rats, is of no great significance.

We know that rats are highly susceptible to plague, and they readily
communicate the infection to other rats. Rats, when suffering from
plague, are peculiar in their habits. They leave their holes. They are
apparently deprived of any sense of fear or danger when near human
beings. They progress with a drunken-like gait. Convulsive movements
cause them to make erratic springs into the air during ordinary
progression. They die suddenly, usually from convulsions. The natives
of Bombay are so frightened by the peculiar appearance of these animals,
and the finding of dead rats in their houses, that they flit at once.
According to Zupitza, the natives of Kisiba, in Ceatral Africa, have the
same dread of rats either suffering from this condition or dead in their
dwellings.

These remarks bring me up to the general consideration of my own
results. Much evidence has been advanced in favour of the rat theory,
and that, when added to the conclusions drawn from my own researches,
ought to bring the epizootic and the epidemic into close relationship to
each other.

Previous to my arrival in Hong-Kong, no attempt had been made to
thoroughly investigate the course of epizootic plague in rats. From
time to time a few rats had been examined for the presence of plague
infection, but these examinations, amounting to a few hundreds only,
cannot be regarded as of much value, beyond establishing the fact that
such a disease as " Rat Plague " existed in the Colony.

(221)

8 QUATERCENTENARY STUDIES IN PATHOLOGY

It was evident, from the investigations of Professor W. J. Simpson,
that much was to be gained by regular and systematic examinations of
all rats, dead or alive, and further, his experience of plague in South
Africa pointed to an intimate association between the rat epizootic and
the human epidemic.

Accordingly, on my arrival in Hong-Kong, arrangements were made
to have as many rats as possible collected from the various Health
Districts of the Colony, and forwarded to me for bacteriological
examination. Liberal assistance for the methodical carrying out of this
research was provided by the Government. Four bacteriologists were
obtained from Japan, and assistants were requisitioned from the College
of Medicine in Hong-Kong. The following bacteriological methods
were employed : — Exact details as to the place where each rat was found
was furnished by the Sanitary Department. The post mortem examina-
tion on each rat was made under antiseptic precautions, and smears,
both of the heart blood and spleen pulp, were made on microscopic
glass slides. These were dried, fixed, stained by the usual tinctorial
methods, and examined microscopically. In almost all cases, plague
infection in rats is most pronounced, the smears usually containing
millions of typically oval, bipolar, plague bacilli. Unless the bacilli
were typical in appearance and present in enormous numbers, a positive
diagnosis was never made. If any doubt existed as to the nature of the
organisms present, Gram's method of decolorisation was employed as a
counter test.

From time to time, cultural tests were also made. Owing, however,
to the enormous number of rats examined daily by such a limited
staff, the microscopic and tinctorial methods of examination were the
only possible means of arriving at a diagnosis. Subsequent to the post
mortem examinations, all the rats were cremated in an apparatus erected
in the immediate neighbourhood.

T\i^post mortem and bacteriological examination of large numbers of
living and dead rats, is a research which must be carried out with
considerable care. The diagnosis of the presence of plague bacilli in any
tissue or organ, by means of the microscope alone, is frequently one of
extreme difficulty. This is all the more so, because, in the tissues of
man and animals, micro-organisms occur, which by the microscope alone
cannot under any conditions be distinguished from the plague bacillus,

(222)

THE RAT THEORY OF PLAGUE EPIDEMICS 9

The research is even comph'cated to a much greater degree in regard to
the diagnosis of the presence of plague bacilli in rats.

Rats suffer from a large number of septicaemia diseases. These
diseases frequently break out amongst them in epizootic form with a
heavy mortality.

Among the many micro-organisms causing these epidemic diseases
may be mentioned B. Danysz, the B. of Schilling, etc. Other micro-
organisms, such as the B, of fowl cholera, B. of swine plague, etc., are also
pathogenic for rats. Morphologically and tinctorially all the above-
mentioned micro-organisms resemble the B. pestis. Therefore, it is
evident that even though plague is prevalent among the rats in any
particular city, one has to be on guard for the possible occurrence of
other epidemic diseases which might account for an increased death rate
at any time among them.

Since the year 1902, this systematic bacteriological examination of
rats has gone on. The following figures give approximately the number
of rats collected and examined, with the total found plague-infected per

annum. Year. No. examined. No. found infected.

1902. 117,839 2015

1903. 101,056 3744

1904. 21,907 993

Having established bacteriologically the existence of Rat Plague in
Hong-Kong, the examinations were carefully continued, and the amounts
of the epizootic and the epidemic noted week by week. In this way it
has been found possible to plot curves of each form of the disease.
Annual charts, representing the weekly fluctuations in the epizootic and
the epidemic, have been prepared. In these, the epizootic is marked by
an uninterrupted line, the epidemic by a dotted line.

In dealing with the results which have been obtained through
following the courses of the epizootic and the epidemic, I propose to
divide my investigations into the following divisions : —

I. The History of the Course and Relations of Epizootic and
Epidemic Plague in Hong-Kong —
{a) during the year 1902.
(b) during the year 1903.
{c) during the year 1904.
H. The Bridging of Epidemic Plague during the intervals i902-'o3
and '04.

(223)

lO

QUATERCENTENARY STUDIES IN PATHOLOGY

I. — The History of the Course and Relations of Epizootic and

Epidemic Plague.

{a) During the year igo2.

The examination of rats was commenced in April. Previous to this
date it was impossible to say more than that casual examinations showed
rat plague to be existent. The extent and fluctuations of the epizootic,
however, were not accurately determined until my series of investigations
commenced.

II o N a K o N a - 1 1) o sa .

X.i^m.

: uji ?.■( 4 111

L

III f^

Hiiniftn Plnjni*

^

r

■1.1

a^t4

Chart I.

IL' l.M iC, •> u. 16 23 30 7 U -n 2S 4 1 I IS L'i L' a 1« i3 30

On examination of Chart I. for the year 1902, certain general
features are at once evident. The initial rise in rat plague is a
rapid one. It commences with 2, 4 or 6 cases per week. This number
is doubled or tripled during the succeeding weeks, and, with gradually
increasing amounts and fluctuations, the maximum of the epizootic is

(224)

THE RAT THEORY OF PLAGUE EPIDEMICS II

reached in about a month or six weeks after its commencement. Again,
it will be seen that the epizootic reaches a higher level than the epidemic.

An important point is that the epizootic appears in advance of the
epidemic, the latter making its appearance a week or a fortnight later.
Further fluctuations in the amount of the epizootic are usually followed
by fluctuations in the epidemic, and a general resemblance of the two
curves to each other is reproduced. It is interesting to note that the
interval of a week or a fortnight between the fluctuations in the two
curves is maintained until the maximum amount of both forms of
plague is developed.

The apex of the curve of the epidemic is reached almost exactly a
fortnight after the maximum amount of the epizootic.

With these general remarks, it will be well to examine the chart for
1902 more in detail.

The epizootic curve reaches its maximum within five weeks. The
epidemic follows closely behind, reaching its maximum in seven weeks
from the date of the commencement of the epizootic, i.e., a fortnight
later than rat plague. If the chart be examined closely, it will be found
that elevations in the rat plague curve are followed by similar exacerba-
tions in the amount of the epidemic. This picture is perhaps not so
graphic in the chart for the year 1902, but, when examined along with
that for the year 1903, the elevations and depressions of the epizootic
curve are found to be followed by similar irregularities in the curve of
the epidemic. Indeed, these variations of the one curve, followed by
similar changes in the other, are so admirably reproduced (especially in
the year 1903), that one would imagine that the epidemic curve in its
rise and fall is too good to be true.

Such an epidemic curve, occurring ten days or a fortnight after that
of the epizootic, and reproducing so admirably the variations in the
amount of the latter, is, in my opinion, strong evidence of some
extremely close relation existing between rat plague and human plague.

In order to bring the epidemic curve into close relation with that of
the epizootic, the defervescence of the curves is of the highest importance.
So far as the ascent and maximal elevation of the curves is concerned,
the two curves follow each other closely. From this alone, however, one
could not claim to have established a connection between the two out-
breaks. All that can be said is that both occur about the same time,

(225) Q

12 QUATERCENTENARY STUDIES IN PATHOLOGY

but probably quite independently of each other. At least, the question
would be asked, that, given these two curves, the one, as it were, lying
within the other, what points in these lead to the conclusion that the
one is the result of the other ?

The history of the outbreaks is useful. It is seen that the epizootic
precedes the epidemic by about a fortnight ; that variations in the
epizootic are followed by variations in the epidemic ; that their maximal
points more or less correspond ; and, in general, a superficial glance at
the curves makes one think of the extremely close relation which must
exist between the two outbreaks. But it must be remembered, that rat
plague may be present without the advent of human plague, and vice
versa, and that occasionally the amount of rat plague bears no relation
to the amount of human plague. These questions, and even others,
must be answered before accepting the statement that in rats we have
the key to the problem of plague prevention.

The general curve for the year 1902 shows the epidemic curve almost
surrounded by that of the epizootic. From this chart alone it could not
be concluded that the one is the cause of the other ; if so, we should
expect the rat plague curve to fall inside that of the epidemic curve.
Shortly, if there is a connection between the epizootic and the epidemic,
the latter should rise according to the elevations in the epizootic, and,
with a fall in the latter, human plague should more or less cease.

The preliminary parts of these curves have been followed. They are
seen to more or less harmonise. The point now to be considered is the
relation existing between the epidemic and the epizootic towards the
close of the outbreaks. If the latter is the cause of the former, it should
fall previous to the decline in epidemic plague. In the general chart for
the year 1902, the rat plague curve has just missed falling inside that of
the epidemic. Here, however, we are dealing with an extremely mixed
collection of results, taken from the City of Victoria as a whole. In
Hong-Kong the city is divided up into a series of Health Districts. By
comparing the relationships between the two outbreaks in individual
districts, results of even a more convincing nature are obtained. By
viewing the results according to this plan, a sufficient complement of
epizootic and epidemic plague is obtained, and more accurate conclusions
drawn. The dearth in the number of cases of the epidemic, and the
excess in that of the epizootic vitiate what would otherwise have been a
positive result. (226)

THE RAT THEORY OF PLAGUE EPIDEMICS

13

In all the districts in Hong-Kong in the year 1902, with three
exceptions (European quarter), the charts showed that the epizootic fades
first, and is followed by the epidemic. Immediately the rat plague
diminishes or disappears, human plague becomes more or less extinct.
The accompanying Charts II., III., showing the condition of the
outbreaks in the Health Districts, Nos. 9 and 10, clearly prove, to my
mind, the close relationship which must exist between epizootic rat

HONGKONO DISTRICT No. 0.-1002.

tiioSSuu 0

H>^&j.i.<

m

A^lJ] n

-1 J : •= h

n i

!,..!

-[Rj

xra_n_

7 U 21 M 1 1 1 IS 3,', 4 1 1 IS JS I .S li 2-i 2U G 13 20 27 3 10 17 U I 8 15 22 29 5 12 li) 2G 2 9 IC 23 30 7 14 21 28 4 1 1 18 2i 2 9 IG 23 30

Chart II.

rir)NGKONG DISTIilCT No. lO.-lOOS

n_rL_

1 2S 4 I1IS2.-, I I1I.S2.-) 1 S 15 22 2',! G 1320 27 3 10 17 24 1 8 l.i 22 20 J I2102G 2 9 ir,;3.iO 7 11212S 4 111835 2 9 1623 30

Chart III.

plague and epidemic plague. These charts demonstrate the fall in the
epizootic, followed by a similar defervescence in the amount of epidemic
plague.

(b) During the year IQOJ.

An important point in connection with the determination of the
relation of the epidemic to the epizootic, is that, during this year, one is
enabled to follow closely the train of events from the commencement of
the year.

(227) Q I

14 QUATERCENTENARY STUDIES IN PATHOLOGY

On the 1st January, 1903, rat plague was existent, and constituted
more or less a definite epizootic. On the other hand, there was no
trace of an epidemic. If reference be made to the curves of the epizootic
and epidemic during the last quarter of 1902, it will be seen that the
former raged more or less incessantly ; whereas the epidemic, apart from
one or two sporadic cases of plague, was non-existent. The epizootic
maintained an average level of from 20 to 30 cases per week. Such a
condition of affairs did not hold good for the commencement of the new
year 1903. Almost immediately the epidemic appeared, and subse-
quently both this and the epizootic ran very characteristic charts.
Within a fortnight of the new year, human plague was present (Chart
IV.). The commencement of the third week of January marked the
advent of the 1903 epidemic of pest. A careful survey of both curves,
will enable one to draw one's own conclusions. Although the rat plague
is in excess numerically, the rises and falls of the epizootic curve are
followed only too closely by similar elevations and depressions in the
incidence of human plague. One rarely observes such a following of one
curve after another. The apices of both curves are found to correspond
with the result obtained during the year 1902. Both reach their highest
points within a fortnight of each other, the epizootic being first.

The course of events, after the outbreaks have reached their highest
figures, is interesting. The epidemic falls in numbers so rapidly, that,
practically within a month, the number of cases of human plague has
fallen to an insignificant figure. But what is even more remarkable, is
the sudden drop in the epizootic. The time occupied by the epizootic in
falling from its maximum to its lowest level is about 15 weeks. The
epidemic took about 13 weeks, namely, about a fortnight less.

The subsequent history of the curves is not so interesting. Rat
plague never disappears. Human plague is practically absent. This
interval between the incidence of human and rat plague appears to be
more or less constant. In the year 1902, the interval between the
occurrence of both outbreaks was from 10 to 14 days. The course of
events during 1903 leads us to the same conclusion in regard to the time
limit.

Comparing, therefore, the curves for both years we find that the
epidemic begins a fortnight later than the epizootic ; it reaches its
maximum a fortnight later ; this maximum is maintained for a similar

(228)

THE RAT THEORY OF PLAGUE EPIDEMICS

15

length of time ; and its fall occupies practically 14 days. Subsequently
it disappears, whereas the epizootic maintains a low but certain level.
One would expect the fall in rat plague to occur somewhat previous to

HONt^KONG-lOOO

U»flCu<|.

nJ

'' -if

20 Ll

^

T U 21 28 4 11 IS 25 4 11 18 25 1 8 15 i'A 29 ti 13 20 27 :j 10 17 2-1

22 29 5 12 19 20 2 9 Ili2;i30 7 1)

•3-«t».uw^ Va.e/L. <i^i^

Chart IV.

that of epidemic. The condition of affairs is, however, a repetition of
what took place during 1902.

(229)

l6 QUATERCENTENARY STUDIES IN PATHOLOGY

Summing up these two curves, the conclusions which one must
necessarily draw, are that human plague commences immediately after
the increased incidence of rat plague, and that variations in the latter
occasion variations in the former.

A close study of these charts is almost bound to lead one to think of
some relation between the two outbreaks. It may be said that both run
independently of each other, only the epidemic appears about a fortnight
later. However, it would be difficult to imagine such a regular system
in the incidence of an infectious disease, when compared with an
epizootic. We may imagine the epidemic of plague commencing
practically at the same season during each year, but that it should
regularly appear within a definite incubation period after the advent of
the epizootic is difficult to get away from, unless we admit of the
existence of a definite relationship. Again, it is found that the plague
epidemic does not commence at a definite period during each year. In
1902 it was not present until the middle of April. In 1903, it began
about the ist of February ; in 1904, as we shall see later, its commence-
ment was even later than during 1902.

If we compare these dates with that of the advent of the epizootic,
the following is found : — In 1902 the epizootic began about the ist of
April ; in 1903 it began about the middle of January, and in 1904 it will
be found that a similar condition of aflfairs was present. Thus it is
evident that, apart from the season of the year, epidemic plague appears
to be preceded by epizootic plague, and that the interval of time
between the outbreaks varies from ten days to a fortnight.

(c) During the year igo/j..

The condition of affairs during this year compares very favourably
with the points already alluded to under the two previous years. The
attached Chart V. itself explains the constant relation between the degree
of severity of the epizootic and the epidemic. During the first week in
January it is seen that rat plague was present to a considerable extent in
the Colony. The epizootic rose step by step, until the beginning of
February, when a rather erratic jump upwards took place. This sudden
increase in the amount of rat plague was followed by the appearance of
human plague after about ten days.

(230)

THE RAT THEORY OF PLAGUE EPIDEMICS
1 9 O 4 .

17

^

HiMiian,riat;w«,
H;il Plague,

Ul

Ln

if

j-

nJ

C-b.

21 28 r. 12 |;i2.i 2 9 1« 2:i Hd 7 1) 21

Chart V.
(230

l8 QUATERCENTENARY STUDIES IN PATHOLOGY

Further remarks in regard to the curve appear unnecessary, as a
comparison of this chart with the foregoing will clear up any doubtful
points.

II.— The Bridging of Epidemic Plague during the intervals
1902.'03 and '04.

In almost all plague-infected countries, the epidemic is found to be
most prevalent during certain seasons of the year. In Hong-Kong, these
epidemics rage from March to July inclusive. In other countries plague
appears during the colder seasons of the year and vice versa. The
reasons for such a seasonal recurrence of plague are by no means obvious,
and a consultation of current plague literature helps one but little. At
the present time, climatic influences would appear to have lost much of
their significance. Plague epidemics occur in Siberia as well as in the
equatorial regions, and carefully prepared records of sunshine, humidity,
and rain shows these factors to have little determining influence on the
occurrence of plague epidemics. All that can be said in regard to
climate is that it may exercise an indirect influence on the course of the
infection.

During the interval between the end of one epidemic and the first
cases of the succeeding epidemic, human plague is, to all intents and
purposes, non-existent. A few cases do occur, but these are of no great
practical significance, apart from the fact that they help us to remember
that with the decline of the last epidemic our sanitarians have, in all
probability, not yet succeeded in stamping out the disease.

Further, the reasons for the outbreak of erratic cases of human
plague are by no means clear.

To begin with, little or nothing is known in regard to the history of
epizootic plague in rats through a number of years. Many authorities
believe that rat plague dies out on the decline of human plague.

Two charts have been prepared, showing the relations existing
between epizootic and epidemic plague during the period, July, 1902, to
June, 1903, inclusive (Chart VI.) ; and during the period, July, 1903, to
May, 1904, inclusive (Chart VI I.). These show clearly the condition of
affairs which obtains in an endemic plague centre like Hong-Kong. In
both charts the epidemic vanishes, but the epizootic persists with a low

(232)

THE RAT THEORY OF PLAGUE EPIDEMICS

19

and oscillating curve. This curve is more regular in its course than
during the epidemic periods.

From these curves, therefore, we find that rat plague is present
throughout the whole year in mild epizootic form. As a result of this,
sporadic cases of human plague appear from time to time. Epidemic
plague, on the other hand, is prevalent only during the acute exacer-
bations of epizootic plague.

iioNaKONa-1002-00.

(fi^.

ru

Lr

^

,.j,„.._ Hum.. 1H.S'-
Rnl fin.....

8 l.=.22?0 S 151920 •> D lfi2S30 7 14 21 1% \ 11 IS 2i 2 '.) IG 23 30 7 H21 2S •( II iSJ.i 4 11 Ig !,i I St522J3 C 132(127 .I 10 1724

July, ■ August. . September. . October. .November.. December. . January. . February. March April. . May . June.

1002., 1000.

Chart VI.

Another point of importance in regard to these two charts is the
following: — A comparison of the epizootic during i902-'o3 with that
during i903-'o4 shows us that the former was much more severe.
Secondly, the amount of rat plague at the end of 1902 was in excess ;

(233)

20 QUATERCENTENARY STUDIES IN PATHOLOGY

and, thirdly, there is the fact that the epidemic of 1902 was a mild one ;
that of 1903 was severe, and during 1904 the epidemic was again slight.
Given these data, coupled with a careful scrutiny of the two charts, it
would appear that a malignant epizootic occurs towards the end of an
epidemic, and its continuance after the disappearance of the latter means
an early recurrence of human plague and a severe epidemic. On the
other hand, a mild epizootic during the interval between two epidemics
would appear to indicate a late recurrence of human plague and a mild
epidemic. The condition of affairs may be stated as follows : —

1902. — Epidemic mild. '
1902- 1903. — Epizootic severe. The epizootic

1903. — Epidemic severe. - is

1903-1904. — Epizootic mild. continuous.

1904. — Epidemic mild.

It has now been clearly shown that in an endemic plague centre like
Hong-Kong, rat plague is present, in varying amount, throughout the
whole year. It becomes epizootic at certain times of the year, and the
reasons for this would not appear, at first sight, to be obvious.

In order to explain this regular appearance of the epizootic, I believe
that a consideration of the question of the variability in virulence of the
B. pestis, and the period when rats are most prolific, will simplify matters
greatly. The B. pestis is an organism which is liable to sudden altera-
tions in its virulence. Two races of the bacillus may be cultivated under
the same conditions for months. One may retain its virulence. The
other may lose its virulence, either partially or completely. This loss of
virulence may take place suddenly. It is stated that the plague bacillus
may remain virulent for a considerable length of time at the temperature
of 95°- 100° F.

Further, the continued passage of a strain of the B. pestis through
rats, leads eventually to the production of practically avirulent stocks of
the organism.

Such cultures are found by experiment to produce in rats and other
animals a condition of chronic and marasmic plague.

Further, chronic rat plague is predominant during the intervals
between epidemics of plague, and from such cases an avirulent strain of
the B. pestis may be obtained.

(234)

THE RAT THEORY OF PLAGUE EPIDEMICS

21

My results show that : —

(i). Acute rat plague is followed by epidemic plague.
(2). Chronic rat plague would appear to bridge over epidemic
plague.
An explanation of the occurrence of chronic rat plague, as a sequel
to an epizootic of acute rat plague, is by no means easy. From what

II O N (t K C) n g

1 003-04

n

a

m KJ

n

ji

J

^^

i..i

■f'

.S 1.5 22;'y 5 121:1 2(1 2 i!) 16 23 30 7 14 21 28 t II
July. August. Scptcnihar. Octobur. Noviiiil

1 o o .3 .

.1 2 9 lfi23.'?0 7 14 21 28 4 11 IS 2,5 3 10 17 24 31 7 14 !
Deccmher. Jivnunry. February. . March. , April.

1 004.

1 28 A 1210 26
. May.

Chart VII.

has already been said, the continued passage of the B. pestis through the
body of the rat tends to render the organism less virulent. Therefore, it
is probable that a strain of the bacillus which excites acute rat plague
during the early passages, will only cause chronic rat plague during the

(235)

22 QUATERCENTENARY STUDIES IN PATHOLOGY

later passages. In the plague bacillus we are dealing with an organism
of the group of hsemorrhagic septicaemiai. Such bacteria, when intro-
duced into the bodies of certain animals, are known to lose their
virulence progressively. They give rise to outbreaks of spontaneous
disease in animals, and, on recrudescence of such epizootics, the
virulence of the organism becomes increased. During the natural
evolution of the epizootic, we find that the microbe becomes
attenuated, and either finally disappears, or, as in plague, persists
in chronic form. The recrudescence of rat plague in acute form would
appear to depend upon the natural infection of fresh generations of rats,
the bodies of which are highly susceptible to plague infection. My
observations and knowledge of the habits of rats lead me to the
conclusion that the obstinate clinging of plague to any particular area,
and its periodic recrudescence, is occasioned through chronic rat plague,
and further, the recrudescence of the epizootic occurs at certain seasons
of the year, namely, during the period of maximal rat prolificacy.

(236)

SOME EXPERIMENTS WITH DISINFECTANTS.
By Andrew Ross Laing, M.D., D.P.H.,

Assistant in the Bacteriological Department, University of Aberdeen,

(237)

Some Experiments witli Disinfectants.

The following investigation was undertaken at the suggestion of
Professor Hay, in order to obtain, at first hand, reliable information as to
the efficiency of certain disinfectants, especially formaldehyde or
formaline, for the guidance of the Public Health Department of the city.
A trustworthy means of gaseous disinfection has been long sought
for by medical officers of health, and it is now the accepted
opinion that formaldehyde is the best. The disinfectant properties
of formaldehyde have been investigated by numerous experi-
menters, but their results have been by no means uniform. While all, or
nearly all, agree as to its undoubted value as a surface disinfector, there
is a wide variance of opinion as to its penetrating power. It is admitted
by all that it is virtually non-poisonous, and does not injure textile
fabrics. Kenyon states that he subjected over 225 samples of different
textile fabrics, hair, fur, leather, etc., to crucial tests, using solutions of
various strengths and a saturated atmosphere of the gas. The results
obtained were in every way satisfactory.

Formaldehyde exists in at least three well-recognised isomeric
states : —

1. Formaldehyde (formic aldehyde), a gas at ordinary temperature,

colourless and possessing slight odour, but having an extremely
irritating effect upon the mucous membranes of the nose and
conjunctiva. At a temperature of about 20° C. the gas
polymerizes into paraformaldehyde, known commercially as
paraform.

2. Paraform, a white substance, unctuous to the touch, soluble in both

water and alcohol. It consists chemically of two molecules of
formaldehyde. It is this substance which is supposed to
(239)

4 QUATERCENTENARY STUDIES IN PATHOLOGY

compose the commercial solutions of formaldehyde, known as

formaline, formol, etc.
3. Trioxymethylene, formed by the union of three molecules of

formaldehyde. It is a white powder, giving off a strong odour

of the gas. It is but slightly soluble in alcohol and water.
Formaldehyde gas possesses about the same specific gravity
as air, which renders it only slightly diffusible, although more so
than sulphur dioxide, in consequence of which it penetrates more
readily to all portions and corners of a room. Formaldehyde
combines with nitrogenous organic matter to form new com-
pounds, A few drops of formaldehyde added to white of egg
will prevent its coagulating by heat, and it is from this faculty of
combining directly with albuminoids forming the protoplasm of the
micro-organisms that the gas is supposed to derive its powers as a
germicide. Formaldehyde also readily unites with the nitrogenous
products of decay, fermentation, and decomposition, forming new
chemical compounds, which are both odourless and sterile. The
commercial solutions known as formaline, formol, etc., are said to contain
40 per cent, of formaldehyde gas. They are not always up to this
standard, and being volatile, there is a certain loss if they are not
properly kept. In winter there is a decided deterioration owing to the
polymerization and precipitation of the insoluble trioxymethylene, the
gas probably existing in solution in one of its polymeric states.
Formaline is probably dissolved paraformaldehyde. This is one of the
reasons why the simple heating or evaporating of the solution does not
always result in driving off the gas, but sometimes results merely in
dissipating the water, leaving a residue of solid paraform. Trillat has
shown, however, that this polymerization is prevented if the solution is
heated under pressure.

A.— GASEOUS DISINFECTION WITH FORMALDEHYDE.

Temperature, as with disinfectants in general, is an important factor
in disinfecting with formaldehyde. Disinfection with this gas should not
be attempted if the temperature is under 10° C, as the gas tends to
condense to paraform at lower temperatures. In cold weather it is

(240)

SOME EXPERIMENTS WITH DISINFECTANTS 5

recommended that the room to be disinfected should be heated by
artificial means, high degrees of heat aiding the disinfectant powers of
the gas. A certain amount of moisture has also been found to be
essential to successful gaseous disinfection, and the maximum dis-
infecting power is got only if the atmosphere be saturated with moisture.
H. W. Hill states that an amount of gas which failed to kill in 6 hours at
42 per cent, humidity, killed in 20 to 40 minutes at 100 per cent.
Another important requirement is that the gas be evolved rapidly, so as
to obtain a sufficient concentration of the gas in the room at one time.
Rideal, in his experiments with Thursfield's lamp, found that a larger
amount of the disinfectant rapidly evolved is of greater value than
prolonged exposure to a smaller quantity of the gas.

Many methods have been given as the most reliable for fumigation
with formaldehyde gas. For this investigation three methods were
chosen, viz. : —

1. Lingner's apparatus, as representing the method of gaseous

disinfection with the atmosphere saturated with moisture.

2. The " Alformant" Lamp, where the method is dependent on the

natural humidity of the atmosphere at the time of the
experiment.

3. The Permanganate Process, as recommended by the Sanitary

Authorities of the State of Maine, U.S.A.

1. — Lingner's Disinfecting Apparatus.

This apparatus, which is made in Germany, consists of a ring boiler
in which steam is generated and driven into a reservoir filled with
formaline or glyco-formol (30 per cent, formaldehyde with 10 per cent,
glycerine). By the pressure of the steam the formaline is ejected in the
form of a fine and very abundant spray through four separate nozzles
out of the reservoir. Two litres of glyco-formol are recommended as
sufficient for a room of 2800 cubic feet capacity. The addition of glycerine
is supposed (i) to hinder polymerization, (2) to hinder evaporation of the
formaline after deposition in the room, (3) to thicken the solution and
favour its adhesion to a smooth surface and its absorption by pores, in
consequence of which its effect is believed to be enhanced. The

(241) R

6 QUATERCENTENARY STUDIES IN PATHOLOGY

addition of glycerine has since been found to be unnecessary, the
employment of the 40 per cent, formaldehyde solution being quite as
efficacious.

Koch, in 1898, from a series of comparative experiments with different
methods of gaseous disinfection, concluded that the Lingner method was
superior to the other methods of gaseous disinfection. He found that,
after ij hours' exposure, Bacillus anthracis and Staphylococcus pyogenes
were killed ; that faeces exposed on a shelf was sterilised, but that faeces
in the pocket of a coat gave positive results. These satisfactory results,
coming, as they did, when plague was prevalent, led many sanitary
authorities on the continent, especially in Russia, to adopt it as a means
of disinfecting plague-infected houses.

In this country, Lingner's apparatus is not yet well known, and rela-
tively few scientific tests have been made of its efficiency. Accordingly,
attracted by the continental reports of its value, Professor Hay, four years
ago, procured the apparatus for use in the work of the Public Health
Department of the city, and the immediate incitement to the present
inquiry was his desire to have it tested under conditions that would
reproduce, as nearly as possible, those under which the work of
disinfection of a room and its contents has to be carried out in
Aberdeen. The following series of experiments with the apparatus
were designed to ascertain not only (i) its power as a disinfector, and (2)
its powers of penetration, but also (3) to test how far these good results
were obtainable where no special precautions were taken to seal up
every aperture in a room, further than roughly stopping up the chimney,
as often happens in ordinary fumigation.

Experiment i.

The first experiment was carried out in an old working-class
dwelling, forming part of a three-storey tenement in the heart of the
city, and consisting of one principal room, with two small rooms or
sleeping closets opening off it. The large room measured 18 ft. by 11 ft.
by 8 ft., and had an open fireplace and two windows, both windows
being loose in their frames. Two panes of glass had been broken, and
were roughly covered wdth paper tacked to the frames. The entrance
door of the room fitted very loosely, there being a considerable space

(242)

SOME EXPERIMENTS WITH DISINFECTANTS 7

between the foot of the door and the floor. The two smaller rooms
measured 9 ft. by 7 ft. by 8 ft, and 8 ft. by 7 ft. by 8 ft. respectively, and
each had one small window but no fireplace. The total cubic space of
the main room and the two closets amounted to 2536 cubic feet.

The object aimed at was the disinfection of the whole dwelling, the
apparatus being placed in the middle of the floor of the main room,
while the doors leading to the small rooms were left open. During the
experiment the four windows and the door were closed in the usual way,
without sealing, but the chimney was roughly stopped with paper.

Technique. — Small blocks of unvarnished wood, each 4 in. long by
I J in. broad, and having a small oblong area marked on the surface, were
sterilized in an autoclave at a temperature of 120° C. for one hour.

The following micro-organisms were used in the experiment : —

1. an actively sporing culture of Bacillus anthracis.

2. 24 hours' culture on agar of Staphylococcus pyogenes aureus.

3. 24 hours' culture on agar of Bacillus coli communis.

4. 24 hours' culture on agar of Bacillus typhosus.

5. 24 hours' culture on serum of Bacillus diphtheriae.

Emulsions of these different micro-organisms were made with
bouillon in sterile test glasses, and spread, by means of sterile brushes,
on the marked areas of the wooden blocks, and allowed to dry in air.
They were then distributed in the following places : —

1. floor of the main room.

2. the mantelshelf above fireplace of main room.

3. ledge above window of main room.

4. shelf, 7 feet from floor, directly opposite the apparatus.

5. floor of one of the closets, in a corner remote from the door.

In this experiment the blocks were protected from direct deposition
of the disinfectant by means of several loosely placed layers of brown
paper, and were exposed to the action of the vapours from the apparatus
for 3 hours. The marked areas on the wooden blocks were then scraped
by means of sterile chips of glass, and the scrapings were placed in
tubes of bouillon, and allowed to incubate for 7 days at a temperature of
37° C. Control blocks made at the time of doing the experiment were
treated in a similar way, except that they were not exposed to the
vapours.

(243) R I

8 QUATERCENTENARY STUDIES IN PATHOLOGY

Within 24 hours a copious growth had taken place in the tubes
prepared from the control blocks. On the other hand, all the tubes
containing the material which had been exposed to the vapour from
Lingner's apparatus remained sterile even after 7 days.

. Experiment 2.

The second experiment was carried out in a small ward of the
City Hospital, Aberdeen. It had 3 close-fitting two-sashed windows
and a stove, the outlet of which was closed by stuffing with paper.
The only other precaution taken was to suspend a wet sheet outside the
door to assist in preventing any fumes reaching the patients in the
adjoining wards. The size of the room was 16 ft. by 17 ft. by 10 ft.,
giving a cubic capacity of 2,720 cubic feet, or just 80 feet under the
maximum allowed by Lingner for the adequate action of the apparatus.
The apparatus was placed in the middle of the floor.

Technique. — Small pieces of glass, i in. long by \ in. broad, were
sterilized in the autoclave. The following micro-organisms were used : —

1. actively sporing culture of Bacillus anthracis.

2. 24 hours' agar culture of Staphylococcus pyogenes aureus.

3. 24 hours' agar culture of Bacillus coli communis.

4. 24 hours' agar culture of Bacillus typhosus.

5. sputum containing the tubercle bacillus in large numbers.

The cultures were made into emulsion with bouillon, and spread
on the sterile glass slips. The sputum was spread on an uncovered
Petri's capsule. All were allowed to dry in air. They were then
exposed in the following manner for three hours to the action of the
vapours from the Lingner apparatus : —

1. One set of the four cultivations and one specimen of the sputum

were placed on the floor, and protected from direct deposition
of the vapour by several loose layers of paper.

2. One set of the cultivations alone were set on the mantelshelf above

the stove.

3. One set of cultivations and one of sputum were placed between

two blankets — that is, having the thickness of two blankets
above and two below.

(244)

SOME EXPERIMENTS WITH DISINFECTANTS 9

4. One set of cultivations and one of sputunn were placed in the
centre of a thick mattress, which had been opened up for
the purpose and re-closed.

Control experiments were at the same time carried out.

After 3 hours' exposure, the slips of glass were transferred direct into
bouillon tubes, without any intermediary process, and incubated at
37° C. Within 24 hours, growth occurred in all the control tubes, as
also in all the tubes from the 4th set ; while the tubes from the ist, 2nd
and 3rd sets remained sterile after 10 days' incubation. The sputum,
from exposures Nos. i and 3, were each mixed up with a small quantity
of bouillon, and inoculated into guinea-pigs — the sputum from No. 4
being rejected owing to the negative results from the cultivations
similarly exposed. After an interval often weeks, both guinea-pigs were
killed, and in neither case was there the faintest trace of a tubercular
lesion.

Conclusions. — From these experiments, conducted as they were under
conditions which severely tested the powers of the apparatus, it will
be seen that Lingner's disinfecting apparatus gives trustworthy results in
the disinfection of dwelling-rooms, and, as shown by the result of sets 3
and 4, the penetrating power of the disengaged vapour, although not
sufificient for mattresses or other thick goods, is apparently quite effective
for lighter articles of bedding and clothing.

In the working of the apparatus, I found it advantageous, in order to
obtain an immediate and full effect, to begin with very hot water in
the boiler. It is also essential to see that all the four jets are working
freely.

2.— The "Alformant" Lamp.

The Alformant Lamp had also been employed for some years in the
Public Health Department of the city, its use being confined to certain
cases where its convenience was an inducement. It was, therefore,
suggested by Professor Hay that I should also test its efficiency, the
more so as its makers ascribed to it a high disinfecting power, and the
lamp was widely advertised and used.

This method of evolving formaldehyde gas consists in the heating of
paraform, one of the polymeric forms of formaldehyde, which first melts
and then breaks up into the two molecules of formaldhyde gas. The

(245)

to QUATERCENTENARY STUDIES IN PATHOLOGY

lamp is cheap and light, and of very simple construction, consisting of a
receptacle or basin for holding the paraform, or, as they are commonly
called, the "formaline tablets," with a small methylated spirit lamp
beneath. The disadvantages of the method are that the gas is given off
without moisture, and tends to polymerize readily, especially on cool, dry
days. The gas is also given off rather slowly, and with little force, so
that it permeates slowly to all the crevices and corners of a room. The
room to be disinfected must have all cracks and crevices carefully
sealed up.

Dr. Kenwood, in a paper read at Leeds, in 1897, stated that he had
succeeded, by means of the lamp, in sterilizing swabs infected with B.
diphtheriae, using 21 tabloids in a room of 2,004 cubic feet, with a four
hours' exposure. More recent experiments, under varied conditions,
seem to have given very mixed results, and Kenwood now advocates
using 25 tablets per 1,000 cubic feet, if disinfection is to be ensured.
Allan, in 1898, in a series of comparative experiments with sulphur
dioxide and formaldehyde, conducted in two ordinary dwelling-rooms,
used 10 tablets for each 1,000 cubic feet, as recommended by the makers,
and got satisfactory results. He found (i) that it was not so much a
question of time as of initial volume of gas liberated, and suggested
using a large number of lamps rather than a large number of tablets in
one lamp ; (2) that the gas had no great power of penetration ; and (3)
that the generating apparatus should be as near to the floor as possible.

The following experiment was conducted with the view of comparing
the Alformant lamp method with the Lingner method. In order that
the comparison might be as fair as possible, the house used in experi-
ment I was employed, and the same conditions were adhered to, viz.,
merely stopping up the chimney. In order that the initial volume of the
gas should be as large as possible, I used two lamps, with 30 tablets in
each, giving approximately 24 tablets per 1,000 cubic feet. The relative
humidity of the air before starting was 86 per cent. ; the inside
temperature was 56° F. The organisms used in this experiment were (i)
actively sporing Bacillus anthracis, (2) 24 hours' culture on agar of
Staphylococcus pyogenes aureus, (3) Bacillus typhosus, and (4) Bacillus
coli communis.

Technique. — Small pieces of glass, i in. long by J in. broad, were
sterilized in the hot air chamber. Emulsions of the different cultures

(246)

SOME EXPERIMENTS WITH DISINFECTANTS II

were made in sterile test glasses, and thereafter spread on the slips of
glass with a sterile ose, and allowed to dry in air. Control slips were
made, and also allowed to dry in air, and then put inside a sterile Petri
dish, and kept for the same time as those exposed to the disinfector.
The other slips were distributed in the house as in experiment i, viz.,
(i) on floor behind the main door, (2) upon the shelf, 7 feet from the
floor, near to the lamp, (3) upon the mantelshelf above fireplace, (4) upon
the ledge above the window, and (5) on the floor of one of the smaller
rooms, in a corner remote from the door. They were then exposed to
the action of the Alformant lamps for 3J hours. The room, on being
opened at the end of the time, had a powerful odour of formaldehyde
gas. The different slips were then placed direct into bouillon tubes and
incubated for 7 days at 37'' C, the control slips being at the same time
placed in bouillon tubes. Within 24 hours, a copious growth was present
in all the control tubes, x^t the end of 7 days, the following results were
observed in the tubes containing the exposed slips : —

In the 1st set (floor of main room) growths were observed in the
tubes inoculated with Anthrax, Staphylococcus, and B. coli, while the
B. typhosus tube remained sterile.

In the 2nd set (shelf near to lamps) negative results were obtained in
all the tubes.

In the 3rd set (mantelshelf above fireplace) growths were obtained in
the Anthrax and Staphylococcus tubes, while both the B. coli and
typhosus tubes remained sterile.

In the 4th set (above window) growths were obtained in all the
four tubes.

In the 5th set (floor of smaller room) growths were obtained in all the
four tubes.

Conclusion. — This experiment shows that, where the gas is present in
sufficient quantity, as in the 2nd set of organisms, disinfection will be
obtained ; but it cannot generally be relied upon in ordinary conditions,
even where the proportion of paraform used, as in the foregoing experi-
ment, is considerably above that recommended by the makers of the
lamp. It also shows that B. typhosus is the most susceptible of the
exposed germs to the action of the gas, three out of the five samples of
the cultures of this bacillus being sterilized.

(247)

12 qUatercentenary studies in pathology

Of course, these indifferent results were obtained in the case of a
room which was by no means hermetically sealed, but it was in the
same condition as when the corresponding test was made with Lingner's
apparatus ; and it is well not to rely on any very careful sealing of
rooms in the large number of fumigations which have to be carried
through by sanitary officers in times of heavy epidemics. The better
results obtained by some other observers were no doubt largely due to
the more careful sealing of the rooms.

3. — The Permanganate Process.

This method is based on the fact that, by the mixture of formaline
and potassium permanganate, a copious liberation of gaseous formal-
dehyde takes place.

The State Board of Maine recommends the following method : — In
the centre of the room is placed a large vessel containing the requisite
amount of potassium permanganate. The proper quantity of formaline
is then added. As much foaming occurs, it is well to place the vessel
within a large pan.

For each looo cubic feet, 13 ounces of potassium permanganate and
2 pints of formaline are used. It may be mentioned that the New
Hampshire Board, who have also employed this method, considers that
smaller quantities of the substances (3J ounces of potassium perman-
ganate and I pint of formaline to the 1000 cubic feet) serve the purpose
equally well.

Experiments were carried out by the Maine authorities on the
following micro-organisms : — B. diphtherias, B. typhosus, B. coli, B.
pyocyaneus, B. anthracis, B. subtilis. Staphylococcus albus, S. aureus,
Streptococcus pyogenes, &c. After an exposure of 3 hours to the action
of the disinfectant, out of 1529 test objects, only 27 showed any growth,
and of these, 21 were Bacillus subtilis.

In these experiments the ordinary openings of the room were closed,
but in no case were precautions taken to close up cracks and crevices.

The following is a description of my own experiments with these
methods : —

Experiment i.

This experiment was carried out in the same house as that used in
experiments with the two last-mentioned methods, and the same

(248)

Some experiments with disinfectants 13

conditions were adhered to. It was conducted in all respects according
to the recommendations of the New Hampshire Board, 2| pints of
formaline and 8J ounces potassium permanganate being employed. The
temperature of the air was 51-5° R, and the relative humidity was 83 %.

The organisms used in this experiment were : —

1. actively sporing B. anthracis.

2. 24 hours' culture on agar of Staphylococcus pyogenes aureus.

3. B. typhosus.

4. B. coli communis.

Technique. — The technique employed was similar to that of the
previous experiments, that is, the exposure of glass slips coated with
dried emulsions of the various organisms. Control experiments were
made at the same time with similar slips kept in a sterile Petri dish in
the laboratory.

The other slips were distributed in the house, as in experiment i,
viz. : —

1. on floor, behind main door.

2. upon the shelf, 7 feet from the floor, near the vessel.

3. upon mantelshelf, above fireplace.

4. upon ledge, above window.

5. on floor of one of the smaller rooms, in a corner remote from the

door.
They were exposed to the action of the disinfectant for 3 hours.
The exposed slips, and the control slips, were then placed direct into
bouillon tubes, and incubated at 37° C. Within 24 hours growth was
present in all the control tubes. At the end of 7 days, the following
results were observed in the tubes containing the exposed slips : —

In the first set (floor of main room) growths were observed in the
Anthrax, Staphylococcus, and B. typhosus tubes, while the B. coli tube
remained sterile.

In the 2nd set (shelf near to vessel) growth was observed in the
Anthrax tube, while the other tubes were sterile.

In the 3rd set (mantelshelf above fireplace) growths were observed in
the Anthrax and B. typhosus tubes, while the Staphylococcus and B.
coli tubes remained negative.

(249)

14 QUATERCENTENARY STUDIES IN PATHOLOGY

In the 4th set (above window) growths were observed in the Anthrax
and B. typhosus tubes, while the Staphylococcus and B. coli remained
sterile.

In the 5th set (floor of closet) growths were obtained in all the four

tubes.

Experiment 2.

The second experiment was carried out in a small ward of the
City Hospital, Aberdeen. It had 3 close-fitting two-sashed windows
and an open fireplace, the chimney of which was loosely packed with
sheets of newspapers. The temperature of the room was 54° F.
The capacity of the room was 2,890 cubic feet. It may be noted that
in this experiment the proportion of formaline was doubled, 5 pints
of formaline and 16J ounces of potassium permanganate being used.

Technique. — This was in all respects similar to that of the last
experiment, he micro-organisms used being : —

(i) Sporing Anthrax; (2) Staphylococcus pyogenes aureus, (3) B.
coli, and (4) B. typhosus. Three sets of these organisms were made
for exposure to the action of the disinfectant, and were distributed
as follows, viz. : —

1st set, on floor of room behind door.

2nd set, on mantelshelf above fireplace.

3rd set, on sash of one of the windows.

They were exposed to the action of the disinfectant for 3 hours.
The room, on being opened at the end of the time, had a fairly powerful
odour of formaldehyde gas, although not quite so marked as in
experiment i. The different slips were then placed direct into bouillon
tubes, and incubated for 7 days at 37° C. Control slips, made at
the same time as those used in the experiment, were likewise placed
in bouillon tubes and incubated. Within 24 hours, a copious growth was
present in all the control tubes. At the end of 7 days, the following
results were observed in the tubes containing the exposed slips : —

In the 1st set (floor of room) growths were observed in the Anthrax
and Staphylococcus tubes, while the B. coli and B. typhosus tubes
remained sterile.

In the 2nd set (mantelshelf over fireplace), growth was present in the
Anthrax tube, while the Staphylococcus, B. coli, and B. typhosus tubes
remained sterile.

(250)

SOMH: EXPERIMEI^TS WITH DISINFECTANTS 1 5

In the 3rd set (sash of window) growth was present in the Anthrax
and Staphylococcus tubes, while the B. coli and B. typhosus tubes
remained sterile.

Conclusions. — These experiments show that this method of disinfection
is quite as efficient as the Alformant lamp, but the results are not
so conclusive as those got by Lingner's apparatus. It has distinct
advantages in that there is no danger from accidents due to fire, and that
it requires no special form of apparatus.

On the basis of my experiments, the Lingner apparatus is, for
ordinary disinfecting work, unquestionably superior to the Alformant
lamp and the Permanganate process.

It ought to be added that, although the Alformant lamp is cheap,
costing only a few shillings, the paraform tablets are much more
expensive than a corresponding amount of formaldehyde in solution.

B.— LIQUID DISINFECTION BY FORMALINE AND OTHER

DISINFECTANTS.

Formaline is a 40 per cent, solution of formaldehyde in water, and, as
met with in commerce, is a liquid of a pale, sea-green tint, with a specific
gravity of about 1,070, and an acid reaction which is due to formic
and acetic acids. In more concentrated solutions, the formaldehyde
tends to pass into the polymeric and comparatively inactive form. This,
therefore, is the strongest solution which will remain permanent. When
exposed to the air the formaldehyde vapour is gradually dissipated
and the solution loses strength. Loew and Fischer, in 1886, were
the first to state that formaline possessed powerful antiseptic properties.
Trillat showed that the presence of a minute quantity of this substance
in urine preserved it from putrefying. Buchner, Aronson, Blum, Cohn,
and others, have investigated the disinfectant properties of formaldehyde
in solution. All are agreed in attributing to it powerful antiseptic
and deodorant properties. Blum pointed out that to kill micro-organisms
it requires a somewhat strong solution. Slater and Rideal found that a
I per cent, solution of formaldehyde is sufficient to kill Staphylococcus
pyogenes aureus in from 50 to 60 minutes, B. typhosus in from 40

(251)

l6 QtJATERCENTENARY STUDIES IN l»ATllOLOGY

to 50 minutes, B. coli in from 30 to 40 minutes, B. anthracis and Vibrio
cholerai in less than 15 minutes; while solutions, varying in strength
from I in 5,000 to i in 20,000, were sufficient to inhibit the growth
of most organisms. Tr^trop considered a 2 per cent, solution of
formaline to be a most efficient antiseptic, and stated that he obtained
positive results with it after other preparations had failed. Bird speaks
highly of formaline as a disinfectant and antiseptic, but finds that
the vapour is more powerful as a disinfectant than the solution.
Schepilewsky confirmed the experiments of Bird. Kenwood suggests
the use of a 2*5 per cent, solution for spraying. Leslie Mackenzie
and Alexander speak of the efficacy of a -5 per cent, solution for
general disinfecting purposes. Park and Guerard find that a 3 per cent,
solution will kill Anthrax spores in 15 minutes, while a i per cent,
solution will kill all other germs in one hour, and most germs within
30 minutes.

Hill and Abram, in their experiments on the disinfection of excreta,
found that faeces when mixed with a i in 20 solution of formaline,
remained sterile on subsequent inoculation. Houston found that formaline
in a dilution of i in 100 did not prevent the growth of B. coli after one
hour's contact, while in a dilution of i in 20 it killed that organism.
Thresh and Sowden have made a series of experiments to ascertain the
strength of solution necessary to ensure sterilisation. Cultures of B.
diphtheriae, B. typhosus, B. pyocyaneus, V. cholerae, M. prodigiosus,
and Staph, pyogenes aureus were spread on wood, on white-washed
surfaces, and on wall paper. The infected surfaces were sprayed with
solutions of formaline varying in strength from '5 to 2 per cent. After
three to four hours' exposure, the sterility of the sprayed surfaces was
tested by means of sub-cultivations, with the following results : —
The '5 solution killed all the organisms on wood and wall paper, but
failed to destroy the B. pyocyaneus on white-washed surfaces. The
I per cent, solution gave similar reactions, but the 2 per cent solution
sterilised all surfaces.

Muir and Ritchie found that to disinfect an organic mixture
containing pyogenic organisms, a 10 per cent, solution, acting for
half-an-hour, is necessary, and that in the case of pure cultures, a
5 per cent, solution would kill the V. choleras in 3 minutes, B.
anthracis in 15 minutes, and anthrax spores in 5 hours. When such

(252)

SOME EXPERIMENTS WITH DISINFECTANTS

17

organisms infect clothing, an exposure to a 40 per cent, solution of
formaline for 2 hours, and in case of anthrax spores, an exposure for
24 hours, is necessary. Silk threads impregnated with B. pestis were
found to be sterile after 2 minutes' exposure. Nils Englund recommended
spraying rooms with a 2 per cent, solution, afterwards closing the room for
24 hours. Walter found that a solution of i to 10,000 arrested the
growth of B. anthracis, V. cholerae, B. typhosus. Staphylococcus
pyogenes, and B. diphtheriae, and that slightly stronger solutions
sufficed to destroy these organisms. Faeces were rendered sterile by
a 10 per cent, solution in 10 minutes.

The lack of uniformity in the various methods of testing the
germicidal power of antiseptics and disinfectants has long suggested
the necessity for the introduction of some standard method. It must, of
course, be conceded that the subject does not lend itself to the exact
treatment of chemical analysis. With a testing material consisting of
living germs always liable to variations of vitality and degrees of
resistance, absolute results are impossible, but, nevertheless, a degree of
accuracy may be attained sufficient to render the results of great
comparative value. It is evident at any rate that isolated observations
upon single antiseptics and disinfectants are of little value. It is only by
comparison of the germicidal action of the various disinfectants under
similar conditions that one can arrive at correct conclusions. With this
object in view, Ainslie Walker has proposed a means whereby a properly
systematized method of estimating the bactericidal value of disinfectants
may be established. He suggests that some well-known disinfectant
be selected as a standard — one known to give regular and consistent
results — such as pure phenol (carbolic acid). Briefly summarised, the
technique of his method is as follows : — To 5 c.c. of a 24 hours' blood-
heat culture in broth of the organism add 5 c.c. of the dilute disinfectant.
Shake, and take sub-cultures at definite intervals in suitable media.
Incubate for at least two days at 37° C. If an agar culture be preferred,
take up part of the growth on the point of a platinum needle, and
distribute it evenly in sterilized water ; the resulting emulsion may be
used in place of the broth culture, the rest of the procedure in both cases
being identical. A similar experiment is carried out simultaneously
with a dilute solution of phenol. From the results thus obtained, the
strength or efficiency of the disinfectant is expressed in multiples of the

(253)

i8

QUATERCENTENARY STUDIES IN PATHOLOGY

volume of carbolic acid performing the same work ; or this multiple may
be expressed as the " Carbolic Acid Co-efficient." For example : —

B. COLI COMMUNIS— 24 HOURS' BROTH CULTURE.

Disinfectant.

Dilution.

2i

Thin Cultures exposed to
Disinfectant (in minutes).

5 7h 10 I2i

15

Period of
Incubation.

Tempera-
ture.

X.

1-1,000

+

+ — — —

—

48 hours.

37 c.

Phenol.

I-IOO

+

= Growth; — = No Growth.

"

48 hours.

37 c.

The multiple here is 10 — i.e., it will require 10 volumes of carbolic
acid to perform the work done by one volume of the disinfectant (X)
under observation.

The following experiments were undertaken to test and compare the
relative actions of Formaline, Phenol, Corrosive Sublimate, Cyllin, Izal,
Sanitas, and Sapo-Cresol.

Phenol. — Carbolic acid, once the most widely used of disinfectants,
has been disparaged in recent years because laboratory tests have
demonstrated that it cannot be depended upon to kill spores. A i or 2
per cent, solution has no certain effect upon anthrax spores. A 3 per
cent, solution requires 7 days, a 4 per cent, solution, 3 days, and a 5 per
cent, solution 2 days, to kill anthrax spores. Weaker solutions are
effective for non-sporing bacteria. Behring found that \ per cent,
solution killed the germs of Cholera, Plague, Typhoid, Diphtheria, and
Erysipelas in one hour, while a i to ij per cent, solution would destroy
these organisms in one minute.

Corrosive Sublimate. — The high disinfectant power of perchloride of
mercury is well known. It is largely used in surgery, but it is little used,
at any rate in this country, for ordinary disinfection. One important
objection to its use being its great toxicity. A solution of i in
1000 is ample for the destruction of non-sporing bacteria in half-an-hour.
Solutions of 1-500 and 1-800 will kill non-sporing bacteria in a short
time. For spores, 1-500 solution, with an exposure for one hour, is
necessary. Andrewes has pointed out that the resistance of the
Staphylococcus pyogenes aureus against mercurial salts is altogether
exceptional, but that this resistance does not extend to antiseptics of

(254)

SOME EXPERIMENTS WITH DISINFECTANTS 19

Other groups, such as carbolic acid, &c. He found that suspensions of
the Staphylococcus pyogenes in sterile water were destroyed by a
1-500 solution of corrosive sublimate in 12J minutes, and by a 1-1,000
solution in 50 minutes, while beef-tea cultures of the organism required
for their destruction with similar dilutions of the disinfectant, 45 minutes
and 4 hours respectively. These results illustrate remarkably well a
point upon which Andrewes has laid great stress — the fact, namely, that
the presence of organic matter exercises a distinctly retarding influence
on the germicidal action of corrosive sublimate.

Cyllin is the new name for creolin, as manufactured by Jeyes'
Disinfectant Co. Its composition, however, is not quite identical with
the fluid sold until recently as "creolin," but has been modified with a view
to obtain the highest germicidal value with the lowest possible toxicity
to higher animals. It is stated to contain no carbolic acid or its
homologues, and to rely solely for its high efficiency on certain members
of another chemical series which have been recently isolated. It forms a
slightly pinkish emulsion when mixed with water.

Klein, in a series of experiments, completed after my own investiga-
tions had been largely carried out, compared the action of cyllin,
formaline, and carbolic acid on B. pestis, and found that a solution of
1-30 formaline did not kill B. pestis. On the other hand, a solution of
1-80 carbolic acid killed B. pestis after 10 minutes' exposure, and a solution
of 1-2400 cyllin killed B. pestis after 10 minutes'. Klein concludes
from these experiments, that the germicidal power of a 1-2,400 solution
of cyllin is equal to that of a 1-80 solution of phenol, and greater than a
1-30 solution of formaline. Therefore, the disinfecting power of cyllin is
about 30 times as great as that of phenol, and more than 80 times that
of formaline. These results are certainly very striking. It ought, how-
ever, to be mentioned that the experiments were apparently undertaken
at the instance of the makers of cyllin. How far my own experiments
agree with these will be seen later. They were undertaken without any
communication with the makers, and the cyllin was purchased in the
ordinary way.

Izal. — A non-poisonous disinfectant obtained as an oil from the by-
products in the making of coke. Izal, as it appears on the market, is a
40% emulsion of this oil. It forms a light fawn-coloured emulsion when
mixed with water.

(255)

20 QUATERCENTENARY STUDIES IN PATHOLOGY

Klein, in a series of experiments, found that solutions of 1-200 killed
all non-sporing bacilli in 5 minutes, and a solution of 1-300 killed the
Streptococcus pyogenes. He found that solutions of i per 1,000
inhibited the growth of all spores and bacilli, except in the cases of B.
prodigiosus and B. typhosus, where a solution of 25 per 1,000 was
required. Fowler obtained the "phenol coefficient" of 5-5 with B.
typhosus and B. prodigiosus in the case of izal, and 8'5 in the case of
cyllin. Kenwood and Hewlett obtained the " phenol coefficient " of 57
with B. coli in the case of izal, and 5 '3 in the case of cyllin.

Sapo-Cresol. — A poisonous disinfectant, with a very pronounced
creosote-like odour. It forms a dirty white emulsion when mixed with
water. So far as I am aware, no bacteriological report has appeared
on this substance, (c.f Wolf Arch, f Hyg., 1894, xx., p. 219,)

Sanitas. — A pleasant odoriferous disinfectant. It contains hydrogen
peroxide, camphor, and thymol. When mixed with water, it forms a
clear fluid.

Thresh, in a recent series of experiments, found that a 20 per cent,
solution of sanitas fluid kills B. typhosus in 5 minutes, and a 40 per
cent, solution kills B. diphtheriae in 20 minutes. Pure sanitas fluid
kills B. diphtheriae in i minute, and B. anthracis in 2 J minutes. A 5-
per cent. " sanitas emulsion " kills the B. diphtheriae in 30 minutes. A
7 J per cent. " emulsion " kills B typhosus in 2 J minutes. A 10 per
cent. " emulsion " kills B. anthracis and B. diphtheriae in 2 J minutes.

The following was the procedure in the first series of the experiments
made by myself (Izal, sanitas, and sapo-cresol were not included in
this series) : —

1. The germs taken were cultures of Staphylococcus pyogenes

aureus, B. typhosus, and sporing anthrax, and they were
grown in bouillon, each tube containing exactly 10 c.c.
bouillon. The first two cultures were of 24 hours' growth, and
the latter of 7 days'.

2. To each of these tubes enough of each disinfectant was added

to give the different strengths of solution of disinfectant
determined upon. The tubes were then thoroughly shaken.

3. At every 2 J minutes, up to 15 minutes, 7 loopfuls of the mixture

were re-inoculated into fresh tubes of bouillon and incubated
at 37° C for 8 days.

(256)

SOME EXPERIMENTS WITH DISINFECTANTS

21

The cultures used had been recently isolated, the Staphylococcus
being isolated from an acute abscess, the B. typhosus from a recently
fatal case of typhoid fever, and the anthrax from a recent case, and in
an actively sporing condition. The formaline was tested just before
commencing the experiment, and was found to contain 32 per cent,
of formaldehyde gas, so that the different dilutions given must be
calculated upon that basis and not the ordinary one of 40 per cent.
The various dilutions of the disinfectants employed were such as are in
common use for disinfectant purposes.

The following results were obtained : —

FORMALINE (Strength 32%).

Organism. ,

Dihition of

Time of Ex

:posure

Period of

Tempera-

)isinfectant.

(i

in minutes).

Incubation.

ture.

Anthrax

24

5

74

10

124

15

spores.

1-30.

+

+

+

+

+

+

8 days.

37° C

1-50.

+

+

+

+

+

+

I-IOO.

+

+

+

+

+

+

B. typhosus.

1-30.

+

+

+

+

—

—

8 days.

37° C

1-50.

+

+

+

+

+

+

.

I-IOO.

+

+

+

+

+

+

Staph, pyog.

aureus.

1-30.

-f

+

+

+

+

+

8 days.

37° C.

1-50.

+

+

+

+

+

+

I-IOO.

+

+

+

+

+

+

( +

rr Growth

>

= No Growth).

Growth was

present in all the 1

tubes with the

exception of B.

typhosus in solution of 1-30

at 12^ and

15 minutes ^

exposure

PHENOL.

Organ:

Dilution

Time of Exposure (in minutes).

ism.

of Disinfectant

t.

2i

5

74

10

124

15

Anthrax spores.

1-20.

+

+

+

+

+

+

1-40.

+

+

+

+

+

+

1-80.

+

+

+

+

+

4-

B. typhosus

.

1-20.

+

+

+

—

—

—

1-40.

+

+

+

+

—

—

1-80.

+

+

+

+

+

—

Staph, pyog

. aureus.

1-20.

+

+

' +

+

+

+

1-40.

+

+

+

+

+

+

I 80.

+

+

+

+

+

-1-

( +

= Growth

; —

-="

No Growth).

Growth was

present in

all the tubes of

antf

irax

; spores and

Staphylococcus pyogenes.

B. typhosus

killed in 10 minutes by i in

20 solution,

in 1 24 minutes

by I in 40

solution,

and in

15 minutes

by I

in 80 solution

of

phenol.

(257)

22

QUATERCENTENARY STUDIES IN PATHOLOGY

CORROSIVE SUBLIMATE.

Organism.

Dilution of
Disinfectant.

Time of Exposure (in

minutes).

2i

5

7h

10

I2i

15

Anthrax spores.

1-1,000.

+

-1-

+

-f

-1-

-1-

1-2,000.

+

-t-

+

+

.+

+

1-5,000.

-t-

+

+

+

+

+

B. typhosus.

1-1,000.

+

—

—

—

—

—

1-2,000.

+

+

+

+

—

—

1-5,000.

+

4-

+

+

+

+

Staph, pyog. aureus.

1-1,000.

+

+

-h

+

+

+

1-2,000.

+

+

+

+

+

-H

1-5,000.

+

+

+

+

+

+

(+=Growth ;

— = No Growth).

Growth was present

in all tubes of anthrax spores and Staphylococcus aureus.

B. typhosus was killed in 5 minutes by

1-1,000,

and in 12^ minutes by

1-2,000.

CYLLIN.

Organism.

Dilution of
Disinfectant.

Time of

Exposure (in

minutes).

2h

5

7h

10

124

15

Anthrax spores.

1-1,000.

+

+

-f-

+

+

+

1-2,000.

+

+

+

+

+

+

1-2,500.

+

+

-1-

+

+

+

B. typhosus.

1-1,000.

+

+

+

+

-t-

+

1-2,000.

+

+

+

+

+

+

1-2,500.

+

+

+

+

-f

-1-

Staph, pyog. aureus.

1-1,000.

-f-

+

-h

+

+

+

1-2,000.

+

+

+

+

+

+

1-2,500.

+

+

+

+

+

+

(-f= Growth;

— = No Growth).

Growth was present in all the tubes,

In my second series of experiments, the procedure adopted by Klein
in his experiments on B. pestis was followed. The same organisms were
used as in my first series of experiments. The dilutions of the

(258)

SOME EXPERIMENTS WITH DISINFECTANTS

23

disinfectants used in this series of experiments were the same as those
employed by Klein in certain of his more crucial experiments. They
were the same for formaline and phenol, as in my first series, but the
dilutions of cyllin and corrosive sublimate were about twice the strength
of those used in the first series. Agar cultures were used instead of
bouillon cultures.

1. Emulsions of the different cultures were made in sterile distilled

water, using one loopful of the culture in each case.

2. 5 c.c. of the different strengths of the disinfectant were put into

sterile tubes.

3. To each tube of disinfectant was added 'i c.c. of the emulsion of

the organism, and the mixture was then shaken thoroughly.

4. At end of intervals of 5, 10, and 15 minutes, 3 loopfuls of the

mixture were smeared over the surface of an agar tube, and
incubated for 8 days at 37° C. The following results were
obtained.

FORMALINE (32 %).

Organism.

Dilution of
Disinfectant.

Time of Exposure
(in minutes).

5

10 15

B. typhosus.

1-20.

+

— —

1-25.

+

+ +

1-30.

+

+ +

Staph, pyog. aureus.

I-I5-

+

— —

1-20.

+

+

1-30.

+

+ +

(+=Growth; — = No Growth).

Growth was present in all tubes, except those of B. typhosus, where the dilutions were
1-20 and 1-25, and the exposures 10 and 15 minutes; and Staph, pyog. aureus,
where the dilutions were 1-15 and 1-20, and the exposures were 10 and 15 minutes.

(259)

S I

24

QUATERCENTENARY STUDIES IN PATHOLOGY

CORROSIVE SUBLIMATE.

Organism.

Dilution of
Disinfectant.

5

Ti

ime

(in

of Exposure
1 minutes).

10

1-500.

+

—

1-1,000.

+

—

1-1,500.

+

-1-

Anthrax.

B. typhosus. 1-500. — — —

1-1,000. — — —

1-1,500. + _ _

Staph, pyog. aureus. 1-500. + -f- —

1-1,000. + + +

1-1,500. + + -f-

( + = Growth ; — = No Growth).
Growth was absent in dilutions of 1-500, after 10 and 15 minutes' exposure in the case of
anthrax spores. With B. typhosus, growth was absent, after 5 minutes' exposure,
in all except in the dilution 1-1,500. With Staphylococcus pyogenes, growth was
present, after 15 minutes' exposure, in all tubes except dilution 1-500.

CYLLIN.

Dilution of
Disinfectant.

Time of Exposure
(in minutes).

5

10

1-500.

+

—

1-600.

+

+

1-700.

+

+

1-500.

—

—

1-600.

+

+

1-700.

-h

+

Organism.

B. typhosus.

Staph, pyog. aureus.

15

+

(4-= Growth; — = No Growth).
Growth was absent in dilutions of 1-500 and 1-600, after 10 and 15 minutes' exposure in
the case of B. typhosus. With Staph, pyog. aureus, growth was absent after
5 minutes' exposure in dilution of 1-500, and absent after 15 minutes' exposure
in dilution of 1-600.

(260)

SOME EXPERIMENTS WITH DISINFECTANTS

25

IZAL.

Organism.
B. typhosus.

Staph, pyog. aureus

Dilution of
Disinfectant.

Time of Exposure
(in minutes).

5

ID

1-700.

—

—

1-800.

-f

+

1-900.

-1-

+

.s.

1-800.

—

-—

1-900.

+

—

1-950.

+

+

( + =

= Growth ; — = No Growth).

Growth was absent in all tubes of 1-700 after 5 minutes' exposure, and in 1-800 after 15
minutes' exposure, in the case of B, typhosus ; in the case of Staphylococcus, it was
absent in 1-800 after 5 minutes' exposure, i in 900 after 10 minutes', and 1-950
after 15 minutes'.

Organism.

B. typhosus.

Staph, pyog. aureus.

SANITAS.

Dilution of
Disinfectant.

Pure.
1-5-

I-IO.

Pure.
1-5-

I-IO.

Time of Exposure
(in minutes).

(+ = Growth

=No Growth).

Growth was present in all the tubes of B. typhosus and Staph, pyog. aureus, except the
tubes in which the Sanitas was applied undiluted.

(261)

15

26 QUATERCENTENARY STUDIES IN PATHOLOGY

SAPO-CRESOL.

Organism.
B. typhosus.

Staph, pyog. aureus.

( + =Growth; — =No Growth).

Growth was absent in dilution of I-750 after 10 minutes' exposure, and in 1-800 after 15
minutes' exposure, in the case of B. typhosus ; while in the Staph, pyog. aureus,
it was absent after 5 minutes' exposure in dilution of 1-400, and, in 1-500 dilution,
after 15 minutes' exposure.

CARBOLIC ACID CO-EFFICIENTS.

Dilution of

Time of Exposure

Disinfectant.

(in

minutes).

5

10

1-750-

+

—

1-800.

4-

+

1-850.

+

+

1-400.

—

—

1-500.

+

+

1-600.

+

+

Disinfectant.

Sanitas.

Formaline.

Cyllin.

Izal.

Sapo-Cresol.

Corrosive Sublimate.

Sporing
B. anthracis.

B.
typhosus.

Staph, pyog.
aureus.

—

•025

•025

—

•25

•18

—

6-0

5*o

—

8-0

9*0

—

9*37

6-25

i6-6

25-0

6*25

The conclusions to be drawn from these two series of experiments on
liquid disinfection — which are summarised in the accompanying table of
co-efficients — appear to be as follows : —

I. Formaline, — In none of the dilutions used — though they are
dilutions recommended for disinfectant purposes — and for the exposure
tested, was formaline found to be completely efficient as a general
disinfectant. The bacillus of typhoid was evidently, however, affected by
it in the stronger dilutions. It was evident that even for non-sporing
micro-organisms the strength of formaline solution should not be less

(262)

SOME EXPERIMENTS WITH DISINFECTANTS 27

than 4 to 5 per cent., and for sporing organisms it ought to be still
stronger. It appeared to have only about a fourth to a half of the
germicidal power of phenol.

2. Cyllin. — In the degrees of dilution with which Klein obtained
satisfactory results, and after the same exposures, cyllin in my hands
has not yielded the same results. At the same time, it is evidently a
disinfectant of great power, being effective against non-sporing organisms
in dilutions of i in 500. My tests would lead me to rank it as about five
to six times as potent as phenol.

3. Izal is another powerful disinfectant, being probably eight to nine
times more active than phenol. Against some organisms, as B. anthracis,
it is less active than corrosive sublimate, but against Staphylococcus
pyog. aureus it is more active.

4. SapO'Cresolxs evidently a very active disinfectant. In my experi-
ments, it was more powerful than izal towards B. typhosus, but less so
towards Staphylococcus ; while, in comparison with cyllin, it was more
powerful against both these organisms.

5. Sanitas showed itself in these experiments to be by much the
least active of the disinfectants tested. Apparently, it can only be
relied upon when applied undiluted, and even then it may fail to kill
sporing organisms.

6. Corrosive Sublimate, in the same dilutions as cyllin, showed itself
to be more powerful than this antiseptic, especially in attacking the B.
typhosus and anthrax spores. In its action on Staphylococcus,
however, it is less effective than cyllin. This relative lack of efficiency
against Staphylococcus is in agreement with the observations of
Andrewes.

(263)

ON ECK'S FISTULA— OBSERVATIONS ON
FOUR DOGS, WITH A REVIEW OF THE
LITERATURE RELATING TO PREVIOUS
WORK ON THIS SUBJECT.

By John James Rickard Macleod, M.B., CM.,

Professor of Physiology^ Western Reserve University^ Cleveland^ Ohio, U.S.A.

(265)

On Eck's fishila.

The great size of the liver in comparison with the other glands of the
body, and the fact that all the blood from the gastro-intestinal tract
must perfuse the liver before being poured into the systemic circulation,
indicate that this organ must be of primary importance in the animal
economy. The absolute exclusion of the liver from the circulation is
impossible unless it be excised, an operation which is followed by such a
degree of shock that any observations made on the animal afterwards
are of little value, and are moreover soon cut short by its death. On the
other hand, the partial removal of the liver from the circulation, by
diverting the portal blood into the inferior vena cava, does not
immediately affect the animal, so that it can be kept alive for some time
and observations made on its metabolism. It is with some of the results
of this latter operation on dogs that the present paper deals.

The first successful operation of this nature was performed on a dog
in 1877 by Dr. von Eck (i) of St. Petersburg. His object in doing the
operation was to see whether it might not be possible, in this way, to
relieve the congestion of the liver which occurs in cirrhosis of the liver.
Of a total of eight dogs, only one survived the operation more than a
few days, sepsis and thrombosis being the cause of death of the others.
One dog, however, lived two-and-a-half months, and then escaped from
the laboratory. No accurate observations were made on its metabolism.
A few years later Stolnikow (2) also performed the operation successfully
on dogs, but otherwise added nothing to Eck's findings.

Although, from these observations, no doubt could exist that the
establishment of a porto-systemic fistula is not immediately fatal, it yet
remained undecided whether there is any ultimate effect on the animal,
and whether its metabolism is affected. To study these questions, Pawlow
and Massen (3) performed the operation on a large series of dogs, taking
every possible precaution to prevent sepsis, and they elaborated a

(267)

4 QUATERCENTENARY STUDIES IN PATHOLOGY

technique for sewing the veins together and making the opening between
them, thus rendering the operation much less dangerous. Many of
the dogs, thus operated on, lived several months, and an accurate
study was made of their general condition and metabolism. In this
part of the investigation M. Nencki and M. Hahn collaborated. The
main results of these observations will be discussed later.

Following Pawlow and Massen's technique, several experimenters
have performed the operation. The more important contributions have
been by Magnanimi f^), von Karltreu (^), Sweet (^) and Herrick (J).

Three of von Karltreu's dogs lived from one to nine months, and post
mortem examination revealed the presence of an extensive collateral
circulation around the liver, as a result of which much of the portal
blood still circulated through this organ. Both Magnanimi and von
Karltreu performed the operation, as recommended by Queirolo, in which
the veins are made to anastomose below the vena pancreatico-duodenalis,
which latter is tied. This operation, being on portions of the veins
which are comparatively free, is much easier than Pawlow's, where the
fistula is made close to the hilum of the liver, above the vena pancreatico-
duodenalis. By ligating the pancreatico-duodenal vein, however, a
collateral circulation to the liver becomes established by enlargement
of the fine blood vessels in the ligamentum hepato-gastroduodenale, so
that the object of the anastomosis is frustrated. Sweet employed a
cautery wire, instead of Pawlow's scissors, in making the communication
between the veins, and thus considerably simplified the technique. The
vena pancreatico-duodenalis was not tied. His results were excellent.

In all these operations, it is most difficult to sew the veins together
without very troublesome haemorrhage. Dr. F. C. Herrick, working in
this laboratory, has shown that the veins involved can be clamped a
few minutes at a time without any permanent injury, and that
while empty of blood they can be quickly and easily sewn together. The
veins were anastomosed just below the opening of the pancreatico-
duodenalis, which latter vein was not tied, so that the portal vein being
ligated above it, its blood found free access to the vena cava. For
making the communication between the veins he used a fine wire snare.

The Results of the Operation. — Apart from the general effect of so
severe a surgical operation nothing unusual is commonly noted in the

(368)

ON ECK'S FISTULA 5

dogs for some time, unless perhaps, as observed by Pawlow, they
become irritable. The temperature immediately after the operation
usually sinks a degree or so, to rise afterwards somewhat above the
normal, at which it stands for about 10 days. The body weight may, or
may not, gradually diminish. In about 10 days after the operation many
of the dogs begin to show symptoms which are considered peculiar to
the condition. At the outset these symptoms are variable in character.
Some animals become very restless and excitable and tear and worry
everything with which they come in contact, others become usually quiet
and sleep most of the time. These conditions may alternate in the same
dog. Sooner or later, more definite symptoms make their appearance ;
the animal (now lying down) becomes depressed and apathetic, and when
he attempts to walk is found to be paretic in the hind legs so that he
continually sits down. The gait is also distinctly ataxic, the legs being
lifted unnecessarily high and then dashed on the ground, and, frequently,
in awkward positions. If the limbs be placed in some unusual position
— e.g.y crossed over one another — the dog leaves them there for some
time. Pawlow considers this an important symptom and calls it
catalepsy.

Supervening on these symptoms a gradual condition of hyper-
excitability is next noted. The dog is continually moving about, but
the movements are purposeless and in-coordinate. During this stage he
becomes blind and anaesthetic, but retains consciousness and hearing. For
example, one may push a needle into the pads of the paws and tramp on
the toes without the dog taking the slightest notice, whereas, if the door of
the room be opened, he will at once start up and try to make for it. That
he is blind is shown by the fact that he does not attempt to avoid
objects in his path but runs against them. Meanwhile, he is gradually
becoming unconscious, and may now show convulsive movements of
greater or less severity. The convulsive stage is followed by one of
coma, in which the animal may die.

Many of the dogs recover from these symptoms, but, in such cases,
can usually be easily caused to manifest them again by rough handling
or excitement. In some cases the above symptoms are only partially
seen, the first thing noted being convulsions. In others, the stages
follow one another so quickly that they are indistinguishable as such.

From the outset, Pawlow, and others, noted that certain of the dogs

(269)

6 QUATERCENTENARY STUDIES IN PATHOLOGY

refused food almost entirely, in consequence of which they lost from
30-40 % of their body weight Sooner or later, these dogs showed
a very acute attack of symptoms which were usually fatal. Other
animals ate sparingly, and were not affected so acutely. Some dogs,
on the other hand, ate plentifully of certain foods and suffered only
mildly. The ingestion of flesh was found to be the immediate cause of
the onset of symptoms. Many dogs, having recovered from an attack,
would afterwards refuse flesh and remain perfectly normal until again
induced to take it, when another attack would result. Other dogs
seemed from the very first to know instinctively that flesh was hurtful
to them, and would absolutely refuse to eat it, and if offered no other
food, starve instead.

Pawlow and his co-workers sum up their result regarding these
observations by stating that "dogs in which the portal blood flows
directly into the vena cava without passing the liver cannot stand
flesh diet without the development of severe nervous symptoms which
often terminate in death."

In the dogs which did not show symptoms, /^j/ mortem examination
revealed a collateral circulation to the liver and a small opening between
the veins. The presence of this collateral circulation in von Karltreu's
dogs undoubtedly explains the absence of the symptoms reported by him.

A careful chemical examination of the urine of these dogs was made
by Nencki and Hahn. A relative increase in ammonia and uric acid,
and a relative decrease in urea, were usually found. The urine was
mostly alkaline in reaction. The most noteworthy abnormal change
however, consisted in the presence of carbamic acid^ and in greatest
amount when the above symptoms were present. This led these
workers to think that poisoning by carbamic acid might be the primary
cause of the symptoms. To test this hypothesis, studies were next made
of the pharmacological action of carbamic acid in dogs. When a five
per cent, solution of carbamate of sodium in physiological saline was
injected intravenously into normal dogs the following effects were noted.
With 0*25 grm. carbamic acid per kilogram body weight the dog
became drowsy and somewhat ataxic in his gait ; with a dose of 0*3
grm. per kilogram the dog became more or less excited, sometimes
excessively so ; the movements were ataxic, and the sense of vision was
lost, but hearing was normal. The skin was anaesthetic, and sometimes

(270)

ON eck's fistula 7

a cataleptic state was noted. With 0'6 grm. per kilogram tonic and
clonic convulsions were produced (simulating epilepsy), the saliva flowed
from the mouth, the pupils were dilated. Anaesthesia and blindness
were present Usually, the dogs recovered. With larger doses tetanus
with opisthotonus and cessation of respiratory movements, followed by
death, were observed.

Exactly the same symptoms were produced in Eck's fistula dogs by
placing carbamate of sodium in the stomach, the contents of which had
previously been neutralized by giving sodium carbonate. On the other
hand, no symptoms, except occasional vomiting, followed a gastric
administration of carbamates to normal dogs.

From these interesting observations it seemed justifiable to conclude
that the symptoms seen in Eck's fistula dogs are due to poisoning by
carbamic acid. The liver is the (chief) site of urea formation, the
immediate precursor of urea is carbamic acid, and large amounts of this
substance are contained in the portal blood (being formed in the intestine,
especially when flesh is being digested). Therefore, when the portal
blood does not traverse the liver before entering the systemic circulation,
the carbamic acid which it contains is retained somewhere in the
organism until present in sufficient amount to induce poisoning.

Accepting these observations as correct, there would appear to be
little doubt that carbamic intoxication is the cause of the symptoms, but
if we consider the chemical nature of carbamic acid, and especially the
extreme rapidity with which, in the presence of water, it becomes
converted into ammonium carbonate, then it would appear that
poisoning by ammonium is more probably the true explanation. When
carbamate of ammonium comes in contact with water the following
reaction ensues : —

ONH4 ONH4

co< + H20 rt C0<

NH2 ONH4

The two arrows in this equation indicate that complete conversion of
carbamate into ammonium carbonate does not occur, but, on account of
reversible action, only until a certain equilibrium between carbamate and
carbonate is established.*

* Haskins and I have recently shown that this decomposition of carbamate is extremely
rapid, and that the relative amount of carbamate which remains undecomposed is relatively high.

(271)

8 QUATERCENTENARY STUDIES IN PATHOLOGY

Ammonium carbamate when dissolved in water will decompose until
there is in the solution exactly the same amount of ammonium
carbamate as if an equivalent amount of ammonium carbonate had been
dissolved.

Consequently, as in the above experiments, when carbamate is
injected intravenously, or when it is given by the stomach, it must have
largely decomposed before gaining the blood. If carbamate is produced
in the intestines after flesh has been ingested it will be largely converted
into ammonium carbonate before getting far in the blood. In other
words ammonium ions will have been produced from it.

Pawlow and Nencki did consider this probability, but thought that the
symptoms produced by injection of carbamate-free ammonium salts were
distinctly different from those which follow carbamate injection. Their
comparison of the two {i.e., ammonium and carbamate), however, was
concerned only with the general symptoms, and these, being entirely
objective, are necessarily difficult to compare. I have found that the
effect on the mean arterial blood pressure and respirations of solutions
of ammonium carbamate and ammonium bicarbonate are identical in
every respect. Now, one of these will contain much more carbamate
than the other, the bicarbonate containing only a small amount of
carbamate, so that on the blood pressure at least the effect can be entirely
ascribed to the ammonium ion.

We see then that the main question regarding these symptoms
concerns not so much the presence of excess of carbamic acid in the
organism as the presence of an excess of ammonia. Even if the
symptoms are due to carbamate intoxication there must be excess of
ammonia. In their more recent publications the St. Petersburg workers
have accordingly confined their attention to the behaviour of the
ammonia content of the organism in Eck's fistula dogs. At the outset, in
doing this, it soon became evident that exact information regarding the
quantitative distribution and the source of ammonia in the normal 2x\\^2\
body had first of all to be obtained.

Nencki, Pawlow and Zaleski (9) and later Salaskin (^o), by the use of
accurate methods for the determination of ammonia in the blood and the
various organs, have recently furnished this information. The following
are among their main conclusions.

(272)

ON eck's fistula 9

In the arterial blood of dogs fed on flesh the amnnonia content is
about 1*5 mg. per icxD cc. It is very constant.

In the portal blood, under the same conditions, from 3 to 4 times as
much ammonia is present, being, however, variable in amount. In the
blood of the branches of the portal vein, coming from the stomach,
pancreas and intestines, the percentage of ammonia is still higher.

In the mucosa of the stomach, the brain and lungs a very high
ammonia percentage exists.

During fasting, the percentage of ammonia in all the above situations
falls markedly (o'4 mg. per 100 cc. in arterial blood).

Concerning the source of the ammonia in the portal blood a large
part of it comes from glandular activity, for it appears in the portal blood
when a fasting dog is merely tempted with savoury food ("sham
feeding.") When food is ingested, still more ammonia appears. Its
source is therefore double — from the active glands and from the food.
Ammonia is produced elsewhere in the body from the katabolism of
proteid food stuffs.

The above observers have also examined the ammonia content of the
blood and various organs in Eck's fistula dogs, and have come to the
following conclusions.

In the blood and urine there may or may not be an increased amount
of ammonia, but in dogs which have died of the above symptoms a large
excess of ammonia is always found in the brain.* Salaskin sums up the
main conclusions of this part of the investigations of the St. Petersburg
workers in the following manner.

After the establishment of the fistula, when the dog is on a flesh diet,
the ammonia which is produced by the activity of the gastro-intestinal
glands and from the food is carried by the portal blood directly to the
systemic circulation. Under normal conditions by far the greater propor-
tion of this ammonia will become converted into urea in the liver. In the
circulation the ammonia produced in the tissues by katabolic processes
is added to it. A part of this excess of ammonia in the blood is filtered
through the kidneys into the urine, some passes via the hepatic artery to
the liver, and is converted into urea f and the remaining excess becomes
stored away in some unknown combination in the organs, especially in

* A table giving all the details of these observations will be found in Salaskin's paper,
t Some may also be converted in urea elsewhere than in the liver.

(273) T

10 QUATERCENTENARY STUDIES IN PATHOLOGY

the brain. When a certain amount of ammonia has accumulated in the
brain, symptoms appear, and if the dog recover from these and receive
no more strongly nitrogenous food for some time, this excess of ammonia
in the brain will gradually be carried away in the blood again, and either
become converted into urea or escape into the urine. It will be evident
from the above that there may or may not be excess of ammonia in the
blood and urine when symptoms are present.

To induce the above symptoms more acutely, Nencki and Pawlow
("), after making an Eck's fistula in a dog, removed as much (5/6
to 7/8) of the liver as possible. The dogs lived on an average from
two to three hours after the operation. One of the animals
remained in a comatose state after the removal. One dog, however, so
far recovered as to run about for some time, but soon showed tetanic
convulsions ending in death. The symptoms were not strictly like those
portrayed above. In two of these dogs the ammonia in the blood and
urine was found to be increased. The urea in the urine was diminished,
whereas the urea in the blood remained normal in amount. This last result
is due to the fact that urea formation is not confined to the liver but may
occur elsewhere in the organism. The increase of ammonia in the blood
was not great enough to account for death. It was consequently thought
that perhaps some toxic substances had been produced by the removal
of the liver. No experimental evidence of this could, however, be
obtained by inoculating a normal rabbit with some of the urine excreted
by the dogs.

In other dogs with Eck's fistula instead of removing the liver, the
hepatic artery was ligated. Some of these dogs lived 12 to 15 hours,
apparently recovering well from the operation, but afterwards passing into
a comatose condition. In one of these dogs, the blood was examined
chemically. For about seven hours after the operation this dog did not
show anything abnormal, but then clonic convulsions appeared, passing
into coma, and ending in death (8J hours after operation).

The ammonia content of the blood was found to be unaffected by
this operation, so that death must have been due to some cause other
than ammonia intoxication.

These observations recall others by Lieblein (^^) in which the liver
cells were destroyed by injecting acid into the hepatic ducts. The dogs
became comatous, and died much in the same way as after removal of the

(274)

ON ECK'S fistula II

liver. Something other than ammonia must have been the cause of
death.

It would occupy too much space to discuss the observations of
Magnanimi and von Karltreu here. They are rendered valueless in this
connection by a collateral circulation which developed around the liver.

I have given a fairly extensive review of the work of the St.
Petersburg observers in this direction, as I believe that, although widely
known in a general way, the exact details of this most interesting
experimental disease are little known. With the recent simplification by
Sweet and Herrick in the technique of Eck's operation, more extensive
observations of the symptoms are rendered possible, and the question as
to whether carbamate or ammonium poisoning is their cause, or whether
they are not due to some other poisonous body (toxin), should not long
remain an open one.

The behaviour of several of the dogs operated on by Sweet was
observed by P. B. Hawk (i3). Only mild, transient symptoms were
noted when flesh powder was given, but they were much more severe
when meat extract or fresh flesh was also given.

It was further noted that the injection of sodium carbamate did not
produce any symptoms similar to those observed in dogs with Eck's
fistula, either when the salt was given to normal or to Eck fistula dogs.

Rothberger and Winterberg (i4) have recently communicated a series
of very accurate observations on twenty-two dogs in which Eck's
operation had been performed. They conclude that the symptoms are
probably not due to carbamate intoxication, but rather to some other
toxic substance which the liver normally destroys. In this connection
they point out the extreme importance of the liver in antagonising the
action of many poisons {e.g., strychnine), when these are absorbed into
the portal blood from the intestine. In comparing the symptoms,
presented by Eck fistula dogs, with those induced by administration
of carbamate or substances which would produce it, they found very
considerable differences.

I shall in this connection offer a brief account of some observations
on the general condition of four Eck's fistula dogs operated on by Dr. F.
C. Herrick. I shall then give the results of a chemical examination of
the urine of three of these animals. Unfortunately, the blood and
organs were not examined for their ammonia content.

(275) T I

12 QUATERCENTENARY STUDIES IN PATHOLOGY

Regarding the condition of the dogs after the establishment of the
fisttda, I have noted undoubted nervous symptoms in two of them
(Nos. 2 and 4), but no definite symptoms in the other two (Nos. i and 3).
The following is an account of the behaviour of the dogs : —

Dog I. — Female, weighing 25 lbs. The operation was performed on
January 21st, 1904. For six days after it no food was taken, there being
slight fever. From January 28th, to February 12th (15 days) a diet of
300 gr. bread and 50 gr. butter was taken, and during this period the
dog increased in weight from 21 to 23 lbs. On two occasions (viz., on
the 4th and 9th of February) meat extract was added to this diet. A
period of six days' fasting followed, and then a bread and butter period
of eight days, after which flesh was given in gradually increasing
amounts (see Table I.).

At no stage were any phenomena noted like those described by
Pawlow and Massen, but marked nephritis with accompanying symptoms
set in, which at one stage looked as if they would prove fatal. On this
account it may be of interest to describe the history more in detail.

From the very outset albuminuria was observed. Until the day after
the first administration of meat extract, however, the proteid was present
only in traces, but on this day it increased distinctly in amount, the
increase continuing daily until, after a few days, about 50% of the total
nitrogen excreted was present as proteid. Fatty, cellular and hyaline
casts were also abundantly present in the urine, which also contained
globular bodies soluble in ether. The high excretion of nitrogen
occurring even when no food was given is accounted for by the presence
of proteid. The animal became very much emaciated. CEdema of the
abdominal wall was very marked, and an interstitial keratitis developed.
On account of the presence of so much proteid, the quantitative estima-
tions of the various urinary constituents were rendered very difficult, and
had to be discontinued because of their inaccuracy. The dog was
carefully treated, being fed for some time largely on milk ; and her
condition gradually improved, the proteid diminishing slowly till it had
almost disappeared. After two months she was removed to the country
where she became perfectly well. After four months she was killed, and
a post mortem examination made. It was found that the ligature had
been applied above the vena pancreatico-duodenalis, and that the opening
between the vena cava and vena porta had closed. Nothing abnormal

(276)

ON eck's fistula 13

was found in the kidneys. A microscopical examination was un-
fortunately not made.

Dog 2, — Male, about 40 lbs. weight. Operated March 8th, 1904.
He recovered well from the operation, and in three days was placed on a
bread and milk diet which was changed to one of flesh (200 gr.) three
days later. For six days the dog took this eagerly, but on the seventh
refused it. During this day he was somewhat restless. Next morning
he was found lying on his side in an unconscious condition, with the head
thrown back and the hind limbs drawn up on the abdomen. The head
was constantly moved backward and forward on an average of about 70
times per minute, but this rhythm was irregular. Each time the head
was drawn back the eyelids closed. On pricking the skin with a needle
no reflex movement resulted ; on placing the limbs in an unusual position
they were not withdrawn (Pawlow's catalepsy). Conjunctival reflex
normal; rectal temperature 39° C. ; respirations 12-14 P^r minute; pulse
1 10-120 per minute, and irregular both in force and rhythm.

In an hour or so the clonic contractions of the head disappeared, and
after about three hours the dog seemed to have recovered and took some
milk. He was kept on a bread and milk diet for some weeks and
gained weight. Flesh was then given him again, but no further
symptoms were ever noted. He lived several months. An opening,
measuring one centimeter across, was found between the veins, and
no collateral circulation.

Dog J. — Male, about 60 lbs. There was a good recovery from
the operation. Nothing unusual in the dog's behaviour was noted
even although he took large quantities of flesh. Three weeks after
the first operation a second operation was performed, when it was found
that a large vein entered the portal above the vena pancreatico-
duodenalis. A second ligature was therefore applied to the portal vein
quite close to the liver. Three days after this second operation the dog
was again placed in his metabolism cage, and the urinary analysis
proceeded with (see Table H.). This dog did not show any definite
symptoms, and died from pneumonia three weeks after the second
operation. He rapidly lost in weight, and was usually feverish, some
sepsis having probably been developed.

Dog 4.. — Operated February 27th, 1905. Food was withheld until
March 8th, when a diet of milk, bread and butter was given, and

(277)

14 QUATERCENTENARY STUDIES IN PATHOLOGY

continued till March 23rd. While on this diet the dog slowly lost weight
(from 28 lbs., to 25 lbs. in 26 days), but appeared normal in every other
respect. Though very quiet while we were in the laboratory, she was
extremely restless during the night time, for on several occasions she
tore her way out of the strongly made metabolism cage in which we
kept her, on one occasion by gnawing through a plank of wood, and
on another by tearing away the stout wire netting on the lid of the
cage.

On March 23rd, the daily diet was changed to one of 400 gr. flesh.
At first, she eagerly ate all the flesh, but in a few days began to
take only a portion of it. Otherwise she appeared quite normal till
April 7th, when it was noticed that she looked uncomfortable, and would
not lie quietly in one place as had previously been her custom, but
wandered aimlessly about the laboratory. On the morning of April 8th,
this restlessness was very marked. At 11 '30 a.m., on this day, her gait
was seen to be ataxic, and at 12 noon she became paretic, particularly in
the hind limbs, so that she could no longer stand, but lay on her side.
At this time, her respirations were, on an average, 48 per minute,
sometimes becoming rapid and gasping ; pulse 140 per minute,
regular ; rectal temperature 39*5 ° C. ; there was very profuse salivation ;
the ears showed fine tremors ; the pupils were very contracted and did
not react to light ; on pricking with a needle between the pads of a paw,
or on nipping with a strong forceps, no reflex movements resulted (skin
apparently anaesthetic) ; the limbs were extended and the muscles rigid,
especially those of the hind limbs ; when a limb was placed in an unusual
position it was kept there for some considerable time (Pawlow's
catalepsy). Some of the saliva was collected and found to be distinctly
alkaline in reaction ; it did not give any reaction for ammonia when
boiled with alkali. From time to time she made futile attempts to
assume the erect posture, and in these attempts moved the fore limbs
as in swimming. She passed urine, which was collected and found to
be alkaline in reaction, with 0*2 % of ammonia, 71 mm. CO2 and 28 mm.
carbamate CO2.

At 3*15 p.m. the head and neck were found to be rigidly thrown
back almost at right angles to the spinal column (opisthotonus) ;
the muscles of the hind limbs were now relaxed, and those of the fore
limbs very rigid ; pulse 160 and of low tension ; respirations slower (26)

(278)

ON ECK'S FISTULA 1 5

but regular and deep ; the pupils still very contracted. She now lay
quietly on her side and made no attempts to assume the erect posture.

For the next 2j hours very little change was noted in her condition,
except that the respirations became laboured and the pulse of lower
tension ; the rectal temperature fell to 38*5 ° C. ; the extremities became
cold to the hand. At 5*48 p.m. she was wrapped in cloths to keep
her warm, and in moving her for this purpose, convulsions were induced,
in which the head and neck were thrown far back on the body and
the fore limbs stiffly extended. The respirations became very rapid and
gasping, and the pulse almost imperceptible. This convulsion lasted
7 minutes, and as she had then become quiet, she was left till next
morning, when she was found dead, lying in the same position as when
last seen. Autopsy revealed the existence of a perfect Eck's fistula
without any collateral circulation.

It will be seen that the dogs observed by us behaved in general
exactly like those of the St. Petersburg workers, and that a flesh diet
seemed to bring on the symptoms. The marked nephritis, which was
noted in dog i, was no doubt due to disturbance in the kidney circulation,
the exciting cause perhaps having been the excess of meat extractives
given the animal on the day before the albuminuria became marked.

The Chennical Examination of the Urine.

This was carried out in the case of three of the above dogs (Nos. i,
3 and 4).

Nencki and Hahn {vide page 270) have shown that in the urine of
Eck's fistula dogs there is a relative increase in the excretion of
ammonia and of uric acid, and relative decrease in that of urea. They
also found carbamates in the urine. In our investigation we therefore
examined total-nitrogen (Kjeldahl-Gunning method) ; urea-nitrogen
(Morner and Sjoqvist and Folin's method) ; purin-nitrogen (Camerer-
Arnstein method) and ammonia-nitrogen (Folin's method). We likewise
tested for the presence of carbamates by the Abel-Drechsel method, and
later, by the method elaborated by us and described elsewhere.*

* Macleod and Haskins : The Quantitative Estimation of Carbamates. American
Journal of Physiol., 1905, xiii., p. 17.

(279)

l6 QUATERCENTENARY STUDIES IN PATHOLOGY

Dog I (Table I.). As already pointed out, this dog developed a
marked nephritis, so that the urinary analysis was rendered extremely
difficult and possibly inaccurate towards the end of the investigation.

Until proteid made its appearance in any considerable quantity the
relative amount of urea and ammonia were about the normal. The purin-
nitrogen excretion previous to the day on which meat extract was given
averaged about 0*04 gr., which is distinctly higher than the average for a
dog of this size. When meat extract was added to the otherwise purin-
free diet, a very distinct increase in purin excretion resulted.*

After albuminuria developed, the relative amount of urea-nitrogen
steadily fell, more than 50% of the total nitrogen being excreted in the
form of proteid. The absolute amounts of urea and ammonia remained,
however, nearly unchanged.

Urine of Dog j. — The results of this investigation are given in Table
II., from an examination of which it will be seen that the daily excretions
of nitrogen, ammonia and urea were very high and to a corresponding
degree, the nitrogen of ammonia being on an average about 7 per cent,
of the total nitrogen, and that urea about 79 per cent. The rapid
emaciation of this dog probably accounts for the high nitrogen content
of the urine.

On March 15th a dose of 10 grams of ammonium acetate was given
by the stomach tube with the result that on the same day and the day
following a marked diuresis was induced, but with no increase of the
ammonia excretion on either day, whereas on the day following there
was a very marked increase in the excretion of urea, its nitrogen being
88 per cent, of the total nitrogen. On the 17th of March 5 grams of
ammonium carbonate (containing, therefore, carbamate) had likewise no
effect on the ammonia excretion, but raised very considerably that of
urea.

Regarding the purin excretion, it will be seen that this was fairly
constant (viz., about 0*04 gms. purin-nitrogen), except on the 17th of

* Burian and Schur (" Uber die Stellung der Purinkorper im menschlichen Stofwechsel,"
Arch. f.d. ges. Physiologic, 1901. Bd. 87, p. 291.) have shown that in normal dogs about 5%
of ingested oxypurin (xanthin and hypoxanthin) reappear in the urine. Our object in giving
the meat extract in the above experiment was to see whether the same integral factor would
hold for a dog with Eck's fistula. The incidence of albuminuria, however, renders the
observed result too inaccurate to draw any conclusion.

(280)

ON eck's fistula 17

March when it was considerably higher. This was the day on which
ammonium carbonate was given, and on which there was a marked
diuresis. The result agrees with those of Burian and Schur * on normal
dogs, and may possibly be explained in the same way as they explain
theirs ; that the circulation through the kidneys had been rendered more
rapid so that more purin was excreted into the urine before it had time
to be destroyed in the liver and other purin-destroying organs.

It is further of interest to note that the purin excretion was not
much higher than normal.

In dogs weighing about twenty pounds the average daily excretion of
purin-nitrogen was found by Burian and Schur to be about 0*035 S^-
Omitting No. 7 in our table, the average for the Eck's fistula dog is 0*039
gr. In Burian and Schur's dogs a flesh diet was given, in ours a variable
diet. The figures are therefore not strictly comparable.

Nencki and Hahn described the uric acid excretion in dogs with
Eck's fistula as distinctly higher than normal. This finding we have
therefore been unable to confirm.

At no period during the investigation was there ever, in this dog's
urine, the slightest trace of carbamate ; the urine was acid or amphoteric
throughout, so that no noteworthy amount of carbonic acid gas was
present. Nevertheless, the dog had been fed with meat, and the post
mortem examination had revealed a perfect Eck's fistula. Not even
when given ammonium carbonate solution (which contains carbamate)
did any carbamate appear in the urine.

Indeed, the urine of this dog, as a result of the chemical analysis,
must be regarded as perfectly normal, the relation both of urea and of
ammonia to total nitrogen being, for a meat diet, about the normal one.

In Dog No, ^, weighing 12J Kg., the urinary analysis was not so
complete, but nevertheless a catheter specimen was daily examined
for total CO2 and carbamate CO2. The difficulty in making a complete
analysis lay in the fact that it was impossible to keep this dog confined
in her cage (see p. 278).

While on a diet of bread and milk, the urine was frequently alkaline
in reaction, and, when so, was found to contain a considerable amount of
carbonic acid and carbamate. The carbamate CO2 did not, however,
bear any constant proportion to the total CO2. The alkaline urine of

* Burian and Schur : Loc. cit., s. 342,

(281)

i8

QUATERCENTENARY STUDIES IN PATHOLOGY

TABLE

Dog's
Weight.

Food.

Urine

—24 Hours

Date.

No.

Flesh

Bread Butter

Total N

Amount

Specific

Reaction to

Total

UreaN

Purin N

in gnis.

in gms. (gms.).

(gms.).

in c.c.

gravity.

litinu.s.

nitrogen
(gms.).

(gms.).

(gms.).

Jan. 21

...

\

Jan. 23

...

Jan. 24

VNo food.

Jan. 25

I

...

Jan. 26

Jan. 27

2

/

Jan. 29

3

182 4

^o•2

2-229

370

+

Acid

3-70

0-0185

Jan. 30

200 4

^o

2-450

Jan. 31

...

...

250 i

p

3-062

Feb. I

4

ISO 2

0

1-837

900

Acid

1-968

1-419

00266

Feb. 2

5

21 lbs.

300 4

0

3-675

500

Alkaline

1-904

1-428

0-0577

Feb. 3

6

300 5

0

3-675

400

Neutral

3-304

2-576

00451

Feb. 4

7

300 5

0

3-675

350

Acid

2-156

1-660

0-0495

Feb. 5

300 5

0

3-675

Feb. 6

*8

22'ibs.

300 5

0

3-675

750

Acid

+

I -606

0-072

Feb. 7

300 5

0

3-675

1 0-0789
\ 0-1067

Feb. 8

9

300 5

0

3-675

700

Acid

3-535

2-825

Feb. 9

10

...

300 5

0

3-675

600

Acid

3-629

+

02142

Feb. 10

300 5

0

3-675

Feb. 11

II

23 ibs.

300 5

0

3-675

650

Acid

4-650

3-I213

*o-o842

Feb. 12

12

230 4

5

2-817

400

Acid

+

+

+

Feb. 13

13

24 Ibs.

0

0

0

350

Acid

2-974

1-700

+

Feb. 14

0

0

0

Feb. 15

14

23 "lbs.

...

0

0

0

600

Acid

2-528

1-394

+

Feb. 16

0

0

0

...

...

Feb. 17

22 lbs.

0

0

0

...

Feb. 18

16

2iilbs.

25*

0

600

Acid

4-032

+

+

Feb. 19

17

21 lbs.

75

0

•918

400

Acid

3-450

I •848t

Feb. 20

18

21 libs.

100

0

1-225

some
lost.

1783%

0-722%

Feb. 21

...

...

100

1-225

Feb. 22

100

1-225

Feb. 23

19

2ii'ihs.

100

1-225

500

Acid

3-556

2-441

Feb. 24

20

100

1-225

355

Acid

3-538

+

Feb. 25

2iiibs.

100

1-225

Feb. 26

21

21J lbs.

200

6-8

550

Acid

3-591

Feb. 27

22

21 lbs.

200

6-8

400

Acid

6-451

...

Feb. 28

23

200

6-8

500

Acid

6-552

Feb. 29

24

20 lbs.

400

13-6

400

Neutral

7-481

(282)

ON eck's fistula

19

Urine —
24 hours.

Carbamic
acid.

NH3 N to
total N.

Urea N to
total N.

Abnormal
substances.

Remarks.

NH3 N
(gms.).

Albumin.

...

...

...

...

Dog (female) Eck's fistula established. Came

out of ansesthetic well.
Dog normal — Falck's operation, no food, no

water.
Unsuccessful catheterization under ether with

no food.
Catheter passed 12:15 — some urine lost in

struggling — NH3 doubtful.
Starved— wasting.
Catheter 12:15— some blood mixed with urine.

0'102

...

Very little urine in basin eight days after opera-
tion.

0*0703

nil.

1:1-39

...

Each dog's urine placed on ice— copious washing.

0-1078

0-I097
oo79»

trace.

i:i7-6
1:30-1
1:27

1:1-28
1:1-299

10 g. gr. beef extract given after catheterizing.

0-372

...

+ +

High purins.

0153
0-2688

trace.

1:23-1
1:13-1

i:i-24

+ +

0-5% (Esbach).

One of the purin precipitates with AgN03,

inaccurate.
20 gms. beef extract given after catheterizing.

0-1483

+

...

1:21

1:1-49

/ Up to U mark
\ in Ksbach.

*High purins and calculation only approximate —
proteids separated by boiling in acid.

(0-113)
0-174

•076

•0537
•048

...

1:17-1

...

1:75
1:72

1:1-74
i:i-8i

1:1 -86
1:2-46

/ 13mm. below
1 mark U.
/ 9mm, below
1 mark U.

Hyalin, fatty and cellular casts in urine ; nucleo-

proteid also present.
Sediment— casts, bladder cells. Globular bodies

like leucin.
i *Given 2 hours before catheterization. Very
\ faint ppt. with HA.

The globular bodies dissolved in ether, not in
hot water, therefore, fat.

•0858
•1215

1:41-4
1:29

i-.i'45

At mark U.

•7%

(tOnly one urea estimation).

0-I355
0-1635
0-1736
0-1344

...

1:26-5
1:39-4

7%o

•5%o
7%o

(283)

20

QUATERCENTENARY STUDIES IN PATHOLOGY

^

is

Z

fo

■2

o

rt

u

"*

'5^

W

M-i

rt

;<

J^

8

.

.c

s

o5

bjO

b

Q

O

(U (U

q3

C C

c

;

o o

o

•

^^

:zi

!5 ui

4-^ OJ J, I (9 3 "43

(u , a

■iki

a

O

s s s
e s s

■Th t^

00

ir^ w

t>»

Tl-U-1

coi^

o^::

b-^ :

• .o P

^

S^

oS

o o

o

o o

o

o o

o o

Nitrogen
urea in

per cent,
of total

nitrogen.

p y^ T*- y^ f^ p p^^ ^ :t rn p p f>^ on «

w vi) 00 o oo Vj- ii-> bsoo 'tJ- r^ t^ i-^ in U-) *Tt
t^r^o t^vo r^oo ilooo onoo oo oo oo oo oo

Grammes

urea

per

diem.

N C^ 11 O 00

HH l-( l-l N

t^ r^ On looo lo -j-j Tf o "^ M fo vnoo r^oo
MD ON CO fO ro mii-1 t^^p J^ io>0 ro pNOO N

w \b Lcoo t^b>b ONroasiobvo b rjoo

HHI-ll-ll-ll-Hi-lCVJl-ITlJ-T^^rnCOl-lWCSC)

Nitrogen
ofNHsin
per cent,
of total
nitrogen.

cnoo iri o O
00 'rl- <^ iri V

t^ r^ HH i^oo ONONiot^vb t^oo vb MD iovb

Grammes

ammonia

per

diem.

<X) vo VO MD t^
VO O "-1 fOOO
t^ U-) lo t^ O

00 1^00 VO ■* ro TfvO N 00 On -1 « vo

MOO c<^T^^rlTt^^ o fnoo moo t^MD n t-,

t^ON-^O M fON 1-1 000\0\OvOOOOO i-i

O O O O M

OOi-ii-i«i-hmNC)i-ii-ii-iOOOi-i

Grammes

nitrogen

per

diem.

vo 1- O t^
^ t^ tJ- C> fO

cJo 00 Vt b «

vO"^ONr^ OMOONOnNOONvOMmvO
^ P ."* ." "^ ." .^ P 9' P .^ .'^ f^ ."^ .'^'P
^^ b b HI w "(^ w Ij^ rn W-oo vb 00 w c) io

Total

quantity

per.

diem.

8 8888

CO N t^ m ■<:^

OCJOOCJOOOOOOOOOOO
CJOOOCJOOOOOOOOOOO

NLOLooOOQ'^OOOThOi^iJivo
M H-i O N rr; u-1 0 O vovo O t^ N vo t^ ''t
rooo Ti- Tf lo "-d- tn 1^00 On>0 --i- CO CO ro u->

1

S >.S -G S -G = = = = r g-o = =
2

1.4

c

2i

3

a

^

1

•?

?

d

S

-% ■-

=

= =

~ vj

. 1J

;s

•s-s

«

o3

s

^s

s

c c E

s a

a a

:

a s

a a

1^^?!

°^^^^

a s s

a a

id

ass

a a

■;b

N 00 VO

00 HH

8;;

HH hHVO

COO

CO

*-l ►H

in

CO

00

: : •

8

i?

; i

O

o

1—5

o

CO

h-(

: : :

s

• M

(— 1

W

: On

H-)

- • •

■^

lO

pq

<

N

N

H

00

•oo

; ;

II ^

: : :

^
t^

il

. co'g
5^b

o

o

vO

.

M

: 00

CO

■ CO

^ S -^

(<t rnrf vo-^ VO r^OO On O

-^ -^ rt N ^*S ?t On O ^ lOvS t^W 0\ w C^ CO ^
-^ rg ^ ^ = = = = = = = = = = = r = = =

^ w cj CO Tj- u^vo r^oo

1;
fe

iil^ On O ^ ^ioOOO

o s

Oh

(284)

ON ECK'S fistula 21

the 20th of March was slowly evaporated almost to dryness, and then
made up with distilled water to the original bulk ; the resulting solution
was found still .to be alkaline in reaction, showing that the original
alkalinity of the urine had not been due to ammonia. On March nth,
however, the ammonia excretion was considerably above the normal, but
became normal again two days later.

While on meat diet, the urine was unfortunately not examined until
the day on which nervous symptoms appeared, when an alkaline urine
was found to contain both CO2 and carbamate CO2.

It would appear from this last analysis as if there were some relation-
ship between the presence of carbamate and the nervous symptoms, but
if, on the other hand, we bear in mind that it was while on bread and
milk diet that this carbamate was found — i.e.^ when the urine was
alkaline — then we can explain the presence of carbamates as due to the
interaction of soluble carbonate in the urine with the ammonia salts.
Haskins and I have shown that whenever the urine contains soluble
carbonates {e.g., after citrate ingestion, or when it is allowed to stand
a few days) carbamate is formed in this way.*

So far as our chemical investigation of the urine went, we could not
detect anything abnormal in either of the three dogs, except the
remarkable albuminuria developed by the first dog.

Our results were in general confirmatory of those of Pawlow, Hahn,
Nencki and Massen, and point to the fact that in future investigations
into the cause of the peculiar nervous symptoms described above, urinary
analysis alone will not suffice, but must be supplemented or even replaced
by careful analysis of the blood and organs.

To Dr. H. D. Haskins the author owes a deep great debt of gratitude
for the valuable assistance which he rendered in conducting the chemical
analysis. To Dr. F. C. Herrick also the author's thanks are due for the
use of the dogs on which he so successfully performed Eck's operation.

Macleod and Haskins , Loc. at., p. 279.

(285)

22 QUATERCENTENARY STUDIES IN PATHOLOGY

LITERATURE.

1. EcK. Arbeiten der naturforschenden Gesellschaft in Petersburg, 1879, X.

2. Stolnikow. " Die Stelle vv hepaticarum im Leber und gesammten

Kreislaufe." Pfluger's Arch. f. d. ges. Physiologie, 1882, XXVIIL,
P- 255.

3. Hahn, Massen, Nencki and Pawloff. " Die Eck'sche Fistel zwischen

der unteren Hohlvene und der Pfortader und ihre Folgen fiir den
Organismus." Archiv. fiir exp. Path, und Pharm., 1893, XXXIL, p. 161.

4. Magnanimi. "Le modificazioni del ricambio azotato dopo I'innesto della

vena porta colla vena cava inferiore." II Policlinico, 1896, III., p. 153.

5. V. Karltreu. " Ueber die Vereinigung der unteren Hohlvene mit dei

Pfortader." Arch, fiir exp. Path, und Pharm., 1901, XLV., p. 56.

6. Sweet. "The artificial Anastomosis between the Portal Vein and the Vena

Cava inferior — Eck's Fistula." Journ. of Exp. Medicine, 1905, VII.,
p. 163.

7. Herrick. "A Contribution to the Technique of making the Eck Fistula."

Journ. of Exp. Medicine, 1905, VII., p. 751.

8. MACLEOD and Haskins. "The Chemistry of the Carbamates." Journ. of

Biological Chemistry, 1906, I.

9. Nencki, Pawloff and Zaleskl " Ueber den Ammoniakgehalt des Blutes

und der Organe und die Harnstoffbildung bei den Saugethieren." Arch,
fiir exp. Path, und Pharm. 1896, XXXVIL, p. 26.

10. Salaskin. "Ueber das Ammoniak in physiologischer und pathologischer

Hinsicht und die Rolle der Leber im Stoffwechsel stickstoffhaltiger
Substanzen." Zeitschr. f. physiol. Chemie, 1898, XXV., p. 449.

11. Nencki and Pawloff. "Zur Frage iiber den Ort der Harnstoffbildung bei

den Saugethieren." Arch. f. exp. Path, und Pharm. 1897, XXXVIIL,
p. 215.

12. Lieblein. " Die Stickstoffauscheidung rach Leberverodung beim Saugethier."

Arch. f. exp. Path, und Pharm., 1894, XXXIIL, p, 318.

13. Hawk. "A Study of the Conditions following the establishment of the

Eck Fistula in Dogs." Amer. Journ. of Physiol., 1905, XIIL, p. 14.

14. Rothberger and Winterberg. "Ueber Vergiftungserscheinungen bei

Hunden mit Eck'sche Fistel." Zeitschr. f. exp. Path, und Therapie,
1905, I.

(286)

ON THE ACTION OF CERTAIN BACTERIA IN
PRODUCING CELL-NECROSIS, WITH SPECIAL
REFERENCE TO THOSE OF THE BACILLUS
ENTERITIDIS (GARTNER) GROUP.

By George Ford Petrie, M.D.,

Assistant Bacteriologist^ Lister Institute of Preventive Medicine.

(287)

On the Action of Certain Bacteria

In producing Cell-necrosis, with Special Reference to those of the
Bacillus Enteritidis (Gaertner) Group*

It is well recognised that certain bacteria have the property of inducing
a necrotic action in the cells attacked by them. These organisms may
be roughly classed under two heads. • There are, first, those which cause
lesions more or less limited to one tissue or organ, for example, B.
diphtheriae, B. mallei, B. necrophorus and B. typhosus. The necrotic
areas occurring in the liver in typhoid in man may be cited as an
instance in which these particular lesions occupy a secondary place in
the pathology of the disease. There is, however, another group of
organisms, including B. pseudo-tuberculosis rodentium, B. pestis and the
tubercle bacillus, which cause widespread focal necroses, presenting a
characteristic pathological picture.

The necrotic effects are generally attributed to bodies — proteins of
Buchner — in the cell-substance of the bacteria. In regard to the nature
of these bodies, however, very little is so far known.

It is proposed in what follows to record some observations of a
guinea-pig disease, recently met with in Bombay, due to an organism
belonging to the B. enteritidis (Gsertner) group, with lesions closely
resembling those of pseudo-tubercle. The attempt will also be made to
correlate its pathological features in respect of the characteristic necroses
with those of diseases caused by other members of the group.

In the Plague Research Laboratory, a considerable number of young
guinea-pigs in the stock died apparently from a common disease during
the monsoon months, i.e., June, July, August and September. The
animals were about three or four weeks old ; the disease was not
observed among full-grown guinea-pigs. The post mortem appearances
of several of those examined are appended.

(289) U

4 QUATERCENTENARY STUDIES IN PATHOLOGY

Guinea-pig 1 : — examined 27/7/05.

Naked-eye appearances —

No enlarged glands ; liver studded with fine nodules ; spleen
enlarged and congested — no nodules seen. Small intestines markedly
congested. Kidneys show minute punctate haemorrhages under the
capsule. Lungs somewhat congested.
Microscopical examination —

Spleen smear shows bipolar organisms resembling B. pseudo-
tuberculosis ; many of these are in the cells. Heart-blood contains
fairly numerous bipolar organisms, some being in the mononuclear cells.
Animal experiment —

The spleen and heart-blood were rubbed into a shaved area of skin
of a guinea-pig which died of haemorrhage from the lungs (probably
traumatic) within 24 hours. No organisms were seen in the heart-blood
of this animal, but a few bipolar organisms were observed in a spleen
smear. There was a cutaneous reaction at the site of infection, and the
inguinal glands were enlarged.
Cultural test —

An organism was isolated from the heart-blood which corresponded
culturally, morphologically, and in certain sugar reactions with B.
enteritidis (Gaertner).

Guinea-pig 2: — examined 31/8/05.

Naked-eye appearances —

Subcutaneous oedema ; no enlarged glands ; spleen very dark in
colour — markedly enlarged and studded with grey nodules exactly
resembling those of pseudo-tubercle. Very fine nodules were present in
the liver. Small intestines extremely congested. A small amount of
pleural and peritoneal effusion ; lungs apparently normal.
Microscopical examination —

Heart-blood — very few bipolar organisms.

Spleen and Omentum — very numerous bipolar organisms.
Cultural test —

An organism similar to that from guinea-pig i was isolated from the
heart-blood.

(290)

ON THE ACTION OF CERTAIN BACTERIA 5

Guinea-pig 3: — examined S/O/^S-

Naked-eye appearances —

Spleen enlarged and studded with grey nodules. Mesenteric glands
enlarged and congested. Haemorrhages in the large intestine. Suprarenals
were congested and showed small haemorrhages.
Microscopical examination —

Heart-blood — no organisms seen.

Spleen — organisms, very like those of pseudo-tubercle, fairly numerous.

Culture from heart-blood gave similar colonies to those previously
obtained.

Guinea-pig 4: — examined 6/9/05.

Naked-eye appearances —

Spleen much enlarged, with yellow and grey nodules — a few also
in the liver. Small intestines congested. Lungs very congested.
Microscopical examination —

Spleen — few, if any, organisms seen.

Heart-blood — nothing definite.

Culture — Same organism as above obtained from spleen.
Animal test —

The spleen was rubbed into a shaved area of the abdomen of an adult
healthy guinea-pig weighing 520 grammes. On the third day the
inguinal glands could be felt to be enlarged. It steadily lost weight
throughout, the last weight, on the loth day, being 400 grammes. The
animal was then killed by chloroform, and the following appearances
were noted : —

Naked-eye, — Extensive purulent ulceration of the skin limited to the
site of innoculation and burrowing into the subcutaneous tissue. Double
inguinal buboes, the surrounding blood-vessels being moderately injected.

Spleen — somewhat enlarged with some yellow and white nodules ; a
few similar modules in the liver. Small intestines very hyperaemic.

Microscopically. — No organisms were seen in the buboes or organs.

Cultures. — The pus from the cutaneous reaction furnished a pure
culture of the organism. The heart-blood proved sterile.

(291) U I

6 QUATERCENTENARY STUDIES IN PATHOLOGY

Guinea-pig 5: — examined 15/9/05.

Naked-eye appearances —

General subcutaneous congestion ; no enlarged glands. Spleen
enlarged and densely studded with grey necrotic foci ; there were very
few in the liver also. The lungs contained a few grey nodules, each
surrounded by a ring of congested lung tissue. The stomach and small
intestines were extremely congested.
Microscopical examination —

Spleen — Fairly numerous deeply stained bipolar bacilli.

Heart-blood — a very few similar organisms.

Cultures —

Heart-blood — pure colonies of the organism.
Liver — pure colonies of the organism.
Spleen — pure colonies of the organism.

Guinea-pig 6: — examined 15/9/05.

Naked-eye appearances —

Inguinal glands slightly enlarged.

Spleen — somewhat enlarged.

Liver — a few very small nodules.

Left suprarenal capsule showed nodules, and on section was seen to
be almost wholly converted into cheesy masses. Lungs extremely
congested. Stomach and intestines apparently normal.
Microscopical examination —

Nothing definite seen in heart-blood or in the organs. Suprarenal
capsule showed fairly numerous bipolar bacilli arranged in pairs.

Cultures — (from heart-blood) furnished the usual growth.

It will readily be appreciated from the details given above how
markedly the lesions simulate those met with in the well-known
pseudo-tubercle of guinea-pigs. The characters of the bacillus isolated
from the organs may be described briefly as corresponding closely to
those of B. enteritidis (Gsertner). Culturally, and in its morphology, it
gave evidence of being not far removed from B. coli. It can, however,
be readily differentiated from the colon bacillus by the fact that it does
not ferment lactose. It also fails to split up cane-sugar, but produces

(292)

ON THE ACTION OF CERTAIN BACTERIA 7

acid and gas in media containing dulcit. Attention may be drawn to
the animal experiments in the case of guinea-pigs i and 4. It would
appear that a cutaneous infection can take place by using the method of
Albrecht and Ghon. In the case of guinea-pig 6, a chronic form of the
disease was produced, although it is possible that by varying the experi-
mental conditions, e.g., by raising the virulence of the bacillus artificially,
an acute condition, resulting fatally in a few days, might be brought
about. Unfortunately, lack of time from pressure of other work
rendered it impossible to carry out further experiments in this direction.
The significance of the cutaneous method of infection will be referred
to later.

It may appear somewhat curious that, although the disease shows
well-marked lesions, very little can be found concerning it on consulting
the literature. Nocard and L^clainche in their classical work on the
microbic diseases of animals (edition 1903) make no mention of it.
MacConkey (i), in a foot-note to a recent paper dealing with lactose-
fermenting organisms, gives a brief description of its main features. He
notes that an organism indistinguishable from B. enteritidis (Gsertner)
seemed to be the cause of a small epidemic amongst the experimental
guinea-pigs at the Serum Institute of the Lister Institute near London,
and refers to the resemblances of the naked-eye appearances to those of
pseudo-tuberculosis rodentium.

Kovarzik (2) has given an account of an epizootic, amongst guinea-
pigs, which he ascribed to a variety of B. coli. Lehmann and Neumann,
however, include the bacillus in the B. enteritidis (Gsertner) group. The
principal points of his paper may be briefly summarised. Post mortem —
there were found greyish-white necrotic foci on the surface of the liver
and spleen. In smears of these organs the bacilli were isolated, but
sometimes occurred in pairs — short rods (Gram-negative) with rounded
ends. Coccal forms occurring in cultures ; giving on agar, greyish-white
glancing colonies, bluish by transmitted light, later white ; gelatine not
liquefied ; acid and gas in glucose ; no clotting of milk nor indol
reaction. Cultures were very pathogenic to guinea-pigs, especially
when inoculated intraperitoneally — an infection was also produced
through the intact conjunctiva. It was pathogenic to rabbits by intra-
venous but not by subcutaneous inoculation, and killed also by injection
into the large intestine. Gray rats were killed by feeding on cultures —

(293)

8 QUATERCENTENARY STUDIES IN PATHOLOGY

— pigeons by intramuscular injections. Kovdrzik considers that his
bacillus is most nearly related to Lochmann's B. caseolyticus.

Lochmann (3) has described a bacillus which, he believes, is closely
related to B. coli, and which he isolated from 4 guinea-pigs dead of
experimental tuberculosis. Culturally and morphologically it resembled
B. coli. It differed, however, from the typical colon bacillus in the
production of alkalinity in milk, by the absence of the indol reaction and
by its pathogenicity for animals by feeding. Greyish-white deposits were
frequently observed in the liver and spleen of experimental animals. It
is very probable that Lochmann's bacillus is identical with those
referred to above, and that it would be more accurate to place it in the
B. enteritidis group, and indeed Lehmann and Neumann, in their hand-
book, have adopted this course.

Before proceeding to discuss the lesions produced by other members
of the group it may be of interest to record a few cases in which a similar
organism was encountered during the ordinary routine work of the
laboratory.

I. — Mus decumanus brought to the laboratory dead.

From the heart-blood of a mus decumanus brought dead to the
laboratory for examination, a culture was obtained of two kinds of
colonies. Subcultures of the predominating colonies yielded a bacillus
corresponding in all the tests applied to the B. enteritidis. The post
mortem examination revealed evidences of an acute disease, the spleen
being very large, soft and deeply congested ; the peripheral glands were
enlarged ; the serous membranes were injected, and there were haemorr-
hages in the epicardium. There were signs of commencing putrefaction
in the organs. No animal test was performed.

II.— Mus decunnanus brought to the laboratory dead.

Naked-eye appearances —

Nothing pathological was seen, except that the lungs were congested,
and showed haemorrhages. The abdominal organs showed signs of
putrefaction.

(294)

ON THE ACTION OF CERTAIN BACTERIA 9

Microscopically —

Nothing definite was seen in smears.

Cultures of the heart-blood on agar gave numerous grey, circular
smooth colonies which, on detailed examination, proved to be of the type
of B. enteritidis.
Animal test —

The spleen of this rat was rubbed into the shaved belly of a guinea-
pig, which, however, remained healthy.

III. — Mus decunnanus brought to the laboratory dead.

Naked-eye appearances —

Putrid ; organs discoloured ; no enlarged glands.
Microscopically —

Preparations of the spleen and heart-blood showed some bipolar
organisms and many putrefactive bacteria.
Animal test —

The spleen was rubbed cutaneously into a guinea-pig weighing 180
grammes. No decrease of weight occurred, but on the tenth day the
inguinal glands were slightly enlarged. At the end of a fortnight the
animal was chloroformed to death, and the following appearances
noted : — The left inguinal gland was somewhat enlarged, but showed no
signs of congestion ; the pelvic glands were also larger than normal.
There was a very slight local reaction. The spleen was somewhat
enlarged, and contained a few grey nodules.

Microscopically —

Preparations from the heart-blood and spleen showed no bacilli.
A culture of the bubo yielded a pure growth of Gaertner's bacillus.

IV. — Liver of a Mus decumanus dead of a natural infection of

Plague.

The liver of a mus decumanus, dead of a natural infection of plague,
was covered with small grey nodules, in which fairly numerous plague
bacilli were found, on microscopical examination. It was rubbed into

(295)

lO QUATERCENTENARY STUDIES IN PATHOLOGY

the shaved skin of a guinea-pig, which died in six days of typical plague.
Although the animal died of undoubted plague, yet from the heart-blood
a pure growth was obtained of numerous colonies, conforming in every
way to the type of B. enteritidis.

MacConkey gives a similar instance in the case of guinea-pigs dying
from experimental inoculations of B. typhosus and B. mallei.

V. — Mus decumanus brought to the laboratory dead.

Naked-eye appearances —

Spleen enlarged and congested ; liver pale, and covered with fine
necrotic areas ; some pleural effusion ; lungs ha^morrhagic.
Microscopically —

Spleen — many bipolar bacilli.

Liver — a few bipolar bacilli.

Heart-blood — no bacilli seen.
Animal test —

The spleen and liver were rubbed into a shaved area of skin of a
guinea-pig weighing 280 grammes. At the end of a week it was
chloroformed to death. Examination showed a small right inguinal
gland, with glairy contents, slight cutaneous reaction, and a nodule in
the spleen the size of a millet seed. Cultures from heart-blood were
sterile, but from the gland and spleen nodule the usual organism was
obtained.

From the foregoing details, it may be asserted that organisms of the
" Gsertner " type occasionally occur in rats, and that if the cutaneous
method in guinea-pigs be employed for purposes of diagnosis, an
infection by this organism may result. Confusion from this cause may
easily be avoided if the possibility of such an infection be remembered.

Additional emphasis may be given to the fact that the organism may
be recovered in pure culture from the heart-blood of a guinea-pig with
well-marked lesions produced by another bacillus. Error in diagnosis
might, in some cases, possibly result if this circumstance were overlooked.

It only remains, in the concluding part of this paper, to compare the
lesions found in diseases caused by other Gaertner-like organisms with
those already described. These are, viz. — the bacillus of swine-pest,

(296)

ON THE ACTION OF CERTAIN BACTERIA II

B. icteroides, B. bovis morbificans Basenau, the bacillus of intestinal
diphtheria of Ribbert, B. psittacosis Nocard, and B. diphtherias colum-
barum Loeffler.

Bacillus suipestifer, Kruse. (Bacillus cholerae suum, Migula).

The following description is abstracted from the work of Nocard
and Leclainche (4).

In the acute form of the disease necrotic foci are seen in the
sub-mucous tissue of the tongue, and consisting of greyish-yellow caseous
masses, which may involve the mucous membrane and finally give rise to
ulceration — the ulcers being covered with diphtheritic deposits ; the
necrosis may extend into the underlying muscles. The tonsils contain
hsemorrhagie foci or cheesy nodules. The lesions in the stomach and
intestine are variable in their intensity, and consist of an exudative
inflammation of the mucous membrane, followed by necrosis of the tissues.
Usually the spleen is normal ; in exceptional cases it contains caseous
foci. Multiple nodules of necrosed tissue often occur in the liver.

In the chronic form, caseous broncho-pneumonia is met with in
addition to the intestinal lesions.

Young animals are especially prone to the disease. The bacillus
is pathogenic to the guinea-pig, the rabbit, the white and grey mouse, the
white rat and the pigeon.

B. icteroides Sanarelli. — From the researches of Reed and Carrol (5),
this organism is closely related to B. suipestifer, and produces identical
lesions in guinea-pigs, mice, rabbits and dogs.

The B. bovis morbificans was isolated by Basenau, in 1893, from the
muscles and organs of a cow killed on account of puerperal metritis.
In several instances the spleen was enlarged, and greyish-white nodules
were found in this organ and in the liver. It is pathogenic by feeding,
with symptoms of gastro-enteritis to guinea-pigs, mice, rats and calves,
and frequently causes, in inoculated animals, numerous miliary purulent
and necrotic foci in the liver and spleen.

B. psittacosis Nocard. — Lehmann and Neumann state that it is
not clear whether this organism should be placed in the hasmorrhagie
septicaemia group, though they point out that it has a similarity to

(297)

12 QUATERCENTENARY STUDIES IN PATHOLOGY

B. enteritidis Gaertner. It is significant that Nocard and L^clainche
class it amongst Coli-bacillary infections. Confirmatory evidence of its
close relationship to B. enteritidis is furnished by MacConkey, who finds
that, in their sugar reactions, the two organisms closely correspond.

Nocard states that the lesions in naturally infected animals (parrots)
are those of an fntense septicaemia. Wolf had previously (1882) described
a similar disease in parrots, and had noted in the liver gray or grayish-
white hard nodules the size of a millet seed or even larger. These were
also present, though smaller and more scattered in the spleen and
kidneys. Nocard and Leclainche admit the possibility of the two
diseases being identical.

B. der Darmdiphtherie des Kaninchens^ Ribbert{^). — This organism was
met with chiefly in pregnant rabbits. Apart from fibrinous peritonitis,
the liver and spleen showed numerous minute foci and nodules, in which
the liver cells were necrotic. Inoculation reproduced all the morbid
changes met with in the spontaneous form of the disease, inclusive of
the diphtheritic condition in the intestine.

B. diphtherice columbarum^ Loeffler. — Lehmann and Neumann
investigated a culture obtained from Krai, and found that in its
morphological and biological characters it exactly resembled B.
enteritidis. In pigeons affected by the disease, yellow, prominent,
fairly adherent spots occur on the mucous membrane of the mouth.
These spots consist of a cheesy material. Caseous pseudo-membranes
partially cover the mucous membranes of the mouth, pharynx and nasal
cavities. The bacillus is pathogenic to the rabbit, when given intra-
venously. Mice are killed with a septicaemia, whereas in the guinea-pig
and dog only an abscess is produced at the site of inoculation.

It is thus evident that organisms belonging to the B. enteritidis
Gaertner group have a common character in the property they possess of
causing necrosis of the tissues selected by them for invasion. Another
fact of interest is the wide distribution of these organisms — a fact which
is a sufficient justification in attempting to trace their relationship to the
type organism of the group, namely, that concerned in a frequent
variety of food-poisoning, and indirectly to the still ill-defined class of
organisms responsible for those obscure cases of continued fever in man
— the paratyphoid bacilli.

(298)

ON THE ACTION OF CERTAIN BACTERIA 1 3

REFERENCES.

1. MacConkey. Journal of Hygiene, 1905, V., p. 333.

2. KovArzik. Centralbl. f. Bakt., 1903, XXXIII., Orig., p. 143.

3. LocHMANN. Centralbl. f. Bakt., 1902, XXXI., Orig., p. 385.

4. NocARD et L^CLAiNCHE. " Les maladies microbiennes des animaux."

Paris, 1903, IlI.e. Edit.

5. Reed and Carrol. Journ. of exp. Med., 1900-01, V., p. 251.

6. Ribbert. Deutsche med. Wochenschr., 1887, No. 8, p. 141.

(299)

THE RELATIONSHIP BETWEEN THE FACTORS
INDUCING HEMOLYSIS AND THOSE
INDUCING THE PHAGOCYTOSIS OF RED
BLOOD CORPUSCLES.

By Robert Donald Keith, M.A., M.D.,

Lecturer on Physiology, Medical College, Singapore ; late Carnegie Research Scholar.
(From the Bacteriology Laboratory, Londou Hospital, London, E. )

(301)

The Relationship between the Factors inducing Haemolysis and
those inducing the Phagocytosis of Red Blood Corpuscles,

A.— HISTORICAL.
Introductory.

The subject of the phagocytosis of red blood cells is an interesting one
from several points of view. It is a phenomenon which occurs in the
animal organism under various conditions, and all these processes in
which certain cells become the prey as it were of others belonging to the
same animal organism must be of much interest to the pathologist.

An interesting example of phagocytosis of red blood cells occurring
in the human subject was recently examined by the writer. The case
was that of a woman who died of puerperal sepsis six days after
delivery. In the large blood sinuses of the uterus could be seen
multitudes of mononuclear cells swollen out with ingested red corpuscles,
and even amongst the muscle fibres such cells could be observed.
Owing to the alteration in form produced by the process of ingestion, it
was extremely difficult to say what the exact nature of these cells was.
They were large and round, the single nucleus was relatively small,
round in shape, and well-stained. Although it is probable that these
were mononuclear leucocytes, yet the possibility of their being connective
tissue cells had to be considered. In any case it was clear that these
cells were acting as scavengers, and were removing the superfluous
erythrocytes.

Exactly similar appearances were presented by sections of a
carcinoma of the breast in a mouse. These were very kindly brought to
the notice of the writer by Dr. F. W. Twort, assistant bacteriologist in

(303)

4 QUATERCENTENARY STUDIES IN PATHOLOGY

the London Hospital. Many of the alveolar spaces were filled with
blood, and lying amongst the red cells could be seen mononuclear
leucocytes closely packed with red blood cells.

Much interest, too, is lent to this subject by the close relationship it
bears to some other phenomena, which are due to the action of certain
body fluids, especially to haemolysis and agglutination. Of late a
considerable amount of discussion has arisen with regard to the relation-
ships of these various phenomena one to another, and in the course of
this, much light has been thrown on various questions in connection with
the immunity problem.

The histories of haemolysis and of phagocytosis, especially of red
blood cells, are so closely bound together that it may be useful to
recount briefly the various stages in the evolution of our knowledge of
these phenomena.

The pioneer in this work was Metchnikoff, and although much of his
work on the subject has been controverted, yet there can be no doubt
that his ideas and observations acted as a great incentive to research on
this problem. His work and that of his pupils will repay examination.

Metchnikoff's Views on Macrophages and Microphages.

In the course of his observations on phagocytosis, Metchnikoff was
led to the conclusion that various atrophic processes, such as progressive
muscular atrophy, senile atrophy, greying of hair, &c., were to be
attributed to what he termed the phagocytic action of certain of the
body cells. The metamorphosis of various insects and amphibians he
considered to be due to the same cause. These processes he attributed
almost entirely to the action of macrophages, a term which he employed
to designate such cells of the fixed tissues as might take on a phagocytic
action for animal cells, and the large mononuclear leucocytes. The
polymorphonuclear leucocytes he termed microphages, and to these he
attributed the power of engulfing micro-organisms.

Metchnikoff's Theory of Cytases and their Origin.

He showed, too, that when one injects blood or other alien cells into
the peritoneal cavity or under the skin of animals, phagocytosis of these
cells ensues, the phagocytic agents employed being almost exclusively

(304)

HEMOLYSIS AND PHAGOCYTOSIS OF RED BLOOD CORPUSCLES 5

macrophages. In the peritoneal cavity, some time after such an injection,
an accumulation of leucocytes takes place. Amongst these are to be
found all kinds of white cells, but the mononuclear macrophages
predominate, the polymorphonuclear microphages playing quite a minor
role. The macrophages take up the alien cells, and proceed to digest
them by means of certain ferments which, according to Metchnikoff,
they contain. These ferments he termed cytases.

Relationship of Cytases to Haemolysis.

Now, the spleen and lymph glands are organs composed mainly of
macrophages, and if one make extracts of such organs with normal
saline solution, one obtains, according to Metchnikoff, an extract of
macrophages. Such extracts or emulsions cause haemolysis of red
blood cells, i.e., they cause the haemoglobin to diffuse out, This
haemolytic function of the macrophage extracts, he held, depended on a
special substance which they possessed. This substance was destroyed
by heat at 56" C. for f to i hour, i.e., it was thermolabile. Now, since
this property of haemolysis was found to be possessed by the blood
serum of animals into which red blood cells had been injected, it
appeared to Metchnikoff that the substance in the macrophage extract
was identical with that in the serum, and he was of the opinion
that in both cases it was an unformed ferment or cytase, which, since it
was derived from the macrophages, he termed macrocytase. The cytase
which the polymorphonuclear leucocytes contained, he called microcytase.

Views of other Researchers on this Subject.

These views on the cytases and their origin did not long remain
unchallenged. It was denied by Korschun and Morgenroth (i) that the
haemolytic action of macrophage extracts was due to a thermolabile
cytase, and in support of this view, they asserted that heating at 56° C
did not cause the disappearance of the haemolytic property of macrophage
extracts, and that they even bore the temperature of the boiling point
before giving up this power. They showed, too, that the haemolytic
substance in macrophage extracts was soluble in alcohol, and quite
different from the true enzymes or cytases, and therefore different from
the thermolabile cytase in the blood serum.

(305) V

6 QUATERCENTENARY STUDIES IN PATHOLOGY

Savtchenko and Berdnikoff (2), as well as Donath and Landsteiner (3)
and Domeny (4), all controverted the view of Metchnikoff that the
macrophages possess a thermolabile enzyme or cytase, which, passing
into the bFood, affords to this its haemolytic power. Metchnikoff, however,
received support from Levaditi (5), who found that such extracts, when
fresh, possessed this thermolabile substance, but when old, lost it, and it
therefore appeared probable to Metchnikoff that the macrophages
contained a thermolabile enzyme which causes dissolution (autolysis)
of the cells of such extracts as contain macrophages, and that in the
course of autolysis, among the newly formed substances, arises a number
of haemolytic substances of a thermolabile nature.

Levaditi further supported Metchnikoff by showing that fresh extracts
of lymph glands contained a macrocytase, and that by means of it
one could reactivate sera which had been rendered inactive by heat.

The position held by Metchnikoff was the following. He considered
that the macrophages of the lymph glands and other macrophage-forming
organs greedily devoured various cellular elements, amongst which were
red blood cells. They proceeded to digest these ingested cells, in which
process the thermolabile enzymes or cytases possessed by the macro-
phages played a prominent part. To this thermolabile enzyme or cytase
he attributed also the power of causing extra-cellular haemolysis of
red blood cells, and held that it gained access to the blood frqm the
macrophages, and thus lent to the serum its haemolytic power in cases in
which it was possessed of such.

These views were destined to be subjected to considerable discussion
and change. Bordet found that in animals into which blood had
been repeatedly injected, the haemolytic action of the macrophage
extracts was not increased, while that of the blood serum was
much augmented if already present, and produced de novo, if not
originally there. He demonstrated, too, that instead of a single
cytase there were two substances concerned in haemolysis, one
of which, alexine, was present to the same extent in normal as
in immune serum. This substance was destroyed at temperatures
between 56° C and 60" C. In addition to this complement, however, he
showed that in haemolytic sera there was another substance, which he
termed the substance sensibilisatrice, from the idea that it made the red
blood cells sensitive to the action of the alexine. Bordet showed that it

(306)

HEMOLYSIS AND PHAGOCYTOSIS OF RED BLOOD CORPUSCLES 7

was much less sensitive to the action of heat and other injurious
influences than was the alexine.

The next advance was made by EhrHch and Morgenroth, who
showed that the substance sensibilisatrice of Bordet could fix itself on to
red blood cells without producing haemolyis, which was brought about
by a thermolabile substance, the alexine of Bordet.

The substance sensibilisatrice of Bordet, they termed the amboceptor;
to the alexine, they gave the name complement, and showed that the
latter did not join on to the red blood cells directly, as Bordet supposed,
but probably fixed itself on to the amboceptor, which acted as a
connecting link between the red blood cells and the complement, and
not as a mordaunt, making the red blood cell more susceptible to the
action of the alexine, which was the view of Bordet.

Revised views of Metchnikoff.

Metchnikoff now accepted the view of the duality of these substances,
but proposed, for the sake of clearness, to use the term fixateur instead
of amboceptor or substance sensibilisatrice, and the term cytase instead of
complement or alexine.

The point, however, which mainly appealed to Metchnikoff was
whether the substances producing haemolysis had any relationship to the
phagocytes, and he turned again to the question of the origin of the
cytase or alexine, formulating the belief that the macrocytase present
in the macrophages of the lymph glands, pancreas aselli and spleen
was also to be found in the mononuclear leucocytes of the blood, lymph,
and exudates, and that it was identical with the hsemolytic cytase of the
blood serum. The grounds for his belief were the following : —

(i) If one immunised a guinea-pig with the blood of a goose, and
then injected some goose blood into the peritoneal cavity of the guinea-
pig, the alien red corpuscles were quickly haemolysed without a definite
phagocytosis taking place. If one examined the exudate resulting
from such a procedure, one found but few leucocytes, and even those
were in a state of inactivity, being non-mobile, clumped together, and
unable to engulf foreign bodies. They were, in fact, undergoing a process
of dissolution, which Metchnikoff termed phagolysis. According to
Metchnikoff, it was possible to prevent the occurrence of phagolysis by

(307) V I

8 QUATERCENTENARY STUDIES IN PATHOLOGY

injecting sterile broth or normal saline solution into the peritoneal cavity
24 hours previous to the injection of red blood cells. One thus inured
the phagocytes to insult, and no phagolysis was produced on the
subsequent injection of red blood cells. As a consequence, when the
erythrocytes were injected, phagocytosis went on readily without the
occurrence of extracellular haemolysis. The red cells were hajmolysed
only in the interior of the phagocytes.

From these observations Metchnikofif concluded that the ha^molytic
substances in the peritoneal exudate were set free from the leucocytes
through the occurrence of phagolysis, and that, as soon as the latter was
prevented, haemolysis ceased to occur in the peritoneal fluid and took
place in the interior of the leucocyte.

(2) Metchnikoff further asserted that the presence of alexine or
complement or cytase, which occurs in blood drawn from an animal into
an ordinary glass vessel, was brought about by injury to mononuclear
leucocytes, many of which were broken down during the process of
clotting, and thus set free their cytases. This opinion was supported by
the experiments of some of his pupils (^) who found that if a vein were
ligatured and then excised and suspended, the plasma, which could be
pipetted off, contained no cytase; and that, further, if blood were drawn
into paraffined glass tubes, no cytase was present, since in neither case had
any destruction of macrophages occurred. These views, which were subse-
quently supported by the experiments of Tarassewitch (J) and Levaditi,
have recently been controverted by Lambottc and Stiennon (S), who
have maintained that even when the methods recommended by
Metchnikofif were carefully carried out by them, the results obtained by
him and his pupils did not ensue. As a result of their experiments,
they denied the intra-cellular nature of the cytases. The question of the
origin of the cytases is a difficult one, and cannot as yet be regarded as
definitely settled.

The Relationship of the Substances in the Serum producing
Haennolysis to the Phenomenon of the Phagocytosis of red
blood cells.

After this short resume of the history of the substances in serum
which produce haemolysis, it will be proper to discuss the relationship of

(308)

HAEMOLYSIS AND PHAGOCYTOSIS OF RED BLOOD CORPUSCLES Q

these substances to the phenomenon of the phagocytosis of red blood
cells. Our knowledge of this subject was considerably enlarged by
Savtchenko (9), who immunised rabbits against the red cells of the
guinea-pig. The serum of the former naturally became haemolytic for
the red blood cells of the latter. Savtchenko showed that if one injected
sterile bouillon into the peritoneal cavity of a guinea-pig, and 24 hours
afterwards the washed red cells of another guinea-pig, along with a
quantity of the heated serum of the immune rabbit, no haemolysis took
place, but phagocytosis of these red cells was marked. On the other
hand, in a control guinea-pig injected with the red blood cells, but without
the immune serum, no phagocytosis took place. When he injected a
fresh non-immune guinea-pig with the red blood cells of another guinea-
pig along with the heated immune serum, but without having prevented
phagolysis by the appropriate measures, haemolysis took place owing to
the cytase or complement being set free by the destruction of the
macrophages, consequent on the injection. On injecting a larger
quantity of the red cells, both haemolysis and phagocytosis occurred.

He also performed experiments in vitro, using the same factors
as before, and found that under the influence of the immune haemolytic
serum, deprived of its alexine by heat, the red blood cells were engulfed
by the leucocytes of the guinea-pig which he used as the phagocytic
agents, and that no extra-cellular haemolysis occurred.

From his experiments, Savtchenko concluded that in the serum of a
rabbit rendered immune towards guinea-pig's blood, some body was
present which made the erythrocytes an easy prey to the leucocytes of
another guinea-pig. This was not the alexine, since the property was
possessed by sera which had been deprived of that body by heat, and
therefore Savtchenko concluded that it was the specific haemolytic
fixateur, the amboceptor, which induced the phagocytosis of red cells.
This body, he asserted, could act in two ways. It could either attach
itself to the phagocyte and stimulate it to engulf the red blood cells,
or, by attaching itself to the erythrocytes, could render them an easy
prey to the phagocytes. Metchnikoff agreed with Savtchenko, Tarasse-
vitch and Levaditi, that the specific fixateur or amboceptor induces
phagocytosis.

(309)

lO QUATERCENTENARY STUDIES IN PATHOLOGY

Relation of Fixateur to Opsonin.

This subject acquired a fresh interest from its bearing on the work of
Wright and Douglas (^o), Bulloch and Atkin (^^) and others, on
phagocytosis.

Wright and Douglas described the power which the blood serum has
of inducing the phagocytosis of micro-organisms as the "opsonic"
property, and the substance possessing this as the "opsonin," which,
in their opinion, was a substance not previously described, and which
acted on the micro-organisms or other bodies to be phagocytosed,
and not on the phagocytes.

It was subsequently asserted by Dean (j^) and others that this was
not a new substance, but that it was identical with the fixateur, using
that term in the sense in which it was employed by Savtchenko, />.,
regarding it as identical with the amboceptor.

Barratt (^3), however, has since shown that the specific haemolytic
fixateur may be absent in an immune serum, and that, nevertheless,
a substance inducing the phagocytosis of red blood cells of the kind
employed for immunisation may be present. This substance he classed
with the "opsonins."

The point at issue, then, is whether the substance which induces the
phagocytosis of red blood cells is identical with the specific haemolytic
fixateur or amboceptor, and the following experiments were carried
out with a view to elucidate this question.

B.— EXPERIMENTAL.

If the process of phagocytosis of red blood cells be due to the action
of the haemolytic amboceptor, as suggested by Savtchenko, then there
are two ways in which it may act.

(i) It might act alone by attaching itself to the red blood cells, thus
rendering them liable to be phagocytosed. This theory is supported by
Savtchenko, Metchnikoff, Dean, &c.

(2) It might, however, as Dean has also suggested, be aided in this
action by a complement, but if the complement were destroyed, might
still be capable in itself of inducing phagocytosis.

(310)

HAEMOLYSIS AND PHAGOCYTOSIS OF RED BLOOD CORPUSCLES II

In dealing with the first point, a question naturally suggesting itself
was how do these two substances, viz., that which induces phagocytosis
and the haemolytic amboceptor behave towards heat ? If the one be
partially destroyed while the other is not, then naturally one would
conclude that the two are not identical.

In order to procure information on this point, the following mode of
procedure was adopted : — A rabbit was immunised against the red blood
cells of the ox. The serum of this rabbit became haemolytic for these
red blood cells, '002 c.c. ultimately producing, when fully complemented
with fresh guinea-pig serum, complete haemolysis of 2 c.c. of a 5 %
suspension of the red cells, after two hours at 37° C. and 12 hours at
room temperature; '0001 c.c. produced a trace of haemolysis under the
same conditions.

The following series of tests were directed towards ascertaining
whether the haemolytic amboceptor (fixateur) was thermolabile or
thermostable.

Two separate portions of the serum were taken, one being heated at
55° C for 15 minutes, the other being left unheated. Two series of
corresponding dilutions were made from these, and to each tube 2 c.c. of
a 5 7o suspension of the washed red cells of the ox were added with
•2 c.c. of fresh guinea-pig serum as complement. Both series were kept
at 37° C. for two hours, and subsequently at 0° C. for 24 hours. The
corresponding dilutions showed exactly corresponding degrees of
haemolysis, even when examined by means of von Fleischl's haemometer.
These results, which are given in detail elsewhere (^'^), and which were
confirmed by repeated subsequent experiments, indicate that there is no
destruction of the haemolytic amboceptor after an exposure of 15
minutes to a temperature of 55° C.

Similar experiments were also made on the substance inducing
phagocytosis of red blood cells.

■ These effects were very different from those obtained by heating the
haemolytic amboceptor.

The method adopted for the performance of the phagocytic tests was
that invented by Wright for micro-organisms. Equal parts of serum,
of the usual 5 7o suspension of red blood cells and of the washed human
leucocytes were mixed in capillary pipettes, and placed at 37° C. for
15 minutes, films being then made, and stained with Leishman's stain.

(3")

12 QUATERCENTENARY STUDIES IN PATHOLOGY

In such tests it is a matter of some difficulty to compare heated and
unheated sera when undiluted, since the haemolysis produced by the
latter is apt to obscure the phagocytosis, but by means of a method of
dilution which was adopted, it was found possible to abolish the action
of the complement, and thus suppress haemolysis, while phagocytosis
could still be observed. By the employment of such a method of
procedure it was possible to demonstrate that in corresponding dilutions
unheated serum gave a higher phagocytic count than did heated.

The following experiment illustrates this : — The unheated immune
rabbit's serum was diluted so as to give proportions of i in 15, i in 20,
I in 30 and i in 60, and of each of these dilutions, one part was mixed
in a capillary pipette with one part of a 5 % suspension of the washed
corpuscles of an ox and the part of washed human leucocytes,
the final dilutions thus resulting being i in 45, i in 60, i in 90
and I in 180 approximately. At the same time a similar series
was made with serum which had been heated at 55° C. for 15 minutes.
A control was made, consisting of i part of "85 % saline solution, one
part of the suspension of red cells and one part of washed human
leucocytes. Each mixture, as soon as it was made, was placed for 15
minutes at 37° C, films being then made, and stained in the usual
manner. In each case, 100 polymorphonuclear leucocytes were counted,
and the degree of phagocytosis estimated from the percentage of the
leucocytes containing red blood cells. This method was found to be
more accurate in the case of red blood cells than that of estimating
the average number of the red corpuscles taken up per leucocyte, as it
was in many instances difficult to distinguish the individual erythrocytes.

In this particular experiment, the results were as follows (the
percentages refer to the leucocytes containing red blood cells) : —

Dilution. Unheated serum. Heated serum.

I in 45 65 per cent. 20 per cent.

I in 60 60 II 6 II

I in 90 21 I) 4 "

I in 180 8 II —
Saline control, 5 per cent.

From these and similar results, it is to be assumed that the substance
inducing the phagocytosis of red blood cells is partially destroyed by
heat, and that, therefore, it cannot be identical with the haemolytic
amboceptor or fixateur.

(312)

HEMOLYSIS AND PHAGOCYTOSIS OF RED BLOOD CORPUSCLES I3

This is corroborated by the results of experiments with normal
haemolytic sera. The serum of a normal guinea-pig is slightly
haemolytic for the red blood cells of the rabbit. The serum of a normal
guinea-pig, the washed red blood cells of a rabbit, and washed human
leucocytes, were employed for phagocytic tests in vitro. It was found
that many of the leucocytes employed as the phagocytic agents
contained blood shadows, that is, instead of red blood cells they
contained clear circular spaces. In most films, if 100 polymorpho-
nuclears were counted, a few contained one or two apparently unchanged
red cells, and a few contained erythrocytes, more or less haemolysed. It
was thus possible to trace the gradations between apparently normal
red cells lying in the polymorphonuclear leucocytes and clear round
spaces representing the erythrocytes in a haemolysed condition.

If, however, with such a serum heated at 50° C. - 60° C. for 15
minutes, phagocytic tests were performed, no evidence of phagocytosis
was observed. Neither blood shadows nor red blood cells were to be
seen in the leucocytes, except in a very few instances, where one or two
polymorphonuclears contained a single unchanged erythrocyte. The
opsonic action of the serum had been abolished. Yet there was
undoubtedly a considerable amount of haemolytic amboceptor present,
since it is not destroyed by heat at the temperature mentioned above,
and therefore we cannot assume that abolition of the opsonic effect of
the blood serum is due to destruction of the haemolytic amboceptor.

That a destruction of the " opsonin " takes place may be seen from
the following experiment : — The opsonic power of the serum of a
guinea-pig unheated on the one hand and on the other heated at 55° C.
for 15 minutes, on the washed erythrocytes of the rabbit, was deter-
mined in the usual manner twice a day for three consecutive days, the
amount of phagocytosis being estimated by the percentage of poly-
morphonuclears containing blood shadows or red blood cells more or
less haemolysed.

:imation.

Unheated serum.

Heated

serum.

I.

37 per cent.

0

per

cent.

2.

47

0

3-

40 II

0

4.

47

0

5-

41 II

0

6.

48 II

0

(313)

14 QUATERCENTENARY STUDIES IN PATHOLOGY

It was evident from these figures as well as from many similar results
that a great destruction of the normal " opsonin " took place at 55° C.

Again the unheated blood serum of the eel is strongly haemolytic for
most red blood cells, yet, when heated, it appears to exert no opsonic
effect on red blood cells for which it is haemolytic. It is impossible to
perform phagocytic tests with eel's serum in the unheated condition,
since it is markedly toxic for leucocytes, and, by dissolving these,
renders such tests futile.

Example, — An eel's serum, of which when unheated '05 cc. produced
almost complete haemolysis of 2 cc, of a 10 % suspension of guinea-pig
corpuscles, after 2\ hours at ly"" C, was used. Such a serum when
heated no longer produced haemolysis, nor had it any power of producing
phagocytosis of the red blood cells of the guinea-pig when tested in the
usual way. In films in which 100 polymorphonuclears were enumerated
none contained any red cells, although we may assume that haemolytic
amboceptor was present in a very large amount.

In diluted immune sera the haemolytic amboceptor may be present,
but without being able to render the corresponding red cells suitable for
phagocytosis.

Experiment. — Into two series of suitable test tubes 'Oi, "005, '003, '002
cc. of the immune rabbit's serum was placed. To each tube was added
2 cc. of a 5 7o suspension of the washed red cells of the ox, and all were
made up to equal volumes with '85 saline solution. One series was
complemented with fresh normal guinea-pig serum, the other being left
uncomplemented. The immune serum was present in the mixtures in
the proportions of I in 220, i in 450, i in 730, and i in 1 100 respectively.
A control consisted of a similar mixture containing the serum in the
proportion of i in 10. All the mixtures were placed for 2 hours at 37°
C. At the end of that time the following results were noted : —

Amount of serum. Complemented series. Uncomplemented series.

•01 CC. Haemolysis almost complete. No haemolysis.

•005 cc. M marked. n

•003 cc. II slight. II

•002 cc. II slight. I!

The tubes of the uncomplemented series and the control were then
well shaken up and centrifugalised. The supernatant fluid was pipetted
off, and replaced by normal saline solution. This was repeated thrioe.

(314)

HEMOLYSIS AND PHAGOCYTOSIS OF RED BLOOD CORPUSCLES 1 5

The deposits, thus freed from all traces of serum, were then mixed with
equal parts of '85 % saline solution, and were rapidly drawn up and
down in capillary pipettes to produce a uniform suspension.

Equal quantities of each deposit and of washed human leucocytes
were mixed in capillary pipettes, which were then kept at 37° C. for 15
minutes. Films were made and stained in the usual manner.

In the control the phagocytosis of the red cells was marked, 93 % of
the leucocytes containing red cells, but in the dilutions of i in 220, etc.,
no phagocytosis could be seen, altough 100 leucocytes in each case were
counted.

It was scarcely to be supposed that the absence of phagocytosis was
to be attributed to lack of the haemolytic amboceptor in these dilutions,
since haemolysis took place, in one case to a marked extent, in the
corresponding complemented tubes under conditions similar to those in
which the uncomplemented series used in the phagocytic tests was
placed.

The natural conclusion to be drawn from these experiments is that
the haemolytic amboceptor is not identical with the substance inducing
the phagocytosis of red blood cells.

As was indicated previously in this paper, the possibility has been
suggested by Dean that the opsonic action may be induced by a
complement acting through an amboceptor.

The complementary part of the action would be suppressed by heat,
while the amboceptor might go on acting in such a way as to still cause
phagocytosis.

It is very difficult to disprove this theory by direct experiment for
several reasons. The complement is destroyed at the same temperature
as the opsonin. If one add fresh complement, i.e.y fresh guinua-pig
serum, it is impossible to exclude the effect of the inherent opsonic
property it possesses. Thus it is impossible to say how far the increase
in opsonic power in such a case might be due to a complementary action,
apart from the opsonic property possessed by fresh sera generally.

To this view of Dean's, however, there are various objections.

In the first place, we have as yet no example of such on action of
immune substances. In the case of haemolysis, which is a typical
example of the action of amboceptor and complement, when we destroy
the complement by heat, no haemolysis takes place, notwithstanding the

(315)

l6 QUATERCENTENARY STUDIES IN PATHOLOGY

presence of a large amount of amboceptor. On the other hand, in
highly immune sera, the opsonic property of the serum, after heating,
may still be present to a marked extent, and in general, as is shown later
on, the more highly immune a serum is the less is the difference in opsonic
effect between the heated and unheated sera. Again, from the dilution
experiments recorded above, it is evident that, after the serum has been
diluted to such a degree that the complement is no longer present in an
amount sufficient to cause haemolysis, yet the serum in the unheated
state exercises a greater opsonic effect than in the heated condition in
corresponding dilutions.

Even if it be admitted that Dean's suggestion is feasible, yet to
explain the action one would have to suppose that the amboceptor
possesses two groups, one of which is a complementophilic, the other a
special function group. Thus, though the action of the complement
were suppressed, such an immune substance, in virtue of the action of
its special function group, might still be capable of inducing
phagocytosis. Even admitting this, the fact would still remain that the
substance is a new one, and not of the exact nature of any immune
substance previously described, and to such a substance the term opsonin
might reasonably be applied.

There are two further points to which reference may be made. The
first is the difficult question of the leucocytic origin of the complement.
Time has not permitted of experiments on this problem, but one
observation may be referred to. If one perform phagocytic tests in the
usual way, with an unheated haemolytic scrum, whether normal or
immune, human leucocytes, and the red blood corpuscles of an ox,
guinea-pig, etc., blood shadows appear in the leucocytes. This
phenomenon appears almost instantaneously at 37° C. In the case of
unheated guinea-pigs' serum and rabbits' corpuscles, it was found that,
after three minutes at 37° C, blood shadows appeared in the leucocytes.
On the other hand, using an immune heated haemolytic serum after one
hour at 37° C, the red corpuscles in the leucocytes were apparently still
unchanged. No haemolysis had taken place. But if, as Metchnikoff
and his followers have stated, the phagocytes contain the cytases or
alexines, then one should have expected haemolysis to take place inside
the leucocytes. The above observations, however, tend to show that
such is not the case.

(316)

HAEMOLYSIS AND PHAGOCYTOSIS OF RED BLOOD CORPUSCLES 1 7

The last point to which reference will be made is the question of the
destruction of the opsonin by heat. Wright and Douglas (loc cit), as
well as Bulloch and Atkin (loc cit.) and Hektoen and Ruediger {^^\
working with normal sera with Wright's method, found that this
substance was almost entirely destroyed by heat at 50° C. - 60° C. Dean
repeated this work not only with normal, but also with immune scrum,
using, however, a somewhat different mode of procedure, and found that
in normal, but especially in immune sera, there was less destruction than
the authors mentioned had supposed.

The following experiment illustrates this point, and shows that the
more highly immune a serum becomes, the less is the apparent
destruction of opsonin. This experiment was only a trial one, directed
to find the relationship of the curves of haemolysis and phagocytosis to
one another. Owing to the fact that the author had perforce to abandon
this work, no further experiments on the point could be carried out, but
from the results it is evident that the higher the immunity becomes the
less was the apparent destruction of the opsonin. The opsonic action is
limited in two ways. On the one hand, a certain number of leucocytes
constantly seem to be incapable of exercising, or exercise only with
difficulty, the property of phagocytosis, and on the other the capacity of
the leucocytes is not unlimited, especially for bodies of the size of red
blood cells. Thus there are two limitations. Suppose, e.g., that in a
given case, 90 ^/^ of the leucocytes used for a test are capable of acting
as phagocytes. Now, a highly immune serum, even in the heated
condition may be so active, and make the erythrocytes so easy to engulf,
that 90 °/^ of the leucocytes may contain red blood cells. But although
the unheated serum may be much more powerful than the heated, yet it
cannot produce more phagocytes, since the limit (90%) has been reached
by the heated serum. The greater the opsonic content of an immune
serum is, the more will this limitation come into play.

The method of experimentation was as follows : — A normal guinea-
pig was taken. Of its serum, one part was mixed in capillary pipettes
with one part of a 5 7o suspension of the washed red blood cells of the
rabbit and one part of washed human leucocytes. Phagocytic tests were
then performed in the usual way. The percentage of leucocytes
containing blood shadows was counted in the case of heated as well as
unheated serum. At the same time serum of a control normal

(3«7)

i8

QUATERCENTENARY STUDIES IN PATHOLOGY

guinea-pig was employed in the same way. Haemolytic tests were
performed by means of Wright's tubes, and without additional
complement.

After testing the serum of the guinea-pig, 3 cc. of the washed
corpuscles of a rabbit were injected intraperitoneally, and 5 days
afterwards 5 cc. of the same material. The serum was examined, as a
rule, daily for three weeks, from the time of the first injection, and the
results are to be seen in the accompanying chart.

"Iffi* :

Ddys

2345 67 8 9 10 U 12 1314 15 16 17 1819 20 21

§

'^

'^A

f

'^ i

1-4 h

1

1

y,

13 ^

1

h

'' \ h

*^^

f

i

1

IT i.

r^

r

if

10 i

/ i

1

I
1

f /

i

•9

•8

^^

'

.-'r

1

j

' i

I

f
1
I _

^

\

•

f_ _

\

"4

^

/

•3

f

/

0 r.

1

(

•——•= Total, opsonin.. „ , ^. ^

0- — ^ = Opsonin remaining after heading to55 C-

Fig. I.

The method of estimation was as follows : —

The opsonic effect of the serum of the immunised guinea-pig when
unheated, and when heated at 55° C. for 15 minutes, was compared with
that of the control animal.

(318)

HAEMOLYSIS AND PHAGOCYTOSIS OF RED BLOOD CORPUSCLES 1 9

In each case equal parts of the serum, of a 5 % suspension of the
washed corpuscles of a rabbit and of washed human leucocytes, were
employed.

In the case of the unheated sera the number of leucocytes in 100 con-
taining blood shadows was taken as a criterion, and in the case of the
heated serum the number of leucocytes containing red blood cells, since,
in the latter, no shadows appear.

The numbers so found for the heated and unheated sera of the
immune animal were divided by the number found for the control, and
thus figures were obtained representing the opsonic index. Thus, if in
the case of the unheated control serum 50 % ^^ the leucocytes contained
blood shadows or red blood cells, while with the immune serum in the
unheated condition 80 "/^, and in the heated condition 60 % ^f the
leucocytes showed erythrocytes or shadows or both in their interiors,
then the opsonic index, in the case of the heated serum, was i*2, and in
that of the unheated i'6.

Testing in this way, a curve (Fig. i) was obtained from which one
deduces that the more highly immune the serum became the less
was the apparent destruction of the opsonin by heat.

This experiment was only a trial one, but the principle involved
seems to be worthy of further exploitation.

I have to express my warmest thanks to Dr. William Bulloch of the
London Hospital for very much kind advice and assistance throughout
my work on this subject.

(319)

20 QUATERCENTENARY STUDIES IN PATHOLOGY

LITERATURE.

1. KoRSCHUN and Morgenroth. "Ueber die haemolytische Eigenschaften

von Organ-extracten." Bed. klin. Wochenschr., 1902, No. 37.

2. Savtchenko and Berdnikoff. Quoted by Metchnikoff in Kolle-

Wassermann's "Handbuch d. path. Mikro-organismen," 1904, IV., p. 353.

3. DoNATH and Landsteiner. "Zur Frage der Makrocytase." Wien klin.

Rundschau, 1901, No. 40, p. 733.

4. DoMENY. " Stammt die wirksame Substanz der haemolytische Blutfliissigkeit

aus den mononuklearen Leukocyten." Wien klin. Wochenschr.,

1902, Oct. 2.

5. Levaditi. "Sur les hemolysines cellulaires." Annales de ITnst. Pasteur,

1903, XVIL, p. 187.

6. Gengou. "Contribution a I'etude de I'origine des alexines der serums

normaux." Annales de ITnst. Pasteur, 1901, XV., p. 68.

7. Tarassewitch. " Sur les cytases." Annales de ITnst. Pasteur, 1902,

XVI., p. 127.

8. Lambotte and Stiennon. "Alexines et leucocytes." Centralbl. f.

Bakteriologie, 1905, XL., p. 224.

9. Savtchenko. " Du role des immunisines (Fixateurs) dans la phagocytose."

Annales de ITnst. Pasteur, 1902, XVI., p. 106.

10. Wright and Douglas. "An experimental Investigation of the role of the

blood fluids in connection with Phagocytosis." Proc. of the Royal Soc,
1903, LXXII.

11. Bulloch and Atkin. "Experiments on the Nature of the opsonic

Substances in the Blood Serum." Proc. of the Royal Soc, 1905, LXXIV.

12. Dean. " An experimental Enquiry into the Nature of the Substance in Serum

which influences Phagocytosis." Proc. of the Royal Soc, 1905.
B. Vol. LXXVI.

13. Barratt. "On the Phagocytosis of Red Blood Corpuscles." Proc. of the

Royal Soc, 1905, B. Vol. LXXVI.

14. Keith. "On the Relationship between Haemolysis and the Phagocytosis of

Red Blood Cells. Proc of the Roy. Soc, 1906, B. Vol. LXXVIL, p. 537.

15. Hektoen and Ruediger. "Studies in Phagocytosis." Journ. of Infectious

Diseases, 1905, II., p. 128.

(320)

AN EXPERIMENTAL ENQUIRY INTO THE
RELATIONSHIP OF LEUCOCYTOSIS TO
THE OPSONIC CONTENT OF THE BLOOD
SERUM.

By John Charles Grant Ledingham, M.A., B.Sc, M.B., Ch.B.,

Assistant Bacteriologist^ Lister Institute of Preventive Medicine;

formerly Carnegie Fellow in Pathology ;

and

WnxiAM Bulloch, M.D., Bacteriologist to the London Hospital.

(From the Bacteriological Laboratory^ London Hospital^ London^ E.)

(321)

W

An Experimental Enquiry into the Relationship of Leucocytosis
to the Opsonic content of the Blood Serum,

I. — Introduction.

The following experiments were undertaken with a view to determine
whether and to what extent the opsonic content of the blood serum,
towards various micro-organisms, is influenced by changes in the number
and character of the circulating leucocytes. The general association of
leucocytosis with natural and experimental bacterial infection, has long
been observed as a more or less constant phenomenon, without any very
exact conclusions having been reached regarding the precise role played
by such leucocytic variations.

It is obvious that this response on the part of the white cells to
infection of the organism, must form a very important link in the chain
of protective mechanisms called into play by the animal body, but, where
opinions have differed and still differ, is in the interpretation of the
comparative rdles played by the cellular elements and the surrounding
fluid medium, and in the rationale of their inter-dependence, if such
exists.

It has been shown by Wright and Douglas (i) and Bulloch and
Atkin (2) that in comparing in vitro the opsonic content towards micro-
organisms, of normal with pathological sera, the variant is the serum,
while the leucocyte may be regarded as a more or less indifferent factor :
that is to say, normal leucocytes may be employed instead of those
corresponding to the serum to be tested, without incurring any
appreciable error.

It may here be observed, however, that in certain diseases of the
haemopoietic system, one of us (Ledingham) has put on record evidence

(323) W I

4 QUATERCENTENARY STUDIES IN PATHOLOGY

showing that the ultimate agents concerned in phagocytosis — the
neutrophile leucocytes— may vary greatly in their phagocytic power,
quite apart from changes in the opsonic content of the serum. But
these results, though interesting, do not affect the main proposition that,
under conditions which do not primarily involve the haemopoietic system,
the serum is the main variant.

If then the phagocytic power of the neutrophile leucocyte remains
approximately constant both in normal and pathological conditions, two
questions arise which we shall endeavour to answer in this paper.

1. Will the presence of a leucocytosis in the circulating blood have

any effect on the opsonic content of the serum, either in the
way of raising or lowering it ? And, in this connection, is
there any evidence pointing to a probable origin of opsonic
substances from products of degenerated leucocytes ?

2. If the leucocyte, normally, has a constant phagocytic property, does

this relationship hold good in the stage of hyperleucocytosis ?

Both of these questions are obviously of great interest from the point
of view of vaccine-therapy the immediate effect of which, at least, is to
cause some disturbance of the blood picture. For the production of
experimental leucocytosis, we decided that it was of great importance to
choose fairly simple chemical bodies such as sodium cinnamate, tallianine,
etc., and to avoid bodies of more complex chemical constitution such as
bacterial proteins and sterile culture emulsions, which, though causing on
inoculation definite leucocytic changes (Romer(3), Schlesinger (4), Holmes
(5), and others), have been shown by Wright and Douglas to raise the
opsonic content of the serum.

This method of attacking the problem by inoculating leucocytosis-
producing substances has been much employed and elaborated since
Buchner's (^) first application of it to the problem of the nature and
origin of the alexine of normal serum.

The work of Nuttall (7), Nissen (8) and Buchner, which went to show
the presence of a thermolabile bactericidal body in cell-free serum,
was followed by numerous researches, notably of Hankin (9), Denys (^o),
Wasserman (^i), Buchner (/oc. czt.), Schuster (^^), Schattenfroh (^^'), etc.,
which had for their object the demonstration of bactericidal substances
elaborated by the leucocytes themselves.

(324)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 5

Hankin and Kanthack (loc. cit), Denys and Kaisin (i^), showed that
the bactericidal power rose or fell according to the variations in the
number of leucocytes whether intravascularly or extravascularly or in
exudates; but from these experiments, involving as they did the continued
vitality of the leucocytes, and thus of continued phagocytosis, no definite
conclusions could be drawn as to the r61e played by each in the process.

Buchner and Schuster employed substances like aleurone emulsions
or gluten -casein, which they injected into the pleural cavity, thereby
obtaining large quantities of fluid rich in leucocytes. By a process
of alternate freezing and thawing the leucocytes . were killed, and the
mixture was found to have stronger bactericidal powers than the blood
serum.

Other methods of obtaining extracts of dead leucocytes were
employed by Lowit (^^) and Schattenfroh, the latter of whom was led to
the conclusion that the bactericidal substance obtained from dead
leucocytes is relatively thermostable, and is not identical with the blood
alexines.

The whole question of the relationship of the bactericidal body
in leucocytic extracts to the alexine is still sub judice, and, further,
the view of Metchnikoff, that alexine is formed only when the leucocytes
are disintegrated intravascularly or in vitro in the process of coagulation,
has been disputed strongly by various authors, and more recently by
Lambotte and Stiennon (^^). These latter observers have recently
employed very careful methods for the preparation of plasma entirely
free from leucocytes, which, however, are not destroyed in the process.
Their results go to show that alexine circulates as such free in the
plasma, for they find that the bactericidal power of the plasma so obtained
is equal to that of the corresponding serum after coagulation.

The above considerations have an interest for us in this connection,
only in so far as the opsonic substance has been regarded by Wright and
Douglas and Bulloch and Atkin (loc. cit.) as a simple thermolabile
substance like the alexine, but which, in marked contradistinction to the
latter, does not require for its efficient action the co-operation of a
thermostable immune body.

The latter proposition has, however, been recently disputed by Dean
(i7)^ who regards the opsonic substance or the substance concerned in the
preparation of bacteria for phagocytosis as essentially thermostable,

(325)

6 QiJATERCENtENARY STUDIES IN PATHOLOGY

and which, though capable of acting to a certain extent independently,
may have its action enormously reinforced by co-operation with a
thermolabile complement. In a recent paper by R. D. Keith (^s),
the view is also put forward that the immune opsonin is partially
thermostable, but it is not identical with the haemolytic amboceptor,
as Dean believed. To explain the residual opsonic action that takes
place when immune serum has been heated, Keith suggests that, in
addition to its cytophilic and complementophilic groups, the hypothetical
opsonic amboceptor may possess a special thermostable opsoniphoric
group, enabling a certain amount of action to proceed in the absence
of complement. Into the problems involved in this highly theoretical
subject, it would be out of place to enter here.

We take as our main premiss, the standpoint, that the opsonin is a
simple thermolabile body residing in the blood serum, whose action
m vitro is exerted solely on the micro-organisms concerned, and is
entirely independent of the leucocyte.

The ultimate origin of the opsonic substance, like that of the
bactericidal and other protective bodies, is difficult to determine with
precision, and it was felt that light might be thrown on the subject,
not so much by Buchner's method of analysing separately the cellular
and fluid portions of the plasma as to their opsonic content (though this
also was attempted), as by estimating what influence, if any, was exerted
by the presence of hyperleucocytosis on the opsonic content of the
plasma, or, more correctly, of the cell-free serum after coagulation.

If the opsonic substance is ultimately derived from leucocytic
degenerative products, one would expect it to be profoundly influenced
by conditions in which the white cells are in great excess in the blood
stream.

The injection of many substances differing widely in chemical
constitution has long been known to be followed by a period of hyper-
leucocytosis, during which the natural resistance to infection is increased.
Indeed, it is probable that in many of the so-called cases of cross-
immunisation or vaccination with sterile emulsions of bacteria, whose
origin is entirely different from those against which they exert a
protective action, the results obtained are attributable solely to the
leucocytic variations produced by them.

It will be convenient here to summarise briefly the protective effects

(326)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 7

that various authors have recorded by the employment of substances
producing leucocytosis.

The first experiments in this direction were made by Loewy and
Richter (^9), who succeeded in protecting rabbits against experimental
pneumococcic infections, by injecting substances like spermin, pilocarpin
and albumoses. These experiments were further extended by Jacob
(20), who also noted that animals inoculated in the leucopenic period,
following the injection of albumoses, always succumbed, while those
infected during the period of hyperleucocytosis were markedly protected.

Hahn (21) employed yeast-nuclein in dogs and tuberculin in man, and
showed that the blood drawn during the stage of hyperleucocytosis had
a higher bactericidal power than normal blood. The period of immunity
was, however, always very short, and in striking contrast to that obtained
by previous inoculation of extracts or sterilised cultures of the organism
employed for infection.

Very complete and convincing experiments were made by Pfeififer
and Issaefif (22) in experimental cholera infection. They showed that the
inoculation of guinea-pigs with nuclein, serum of cholera convalescents,
serum of healthy persons, bouillon, urine, and salt solution, gave a
protecting power which diminished in the order of the substances named.

Other subtances as tissue-fibrinogen (Wooldridge (^3) )j testicle
emulsion (Zacharofif (^4) ), thymus and lymph-gland extracts (Brieger,
Kitasato and Wassermann (^5) )j gluten-casein, legumin (Buchner, loc.
cit), were found, on injection, to give varying degrees of protecting power,
attributable to the hyperleucocytosis produced by them.

With these and many other substances, different observers have,
however, frequently obtained different results, a fact which is not
surprising when one considers the well known differences in the
individual response to such substances, the mode of administration, and
the amount inoculated.

In all these instances of increased natural resistance associated with
hyperleucocytosis, it would be exceedingly interesting to ascertain with
exactitude how the two bodies amboceptor and complement — concerned
in the destruction of micro-organisms — were respectively influenced by
the leucocytic variation. Very few results, however, are forthcoming in
this direction, but it appears from the work of Nolf (^6), Muller (^7)^
Wassermann (Joe. cit.) and Longcope (^s) that it is the complement and

(327)

8 QUATERCENTENARY STUDIES IN PATHOLOGY

not the amboceptor which is most markedly influenced by the injection
of leucocytosis-producing substances.

Bulloch has also shown that the inoculation of sodium cinnamate
raises the complement content, but leaves the haemolytic amboceptor
quite uninfluenced.

We shall refer later to the bearing of these facts on the experimental
results obtained by us, and which are presently to be detailed.

Technique of Experiments.

The effect of all substances employed in the production of leucocytosis,
was tested both on rabbits and guinea-pigs, the inoculation being made
intraperitoneally in the case of the latter, and either intravenously or
intraperitoneally in the case of the former. In the former, also, the
subcutaneous method was once or twice employed. Immediately before
the inoculation, samples of blood for counting purposes (total and
differential leucocyte counts), and estimation of opsonic content of serum,
were drawn from the posterior auricular vein. In the guinea-pig one
occasionally has a little difficulty in obtaining satisfactory amounts of
blood from the ear, but, with some practice, one can usually obtain quite
a large supply by making a smart puncture in the vein situated close to
the margin of the ear, and exercising a little pressure. At the same time
blood was also drawn from another animal, to act as control in the
estimation of the opsonic index in the experimental animal. Whenever
a sample of blood was drawn from the experimental animal during the
period of observation, a similar sample was drawn from the control
animal. In many cases this was found to be unnecessary, as we shall
show {cf. infra).

The blood counts were made by the usual haematological methods.
For the opsonic determinations the technique recommended by Wright
was carried out as follows : —

1. The blood sample was collected in a glass capsule with a recurved

limb. When coagulation had set in, the tube was hung in
a centrifuge, which brought about the separation of the clear
serum.

2. In most cases the leucocytes employed were of human origin, and

were obtained by allowing the blood to run into a glass capsule

(328)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 9

containing sodium citrate (i % dissolved in '85 % NaCl.
solution). The citrated blood was then centrifugalised, the
supernatant citrated plasma being removed. The deposit
of corpuscles was subsequently washed in a large volume of
saline solution, and finally, the corpuscles were brought down
as a deposit in the centrifuge, the supernatant saline solution
being then removed.
3. A uniform emulsion of the microbe to be tested was made in
saline solution. The serum, corpuscles and microbe emulsions
were then introduced in definite volumes into capillary pipettes,
and, after thorough mixing of the contents, the latter were
sealed, placed in the incubator for 15' at 37° C, at the end
of which time films were made of the mixtures, and after
suitable staining (Leishmann, or Ziehl-Neelsen), the number of
bacteria ingested by a number of leucocytes was counted. In
comparison with the number ingested in the presence of a
normal serum this yielded the phagocytic index :

Number of bacteria ingested by leucocytes in the presence of given serum opsonic
Number of bacteria ingested by leucocytes in the presence of normal serum "^ index.

It is manifestly impossible for one person satisfactorily to carry out
experiments of the kind detailed in this paper. We accordingly arranged
it so that one of us (L) performed the whole of the blood counts, the
other (B) devoting himself to the determinations of the opsonic content.
Each worked separately, and at the end of the experiment the results
were compared.

II. — Description of the normal blood-picture in the Rabbit and
Guinea-pig and its physiological variations.

Rabbit.

The total leucocyte count varies greatly in different rabbits.
Goldscheider and Jacob (^9) found a variation of 8000 to 14,000 leucocytes
per cub. mm., while Schlesinger {loc. cit.)^ from a series of 100 rabbits,
obtained an average count of 10,800, with a variation of 5000 to 17,000.
Schulz's (3o) average was 9900. Ricder (3i), Lowit (^s), Ruzicka (33),

(329)

10 QUATERCENTENARY STUDIES IN PATHOLOGY

obtained averages of 88449, I5>998"6, and 14,6777 respectively.
Brinckerhoff and Tyzzer (34) found a variation of 4000 to 12,000.

In our own experiments we obtained from a series of 25 normal
healthy rabbits an average leucocyte count of 10,533 with a variation of
4500 to 17,200.

Far more important than these large variations in different rabbits, is
the consideration of those minor variations which may occur in the
counts of individual animals independently of experiment, and which, if
neglected during experiments on leucocytosis, may give rise to fallacies.

It would appear sufficient to ascertain the leucocyte count of the
animal on three or four days prior to the experiment, such counts being
made at the same hour every day, and under similar conditions as to
feeding or fasting.

Brinckerhoff and Tyzzer {loc, cit.) recommend more elaborate
precautions. Before experiments on leucocytosis, they make careful
observations of the influence of fasting or feeding on the leucocyte count,
and finally select those rabbits only whose leucocytes decrease during a
period of fasting. For a short experiment, the animal should be allowed
to fast for 12 to 20 hours, and if during this time the leucocyte count
does not rise, the experiment may be carried out with the assurance that
for a time the variations in the leucocyte count will be unworthy of
consideration. Other influences to be reckoned with, such as loss of
body heat, shock and pregnancy, are obvious. In our experiments it did
not appear to us either advisable or humane to keep the animals without
food even during short experiments lasting six to eight hours. After the
inoculations the animals invariably showed a loss of appetite throughout
the course of the experiment, so that any small errors due to digestion-
leucocytosis might be neglected. Further, as we were able to demonstrate
in many control animals, the slight rises, due to feeding, came well within
the experimental error with our available counting methods,

Pohl (35), who investigated this question of digestion-leucocytosis,
came to the conclusion that the phenomenon was marked only in those
animals such as the dog, which received meals once or twice a day, and
was quite inconsiderable in animals like the rabbit, whose stomachs were
continually full.

(330)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM II

Regarding the leucocyte types met with in rabbit's blood, the classi-
fication usually adopted is as follows : —

(i) Lymphocytes.

(2) Large Mononuclears.

(3) Amphophiles or Polymorphonuclears. These correspond in

general morphology and function to the polynuclear neutro-
philes of man, but their granules are smaller and more ovoid,
and, unlike those of man, give a distinctly oxyphile reaction.
For the sake of convenience and uniformity of terminology,
we shall, in the course of this paper, refer to these cells as
neutrophiles, and apply the term neutrophilia to the condition
in which these cells are in excess.

(4) Eosinophiles with large ovoid oxyphile granules.

(5) Mast cells.

As to the relative numbers of these cells in the circulating blood,
Lowit {loc. cit.) found 797 7o polynuclears and 20*3 % niononuclears,
while Hirschfeld (36) found the lymphocyte percentage equal to or even
greater than the polynuclear percentage.

Schlesinger {ioc. cit.) obtained an average polynuclear percentage of
49, with a variation of 40 to 60.

Brinckerhoff and Tyzzer (loc. cit.) give more complete numerical
relations, which will be seen to correspond closely with our own results
obtained from a series of 25 rabbits —

Brinckerhoff and Tyzzer. Ledingham.

Lymphocytes ' 45-55 per cent. 44-8 per cent.

Large Mononuclears 2-8 w 6 n

Amphophiles 40-50 " 42*4 n

Eosinophiles o'5-i n it?

Mast cells 4-8 .1 5-8 !i

Does any marked variation in the differential count take place during
the day ? The question is important in view of the fact that in many of
our experiments the leucocyte response took the form not so much of a
hyperleucocytosis, as of a relative neutrophilia. It was therefore
important to ascertain whether control animals exhibited such variations
during a period of six to eight hours.

(331)

12

QUATERCENTENARY STUDIES IN PATHOLOGY

From a large series of estimations in control animals we shall quote
the following to shew that any variation which occurs in the differential
blood picture is exceedingly slight. The numbers are percentages : —

21 F'eb.

22 Feb.

23 Feb

'Small Lymphocytes

loa.m. 3-30 p.m. 5-3op.i
43 39 47

Ti. 4 p.m.
49

II a.m
49

Large Mononuclears

2

2

0

3

0

Rabbit I.

Amphophiles

50

57

48

45

46

Eosinophiles

0

0

0

0

0

.Mast cells

5

2

5

3

5

Small Lymphocytes

II a.m.
51

5 p.m.
42

Large Mononuclears

0

0

Rabbit II. '

Amphophiles
Eosinophiles
Mast cells

46
0
3

55
0

3

Small Lymphocytes

io*45 a.m.
60

3p.m
50

5-30 p.m.
49

Large Mononuclears

0

0

I

Rabbit III. ^

Amphophiles

35

46

43

Eosinophiles

0

0

0

.Mast cells

5

4

7

'Small Lymphocytes

II '30 a.m.

47

3"i5P.^
43

n.

5 '20 p.m.
52

Large Mononuclears

5

7

6

Rabbit IV. ^

Amphophiles

40

39

35

Eosinophiles

0

0

0

^Mast cells

8

II

7

'Small Lymphocytes

10 a.m.
28

I •30 p.m.
39

Large Mononuclears

14

9

Rabbit V. -

Amphophiles
Eosinophiles
.Mast cells

50
0
8

44
0
8

Small Lymphocytes

io'45 a.m.
34

i'45p.r
21

n.

5 p.m.
36

Large Mononuclears

16

28

II

Rabbit VL -

Amphophiles

46

48

54

Eosinophiles

0

0

0

.Mast cells

4

3

9

(332)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 13

In the following experiment the control animal was inoculated
intraperitoneally with 2 cc. of a 5 % solution of carbolic acid {vide
Exp. 17):—

24 Feb.

25 Feb.

10-30 a.m.

3P.m-

5*30 p.m.

1 1 "30 a.m

Small Lymphocytes

39

23

21

32

Large Mononuclears

2

0

0

0

Amphophiles

56

75

77

64

Eosinophiles

0

0

0

0

Mast cells

3

2

2

4

Total Leucocytes

5.950

5.500

5.900

5.400

Rabbit VIL

It will be seen that in this last case a slight relative neutrophilia
occurred, but in all the preceding cases the variations in the differential
count throughout the day were quite inconsiderable.

Guinea-pig.

Much less attention has been paid to the leucocytes of the guinea-pig
than to those of the rabbit.

Hirschfeld (loc. cit.) gave the following classification, but did not
record any numerical data : —

(i) Eosinophiles with large ball-shaped granules.

(2) Indulinophiles with small ball-shaped granules.

(3) Polynuclears without granulation (scarce).

(4) Mast cells.

(5) Large granulated lymphocytes.

(6) Small non-granular lymphocytes.

Hirschfeld's indulinophile cells, whose granules take up the indulin
from the eosin-aurantia-indulin mixture, correspond to the pseudo-
eosinophile cells of Kurloff, and in function to the poly nuclear cells
of man.

Runoff's (37) classification was as follows : —

(i) Polynuclear cells with pseudo-eosinophile granulation.

(2) Typical eosinophile cells.

(3) Nigrosinophile cells, differing from the true eosinphiles in this

respect only, that, in the eosin-aurantia-nigrosin mixture, their
granules take up the nigrosin tint while the eosinophile
granules are red.

(333)

14 QUATERCENTENARY STUDIES IN PATHOLOGY

(4) Typical lymphocytes.

(5) The vacuolated large mononuclear cells. These latter cells were

first noted by Kurloff in the blood of the guinea-pig. In their
morphology, they correspond to the large mononuclear
leucocytes of man. The protoplasm of these cells, however, is
highly vacuolated, and included in the vacuole is a round or
oval granular mass, staining a dull-red with Leishman's stain.
It is impossible here to discuss the question of the nature of
these bodies enclosed in the cytoplasm of the vacuolated cells.
The whole subject has recently been treated in detail by one
of us (Ledingham) (^9).
The following are the average counts (total and differential) recorded
by Kurloff, Burnett (38), compared with our own observations : —

Burnett. Kurlofif, Ledingham.

Small Lymphocytes

47

32

30

Large Mononuclears

10

17

21

Polynuclears

31

45

39

Eosins and Nigrosins

10

10

8

Mast cells

•3

—

•6

Total Leucocytes

10,897

12,600

12,312

In the control guinea-pigs, as in the control rabbits, the variations in
the total and differential counts throughout the day were found to be
quite trivial.

III. — Fluctuations in the opsonic index in healthy animals
during short periods.

Does the opsonic index in healthy animals vary throughout the day
or from day to day? To settle this important question numerous
investigations were made, some of which we may here quote.

Opsonic indices of two healthy rabbits, C and E, on five successive
days. (C taken as control.)

Date. Opsonic Index.

29/1/05 C=i*o E=i'o

30/1/05 C= I'o E = 1-2

31/1/05 C=ro E- -98

1/2/05 C=i'o E= '98

2/2/05 C = i'o E= '98

(334)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 15

Opsonic indices at different times during the day :

ii-30a.m. 3-15 p.m. 6-15 p.m.

Rabbit I. I'o -91 i-o

10 a.m. 3 p.m.
Rabbit II. I'o i*o

11 a.m. 5-20 p.m.
Rabbit III. ro -95

10 a.m. 3-30 p.m. 5*30 p.m.
Rabbit IV. i*o I'o -94

10-30 a.m. 3 p.m. 5*30 p.m.

Rabbit V. ro -98 1*3

(N.B. — This rabbit received 2 cc. of a 5 % solution of carbolic acid. Vide Exp. 17.)

11*30 a.m. 3-15 p.m. 5*20 p.m.

Rabbit VI. I'o I'o j-o

11 a.m. I p.m.
Guinea-pig I. I'o i-o8

ii'5oa.m. 3 p.m. 4*45 p.m.

Guinea-pig II. I'o 1-2 1-07

From the above results it will be gathered that during periods from
three to twenty-four hours the opsonic index shows very little variation.
In fact, the figures 12 and g may be taken as the upper and lower limits
of what one may either call " physiological opsonic variation " or more
probably " experimental error."

IV. — Details of experimental results.

I. — Sodium Cinnamate.

In Charts I. to VII., are represented, graphically, the leucocytic and
opsonic variations obtained by the inoculation of sodium cinnamate in
rabbits and guinea-pigs.

The various leucocyte types are expressed in absolute values per cub.
mm. The non-granular cells in which are included the small and large
mononuclears and transitional cells are classed together for convenience,
as it was invariably found that these cells reacted in the same direction
to the chemical agents employed for inoculation.

(335)

l6 QUATERCENTENARY STUDIES IN PATHOLOGY

In the following charts the polymorphonuclears are represented as "P,"
lymphocytes as " L," mast cells as " M," large mononuclears as " L M,"
and the eosin-nigrosinophiles as " E N." The total number of leucocytes
is represented as " T." " ST " and " TB " denote that the opsonic power
of the serum was tested against staphylococcus and tubercle bacillus
respectively.

Regarding the granular series, our attention must be primarily
directed towards the polymorphonuclear cells. Mast cell curves and
eosinophile-nigrosinophile curves are also entered in the rabbit and
guinea-pig experiments respectively.

The number of eosiiiophile cells in the rabbit, and of mast cells in the
guinea-pig, is too small, and presents too slight variations to be
conveniently charted. It has to be noted also that, in all cases, the
inoculation was made either immediately or very shortly after the first
sample of blood was drawn for purposes of total and differential count
and estimation of opsonic content.

The quantity of sodium cinnamate (Merck) inoculated, varied from
•5 to I grm.

Exp. I. (Chart I.) Rabbit. Sodium cinnamate i grm. dissolved in
10 cc. of sterile normal saline, inoculated intravenously.

A very considerable hyperleucocytosis (22,900) is apparent three to
four hours after inoculation, and is also marked at the sixth hour. The
rise was due entirely to the polynuclear series, the percentage of which
rose from 49 before inoculation to 88 and 84 at the third and sixth hours
afterwards, respectively, while the non-granular cells fell from an initial
percentage of 40 to 10 and 14 at the same periods thereafter.

The mast cells also fell both relatively and absolutely.

The opsonic variation during the period under consideration was
scarcely appreciable.

Exp. II. (Chart II.) Rabbit. Intravenous inoculation.

The leucocytosis obtained was very slight at the third to the fourth
hour after inoculation, and at the sixth hour the blood-picture had almost
returned to the normal. The very considerable rise in the relative

(336)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM

17

percentage of poly nuclear cells is striking, even in jthe absence of any
great absolute rise.

The opsonic content showed practically no variation.

£ucocybes

per
cub. m.m
22,000

21,000

f\

20,000
I9,000
18,000
17,000
16,000
15,000
14,000
13,000
12,000
1 1,000
10,000
9,000
8,000
7,000
6,000
5,000

I ^

V

/

\

/

/

1

P

,

1

i

\

s

•^^

L

2p0O

1,000

0

V-

"^

^-,

-0-..

M
— 0

Q.C I-O
0- .9

^^-^

-^ST.

Time

n-3o

A.M.

3-15

P.M.

5-20

P.M.

SODIUM CINNAMATE, Igrm.
RABBIT, (Introi-venous) "^

A

A

A

\

1

\

I

N

L

'N-

.-•

0- .

— 0— ,

M

^^•^

--JiT.

10 -45

A.M.

1-45
P.M.

5

RM.

SOD. CI N NAM., Igrm.
RABBIT,(inbravenous)

A

s

A

n/

r

V

V

L

«^*,

""^o^

.,-M

.ST.

n

A.M.

2-15

P.M.

5-15

P.M.

50D.ClNNAM.,-5gTm.
RABBIT,(intrd,venous)

^

\

/'

/

\

/

•

O-.

■~oM

.ST.

*

II

A.M.

5-20

RM.

S0D.CINN.-5grm
RABBIT, "
inbrdferitonea''

ft

/\

r

\

'

/

J

A

>

V

\^

— o—

M
.-- 0 —

.^T.

II

A.M.

3

RM.

5-30

RM.

SOD.CINNAM.,igrm.
RABBIT, (intraperit.)

II.

III.

IV.

(337)

1 8 QUATERCENTENARY STUDIES IN PATHOLOGY

Exp. III. (Chart III.) Rabbit. Intravenous inoculation.

Half the usual quantity of sodium cinnamate was inoculated. No
appreciable rise in the total count resulted, but the polynuclear percentage
again rose from 34 to 69, while the non-granular and mast cells fell
markedly.

No change took place in the opsonic content.

Exp. IV. (Chart IV.) Rabbit. Intraperitoneal inoculation.

In the peripheral circulation, no hyperleucocytosis took place, as has
so frequently been observed after intraperitoneal inoculation, but the
relative neutrophilia in the circulating blood is evidence of a possibly
much greater quantitative leucocytic reaction in the peritoneal area.

No change took place in the opsonic content after six hours.

Exp. V. (Chart V.) Rabbit. Intraperitoneal inoculation.

A very slight absolute rise in the leucocyte count was evident in the
peripheral circulation, but the relative neutrophilia was even more
pronounced than in the last experiment. The percentage rose from 44
to 80 in four hours, and was still high (76 %) ^^ the sixth hour.

No appreciable variation was noted in the opsonic content during the
period of observation.

Exp. VI. (Chart VI.) Guinea-pig. Intraperitoneal inoculation

(•5 grm.).

In spite of the fact that the inoculation was intraperitoneal, an
enormous rise of the total leucocytes in the peripheral circulation took
place four hours after injection. The period of hyperleucocytosis was,
however, very transitory, the count returning to its initial level at the 6th
hour. The rise affected solely the polynuclear series. It is notable, also,
that the relative neutrophilia persisted for some time after the hyper-
leucocytosis had disappeared.

No variation took place in the opsonic content.

Exp. VII. (Chart VII.) Guinea-pig. Intraperitoneal inoculation.

The guinea-pig employed, presented a slight initial leucocytosis with,
however, no derangement of the differential count, so that the figures,
though high (19,200), must be taken as normal for that animal.

(338)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM

19

The total leucocytes rose in four hours to 27,200, and this rise was
accompanied by a marked neutrophilia.
No opsonic variation was evident.

Remarks on Sodium Cinnamate
— Leucocytosis.

The intravenous inoculation of
sodium cinnamate in rabbits pro-
duced the most marked hyperleu-
cocytosis, while the intraperitoneal
inoculation produced little or no
rise in the total leucocytes of the
peripheral circulation. It must be
insisted, however, that the relative
neutrophilia, which is a constant
feature even in the absence of
hyperleucocytosis, is as much an
expression of leucocytic stimula-
tion as the latter, and differs from
it only in degree.

The duration of the period of
hyperleucocytosis after a single
inoculation was always short, in
some cases lasting only a few
hours. On examination, twenty-
four hours after inoculation, the
blood picture was invariably
normal.

The results obtained by us,
in the case of sodium cinnamate
leucocytosis, are in entire accord-
ance with those recorded by
Richter(39), Richter and Spiro(4o)
and Batty Shaw (^i), and in entire
disagreement with those of Char-
teris and Cathcart (^s), who evi-
dently missed the period of
VII. hyperleucocytosis entirely, by

(339) ■ XI

Leucocytee

per
cub. mm
22,000

21,000

20P00
19,000
lft,00O

17,000
16,000
15,000
14,000
13,000
12,000

11,000

10,000

9,000

8,000
7,000
6,000
5,000
4,000
3,000

2000

\

V

\

1

u

1

l\

1

k

1

\

X

1

\

\

'>

\L

1,000

0

^

^M.^

•-®LM

CD—

.-0...

>rfN

N

•-—

■--^

ST

Time

II

A.M.

3

RM.

5-30

P.M.

SODIUM CINNAMATE., 0-5 grm.
GUINEA-PIG , (intraperifconedl)

f

\

1

A

V

1

A

\

i

\

\

1

*L

H^

-^3E1

^

®^"

-®L>

I

O"

..flM

-^.S'l

II

A.M.

P.M.

SODIUM ClNNAMATEp-Sgrm

GUlNEA-PIG,(inCrdperitoneaJ)

VI.

20

QUATERCENTENARY STUDIES IN PATHOLOGY

taking their blood samples at 24-hour intervals. The assertion of
these latter observers that the leucocytosis produced is a mononuclear
one is evidently incorrect, and due to the same cause.

2. — Tallianine.

This substance, recently introduced by Stassano and Billon ("^3), is

formed by the action of ozone on essence of turpentine, the action being

stopped when there has been absorption of a quantity corresponding to

four volumes of ozone. Tallianine, which is supplied by Brigonnet et

Leucocybec

per
cub.m.m
18,000

I6p00

14,000

I2pOO

10,000

8pOO
6pOO
4P00
2pOO
0

/T

y

f

r

V

P

y

/

^

X

.^

V

y

o-^

*>^

L

C-.

.

^/

'^

0—

— 0—

-•^

.- oM

Q.C -9

0- .8

_^^

^«

*

^ST.

Time

I0-50
A.M.

n3o

A.M.

12- 50

P.M.

150

P.M.

6

P.M.

TALLIANINE, BABBIT, 2C.C. (jnbra-
venous)

/^

.^^^

— •T

**^

-^

-oP

0^

<■

^^

-^.

-^L

&.-.

— o—

— o—

— o

-K,M

,/^

^<^

Sw

— —

^ST

II
A.M.

12
noon.

I

P.M.

2

P.M.

3

P.M.

TALLIANINE, RABBIT , 4-5 c.c.

(intravenous)

VIII.

IX.

Gaubert (Paris),* in sealed tubes, is a clear watery fluid with a faint
terebene odour.

Exp. VIII. (Chart VIII.) Rabbit. Intravenous inoculation of 2 cc.

of Tallianine.

In view of the fact that Stassano and Billon, in their brief note on
the subject, asserted that the leucocytosis occurs very quickly after
* Brigonnet p^re et fils et Gaubert, 15 rue du Landy. Plaine St. Denis (Seine).

(340)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM

inoculation, and that its duration

per
cub.m.m
46,000

44,000

42,000

40^000

38,000

36,000

34,000

32,000

30,000

28,000

26,000

24000

22,000

20,000

18,000

16,000

14,000

12,000

10,000

8,000

6pOO

4,000

2,000

^

\

\

\

\

1/

ll

1 1

I

\

Ut

A

\

A

\

\

If

4

w^

^■^■B,

^^L

0- .

-- 0— .

~~rw—-

,-nM

1.-

1N_

^Sl

Time

n-50A.M

yApril

3 '

RM.

4-45

P.M.

UA.M.

6 April

10 April

rALUANINE, GUINM-HG, 5cjc(intrapent.)

X.

is always transitory, we made complete
blood and opsonic estimations every
hour during the first three or four hours.
Blood drawn 40 minutes after injec-
tion showed practically no change
either in the absolute or differential
count. Two hours after inoculation,
however, the total leucocytes had risen
from 8,700 to 14,500, and there was a
marked relative neutrophilia. At the
third hour the neutrophilia had still
further increased, though the total count
had slightly fallen. A final estimation
at the seventh hour showed that, in the
interval, the leucocytes had again
mounted upwards to 17,000, while the
neutrophilia was well maintained.
During the seven hours that the animal
was under observation, no change at
all took place in the opsonic content of
the serum.

Exp. IX. (Chart IX.) Rabbit. Intra-
venous inoculation (4*5 cc).

Hourly examinations were made
after inoculation, but practically no
change took place in the blood-picture
either in the direction of hyperleu-
cocytosis or relative neutrophilia.

The fluctuations in the opsonic
content were also trivial during the
period of observation.

Exp. X. (Chart X.) Guinea-pig. Intra-
peritoneal inoculation of 5 cc. of
Tallianine.
Three hours after inoculation, an

enormous rise had taken place in

(341)

22

QUATERCENTENARY STUDIES IN PATHOLOGY

the total leucocytes (46,300). The rise affected solely the polynuclears.
At the fifth hour the hyperleucocytosis and neutrophilia were maintained,

Lfiucocytes

per
cub.m.m
20,000

19,000

18,000

17,000

16,000

15,000

14,000

13,000

12,000

11,000

10,000

9,000

8,000
7,000
6,000

5,000

4,000

3,000

2,000
1,000

0

V

N

1

i

l^

/

f "•>•

-"••.^

/

/Ip

V

0 OJ i-o

•rt

^

Time

U45A.M
I0-4O5

3-15
P.M

530
RM.

5-30P.M
U-4-05

TALLIANINE, GUINEA-PIG, 5C.c.

(inbraperiboneal)

TALLIANINE, G-PIG, 5c.c

XI.

(intiraperiboneal) | G-PIG,(inCraper)
XIII.

XII.

and even at the end of 24 hours very little change had taken place.

(342)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 23

Unfortunately, no count was made at the end of 48 hours, but, at the end
of 72 hours, there was still a very slight relative neutrophilia, although
the total count had returned to practically normal limits.

Estimations of the opsonic content during the first 24 hours revealed
no change either towards tubercle bacilli or staphylococci.

Exp. XL (Chart XI.) Guinea-pig. Intraperitoneal inoculation
(5 cc. of Tallianine).
In this animal a fairly well-marked hyperleucocytosis and neutro-
philia were again registered, but the duration was much shorter, the
blood-picture returning to the normal at the end of 24 hours.

The opsonic content during the period of hyperleucocytosis was not
appreciably affected.

Exp. XII. (Chart XII.) Guinea-pig. Intraperitoneal inoculation
(5 cc. of Tallianine).

On this occasion, a slight leucopenia was apparent at the end of
the first hour, the fall affecting mostly the polynuclears. At the end
of the third hour, however, a very marked hyperleucocytosis and neutro-
philia were evident. At the fifth hour this condition was still present
though the neutrophilia had slightly diminished. At the end of 24 hours
there was still a well marked hyperleucocytosis, but the neutrophilia had
entirely disappeared. A late response on the part of the mononuclears
was now evident. The opsonic content, during the observation period,
presented only slight fluctuations.

Exp. XIII. (Chart XIII.) Guinea-pig. Intraperitoneal inoculation
(5 cc. of Tallianine).
At the end of three hours, the leucocytes had mounted from 7000 to
19,800, and relative neutrophilia was conspicuous. The opsonic content
was not estimated in this experiment, advantage being taken of the
hyperleucocytosis, to test the phagocytic power of the newly-formed
polynuclears (vide infra.).

Exp. XIV. (Chart XIV.) Guinea-pig. Intraperitoneal inoculation

(5 cc).
In this experiment, the animal was kept under observation for six
days after inoculation, and blood examinations were made every day.

(343)

24

QUATERCENTENARY STUDIES IN PATHOLOGY

XIV.

Three hours after
inoculation, the leuco-
cytes had risen to
26,600, and the poly-
nuclear percentage
from 37 to 74.

At the sixth hour,
the total count was
still high (20,000), and
the neutrophilia main-
tained.

At the end of 24
hours a count of 28,100
was registered, with a
corresponding neutro-
philia.

At the end of 29
hours the count fell to
17,600, while the neu-
trophilia increased.

At the end of 48
hours the total leuco-
cytes had again risen
to their highest level
during the experiment
(33,000).

At the end of 53
hours the total had
fallen slightly, with
little change in the
neutrophilia.

At the end of 72
hours the count had
fallen to 14,500, and
the neutrophilia
showed signs of dis-
appearing.

(344)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 25

The final estimation was made at the end of 120 hours, when the
leucocytes were found to have fallen to the normal level, and the
neutrophilia had completely disappeared.

The long duration of the period of hyperleucocytosis in this
experiment is remarkable, and, taken in conjunction with the previous
experiments, would tend to show that in the guinea-pig, at least, the
effect of the tallianine on the haemopoietic organs is powerful and
prolonged.

Remarks on Tallianine — Leucocytosis.

The leucocytosis due to tallianine is variable, and may be entirely
absent when intravenous injection in rabbits is employed.

Intraperitoneal inoculation in guinea-pigs invariably resulted in
marked hyperleucocytosis, which might persist over three or four days.

Neutrophilia is invariably a conspicuous feature. We did not find so
rapid a response to tallianine, on the part of the white cells, as did Stas-
sano and Billon. As a rule the response did not make itself apparent
before the second or third hour.

No injurious effects on the animals in question were ever noted.
' In one rabbit experiment (intravenous inoculation of 5 cc), not
charted, there resulted no change whatever either in the direction of
hyperleucocytosis or relative neutrophilia, though estimations were made
hourly for six hours after inoculation. It would seem that in the rabbit,
at least, tallianine must disappear very rapidly from the circulation
after intravenous inoculation, while the absorption proceeds more slowly
from the guinea-pig's peritoneum.

3. — Bouillon.

Bouillon has long been known to produce a local leucocytosis when
injected into the peritoneal cavity. (Pfeiffer and Issaeff {loc. cit.) and
Funck (44) ).

Exp. XV. (Chart XV.). Rabbit. Intraperitoneal injection of
10 cc. of a I °l^ sterile peptonised bouillon.

Before the inoculation, the animal was found to have a very low
leucocyte count (4,500), but four hours afterwards it had risen to 7,350,

(345)

26

QUATERCENTENARY STUDIES IN PATHOLOGY

and there was a fairly well-marked neutrophilia which was maintained at
the seventh hour after inoculation.

During the period of observation no rise took place in the opsonic
content.

4. — Filtrates from autolysed staphylococcus cultures.

Exp. XVI. (Chart XVI.) Rabbit. Intraperitoneal injection of i cc. of
autolysed staphylococcus culture ( = filtrate from 70 million cocci).

Five hours after injection there was no rise in the total count, but
there was a marked relative neutrophilia. At the seventh hour, there was

LeucocyCes
per m.m*
14,000

12,000

10,000

8,000

6,000

4,000

2,000
0

yv

>

/^

St

/

/^

-^p

/

rv--

— 0—

flD 1-6
"? 1-5
- 14
.9 1-3

0 1-2

<0 II

0 I-g

-^ST

Time

1045

A.M.

P^..

5-30

P.M.

RABBIT, PEPTONt-BOUILlDN, loc.c
(intraperiDoneal)

/s

T

/

0

y

A

V

/

>«::

-'-L

^

^

•V

/'^j^

-^P

0- .

— 0 —

^-o-«.

— -0—

-oM

A

/\

/ \

V

1

\

\

^

1

ST.

■^^V

10 A.M.
21-205

3-30

P.M.

530

P.M.

4 P.M.
22-205

IIA.M.
23-2-C)5

RABBIT, SrAPHYL0C0CCU5,autX)iysecl
culC. Ice.

A

>

Os

V

A

V

/

\

\t

V

/

V

/>

\ .

^^^

-\p

0—

— 0—

-~oM

/ST.

/

f

/

/

-•-

^

I0-30A.M
24-205

3

P.M.

5-30

P.M.

nao

A.M.

RABBIT, STAPHYLOCOCCUS,

auColysed culd.Zcc.

XV.

XVI.

XVII.

a slight rise in the total count and a still more marked neutrophilia. At
the end of 24 hours, the polynuclears had fallen to a percentage of 42,
and at the end of 48 hours they had returned to their initial level.

Regarding the opsonic content, no change took place in the first five
hours, but, at the seventh hour, the opsonic index rose to 17. The high
opsonic tide was only of short duration, and had completely disappeared
at the end of 24 hours.

(346)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 27

Exp. XVII. (Chart XVII.) Rabbit. Autolysed staphylococcus

culture (2 cc.)

A relative neutrophilia was apparent at the end of four hours, but no
hyperleucocytosis. At the end of 6| hours there was a very moderate
hyperleucocytosis of neutrophilia type, which had completely disappeared
at the end of 24 hours. On this occasion the opsonic index did not rise
at all during the first seven hours, but a sample taken at the end of 24
hours showed a fairly high opsonic index (1*6).

5. — Pilocarpine.

Ruzicka (loc. cit.) states that the immediate effect of an intravenous
injection of pilocarpine, in doses up to 75 mg., in the rabbit, is the
production of an enormous leucocytosis. In one case the leucocytes
rose to 235,000 almost immediately after inoculation of 50 mg. intra-
venously. He lays stress on the fact that the hyperleucocytosis is never
preceded by a period of hypoleucocytosis.

Ruzicka's results have not been generally confirmed. Spiro (^5)
found an initial leucopenia with pilocarpine, followed later by a period of
hyperleucocytosis of neutrophile type. In the two following experiments,
this phenomenon was strikingly brought out.

Exp. XVIII. (Chart XVIII.) Rabbit 60 mg. injected
subcutaneously.

Ten minutes after injection the leucocytes fell from 13,400 to 10,900,
and one hour afterwards to 9,100. The count then commenced to rise.
Three hours after inoculation, the leucocytes had reached their initial
level, and at the sixth hour, a marked hyperleucocytosis of neutrophile
type was evident.

A slight rise in the percentage of the lymphocytes during the first
half-hour was the only evidence of an initial mononucleosis.

During the experiment, no appreciable change in the opsonic content
took place.

Exp. XIX. (Chart XIX.) Rabbit. Intravenous inoculation (50 mg.).

Ten minutes after injection practically no change had taken place
either in the total or differential count. During the next twenty minutes,

(347)

28

QUATERCENTENARY STUDIES IN PATHOLOGY

however, a marked relative mononucleosis was evident with little absolute
change. In the course of the next two-and-a-half hours the total count
fell slightly, but this was accompanied by a very marked neutrophilia, the

-eucocytas

cub.m.m
19,000

18,000

17,000

16,000

15,000

14,000

13,000

12,000

11,000

10,000

9,000

8,000

7,000

6,000

5,000

4,000

3,000

2,000

i,ooo

0

.T

/

/

\

\

/

/

\

V

/

\

/

\

, /

V

„P

••«.

/

■"N

/

t

^

N

y\

\^

\

•l

0-

— EK

-^ttr

-^E

N.

0 '-^

E X 1-2

04) M
S.CI-0

./N

J

V^ST.

-V

Time

4-25

P.M.

4-52
P.M.

5-15
PM.

7

PM.

9-45

P.M.

RABBIT, PILOCARPINE, 60 m^ ^
Csubcuta-neous)

XVIII. XIX.

percentage rising from 22 to 84. At the end of the next four hours, i.e.,
eight hours after injection, hyperleucocytosis, with relative neutrophilia,
was well established.

(348)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 29

The fluctuations in the opsonic content during the experiment were
quite trivial.

In similar experiments, not here charted, we never succeeded in
obtaining a marked hyperleucocytosis immediately after intravenous
injection as Ruzicka did. In fact, the hyperleucocytosis appears to be a
late phenomenon.

In the foregoing experiments, chemical substances of more or less
simple chemical constitution have been employed, to produce artificial
leucocytosis. We have seen that in the presence of marked hyper-
leucocytosis there has been no corresponding rise in the opsonic content.

6. — Nuclein.

We have already referred to the employment of nuclein by Issaeff
{loc. cit.) in experimental cholera infection. Similar results were obtained
by Miyake {^) working in Mikulicz's laboratory, on experimental coli
infection in guinea-pigs. By preliminary intraperitoneal inoculation of
nuclein, he was able to increase enormously the resistance of the
peritoneum towards a subsequent injection of virulent coli, made during
the period of hyperleucocytosis. The same principle was adopted by
Mikulicz in surgical cases in man, involving laparotomy, and it was
found that previous injection of nuclein acted as a good safeguard
against a possible subsequent infection. He stated that a hyper-
leucocytosis in the circulating blood regularly followed the injection of
nuclein in man. The stage of hyperleucocytosis was usually preceded in
animals by a period of hypoleucocytosis occurring during the first two
hours after inoculation.

Recently, Huggard and Morland {^'^) have tried the effect of yeast
administration on the opsonic index in tuberculous patients. The dose
was usually 10 grm. of ordinary German yeast, and was administered for
at least a month in each case. In a series of 25 unselected cases, they
found a mean rise in the opsonic index of "15 after a mean course of
41 days' treatment.

Two days after the administration of a single dose there was a slight
fall in the opsonic index, followed after four or five days by a rise. The
opsonic variations recorded by these authors are, however, very slight,
and might well come within the experimental error. Thus, after a dose

(349)

30

QUATERCENTENARY STUDIES IN PATHOLOGY

of yeast, there might be a fall in the opsonic index, from "8 to 7, with a
subsequent rise to 'g. Marked changes also occurred in the leucocyte
count in certain cases. During the first half-hour after a dose of 10 grm.
the leucocytes were more than doubled. In the second half-hour there was
a fall equal to half the previous rise. In three hours the leucocytes

sank below the initial level. It has to be
noted, also, that some of the patients
were taking drugs like benzsosol quinine,
terpin hydrate, calcium glycerophosphate,
etc., along with the yeast, so that the pure
effect of the latter must have been somewhat
masked. It seems highly probable, however,
from our nuclein experiments presently to
be detailed, that the leucocytic and opsonic
response (though the latter was undoubtedly
very slight) were attributable to the nuclein
contained in the yeast.

Exp. XX. (Chart XX.) Rabbit. Intraperi-
toneal injection of 8 cc. of a 5 7o solu-
tion of Nuclein.*

During the first six hours very little
variation took place in the leucocyte count
apart from a slight neutrophilia. The op-
sonic index, however, reached a level of rp
at the end of three hours, and of 2*6 at the
XX. end of 5 J hours.

per m.m.

9,000
6,000

7,000
6,000
5,000
4,000
3,ooo
2,000

1,000
0

T

y^

^^

X

•

L

r

A

f

5/

y

\p

/s

s^

^

0— .

G—

— 0-—

— oM

. 2-6
S 2-4
-D 2-2

S 20

1-8

.y 1-6

§ 1-4

0 i-o

p

/ST.

^

y

/

/

i

f

/

/

Time

I045A.M
I0-2O5

2-10 p. M

5-10 RM

11 A.M.
I1-205

RABBIT, NUCLEIN, 5% sol. 8c.c.
(intraperiboneal)

Exp. XXI. (Chart XXI.) Rabbit. Intravenous inoculation (2 cc).

On this occasion, frequent samples were taken with a view to the
demonstration of a negative phase. Very little change took place in the
leucocyte count apart from a transient neutrophilia and a slight leucopenia
following inoculation.

*The nuclein used in the experiments was Nuclein Solution No. i, containing 5 % Nucleinic
acid (from yeast), manufactured by Parke, Davis & Co.

(350)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 31

The opsonic index to staphylococcus did not rise very high on this
occasion (I'sS). At the fourth hour, the opsonic index to tubercle rose
to 1*5, but at the end of 24 hours had returned to unity.

Time n45A.M. 12-45 3-30 6-15 io-30p.m

19-605 P.M. P.M. P.M. 20-605

RABBIT

NUCLEIIN, 5%soI.2c.c.(intniv-
___^ enous)

XXI.

T

A

/

A

f

J

>

V

1

V

J

,L

y\

/

r^

^

\

/

f

f

\

/
/

\

1

\

/

V

\

/

\

\

^P

\

1

1

>

V

V-

' i

0—

.-o~.

-«©.—

— .©-^

-0-

.-^M

1

/^

.^

/

Vc

>

?^

k

/

c

\^

/

al

^d

11-30 A.M
20-605

I-I5

P.M.

3-15

RM.

6-15
RM.

IO-3OA.M. II-3OA.M
2I-6-05 22-6-05

RABBIT, NUCLEI N, 5% 50l. 3 c.c. ( mt ra-^^

XXII.

Exp. XXII. (Chart XXII.) Rabbit. Intravenous inoculation (3 cc).

A slight hyperleucocytosis, with a very well-marked neutrophilia,
appeared four hours after inoculation. At the seventh hour, the hyper-
leucocytosis had gone, but the neutrophilia persisted. Twenty-four
hours after inoculation, the blood presented the initial leucocytic formula.

The opsonic variation was very remarkable on this occasion. Less

(350

32

QUATERCENTENARY STUDIES IN PATHOLOGY

^than two hours after inoculation, the opsonic power had doubled its
original value, and, at the seventh hour, it was still very high (r8). At this
point a further inoculation of 4 cc. was given intravenously, but any
cumulative effect that may have occurred had quite disappeared at the
end of 24 hours after the first injection.

1

LeucecyDes
per m.m?

15,000
I4POO
13,000
izpoo
11,000
10,000
9,000
8,000

7,000

6,000
5,000

4P00

3,000
2,000

1,000
0

s

\

/s

\

r

\

1

V

^

A

i

St

^

V

/N

V ,'

/'

V

\

\r

r

"^

\

i

1

f

\

V

-^

-^

s/

^

,L

N

^>^X

A

/'

v^

.A

\

/

A

/

' \

V

/

\

/

\p

V.

J

V

^^

--^

^.-r'*'

^^

J

"yr

0—

-^..

••• G—

— Oi^-

— 0—

^.G^«

_oM

. 1-4
S 1-3

0 10

1 -.1

r4v

/y

v>

/

r

♦,

s.

* /

^\«

/^

'"^V)

,_nTi4

erce.

^^

N,^

/

\7

^\,

,?5b.

JohHO-

^,^^

-V

.^

V

\^y

" a

iccus.

^^•

V

Time

n-33A.M.
2Z-605

12-3

P.M.

12-33

P.M.

P.M.

1-33

P.M.

2-3

P.M.

2-33

P.M.

3-3

P.M.

3-33

RM.

4-33

P.M.

n30A.M

23-605

RABBIT, NUCLEIN, 5% sol. 3c.c.(inbra,venous)

XXIII.

Exp. XXIII. (Chart XXIII.) Rabbit. Intravenous inoculation (3 cc).

Samples were drawn on this occasion every half-hour in the hope of
demonstrating a leucocytic or opsonic negative phase.

(352)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 33

During the first hour there was h'ttle change in the leucocytic formula,
although the total count diminished.

During the second hour, while the leucopenia persisted, the poly-
nuclear percentage rose considerably (69-5). At the end of the third
hour it had reached 70, where it remained till the end of the fifth hour.

The total count during the latter period varied slightly, but there was
no distinct period of hyperleucocytosis. At the end of 24 hours the
leucocytic formula had returned to the normal.

The opsonic index exhibited a marked negative phase during the
first hour after inoculation, and was succeeded by a moderate rise during
the second hour. The maximum opsonic power, both to staphylococcus
and tubercle, was registered in the course of the third hour, but was
exceedingly transitory. It fell to normal again in the fourth and fifth
hours.

The negative phase in the case of tubercle was somewhat more
pronounced than in the case of staphylococcus. It is notable that the
high opsonic level coincided fairly exactly with the attainment of the
high neutrophilic phase, but this latter persisted much longer.

Exp. XXIV. (Chart XXIV.) Guinea-pig. Intraperitoneal
inoculation (i cc).

A well marked hyperleucocytosis of neutrophilic type took place, and
the opsonic index rose to double its original value in four hours.

Exp. XXV. (Chart XXV.) Rabbit, (i cc).

Four hours after inoculation, the leucocytes showed a very slight rise,
v/hich was accompanied by an enormous neutrophilia (80 7o)- The
opsonic index to staphylococcus also rose in four hours to i'6.

Exp. XXVI. (Chart XXVI.) Rabbit, (i cc).

The inoculation was made at 5-30 p.m., but by next morning, any
effect that the nuclein may have had on the leucocytes and opsonic
content had completely disappeared. Accordingly, another inoculation
was given at 10*45 a.m. A sample drawn at 5*15 p.m. on the same day

(353) Y

34

QUATERCENTENARY STUDIES IN PATHOLOGY

showed a slight neutrophih'a, but a very marked rise in the opsonic index,
both towards staphylococcus and tubercle (r8 and 1*31 respectively).

Exp. XXVII. (Chart XXVII.) Rabbit. Intravenous inoculation

(2 cc).

The opsonic power was tested both towards staphylococcus and
tubercle. A marked rise took place at the end of five hours, but at
the end of 24 hours the opsonic index had returned to normal.

Lsucocytea
per mm'

18,000
le^ooo
14,000

f7 Ctr^ri

-T

/

/

j>?

10,000
8,000

6,000

4POO

Z,000
0

\

)

1

\

/

\

^L

■•»

'^-^M

1-8

^ 1-5
0 1-2

^ T.T
0 10

•9

At

/

/

J

/

/

J

Time

U

AM

330

PM.

GUINEA-FIG, NUCLElN,lc.c

M^

/

/

r

i

/

/

6

=<

^N

yST

/

J

/

12
noon.

4

RM

RABBIT,
NUCLElN,lc.c.

s

Nw

\

^

S^T

V

^P

-^

N^

/ST

/

/

T.H

/

P

L

/

y

5-30

RM.

9-45
P.M.

515

P.M.

RABi

NUCLE

lit

N, Ice.

XXIV.

XXV.

XXVI.

Exps. XXVIII. and XXIX. (Charts XXVIII. and XXIX.).

The opsonic estimations are referred to under Exps. XXIV. and
XXV., but are re-charted alongside Chart XXVII., to show the varying
degrees of nuclein effect on the opsonic power.

(354)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 35

Exp. XXX. (Chart XXX.) Guinea-pig. Intraperitoneal inoculation

(4 cc).

Complete blood and opsonic estimations were made every hour after

injection. At the end of the first hour, the total leucocytes had fallen

slightly, but a relative neutrophilia was already conspicuous (70%).

At the end of the second hour, the leucocytes had risen to 14,900, and

an enormous neutrophilia was present (90 °/J.

At the end of the third hour, the neutrophilia was still high, though the
total leucocytes had fallen.

1-8

1-7

0 1-6

C

- 1-5

1-2

l-I
10

A\

I

• TB>,

ST.

\

1

\

Time

ID A.M.
16505

3 P.M.

loA.M.
I7-506

RABBIT, NUCLEIN, 2 CC.

ST.

f

/

/

/

/

noon

4 P.M.

RABBIT,
NUCLEIN, ICC 1

ST.

/

/

/

/

/

/

10 A.M.

3-30RM.

GDINEA-PIG,
NUCLEIN, ICC

XXVII.

XXVIII.

XXIX.

During the next hour a well marked hyperleucocytosis occurred
(26,600) which had quite gone in the course of an hour. The neutro-
philia was persistent.

A very marked rise in the opsonic index, both to staphylococcus and
tubercle, took place between the fourth and fifth hours after inoculation. In
both cases the rise was very transitory. There was no apparent negative
phase. It is notable also that the relative neutrophilia reached its
maximum two hours before the period of hyperleucocytosis.

Exp. XXXI. (Chart XXXI.) Guinea-pig. Intraperitoneal inoculation

(3 cc).
At the end of the second hour, there was a marked hyperleucocytosis
with very slight relative neutrophilia. The opsonic index was double its

(355) Y I

36

QUATERCENTENARY STUDIES IN PATHOLOGY

original value at the end of the third hour, thus coinciding with the
maximum of hyperleucocytosis.

LfucocyCes

cub. mm
28,000

26,000

24,000

22,000

20,000

18,000

16,000

14,000

12,000

10,000

8,000

6,000

4,000

2,000
0

11

i

I

/I

\

11

<

^

/

vl

1

s

J

^7

V

//

0

i

1

V

\

,P

v

J

\

^^

/

\

/

•L

s

f^^

\

^

r

20

IS

c

I-

0 i-o
0-8

/

V

.y

Ki

y

V

.T.B

^

.>

\

/>

^

>

ST

Time

^,Tl^^n^f.^fAf'!r

p t3 G-PIG, NUCLEIN, 4CC.

"" (Intraperiboneal)

/

'T

f

i

/

(

I

/

i

,P

\

^

/

(

I

j

\

1

yy

L

\

(/

/

^

f

ST

i

/

/

/

1

^

\

/

\

/

II 12 ]

^ .A.M noon p

2

M. P.M.

G-PIG, NUCLEIN, 3CC.
(inCraperitoneal)

XXX. XXXI.

Remarks on the effects of the injection of Nuclein.

It will be seen that the leucocytoses produced by nuclein have never
been marked, and that in some cases a distinct leucopenia has resulted.

(356)

LEUCOCYTOSIS AND Ol'SONIC CONTENT OF SERUM

37

In spite of the slight absolute changes, however, the relative changes
in the direction of neutrophilia have been quite conspicuous, though more
transitory than in the case of tallianine.

The leucocytes of the guinea-pig react much more vigorously to
nuclein than those of the rabbit.

In sharp contradistinction to substances like sodium cinnamate and
tallianine, nuclein invariably produces a marked rise in the opsonic
index, probably of a non-specific nature.

An experiment was made to determine whether the direct addition of
nuclein to normal serum altered the opsonic index of the latter. The
result was negative.

it's

serum i vol

1 II

+ Salt sol. I vol. + Co
+ Nuclein sol. (50 %) i vol. +

cci + Leuc

- +

Cocci per
Leucocyte.

ocytes = 6-9
=.67

1 It

4- Nuclein-salt sol. (i in 2) i vol. +

. +

= 7'2

1 II

+ Nuclein-salt sol. (i in 4) i vol. +

. +

= 6-5

t II

+ Nuclein-salt sol. (i in 8) i vol. +

. +

= 70

1 II

+ Nuclein-salt sol. (i in 16) i vol. +

1 +

-6-6

1 II

+ Nuclein-salt sol. (i in 32) i vol. +

1 +

= 6-8

1 II

+ Nuclein-salt sol. (i in 64) i vol. +

1 +

= 6-4

7. — Papayotin-albumose.

The papayotin-albumose was prepared by digesting washed fibrin
with papayotin at 37° C for two days, filtering and precipitating the
albumose of the filtrate by alcohol. The resulting precipitate was then
dried over H2 S04 in vacuo. Intraperitoneal injection of | gm. of this
substance in rabbits produced a marked relative neutrophilia, but did
not affect the opsonic index.

8. — Aleurone Emulsion.

The following experiment was made with aleuronat emulsion.*
A rabbit was inoculated intraperitoneally with 5 cc. of a 5 %
emulsion of aleuronat. At the end of 24 hours the opsonic power of the
serum was unchanged. At the end of 36 hours the animal was killed.
On opening the abdomen a yellowish cheesy mass was found, adherent
to the parietal pertitoneum. It weighed 1*5 gm., and was composed of

* The aleurone used was aleuronat purissimum, prepared by R. Hundhausen. Hamm, i. W.

(357)

38 QUATERCENTENARY STUDIES IN PATHOLOGY

aleuron granules and leucocytes, a differential count of which showed,
polynuclears 62 %> lymphocytes 22 7o- The mass was frozen and
thawed according to Buchner's method, mixed with an equal weight of
•85 7o NaCl., and ground with sand in a mortar. It was then
centrifuged. The opsonic power of the fluid was then tested along with
the serum of the same animal and the serum of a control animal.
The result was as follows : —

Control serum, 5 cocci per leucocyte.

Serum of aleuronat rabbit, 4 cocci per leucocyte.

Aleuronat exudate of rabbit, o cocci per leucocyte.

No opsonin, therefore, was present in the aleuronat exudate in this
experiment, but without further experiments of a similar nature it would
be unsafe to generalise as to the opsonic content of these leucocytic
exudates.

9. — A Ibumen-metaphosphate.

In view of the fact that the opsonic index was so markedly influenced
by nuclein, it was considered of interest to determine whether the so-
called pseudo-nucleins, formed by precipitating albumen solutions with
metaphosphoric acid, would produce analogous effects. The method
employed for the preparation of the substance was one of the many
recommended by Fuld i^^'). To a certain volume of a 10 % solution of
Qg^ albumen, an equal volume of a 10 % solution of metaphosphoric
acid was added. After filtration the precipitate was suspended in
distilled water, and finally centrifuged. The process was repeated for
five or six times. The precipitate of albumen-metaphosphate was then
dried over sulphuric acid in vacuo, and finely powdered. Of this
substance -5 grm., dissolved in 2 7o sodium carbonate, was inoculated
intraperitoneally in guinea-pigs and i grm. in rabbits.

From a long series of experiments we may quote the following,
to show that no very marked changes took place in the opsonic index : —

Exp. I. Guinea-pig. (-5 gm. intraperitoneally at 3 p.m.).

1st day,

2ncl day,

3 p.m.

4 p.m.

5 p.m.

6 p.m.

1 1 p. m.

1 1 a. m.

II p.m.

Opsonic Index.

I'O

•84

7

•64

I '4

1-3

I'2

(358)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 39

Exp. II. Rabbit. (-5 gm. intraperitoneally at 1*15 p.m.).

i-i5p.m. 2-30 p.m. 3-30 p.m. 5 p.m. 9 p.m.

Opsonic Index. I'o '85 i-i -96 -^^

A notable feature was a rather prolonged negative phase (as in
Exp. I.), which was not, however, followed by a high opsonic tide.

v.— On the reduced phagocytic power of the neutrophile
leucocytes during the period of hyperleucocytosis.

We have seen that, under conditions, in which the polynuclear
leucocytes are greatly in excess in the blood stream, there may be
no appreciable change in the opsonic content (sod. cinnamate and
tallianine experiments) as measured by normal leucocytes.

It seemed to us, therefore, a question of importance to determine
whether the opsonic index would be altered by employing leucocytes
drawn during the period of hyperleucocytosis as phagocytes in the
in vitro test. The results of a series of experiments in this connection
will be found on the next page.

It will be seen that the employment of the newly-formed polynuclear
leucocytes, as phagocytes in the opsonic test, invariably lowers the opsonic
index, and sometimes very considerably. The new leucocytes, in fact,
appear to have their phagocytic powers very feebly developed. Now, it
is obvious that, what one may call the " real opsonic index," must be
measured by the serum and leucocytes of the same animal taken at
the same time. In those cases, therefore, where the opsonic content
of the serum remains fairly constant (sod. cinnamate and tallianine), the
real opsonic index will be very much lowered during the period of
hyperleucocytosis or of relative neutrophilia. Again, in experiments
with nuclein and autolysed cultures, the real opsonic index will still be
much higher than in the control {vide Exps. III. and IV., Table), but
not so high as when normal leucocytes are employed in the test. The
above results have obviously a bearing on the question of vaccine-therapy
in man, where possible variations in the phagocytic power of the
leucocytes during vaccinal leucocytosis may influence considerably
the real opsonic index. Any increase in the opsonic content of the
serum might be neutralised, or rendered unavailable to the organism,
by deficient phagocytic power on the part of the leucocytes.

(359)

40

QUATERCENTENARY STUDIES IN PATHOLOGY

Experiments demonstrating the Reduced Phagocytic Power of the Neutrophilic
Leucocytes during the period of Hyperleucocytosis.

No.

Experiment.

Animal,

Hyper-
leuco-
cytosis.

Relative
Neutro-
philia.

Leucocyte count at
time of sample.

Scheme of Experiment.

Cocci per
leucocyte.

cytfc'

Sodium
Cinnamate.

Rabbit.

Slight.

Marked.

12,300

Control serum -1- control leucocytes.

II It -1- cinnamate h

Cinnamate n -f control n

It It -1- cinnamate n

47

1-3

2 2
08

%

2.

3-

Sodium
Cinnamate.

Guinea-pig

Marked.

Marked.

27,200

Control serum + control leucocytes.

It It -f cinnamate n

Cinnamate n + control .,

It It + cinnamate n

I2-0

6-0

no

7-8

Nuclein
Solution.

Rabbit.

Slight.

Marked.

11,100

Control serum -f- control leucocytes.

It It + nuclein i.

Nuclein n + control n

It It -f nuclein n

67
3-0
8-6
5-6

4-
5-

Nuclein
Solution,

Guinea-pig

Moderate

Marked.

18,000

Control serum -f control leucocytes.

It It -f nuclein n

Nuclein n -f control n

M It -1- nuclein n

9-8

5-6

not tested

7-6

Tallianine.

Guinea-pig

Moderate

Marked.

17,400

Control serum + control leucocytes.

II It -f tallianine n

Tallianine n -f control n

11 It + tallianine n

Rabbit's .. -f control t,

It II + tallianine n

37
3-0
4-8
2-9

6.
7-

Tallianine.

Guinea-pig

Marked.

Marked.

Three samples.

Before inoc'n, (10,200) = Pi

3-15 (24,800) = P2,
5-30(2i,400) = P3.

Tallianine serum (3 15) -f Pi leuc'ytes

-t-P2 1,

Control It 4- Pi „
-I-P2 „

Tallianine t, (5*30) + Pi „
+ P3 It

Control II -fP3 „

14-5
99
i8-6
11-4
17-8
8-0
4 -2

Tallianine.

Guinea-pig

Marked.

Moderate

Two samples.

Before inoc'n (7000) = Pi.

2-15 p.m. (i9,8oo) = P2.

Tallianine serum + Pi leucocytes.

.. +P2

Control It +Pi It
It II -f P2 It

...

6-0

l\

3-8

8.
9-

Tallianine.

Guinea-pig

Marked.

Marked,

Two samples.

Before inoc'n (9500)-- Pi.

3-13 p.m. (26,600)= P2.

Tallianine serum + Pi leucocytes.

It It -fP2

Control It (a) -f Pi
ti 1. (a)-t-P2
Control ,t(b) + Pi .,
ii(b)-j-P2 M

142

1 3 '9
i8-o
II-9
i8-o
127

Tallianine.

Guinea-pig

None,

Slight,

11,600

Control serum -f control leucocytes.

It It + tallianine n

Tallianine n + control n

It II + tallianine n

II -2

lO'O

IO-6

8-5

(360)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 41

VI. — Summary and Conclusions.

1. The opsonic content is uninfluenced by the presence of hyper-
leucocytosis, or relative neutrophih'a in the circulating blood, when these
conditions are brought about by the injection of simply-constituted
chemical bodies, like sodium cinnamate, tallianine, pilocarpine, etc.

2. The opsonic content is very markedly influenced by the
inoculation of substances like nuclein, autolysed cultures, etc., the
chemical constitution of which is exceedingly complex, and probably
closely akin to that of the bacterial proteines. These substances at the
same time produce decided changes in the blood picture, though these
are, as a rule, less marked than in the case of the simpler bodies above
mentioned.

3. During the period of hyperleucocytosis, the phagocytic power of
the neutrophile leucocytes is considerably reduced.

Naturally, we have considered whether any conclusions can be drawn
from the above-mentioned experiments, in reference to the important
questions of the origin, and probable nature, of the opsonic substances in
the serum.

In the first place, it is improbable that the substance is of the nature
of complement, otherwise one might have expected an increase of the
opsonic content of the serum, after the inoculation of substances like
sodium cinnamate, which are known to raise the haemolytic complement
content concurrently with the production of a neutrophilic leucocytosis.

On the assumption that the opsonin is of the nature of an interme-
diary body which manifests its full action when co-operating with a
complement, the results might be more readily explained. Thus, even
with complement-increasing substances, like sodium cinnamate, it might
be supposed that there is always just sufficient normal complement to
neutralise the available amboceptor, and hence no opsonic rise would be
recorded. The injection of substances like nuclein, of highly complex
chemical constitution, and possibly allied to bacterial proteins, might
produce an increase of intermediary body. The effect on the leucocytes
might at the same time produce an increase of complement, and thus the
opsonic content might rise. A third possibility is that the opsonin is a

(361)

42 QUATERCENTENARY STUDIES IN PATHOLOGY

simple substance, which is formed after inoculation of nuclein and
bacterial proteins.

The experiments we have detailed above do not, however, admit of a
solution of these problems, which, most likely, must be settled by other
modes of experimentation.

(362)

LEUCOCYTOSIS AND OPSONIC CONTENT OF SERUM 43

BIBLIOGRAPHY.

1. Wright and Douglas. Roy. Soc. Proc, 1903, LXXIL, p. 357.

2. Bulloch and Atkin. Roy. Soc. Proc, 1905, LXXIV., p. 379.

3. RoMER, F. Virchow's Archiv., 1892, CXXVIII., p. 98.

4. ScHLESiNGER. Zeitschr. f. Hygiene, 1900, XXXV., p. 349.

5. Holmes. Guy's Hosp. Reports, 1905, LIX., 154.

6. BucHNER. Archiv. f. Hygiene, 1890, X., p. 84.

7. NuTTALL. Zeitschr. f. Hygiene, 1888, IV., p. 353.

8. NissEN. Zeitschr. f. Hygiene, 1889, VI., p. 487.

9. Hankin. Centralbl. f. Bakt., 1892, XII., p. 777.

10. Denys. La Cellule, 1894, X., p. 463.

11. Wassermann. Zeitschr. f. Hygiene, 1901, XXXVIL, p. 173.

12. Schuster. Inaug. Diss., Miinchen, 1894.

13. Schattenfroh. Archiv. f. Hygiene, 1897, XXXI., p. i, and 1899, XXXV.,

P- 135-

14. Denys and Kaisin. La Cellule, 1893, IX., p. 335.

15. Lowit. Ziegler's Beitrage, 1897, XXII., p. 172.

16. Lambotte and Stiennon. Centralbl. f. Bakt., 1905, XL., p. 224.

17. Dean. Roy. Soc. Proc, 1905, B. LXXVL, p. 506.

18. Keith. Roy. Soc Proc, 1906, B. LXXVIL, p. 537.

19. LoEWY and Richter. Deutsche med. Woch., 1895, No. 15, p. 241.

20. Jacob. Zeitschr. f. klin., Med. 1896, XXX., p. 447.

21. Hahn. Archiv. f. Hygiene, 1897, XXVIII. , p. 312.

22. Pfeiffer and Issaeff. Zeitschr. f. Hygiene, 1894, XVIL, p. 355.

23. Wooldridge. Archiv. f. Anat. u. Phys. Phys. Abt. Jahrg. 1888, p. 527.

24. Zacharoff. Centralbl. f. Bakt., 1894, XVIL, p. 331.

25. Brieger, Kitasato and Wassermann. Zeitschr. f. Hygiene, 1892, XIL,

P- 137.

26. NoLF. Annales de I'lnst. Pasteur, 1900, XIV., p. 297.

(363)

44 QUATERCENTENARY STUDIES IN PATHOLOGY

27. MiJLLER. Centralbl. f. Bakt., 1901, XXIX., p. 175.

28. LoNGCOPE. Journ. of Hygiene, 1903, III., p. 28.

29. GoLDSCHEiDER and Jacob. Zeitschr. f. klin. Med., 1894, XXV., p. 373.

30. ScHULZ. Deutsch. Arch. f. klin. Med., 1893, LI., p. 234.

31. RiEDER. Beitr. z. Kenntn. d. Leukocytose, Leipz., 1892.

32. LowiT. Studien z. Phys. u. Path, des Blutes, &c., Jena, 1892.
$^. RuziCKA. Allg. wien. med. Zeitung, 1893, P- 345-

34. Brinckerhoff and Tyzzer. Journ. of Med. Research, 1902, VIL, p. 173.

35. PoHL. Arch. f. exp. Path. u. Pharm., 1889, XXV., p. 31.

36. HiRSCHFELD. Virchow's Archiv., 1897, CXLIX., p. 22.

37. KuRLOFF. Quoted in Ehrlich's "Die Ansemie," Wien, 1898.

38. Burnett. Journ. of Med. Research, 1904, XL, p. 537.

39. RiCHTER. Virchow's Archiv., 1893, CXXXIIL, p. 376.

40. RiCHTER and Spiro. Arch. f. exp. Path. u. Pharm., 1894, XXXIV., p. 289.

41. Batty Shaw. Journ. of Path, and Bact, 1902., VI 1 1., p. 70.

42. Charteris and Cathcart. Journ. of Path, and Bact., 1904, X., p. 56.

43. Stassano et Billon. Compt. rend, de la Soc. de Biol., 1903, No. 5.

44. FuNCK. Centralbl. f. Bakt., 1900, XXVIL, p. 670.

45. Spiro. Arch. f. exp. Path. u. Pharm., 1897, XXXVIIL, p. 113.

46. MiYAKE. Quoted by Mikulicz., Langenbeck's Arch., 1904, LXXIIL, p. 347.

47. HucGARD and Morland. Lancet, 1905, June 3, p. 1493.

48. FuLD. Hofmeister's Beitrage, 1902, 11.

49. Ledingham. Lancet, 1906, June 16.

(364)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS.
By George Grant Macdonald, M.A., M.D.

(Bacteriological Laboratory^ London Hospital, London, E.).

(365)

Immunity m Pneumococcal Infections.

Congenital immunity or unsusceptibility towards pneumococcal infec-
tion is possessed in different degrees by animals of different species,
pigeons and domestic fowls being highly immune, the human species,
guinea-pigs, rats, dogs and sheep being more susceptible, while rabbits
and mice fall an easy prey to pneumococcal disease. Acquired immunity
is that attained naturally, by passing through an attack of pneumococcal
disease, or artificially, by the inoculation of the microbe itself or its
products — active immunisation — or of the antibodies produced in the
actively immunised organism — passive immunisation. But on the
quantity and quality of the immunity, and on the mechanism by which
immunity is conferred, unanimity of opinion has not been attained,
though investigated from many directions and by widely varying
methods.

Acquired Immunity. — The immunity naturally acquired by passing
through an attack of the chief pneumococcal infection — pneumonia —
does not appear to be very durable. Thus Rieselle {}) in lOO cases of
pneumonia found that only 50 had it for the first time, that 32 had it for
the second time, and the rest for the third time or oftener. Mollmann
(2) in 832 cases noted that only 13 per cent, were in their first attack of
pneumonia. In 1800 cases collected from the literature on this subject
420 had previous attacks of pneumonia {i.e., 23*3 per cent). The
frequent occurrence, however, of empyaema, meningitis, arthritis, endo-
carditis, etc., following pneumonia, would suggest that no immediate
immunity is conferred. Yet, in the majority of cases, individuals who
have passed through an attack of pneumonia enjoy immunity for a short
time at least, and animals which have been inoculated with pneumococcus
cultures, and which have recovered, are more or less resistant for some
time.

(367)

4 QUATERCENTENARY STUDIES IN PATHOLOGY

Active Immunity. — The methods employed in actively immunising
animals have been very numerous, attenuated cultures, small doses of
virulent cultures, filtrates of fluids containing the pneumococcus, cultures
killed by heat or chemicals, etc. Many investigators, disappointed with
the varying results obtained by these methods, endeavoured to concentrate
the immunising substance. Thus, the Klemperer (3) brothers produced
from filtered cultures a yellowish-white powder, which they regarded as
the specific poison of the pneumococcus — the pneumotoxin — and they
tried to obtain immunity by inoculating it on animals. Foa and Scabia

(4) produced by similar means a " pneumo-protein."

Passive Immunity. — In 1891, researches on the subject of passive
immunity were published almost simultaneously by Foa and Carbone,
Emmerich and Fawitzky, and the Klemperers. Emmerich and Fawitzky

(5) demonstrated that by injection of the blood serum or of pieces of
muscles or organs of immunised rabbits, animals could be protected
against a dose of culture injected simultaneously or a short time before.
Foa and Carbone were pioneers in the therapeutic employment of serum
in pneumonia in man, and in the four cases treated the favourable issue
was hastened, the critical fall of temperature occurring in 24 to 48 hours
after injection. Serum from pneumonia patients, in different stages of
the disease, was inoculated into animals, but with no beneficial result —
in fact, in several cases the fatal issue seemed hastened ; if used 24 hours
after the infection no beneficial result was obtained in any case.

G. and F. Klemperer used therapeutically the serum taken from
pneumonic patients by leeches or blisters in the treatment of artificial
pneumococcal infection in rabbits, and with apparently favourable results.
In the 12 cases treated thus, the crisis occurred immediately in 5, and
there was a transient fall of temperature in 7.

Washbourn (6), however, noticed that the serum of immunised animals
often completely failed to protect other animals from infection, and,
to attain still higher degrees of immunity, injected a pony with
broth cultures killed by heat, then agar cultures, and subsequently living
and highly virulent broth cultures. Better results were obtained, the
serum protecting animals against infection, but, on its employment in
two cases in man, whether the cure obtained was due to the serum or
not, Washbourn was uncertain.

The Italian investigator, Pane (7), immunised cows and donkeys.

(368)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS 5

For this purpose, he employed pneumococci of the highest virulence

such that one-twenty-millionth part of a cubic centimetre of a culture
killed rabbits of any weight in six days. The asses' serum gave the best
results, for an injection of 075 cc. into the auricular vein of a rabbit
protected it against twenty times the minimal lethal dose, even when this
was injected thirty to sixty minutes before infection ; while, from the
serum of a rabbit, i cc. was required to attain the same result. Quantities
of 3 cc. of the asses' serum protected against 20,000 times the minimal
lethal dose of pneumococcus culture.

Pane then applied himself to the treatment of pneumonia in man with
his highly active anti-serum, but the results did not justify his expecta-
tions. He treated 32 cases, with three fatal issues, and is convinced that,
in cases which are not severe, a fatal issue may be averted by doses of
20 cc. given early and frequently. The behaviour of Pane's serum with
different races of pneumococcus was tested by Eyre and Washbourn (S),
and they found that it was active for four strains out of five, and was a
valuable means of separating different strains.

The serum most widely used at the present day is that of Romer (9).
What he attempts to do is to produce a highly bactericidal anti-
pneumococcic serum, starting with the assumption that the antibodies
produced in the serum were of the nature of amboceptors. The specificity
of the immune bodies may be so great that in infection by one strain of a
particular bacterium, the amboceptors formed are not capable of attaching
themselves to the bodies of the bacteria of another strain, and, therefore,
to have a universally active anti-serum, amboceptors suitable to all
variations of the bacterial species pathogenic to man must be inoculated.
Since the labile complement cannot be preserved, we can have only
the amboceptors to deal with in a bactericidal immune serum.
There are great difficulties connected with the production of fresh
complements, because the inoculation of fresh serum may give rise
to the production of anti-complements in the organism (Ehrlich and
Morgenroth). To overcome these difficulties, Romer employs the
mixture of immune sera from various animals (horse, bovines and sheep) ;
each of these animals has been immunised against different strains of
pneumococcus (according to the principle Tavel employs in the production
of antistreptococcic serum) and mixes them. Thus Romer's serum
is polyvalent in a double sense, in that it contains different immune

(369) Z

6 QUATERCENTENARY STUDIES IN PATHOLOGY

bodies suitable to the most varying complements, and, at the same time,
suitable to being anchored to a great variety of strains of pneumococcus.
The practical value of this serum has been favourably extolled by Romer,
Axenfeld, and Kriickmann, in the treatment and prophylaxis of ulcus
serpens cornecB, but of its use in acute pneumonia the results have not
been of such signal success. Thus, Passler {^^\ in 24 cases, had a
mortality of 4 (i6*6%) ; he does not advise the general use of the serum,
but affirms that, in certain cases of danger, it may be of some service,
as the general condition seems to improve after each injection.

Of the value of Pane's serum the most conflicting results have been
published, Fanoni (i^) regards Pane's serum as absolutely specific, and
Dasaro-Cao (^2) says that the serum acts promptly as a cure, and the
mortality has fallen one-third. Banti and Pieraccini (^3) came to the
conclusion that Pane's serum gave no beneficial result in any of their
cases, even in large doses. Reports on the value of anti-serum have been
given by de Renzi {^^\ Concetti (i^), Righi (^6), Spolverini (^'^\ and other
Italians ; by Goldsborough (^s). Sears (^^), Snively (2°), Passler, etc.. etc.
The largest number of cases collected from the literature is that of
Goldsborough, who published a record of 447 cases with a mortality
of i6*5 per cent. Up to the present time (end of 1905) 515 cases have
been recorded with a mortality of 15 per cent, which does not show a
great deal of improvement on the 17*9 per cent., the average mortality in
untreated cases of pneumonia (Fussell).

Treatment was also tried by use of the serum of pneumonic con-
valescents by the Klemperers, Spolverini, Neisser (21), Audeoud (22), etc.,
but with most indifferent results.

Mechanism of Immunity. — As to the mechanism by which immunity
has been acquired, different opinions have been entertained by the
different investigators. Thus the Klemperers, thinking that an antitoxin
(antipneumotoxin) was produced in the blood as in diphtheria and
tetanus, regarded the action as not directed against the killing of the
cocci, but against the neutralisation of the toxin. A substantiation of
their opinion they found in the fact that a filtered bacterial-free broth-
culture when mixed with immune serum caused no rise in temperature,
or only a temporary rise, but alone it killed an animal with the
manifestations of high fever. They ascribed the cure of pneumonia in
man to an assumed antipneumotoxin production in the blood of the

(370)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS 7

patient, in such quantity as to neutralise the toxin completely. The
organism could then quite easily destroy the toxin-free cocci, and
when this was complete the crisis took place. If one supplies to the
serum, therefore, the serum of immunised animals containing much
antitoxin, the individual will be freed from pneumotoxin much more
rapidly ; hence the curative action of the immune serum.

Mosny was of the opinion, ascribing to the immune serum a
" toxinicidal " action, since the diplococcus in this serum preserved its
vitality and virulence for a month, while it was destroyed in non-
immunised animals in four days. In connection with this, Tizzoni and
Panichi (23) have recently showed that it sometimes required several
months for the pneumococci to be destroyed by the serum even of highly
immunised rabbits.

Bonome (^4), however, attributes the immunity acquired during an
attack of acute pneumonia to the bactericidal property of the blood,
and Emmerich voices the same opinion, though he supposed
that in the blood of immunised animals there is a substance, arising
from the union of the globulin with the toxin shed, or with the
toxin in the bacterial cell itself. This substance, which he calls the
" Immuntoxinproteidin," has great difificulty in acting on the tissue cells,
and is therefore innocuous for those, but it can easily penetrate the
bacterial cell, where it is split up into toxin and immunprotein, which,
being both in the nascent state, act as virulent poisons for the cocci and
rapidly destroy them.

The French school of workers, headed by Metchnikoff, ascribe the
immunisation against pneumococcus entirely to phagocytosis. Issaeff (^S),
after a series of most careful experiments, is convinced that there is no
bactericidal and no antitoxic power in the serum or tissue-juices, and
he ascribes the whole r61e to the leucocyte. Pane is also an active
adherent of this theory, and explains the action of his serum on the
supposition that, by the introduction of the immune serum, the leucocytes
are excited to secrete materials which protect animals against the
pneumococci.

In a series of extensive researches, published by Tchistovitch, of
St. Petersburg (26), the exact conditions in the lungs of pneumonic
patients and of dogs, in which pneumonia had been artificially induced,
were carefully studied. In imitation of Patella's work (1888), he made

(371) Z I

8 QUATERCENTENARY STUDIES IN PATHOLOGY

lung punctures with a Pravaz's syringe before, during, and after the
crisis, and inoculated the exudate thus obtained into rabbits and mice.
He thus was able to demonstrate that, even after the crisis, one could
obtain from the lung an exudate containing not only viable, but even
virulent, diplococci, killing the animals rapidly by septicaemia. Thus the
crisis does not depend on the killing of all the diplococci in the lungs,
for they may be found in considerable abundance, yet on culture the
number of organisms was shown to be comparatively few.

Pneumonia was then induced in dogs by intra-tracheal injection of
pneumococcus culture, which was allowed to flow down by gravity into
the lower pulmonary lobes. Graduated doses were used, and in several
cases the most intense pneumonia was set up. The animals were killed
at different intervals, the lungs hardened, sections cut, and stained by the
Gram-Weigert method. Thus he was able to demonstrate —

1. That in the cases going rapidly to a lethal issue, the phenomena of
phagocytosis by the leucocytes lying in \he alveoli were entirely absent.

2. That in cases going on to cure, but killed during the course, there
was a large exudate of phagocytes in the alveoli, with numerous cocci
lying within them.

By this method he was also able to show that, in non-resistant animals,
the diplococcus causes only a very feeble local inflammatory reaction,
and little phagocytosis is present ; the leucocytes do not engulf the cocci,
which can therefore go on rapidly multiplying, and quickly cause a fatal
issue. In resistant animals, the inoculation of pneumococcus causes
an inflammatory process of varying intensity, with dense cellular
infiltration and a more or less pronounced phagocytosis. In a further
series on animals inoculated with a weak pneumonic virus, Tchistovitch
showed the close parallelism between the course of the disease and the
leucocytosis, and considered the crisis as mostly brought about by
phagocytosis, the intoxication ceasing with the taking up of the cocci.
The protective bodies occurring in the serum he regarded as adjuvant
factors, the hypothetical stimulins stimulating active englobation by the
leucocyte, and the agglutinins causing the massing together of the cocci,
and thus rendering their rapid ingestion by the phagocytes more easy.

But it was destined that a severe blow should be inflicted on the
pre-eminence of the leucocyte. The first indication of this was the
work of Denys and Leclef (27) on the mechanism of streptococcal

(372)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS 9

immunity, in which they demonstrated that, besides the bactericidal
action of the serum, to which they faintly adhered, there was a more
important factor at work — phagocytosis brought about under the
influence of the serum. " Chez le lapin vaccina, le leucocyte tient du serum
son pouvoir d'englober et dHruire le streptocoque" A very short time
later Mennes (28), working under the direction of Denys, extended these
researches to the pneumococcus, and was able to show that the primary
immunising element is in the serum and not in the leucocyte. Highly
immune sera were taken and carefully tested for their bactericidal and
bacteriolytic power, and in no case was either of these apparent, even
after adding quantities of normal serum as a complement. But when
leucocytes were added to such sera, the cocci were rapidly picked up by
them and underwent a gradual degeration and lysis within them. In
many cases the cells were packed full of bacteria, whereas with normal
sera the number of bacteria in the phagocyte was comparatively small

The chief question then arose— Has the serum, as Metchnikoff" and
his school believe — a mere stimulating action on the cells, or does it act
directly on the bacteria themselves ?

In 1903- 1904, Wright and Douglas (29) found that the phagocytic
property of leucocytes suspended in a serum which has been subjected
to a temperature of 60-65° C. is entirely in abeyance, and, on the
addition of bacterial emulsions, no phagocytosis takes place. If,
conversely, the unheated serum is mixed with bacteria, and digested for
fifteen minutes at 37° C, and then subjected to 60° C. for a quarter of an
hour, phagocytosis takes place on adding leucocytes. From these
fundamental facts it is manifest that some definite substance has passed
out of the serum, and attached itself to the bacteria, rendering them
suitable pabulum for phagocytosis by the leucocytes. This thermolabile
substance they named " Opsonin."

During the process of active immunisation with staphylococcus,
micrococcus melitensis, bacillus tuberculosis, and bacillus pestis, they
have been able to show that for these infections the opsonic content of
the serum is increased, and have thus demonstrated that, in addition to
anti-toxic and bactericidal immunity, there is a third type — opsonic
immunity, which is a common phenomenon in a number of infections,
and brought about by a coalition and inter-action of the body humors
and cells.

(373)

10 QUATERCENTENARY STUDIES IN PATHOLOGY

Recently, in examination into the phenomena of the immunity
acquired in passing through an attack of croupous pneumonia, I made
an investigation into the course of the opsonic content of the serum
during this disease and in artificial infections in animals. Fifty-five
cases were at my disposal, twenty-five of these being more or less typical
cases of acute croupous pneumonia ; nine others were in children with
the clinical diagnosis of acute croupous pneumonia, seventeen cases were
definitely broncho-pneumonia, and the remaining four were cases of post-
pneumonic empyaema from which the pneumococcus was isolated in
more or less pure culture.

The technique employed was that of Wright and Douglas. Three
parts of the serum of the patient were added to three parts of my own
citrated and washed corpuscles and an emulsion of pneumococci. These
were thoroughly mixed and incubated at body temperature for fifteen
minutes. Films were then made of the contents of the mixing pipette,
dried and stained by Leishman's dye. The number of pneumococci in
thirty to fifty polymorphonuclear leucocytes were counted and an
average per leucocyte determined. Simultaneously with each experiment
a control, or series of controls, was made with the sera of normal healthy
individuals. The result is expressed in terms of the "opsonic index,"
the average number of cocci per polymorphonuclear leucocyte when the
serum added is that of a normal individual being regarded as unity, and
the opsonic index of the pneumonic patient is expressed in relation to
the normal opsonic index of a healthy person.

The strains of pneumococci used in these experiments were seven in
number, being isolated from empyaemata and other pneumococcal
conditions, and many of them were employed a few days later in testing
the sera of the patient from whom they were isolated. All the strains
reacted similarly to variations in the opsonic index. The great essential
in the preparation of the pneumococcal emulsions is the securing of
homogeneity. One strain of pneumococcus had to be discarded owing
to the cocci adhering together in clumps, and was therefore quite unsuit-
able for the determination of the average number of cocci in a number of
leucocytes. The best means of attaining uniformity is to thoroughly
shake up the emulsion and then centrifugalise for a few minutes. The
masses and clumps are then deposited and the uniform upper layer of
emulsion employed in the tests. The cultures were twenty-four hours

(374)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS n

old ; in no case were cultures older than forty-eight hours used. The
culture medium employed was glycerine-agar.

Of the twenty-five cases of acute croupous pneumonia, fifteen were
examined regularly from day to day throughout the course of the disease,
and the opsonic index of the serum determined before, during, and after
the crisis. In the other ten cases it was only possible to obtain the serum
at less frequent intervals, but the results in those in no case disagreed with
the opsonic curve determined for the others.

It will be convenient to discuss the relation of the opsonic index to :

(i) The chill period.

(2) The course of the disease.

(3) The final result, whether going on to rapid or tardy resolution, to
a fatal issue, or to complications.

(i) At the chill period I have only been able to examine the serum
of one patient. This was obtained two hours after the initial rigor, while
the patient was feeling very cold and shivery. The opsonic index stood
at -65 ; the original index was unknown, but in a healthy girl, such as the
patient, was probably near that of a normal person.

(2) During the evolution of the disease the opsonic index of the
disease is considerably below unity, affording indices varying from 0-45
to o*8. The lowest index observed was 0*45, in a poor under-nourished
woman, who rapidly succumbed. In fact, all the cases in which the
opsonic index fell below 0*5 terminated fatally. Considerable daily
variations are present in the development of the disease. A progressive
fall in the opsonic content of the serum does not seem invariably to
indicate a fatal issue, though such a fall was observed in most of the fatal
cases (see Chart I.).

(3) On the incidence of the crisis the greatest changes take place
in the opsonic index. For some twelve hours the amount of opsonin
in the serum is vastly increased, and with the increase the patient's
condition seems to improve very greatly, and the critical phenomena
occur. The amount of opsonin began to increase before any fall
in the temperature was noticeable in many cases in which, with a
high temperature in the morning, but a normal or higher opsonic index,
one could almost predict the critical fall to take place during that day.
A pseudo-crisis is not associated with increase in the opsonic index.

(375)

12

QUATERCENTENARY STUDIES IN PATHOLOGY

In cases undergoing rapid resolution the opsonic index rises suddenly
to above normal, reaching i*i to i"6. When the resolution is more tardy
in nature, the rise is, as a rule, not so high, but a slight increase to the
normal comes on and persists, the temperature and the general condition
improving gradually a day or two later (Charts II. and III.). In the
cases of rapid resolution, the index again falls to normal in periods
varying from one to three days.

In fatal cases the opsonic index goes on diminishing till the exitus
lethalis. This was constant in the five cases with fatal issue. In one
case the temperature fell to normal, after a typical pneumonia, but the
opsonic index did not rise. He died on the following day.

Chart I.

lO

A fatal case of pneumonia. The thicker line and numbers represent the
opsonic curve ; the other, the temperature.

Of complicated pneumonia, I have examined two cases of empyaema,
one of meningitis, one in which the opposite lung was attacked after the
first crisis. The meningitis case died in three days with falling opsonic
index. (The pneumococcus was isolated from the cerebral membranes
post mortem). One of the empyaema cases died, the other survived.
In the fatal case the pneumonia seemed to crisis, but not satisfactorily,
for, two days later the opsonic index rose, and continued rising ; the

(376)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS

13

T . T 1f\A°

;

z

Chart I

[.

6

7

s s

---^^^

\

A

/^

\

K

A

J

\

A

1

•3 '<-'•*

y ,

h

\/

\ ,

1

V

/

.ft ini°

V

V

V

/

V

%

1

A

\

/

/

\

v-A ^

\

/

s^^s

\

<^«

^ .

J

5 I

. R o,ffi

•5 30

.A. C,V°

From a case of pneumonia undergoing gradual resolution, and showing the gradual
rise of the opsonin index (indicated by the thicker curve) to unity.

1-7

1-6 /03°

1-5

1-4 /o;2°

13

1-2 10 f

I- 1

I-O 70(f

09

0-6 SS>

07

0-6 S6

05

97°

Chart III.

Dec'. 15 16 n :s 19 zo zi zz Z3 zt

1

h. A-ie

fw fv

1 r^ fcX

^ ^ X^>^

^ Zi 1 A ^

17 r ^ s.^^^^^-

^"^ t ^4

J V3

7 ^^A ^ ^^

y^ V V^^ ^v^

•■ — ^"^

— 1 1 1 1 1 1 1

From a case of pneumonia undergoing rapid resolution. The thicker line and numbers
represent the opsonic curve, the others that of the temperature. The opsonic mdex
rises above the normal, and falls to normal two days after the crisis.

(377)

14 QUATERCENTENARY STUDIES IN PATHOLOGY

pleural cavity was then aspirated and a large quantity of pus containing
pneumococcus withdrawn. In spite of this, the patient died with an
index above normal. In the other case the opsonic index rose to i'5,
but on opening the pleural cavity it fell to normal and healed quickly.

The amount of opsonin present in the serum and exudate in post-
pneumonic empyaema was examined in four cases. Case i gave an
opsonic index of i'4. Twenty-two ounces of pus were aspirated from
his chest (pure culture of pneumococcus). The opsonic index of the
liquor puris compared with the blood serum was o, the leucocytes refusing
to pick up a single coccus. Two days later the empyaema was drained ;
the blood-serum at time of operation had opsonic index of i'35, and the
liquor puris of the exudate 0'5. Next day, that of the serum was 1*37,
and of the exudate i '36. The progress of the case was rapid, the opsonic
index of the serum and fluid of the exudate were tested successively
for ten days, and both adjusted themselves gradually till both were
approximately normal. The exudate always gave a slightly lower
reading than the blood-serum. The blood-serum of Case 2 gave an
opsonic index of ri on the day subsequent to draining operation on his
pleural cavity, while the exudate gave an index of 099. The infection
was a pure pneumococcus one, and the discharge ceased within a week.
Case 3 was a child, and fatal. Along with pneumococcus. Staphylococcus
albus was present. Two days before death the opsonic index of the serum
was 1*2, and that of the exudate 0"3. Case 4 had a discharging empysema,
subsequent to pneumonia, for two years. The organisms isolated were
Bacillus coli^ Staphylococcus albus, etc. The opsonic index of the serum
of the blood was 0*55, and that of the discharge 0*52. Treatment by
continuous negative pressure was continued for two months, during
which he improved greatly. At the end of the time the opsonic indices
in the blood-serum and exudate fluid were i*o8 and 0*85 respectively.

The seventeen cases definitely diagnosed as broncho-pneumonia came
from the children's wards. These were similarly tested with pneumococcus
and their indices determined. The results in these cases are very much
less definite than in genuine croupous pneumonia in adults. In some of
the cases with a high temperature and signs of a severe infection the
opsonic index was normal throughout. In most of the cases the results
are so varying that a definite curve for the course of opsonin in this
aflection could not be attempted. Of the nine cases in children diagnosed

(378)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS 15

clinically as acute croupous pneumonia, three gave charts very similar to
those of adults ; the others were more or less irregular.

The significance of the variations in the opsonic content in the serum in
acute pneumonia.— In the majority of patients suffering from staphylococcic
infections, as furunculosis, sycosis, acne, etc., or tuberculous infections, as
phthisis, lupus, Addison's disease, etc., the opsonising power of the serum
has been found to be below that of a normal individual. The same
is true of pneumococcal infections and in the pre-critical period of acute
pneumonia. The critical changes in the course of pneumonia must be
regarded as phenomena of immunisation. From the nature of the
opsonin it is manifest that, with the rise in the amount of opsonin in the
serum, and accordingly in the serum bathing the leucocytic exudate
in the pulmonary alveoli and capillaries, more pneumococci are prepared
for the leucocytic meal, and they are taken up in greater number, the
intoxication of the individual rapidly ceases, and the disordered
metabolism is swiftly restored.

Should, however, the amount of opsonin in the serum fall instead of
rise, the quickly proliferating cocci are not taken up in such numbers by
the phagocytes ; the pulmonary alveoli and capillaries thus contain many
free cocci, a condition of bactcriaemia results, and with an aggravation of
this condition intense intoxication and death.

In the case of empyaema developing from pneumococcal pleuritis, the
numerous cocci present and proliferating between the layers of pleura
absorb all the opsonin from the exudate effused. On opening such a
pleural cavity fresh serum laden with opsonin is poured out into the
cavity, the cocci are taken up rapidly by the leucocytes in the granulating
zone and exudate and destroyed. The condition, therefore, tends to go
on to healing should the opsonic content of the serum be sufficiently
high. If it be low, it is manifest that the serum poured into the opened
pleural cavity would also be low, the cocci would be less efficiently
picked up by the leucocytes, and the disease would tend to continue for
a longer period and become an essentially chronic discharging empyaema.

Analogous cases are tuberculous and staphylococcic abscesses, in
which, as Wright and Bulloch have demonstrated, the fluid of the pus of
a closed abscess possesses no opsonic power, although the blood of the
patient exhibits it in a considerable measure. If these be opened, the
lymph afterwards flowing from the wound has high opsonic power.

(379)

1 6 QUATERCENTENARY STUDIES IN PATHOLOGY

The high opsonic index registered in cases of closed empyaemata may
be regarded as an expression of immunisation from resorption of
immunising products from the pleural cavity. When opened, and the
conditions favouring resorption removed, the opsonin in the serum is
restored to normal in a shorter or longer period.

That the variation in opsonin in acute croupous pneumonia is a
specific variation for the pneumococcus is shown by testing the opsonic
content of the blood for another bacterium — e.g,^ the staphylococcus.
Thus it is shown that while the pneumococcal opsonin undergoes great
changes in quantity during the course of acute croupous pneumonia,
the opsonic content for staphylococcus varies from the normal in only a
slight extent. Several of the cases were tested throughout, and in each
case this was verified. Thus, one day three patients gave indices for
pneumococcus of 0'5, 0*65, and 0*95, and staphylococcus 1*09, 0*85, and
i*o respectively. Two days later, the same cases were tested for both
organisms, giving indices of 0*85, 1*3, and 1*03 respectively, while the
staphylococcic indices were i*o, 0'86, and I'oi respectively.

Another method of demonstrating the specificity of opsonin in the
serum is by an absorption experiment. Serum is taken and the mixture
digested with a measured quantity of thick pneumococcal emulsion in.
normal salt solution for fifteen minutes at 37° C. A similar measured
quantity of normal saline solution is added to the control. The cocci
are then removed by centrifugalisation and the supernatant fluid tested
against the control. The result is: Control for pneumococcus, I'oo;
supernatant for pneumococcus, 0*12; control for staphylococcus, i*oo;
supernatant fluid for staphylococcus, 0*95. The remainder of the
supernatant fluid is divided into two equal parts, and a quantity of thick
staphylococcus emulsion added to one while an equal quantity of saline
is added to the other half Both are digested for fifteen minutes at 37°
C. The cocci are deposited by the centrifuge. The supernatant fluid is
tested with pneumococcus, stahpylococcus, and tubercle bacillus.

For Pn. For Staph. For T. B.
Opsonic content of control . . i i i

II II supernatant fluid o'lo 0*13 076

From these experiments it may be deduced that there is an opsonin
specific for each organism or group of organisms (the opsonin for
Staphylococcus aureus does not seem to be different from that of Staphylo-

(380)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS 17

COCCUS albus), and the changes in the opsonic content in a particular
infection concern only the opsonic content with reference to the infecting
agent.

In the course of active immunisation of animals against the pneu-
mococcus, the opsonic content of the serum follows a fairly definite course.

On inoculation of a rabbit with 1,000 millions of pneumococci killed
by heat (counted by Wright's method), the opsonic index fell to 7 and
•65 on the two succeeding days; on the third day, it rose to 1*3, and
remained above normal for ten days.

A second rabbit was inoculated with 30,000 millions of pneumococci.
The negative phase persisted for seven days, after which a very marked
positive phase, giving indices of v6, r8, 1-9, followed, the serum not
giving the normal index for four weeks after inoculation.

Many similar experiments with varying dosage gave results which
would point to the conclusion that the smaller the dose the less pro-
nounced and the less durable was the negative phase. The amount and
persistence of the positive phase was not so constant, as in many cases,
after a long negative phase from a large dose, the positive phase was
entirely absent, the opsonic index gradually rising to unity.

The specificity of the opsonin was also apparent in these experiments
on animals, for in some cases the serum was tested throughout against
both pneumococcus and staphylococcus, and showed that while the
opsonic index for pneumococcus was passing through wide variations,
that for staphylococcus remained almost unchanged from day to day.

In the lower animals there are great differences in opsonic index
observed between different species, but the individuals of any given
species do not appear to vary much. For pneumococcus the indices of
the laboratory animals are in comparison with man 0*4, 0*6, and o*8 for
rabbits, guinea-pigs, and rats respectively. On the other hand, their
opsonic index for the tubercle bacillus is found to be 0'6 and 0*4 for
rabbits and guinea-pigs respectively. Their opsonic index seems to
correspond very closely with their susceptibility to infection.

Nature of the Opsonic Substance in the Serum. — Before making any
systematic enquiry into variations in the opsonic content of the serum,
it is necessary to enquire into the effects of preservation, effects of light,
of dilution, of addition of salts, etc.

Effect of preserving the 5^^^^;;/. —Samples of blood were drawn from

(381)

iS QUATERCENTENARY STUDIES IN PATHOLOGY

the finger each day for 12 days, and kept in the dark in unsealed
capsules. The sera thus obtained were compared with fresh serum for
opsonic content (in all the experiments my own washed blood corpuscles
were used), and the experiments on normal serum were conducted on
my own serum or that of healthy normal individuals.

Cocci per leucocyte.

Fresh serum, 2 hours old - - - 17*6

Serum, 24 hours old - - - - i6"8

M 48 M - . . - 15-4

II 3 days old . . . . 14-3

II 5 M - - - . 14-2

II 8 II - . - . io*3

II 10 II . - . - 8*5

II 12 II - - . - 6'5

Serum over 3 weeks old - - - 2*5

Thus it is seen that there is a gradual diminution from the first day
onward, and that it would give quite erroneous results if we were to
compare sera of different ages with regard to opsonic content.

What are the changes the serum undergoes within the first few hours
of its withdrawal from the body ?

Blood was drawn from the finger at different times during the day,
and left exposed to daylight when all the samples were tested
simultaneously.

Blood drawn at

Tested at

Exact age.

Cocci per leucocyte.

11*55

3*51

3

hrs.

56 mins.

137

12-55

3'49

2

II

44 "

13*1

1-25

3'47

2

II

22 M

12-9

i'55

3*44

I

II

49 "

13-1

2'IO

3-38i

I

ti

28J II

11*3

2-25

3-36

I

II

II II

10-4

2 -40

3*29

49

mins.

9*9

2'55

3*27

32

II

97

3-10

3*25

15

"

9-0

330

3*42

12

II

9*5

Thus it is shown that there is a gradual increase up to about two
hours in the quantity of opsonin in the serum. It may be asked,
however, does the increase in opsonin depend on a change in the serum
itself, or is it due to certain substances squeezed out of the clot ?

(382)

I

IMMUNITY IN PNEUMOCOCCUS INFECTIONS 19

The following experiment solves this question. A large quantity of
blood was drawn, and the corpuscles driven down to the bottom of the
capsule by centrifugalisation. Quantities of the serum were then drawn
off at intervals, and stored in sealed glass pipettes. The contents of
these pipettes were all tested simultaneously with normal corpuscles and
bacterial emulsions.

Serum removed from clot after 5 mins. - - - 6-7

- - - 6-9

7'5

- - - 7-8
8-8
97

lO'I

10*7
10-3

lO'I

- 10-5

IO*2
107

Thus, contact with the clot seems to be necessary for increase in the
opsonising power. This increase was observed in every instance, and it
seems a constant law that the opsonic power of the serum increases up
to a period of two hours, after which it gradually diminishes so that, by
the seventh or eighth day, it has lost 50 per cent, of its opsonin.

A probable explanation of this property is that the clot acts as a
permeable filter, which, as it contracts, squeezes out constituents in a
definite order, according to the size of the molecule :

1. Sodium chloride and salts ;

2. SemicoUoids ;

3. True colloids, to which there is much evidence that substances of
the opsonin type belong. This hypothesis is substantiated by the fact
that the serum squeezed out last from the contracting clot contains more
opsonin per volume than that squeezed out at any earlier period in the
contraction.

From these figures, it is evident that, before the sera of two indivi-
duals can be compared accurately with regard to their opsonic indices,
they must be subject to the same physical conditions for similar periods.

(383)

15

30

45

60

75

90

2 hrs.

4

hrs.

3 hrs.

3i

hrs.

4 hrs.

4i

hrs.

4

+

6

+

lO

+

15

+

20

+

30

+

40

+

50

+

60

+

20 QUATERCENTENARY STUDIES IN PATHOLOGY

Effect of Temperature. — If the serum of a normal healthy individual
be exposed to a temperature of 55 to 60° C. it loses its opsonic power
almost entirely ; never, however, completely, for even after heating to
60° for one hour there is always a small irreducible minimum left.

Cocci per
leucocyte.
Normal serum + Normal corpuscles -f Cocci 18*2

Normal serum, heated to 60" for 2 mins. + n +11 15 "3

II + M 13*2

II 4- M 107

+ " 5*5

" + " 5*3

II + M 37

+ M 3-5

II + M 3*3

M + II 3*1

II + II 2'9

Heating to 60° for 60 minutes destroys 83 per cent, of the opsonin
contained in the serum, and therefore the opsonin in the serum of normal
individuals may be regarded as thermolabile.

That the irreducible minimum after heating normal serum is still
opsonin is shown by an experiment in which a large quantity of cocci
are passed through the heated serum and afterwards removed by the
centrifuge.

A quantity of normal serum is heated for 30 minutes in the water-
bath at 60° C. A thick emulsion of cocci is added, and the mixture
allowed to digest for 15 minutes. The cocci are then removed by
centrifugalising for two hours. Result : —

Cocci per
leucocyte.
Normal serum - - 15-3

Normal serum heated to 60° for 30 mins. - - - 2*4

Serum after treatment as above - - - - - 0*14

The substance has evidently combined with the cocci, and is thus also
opsonin.

The role of the Leucocyte. — The comparative indifference of the
behaviour of the leucocyte in phagocytic phenomena was shown by the
early experiments of Wright, and later by Bulloch and Atkin (3°). In a
comparison of normal serum and that of a lupus patient, Bulloch and

(384)

\

5*3
5*2

2-5

2*4

IMMUNITY IN PNEUMOCOCCAL INFECTIONS 21

Atkin found that the leucocytes of a lupus patient and those of a normal
individual were approximately identical in action. To quote their
experiment : —

I. Serum of a normal individual + T. B. Emulsion + leucocytes of a normal

individual - - 5*7

2- " '• + 11 + leucocytes of a lupus

patient- - - 5*4

3- " •• + „ + „ „

4- " " + ri + „ „

5. Serum of a lupus patient + T. B. Emulsion + leucocytes of a normal

individual

6. It II + „ 4. „ „

7- " n + II + II M 3*2

I have determined the behaviour of the leucocytes in acute croupous
pneumonia, a disease in which there is pronounced changes in the
number of leucocytes at the different periods.

Serum of Normal individual + Pneumococci + normal leucocytes - - 13*2

Serum of Pneumonia patient No. i + Pneumococci + normal leucocytes 6*5

Serum of Pneumonia patient No. 2+ m + m 9-7

Normal serum + Pneumococci + leucocytes of Pneumonia patient No. i 9*5

Normal serum + n + n n No. 2 8*4

From these results it would appear, at first glance, that the leucocytes
were not so indifferent, and that the leucocytes of a pneumonia patient
were not capable of taking up so many cocci in the same period of time.
But the number of leucocytes in the experiments was here variable, for,
in the normal blood, on the day in which the experiment was performed,
there were 8,200 leucocytes per cm.m., in pneumonia patient No. i, 18,000,
and in No. 2, 19,200. Many more leucocytes were present, and actively
phagocytic, in the last two cases.

Total in
50 leucocytes.

Normal serum + Normal leucocytes + Pneumococci . - - - 660
!i + Leucocytes of Pneumonia patient No. i 4- Pneumococci 475

It + M II No. 2+ II 420

But if numbers in proportion to the number of polymorphonuclears of
patients with pneumonia were taken, the results show that the poly-

(385) A A

22 QUATERCENTENARY STUDIES IN PATHOLOGY

nuclears of pneumonia patients are equally active to those of normal

blood.

Total number of cocci in 40 leucocytes of normal blood - - - 540
») It 90 II Pneumonia patient No. i - 700

ti II 85 It M No. 2 - 720

In some of the cases it was very remarkable how wonderfully the
counts in relative numbers totalled up. In a case of pneumonia in which
there were, on the days of estimation, 12,500 leucocytes to normal 8,700,
the results were : —

Cocci in
50 leucocytes.

Normal serum + Normal leucocytes + Pneumococci - - - 473

•1 + Pneumonia patients' leucocytes 4- Pneumococci - 376

But the number of polymorphonuclears of normal blood to that of the
patient's blood was approximately i : 1*4, and, if one adds two-fifths to
the pneumonia patient's count, we have 470, as compared with 473 in the
normal man.

Effect of dilution of the Serum. — Many experiments were performed

on the effect of diluting the serum of the individual examined by normal

saline solution (0*85 per cent.). Serum of a normal individual was

diluted with definite quantities of saline solution. Three parts of the

normal serum and three parts of each of the diluted specimens were

added to the leucocytes of a normal individual and an emulsion of cocci.

The mixtures were incubated at 37° C. for 15 minutes, films made, and

the total number in 80 leucocytes estimated and an average determined.

Result :—

Cocci per
leucocyte.

Normal serum 3 parts 10*9

Normal serum 2 pts. + Normal saline i pt. - - 107

Normal serum i pt. + n 2 pts. - - 10*2

Normal serum J pt.+ m 2 J pts. - - 5*9

(Normal serum : Normal saline = i : 8) 3 pts. - - 2*0

( II II =1 : 10) 3 pts. - - 1*9

The conclusion drawn at first glance from these figures is that the
opsonic index is not an index of the amount of opsonin in the serum,
seeing that the serum can be diluted to as much as one in three with

(386)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS 23

very little change in the opsonic index. The time given for the inter-
action of the fluids, bacteria and leucocytes, did not allow either the
cocci to be thoroughly opsonised or the leucocytes to take up the cocci
that were opsonised. If the interaction is allowed to go on for 45
minutes, the result is somewhat different : —

Cocci per
leucocyte.
Normal serum 3 pts. 16-7

Normal serum 2 parts + Normal saline i pt. - - 117

n I part 4- II 2 pts. - - 5*9

II J part + II 2J pts. - 3*4

(Normal serum : Saline = i : 8) 3 parts - - - 1-9

( n M =1 : 10) 3 parts - - i'8

The numbers in this latter experiment fall almost exactly proportional
to the dilution.

From this and the preceding experiment the conclusion to be drawn
is that the most accurate means of determining the opsonic content and
opsonic index of any serum is to allow either the serum, leucocytes and
bacterial emulsion, to interact for at least 30 to 45 minutes, or to dilute
the serum one part in 3 parts with normal saline, if the period of
interaction allowed is to be 15 minutes.

Effect of dilution with Na CI. solutions stronger than 0'85 per cent.

Cocci per
leucocyte.

Normal serum 3 volumes - - - - - - 10*5

Normal serum i vol. + '85 °l^ Na CI., 2 vols. - - - 4*3

M +I'0% II - - - 2-0

+ 1-5 7o " - - - 1*7

u + 2 % II - - - 1-6

+ 5 7o " - - - 1-3

(These were incubated with leucocytes and coccal emulsion for 30
minutes.) Increasing the concentration in salt seems to diminish the
amount of opsonising power of the serum.

It is affirmed by Buchner that the destruction of alexine on heating
to 55° C. can be paralysed by adding salt. Whether the same was true
of opsonin was tested. Normal serum was diluted in the proportion of

(387) A A I

24 QUATERCENTENARY STUDIES IN PATHOLOGY

two parts of serum and one part of different strengths of salt solution,
and then subjected to 55° C. for varying periods.

I. Normal serum (2 vols.) + '85

7c

, NaCl

.(

I vol.) unheated
heated to 55° for 5 mins.

Cocci per
leucocyte.

17-9

io'6

11 for 10 II

6-9

ti for 15 II

3*1

II for 20 M

2*0

II. Normal serum (2 vols.) 4- 1

7o

NaCl.

(i

vol.) unheated -

12*5

heated to 55" C. for 5 mins.

7 '9

II for 10 II

2*1

M for 15 II

I 3

II for 20 II

0-8

III. Normal serum (2 vols.) + i

•5

7oNaCl.

(i vol.) unheated

II-3

heated to 55° for 5 mins

21

II for 10 II

12

It for 15 II

0-9

II for 20 II

o-S

In the experiments with 2 % and 5 "/^ NaCl. the opsonin seemed to
be entirely destroyed by heating for five minutes to 55° C.

Far from preserving the opsonin from destruction by heat, the
increasing concentration of salt seems to favour its destruction very
greatly.

The work of Wright and Douglas, corroborated also by Bulloch and
Atkin, has shown that the opsonin disappears from the serum when the
serum is digested with bacteria at 37° C. or at 0° C. We have seen that
serum heated to 60° C. for some time loses the great part of the opsonin
present. Is this really a destruction of the opsonin, or does it pass into a
modification (opsonoid), in which it does not fully prepare the bacteria
for phagocytosis? Can a complement-like body be isolated as in
haemolysis at 0° C. which would give a reaction with heated serum ?

An emulsion of staphylococci (pneumococci gave a similar result) was
added to normal serum and digested for 30 minutes at 0° C. The
centrifuge was then applied, and a supernatant clear fluid obtained with
a deposit of staphylococci. Serum was also heated to 60° C. for 30
minutes.

(388)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS 2$

Cocci per
leucocyte.
Normal serum + Normal leucocytes + Normal cocci - 19-1

Normal serum heated to

60° for 30 mins. + n 4. „ _ ^.3

Supernatant fluid i pt. \
Heated serum 2 pts. / "^ " + <- - 2-9

Supernatant 2 pts. \

Heated serum i pt. / "^ " -f m - 3-2

Supernatant fluid 3 pts. + „ + „ - 2*2

The deposit was then made into an emulsion with normal saline, and
tested with normal leucocytes, which showed that the cocci were
opsonised, and, therefore, that the opsonin had passed from the serum
into the cocci. In haemolytic experiments, if a similar procedure is
undertaken at 0° C, there can be separated from the supernatant fluid a
complement which would give a reaction with heated serum. This is
evidently not the case with opsonin. It has been alleged by Crofton (3i)
that there are two bodies in normal serum contributing to opsonisation
of bacteria, one thermolabile corresponding to the complement, and the
other thermostable corresponding to the immune body. His experiments
do not give much support to this deduction.

When bacteria are opsonised, heating for prolonged periods to 60° C.
is unable to destroy their power of being taken by the polymorphonuclear
leucocytes without the intervention of serum. When once opsonised it
is impossible to opsonise further, and opsonised cocci do not appear to
be capable of removing opsonin from normal serum.

Cocci were opsonised by digestion at 37° C, and the mixture of
serum and cocci was heated to 60° for 30 minutes. The cocci were
centrifugalised down, and the deposit added to normal serum, and
allowed to stand for 30 minutes. The centrifugi was again applied, and
the supernatant clear fluid tested against normal serum.

Normal serum + Corpuscles 4- Cocci = 1 1 '8
Supernatant fluid + n + " =11*2

Thus, passing opsonised cocci through the serum does not rob it of any
of the opsonin contained.

Is the increased opsonic power developed during immunisation of the
same nature as that of normal present in the serum ?

(389)

26 QUATERCENTENARY STUDIES IN PATHOLOGY

In the investigation of this question, sera from several sources were
examined, serum from pneumonia patients at the crisis with high
opsonic index, serum from a patient suffering from acne vulgaris who
had been treated with injection of staphylococcus vaccine, and sera of
rabbits actively immunised against staphylococcus and against pneumo-
coccus.

Sera from each of these sources gave the same reaction towards
physical agents. They shall be designated " Immune sera."

The only great difference in the behaviour of immune sera and normal
sera towards chemical and physical influences is in relation to the effect
of heat. Heating to 60° destroys 80 to 90 per cent, of the opsonising
power of normal serum. In immune sera heat does not exert such a
great influence.

The serum of a patient who had received repeated injections of
staphylococcus albus was tested. His opsonic power, with relation to a
normal individual, was i*8. This was subjected to heat for 30 minutes.

1. Control serum i4'8

2. Patient's serum ------- 237

3. No. I heated to 60° for 30 mins. - - - - 2*9

4. No. 2 II ----- 6*9

From this experiment the opsonin does not seem to be destroyed to
such an extent by heat. Is the residue in the serum of the nature of
opsonin, or is it another substance ?

If it is opsonin, then it would be able to opsonise fresh cocci passed
through it. To demonstrate this, several experiments were performed of
which a type may be cited.

Immune serum and normal serum were heated to 60° C. for 30 min.,

the thermolabile opsonin being thereby destroyed. To the heated sera

were then added equal parts of an emulsion of staphylococcus, and the

mixtures were digested at 37° C. for 30 minutes. At the end of this time

the mixtures were centrifuged, the supernatant liquid and the cocci

being separately tested with the following result : — Cocci per

leucocyte.

1. Heated normal serum 2*9

2. Heated immune serum ------- 6*9

3. Supernatant fluid after digestion with normal heated serum - 0*9

4. Supernatant fluid after digestion with immune heated serum 6'o

5. Cocci after digestion with normal heated serum - - - 2*3

6. Cocci after digestion with immune heated serum - - 2-1

(390)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS 27

Similar experiments on immune sera from various sources gave a
similar result, though the amount of thermostable element varied. greatly
in quantity. It is not proportional to the amount of opsonin present, or
to the opsonic index.

The opsonic indices of two individuals may be the same, though one
has been subjected to active immunisation. Both may be unity, yet,
when subjected to heat, the difference between the two becomes
apparent.

Example:— Cocci per

leucocyte.

Immune serum -------- n-g

Normal serum -------- 8*8

Heated immune serum - 3-2

Heated normal serum - - - - - - - . i-i

Supernatant fluid after digestion — immune - - - 3-4
i» II — normal - - - o'8

Normal corps. + deposit from "immune ".digestion - 2*0
II + II " normal " digestion - - 2*1

It would appear, therefore, that the serum of an individual which has
been raised artificially to the normal standard differs considerably as
regards thermolability from that of a normal individual.

With regard to the nature of the residual body after heating, it is
present in the serum, not destroyed by heat, it does not attach itself to
fresh bacteria added, and yet it causes phagocytosis. Most probably it
acts as a stimulant to the leucocytic amoeboid movement — a true
stimulin.

In the active immunisation of animals, this residual substance appears
previous to the increase in opsonic index, and disappears for a short time
before the opsonic index has again fallen to unity.

Relation of the Opsonin to the Leucocytosis. — In the examination of
typical cases of acute croupous pneumonia, it would appear as if the
leucocytosis curve and that of the opsonic content were inversely
proportional, and this is so close that prima facie one might be led to
trace phenomena of cause and effect, either that the rise in opsonic
content caused the fall in the number of leucocytes or vice versa.
The best method of determining whether any causal nexus exists between
the two factors is by animal experiment. Thus, a rabbit was inoculated

(391)

28 QUATERCENTENARY STUDIES IN PATHOLOGY

with I cc. of a killed agar emulsion of pneumococci subcutaneously at

1 1 a.m, when the leucocyte count stood at 8,000 per c.mm. The result

was : —

At I p.m. 8,800 per c.mm.

At 2*15 p.m. - - - - 5,600 It

At 3*15 p.m. - - - - 5,060 I.

At 4*15 p.m. - - . - 7,800 M

At 5*15 p.m. . - - - 8,400 II

At 6 p.m. ----- 9,800 II

The next morning, the leucocytosis was 8,600, and on subsequent days
6,800, 8,900, and 8,350. The quantity of opsonin in the serum during
the first day when any leucocytosis present occurred (leucocytosis from
bacterial inoculations usually attaining a maximum in about 12 hours if
injected subcutaneously, and 8 hours if intravenously), there was only a
slight fall in the opsonin (at 4'i5 p.m.); on the following day, when
the leucocytes were normal in number, the opsonic index was 07 ; then
o*5 the day after; two days later i*i, and then on the second day after
that 17. Thus, in the case of a single inoculation of a bacterial irritant,
the leucocytosis does not run a similar inverse course to that of the
opsonic content. Is there then any physiological nexus between the
two?

The course of leucocytosis in acute croupous pneumonia may, like
the opsonin course above described, be conveniently treated under

the headings —

1. Chill period.

2. During the evolution of the disease.

3. Terminal leucocytosis.

Chill Period. — Lceper examined two cases in this stage, with the teeth
chattering, and found the total number of leucocytes raised to 24,000 and
18,000. This fact would indicate that the chill does not indicate the
beginning of the pneumonia, but most probably, as in the inoculation
above described, there was a period of hyperleucocytosis, as was some
time ago suspected by Werigo, who affirms that the leucocytes rush
in virtue of chemotaxis to the focus of infection, the haimopoietic organs
being simultaneously excited to send out the reinforcements necessary.
By this hypothesis, the hyperleucocytosis would, like the chill, interpret

(392)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS 29

the general reaction of the organism and stimulation of the defensive
apparatus by the bacterial irritant.

During the Evolution of the Disease. — The usual number is about
18,000 to 24,000 per c.mm. The maximal leucocytosis is always about
the first day or two, and it diminishes slightly afterwards.

Terminal Leucocytosis. — In cases undergoing rapid resolution there is
usually a sudden rise in the number of leucocytes just on the eve of the
temperature-fall, by about 2000 to 6000 : on the day of the crisis the
number is much reduced, and the day following is almost always nearly
normal. In cases terminating by lysis, the rise in leucocytosis just before
the crisis usually fails to occur. In complicated or fatal cases the
leucocytosis may not fall to normal, but go on increasing until the fatal
end. With very severe infections, however, or in patients with poor
resistance, alcoholic patients, etc., acute pneumonia may run its course
without any leucocytosis whatever.

On a single intravenous inoculation of pneumococci, there is a
pronounced leucocytosis in five hours. In acute lobar pneumonia the
researches of Prochaska (32), Cole (33), and Frankel (34), show that the
pneumococcus can be very frequently isolated from the blood of patients
suffering from typical acute croupous pneumonia, if sufficient quantity of
the blood is removed, at least 10 cc. being required. It would seem as if
there were discharges, slight and at frequent intervals, of pneumococci
into the pulmonary capillaries and veins from the morbid lung, a
condition of bacteriaemia excited with pronounced stimulation of the
haemopoietic organs and a high degree of leucocytosis. The leucocytosis
is to be regarded as the cellular reaction product of the organism on
infection, the anti-bodies produced in the serum being the humoral
reaction products.

Agglutination Phenomena in Acute Croupous Pneumonia. — In this
affection there are two separate agglutination phenomena witnessed —
agglutination of the red blood corpuscles, and bacterial agglutination.
The commoner and more uniform is the chromocytic agglutination, of
which some mention may be made.

Wharton Jones (35) stated that in the blood of inflammatory diseases
" the rouleaux are formed more rapidly, and run into masses which have
large spaces between them, and thus the buffy coat is produced by this
close aggregation and more rapid subsidence therefrom resulting." If

(393)

30 QUATERCENTENARY STUDIES IN PATHOLOGY

one adds a small quantity of the blood serum of a patient sufifering from
acute pneumonia to a small quantity of the corpuscles of a normal
individual and examine in hanging-drop, the red blood corpuscles are
clustered into large-meshed masses. Dilution with normal saline in
sufficient quantity abolishes this reaction. A better result was obtained
in the later cases of acute pneumonia, in which I used long sedimentation
tubes used for Widal's macroscopic typhoid reaction. The dilutions
borne were not very high. Many refuse to become agglutinated in i : 4
or I : 10. In the case of empyaema, which had been discharging for two
years, the blood serum agglutinated my corpuscles in dilution i in 20,
and the fluid from the pus i in 4. The agglutination usually began one
or even two days before the crisis, and continued for a week after. The
agglutination of the red corpuscles does not seem to stand in close
relation to bacterial agglutination, as the latter was very slight, if at all
present, in cases where the red cells were very markedly agglutinated.
That the property lies in the serum is of easy demonstration, since the
haemagglutinin is thermolabile, being destroyed by 60° C. like the
opsonin and also the haemolysin. The addition of normal sera to the
hsemagglutinating serum is, as a rule, not capable of inhibiting
hsemagglutination.

Bacterial Agglutination. — In 1891, a phenomenon very close to
agglutination was observed by Metschnikofl* (36), even before the
publication of the papers of Gruber and Durham in 1896. He noted
that in the serum of rabbits immunised against the pneumococcus the
growth was in long coiling chains, while the same organisms in normal
rabbits' serum have the usual diplococcic form. These facts were amply
verified and more closely investigated by Mosny, Kruse and Pansini (37),
Washbourn, Bezangon and Griffon (^8), and others.

The agglutination of the pneumococcus in the sense of the Gruber-
Widal reaction with the microbes in bouillon culture after addition of
quantities of specific serum was first described by Neufield, who comes to
the conclusions : —

1. The occurrence of a strong agglutinating serum in immunised

rabbits is entirely independent of the degree of immunity.

2. The agglutination phenomena are specific for the pneumococcus.

3. The agglutination of the pneumococcus shows many peculiarities,

among them the great resistance of the bacteria against high
degrees of heat. (394)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS 31

In none of the rabbits immunised by pneumococcus did I obtain
an agglutinating serum. The serum of several cases of pneumonia
examined agglutinated the pneumococcus markedly, but in none were
such high dilutions as i in 40, i in 50, etc., obtained. In many cases the
addition of equal parts of a fine pneumococcal emulsion to serum, and
incubation for J-2 hours at 37° C, succeeded in demonstrating that
agglutination had taken place. On examining this microscopically, the
individual cocci seem to become swollen and run together into masses,
thus verifying the truth of Gruber's assumption of the swelling and
cohesion of the outer bacterial coats. The dilutions are best performed
by Wright's sedimentation pipettes, in which they can be quickly and
accurately performed and incubated at body-temperature.

If an agglutinating serum be diluted i to 4, the agglutination
is not so marked as in the undiluted serum, and microscopically
also the appearance differs ; the deposit is not so granular. The
cocci do not appear so much swollen, and are joined together into
coiling chains — " the Fadenreaction." As shown by Neufeld, the
agglutinin is thermostable, but is destroyed by 100° C. It can withstand
60° C. for more than one hour. On healing the agglutinated cocci (in
the undiluted experiments) at 60° C. for an hour, the clumping and
swelling of the bacteria disappeared.

As regards agglutination, in the Gruber-Widal sense of the term, the
cases examined lead to the conclusions : —

1. That it is not present in every case of acute croupous pneumonia :

it is very frequently absent altogether in cases which " crisis "
before the fifth day, and the longer the time till the crisis
occurs, the better is it marked.

2. That the serum is very insusceptible of dilution, i 14 being

considered as a high dilution.

3. That it is essentially a critical phenomenon, appearing at, and

disappearing soon after, the crisis.
The most recent addition to our knowledge of immunity has its origin
from the workers in the German University of Prague. Bail P) has
observed the great sensibility of tubercular guinea-pigs to fresh injections
of living tubercle bacilli. On intraperitoneal injection, the animals die
within a few hours, with an exudate in the pleural cavity, the cells

(395)

32 QUATERCENTENARY STUDIES IN PATHOLOGY

of which showed no phagocytosis whatever. Animals which had not
attained this stage of hypersensibility either do not die at all, or only
after some days or weeks, and then with well-marked phagocytosis in the
exudate. He has shown that in the blood-serum of tuberculous animals
or tuberculous exudate there is a substance which he names " Aggressin,"
which has the power of holding active phagocytosis in abeyance in
anrmals in which the tubercle bacillus is a true parasite ; and he has
been able by successive inoculations of these exudates to attain high
degrees of immunity. Hoke (^) has applied himself to the investigation
of the "aggressive" action of diplococcic exudates, and comes to very
remarkable and important conclusions. The rabbit is employed,
being an animal in which the pneumococcus is a true parasite.-
By intrapleural injection of cultures of pneumococcus isolated from a
lung abscess secondary to acute croupous pneumonia, a rabbit was killed
in 8 to 10 hours, and, post mortem, 20 cc. of exudate was obtained from
the pleural cavity. In this exudate numerous cocci were found, but no-
phagocytosis. The exudate was centrifugalised for 12 hours or longer,-
and all the cocci removed and subsequently carefully sterilised with
toluol. -By prolonged standing in the incubator the toluol separates and
forms a distinct layer, which can easily be pipetted off the top of the
serum.

Should such a pneumococcal exudate possess " aggressive " properties,
it must be capable of shortening the duration of the disease,
because it holds in abeyance the natural protective power of the body,
and allows the bacteria to multiply rapidly, without being engulfed by
the phagocytes. Intrapleurally and intravenously were injected in one
series of rabbits fresh pneumococcus emulsions in quantities of 0'2 to 0*5
cc, along with a quantity (2 cc.) of exudate : into the control series,
instead of the exudate, normal saline was injected. The first series died
in every case earlier than the control.

A remarkable property of ^' aggressive " exudates is that the aggressin
action can be inhibited by the addition of leucocytes. This was also
shown by Bail (^9) for cholera, and by Hoke (^) for staphylococcic
exudates, though first demonstrated by Kikuchi (^i). Aleuronat was
injected into the pleural cavity twelve hours before an injection of
aggressin, and pneumococcal emulsion, and in that case death was much
later than in the controls to which no aleuronat had been administered.

(396)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS 33

By treating animals with repeated injections of aggressin, great
protective power is produced. An anti-aggressin is developed, by which
phagocytosis is enormously stimulated. Thus, after three inoculations
of an actively " aggressive " pleural exudate at intervals of five days, a
rabbit remained alive with a dose of pneumococcus emulsion which
killed another in 18 hours. Passive immunisation with the serum of
rabbits treated with aggressin seemed also to protect against pneu-
mococcal infection.

In the mechanism by which the different anti-bodies act in the
production of immunity against pneumococcus, it will be seen that the
great immunity agent is phagocytosis, in which the leucocyte, in itself
a passive factor, is made capable of rendering great service to the
individual by engulfing the pneumococci which have been massed
into suitable clumps by the agglutinin, and rendered suitable pabulum
by the opsonin, while the leucocyte itself is excited to labour by the
stimulin in the serum. Aggressin also seems to be at the same time
capable of exciting or paralysing phagocytosis, according to whether
inoculated before or after infection.

(397)

34 QUATERCENTENARY STUDIES IN PATHOLOGY

LITERATURE.

1. RiESELi.E. Vierteljahrsschr. f. ger. Med., 1889.

2. MoLLMANN. Berl. klin. Woch., 1887, p. 729.

3. Klemperer, G. and F. Berl. Klin. Woch., 1892, p. 421.

4. FoA and Scabia. La rif. med., 1892, XL, p. 277.

5. Emmerich and Fawitzky. Munch, med. Woch., 1891, No. 32.

6. Washbourn. Journ. of Path, and Bact, 1896, III., p. 214.

7. Pane. La rif. med., 1897, II., p. 40.

8. Eyre and Washbourn. Brit. Med. Journ., 1899, II., p. 1247.

9. Romer. Arch. f. Ophthal., 1902, LIV., p. 99.

10. Passler. Deutsch. Arch. f. klin. Med., 1905, LXXXIL, p. 361.

11. Fanoni. New York Med. Journ., 1898, May 7, p. 647.

12. Dasaro-Cao. Baumgarten's Jahresb., 1902, p. 64.

13. Banti and Pieraccini. Baumgarten's Jahresb., 1899, p. 58.

14. DE Renzi. Gaz. degli Ospedali, 1896, p. 1476.

15. Concetti. Baumgarten's Jahresb., 1899, p. 59.

16. RiGHi. La rif. med., 1895, IIL, p. 566.

17. Spolverini. La rif. med., 1899, III., p. 598.

18. GoLDSBOROUGH. Joum. of the Amer. Med. Assoc, 1901, June.

19. Sears. Boston Med. and Surg. Journ., 1901, CXLV., p. 639.

20. Snively. Therap. Gaz., 1901, XXV., p. 382.

21. Neisser. Deutsche med. Woch., 1892.

22. Audeoud. Rev. m^d. de la Suisse rom., 1893, XIIL, p. 130.

23. TizzoNi and Panichi. Centralbl. f. Bakt., 1905, XXXVI., Ref., p. 25.

24. BoNOME. Fortschr. der Med., 1891, IX., No. 18.

25. IssAEFF. Ann. de I'lnst. Past, 1893, VII., p. 260.

26. TcHiSTOWiTCH. Ann. de I'lnst. Past., 1904, XVI IL, p. 304.

27. Denys and Leclef. La cellule, 1895, IX., p. 178.

28. Mennes. Ztschr. f. Hyg., 1897, XXV., p. 413.

29. Wright and Douglas. Roy. Soc. Proc, Vols. 72 and 73.

30. Bulloch and Atkin. Roy. Soc. Proc, Vol. 74.

(398)

IMMUNITY IN PNEUMOCOCCAL INFECTIONS 35

31. Crofton. Journ. of Hyg., 1905, V., p. 444.

32. Prochaska. Deutsch. Arch. f. klin. Med., 1901, LXX., p. 559.

33. Cole. Johns Hopkins Hosp. Bull., 1902, p. 136.

34. Frankel, a. Int. Beitr. z. inn Med. Leyden Festschr., 1902, II., p. 103.

35. Wharton-Jones. Brit, and For. Med. Rev., 1842.

36. Metchnikoff. Ann. de I'lnst. Pasteur, 1891, p. 371.

37. Kruse and Pansini. Ztschr. f. Hyg., 1891, XL, p. 279.

38. BESAN90N and Griffon. Ann. de I'lnst. Past., 1900, XIV., p. 449.

39. Bail. Miinch. med. Woch., 1905, p. 1865.

40. Hoke. Wien. klin. Woch., 1905, No. 13.

41. KiKUCHi. Berl. klin. Woch., 1905.

(399)

PRELIMINARY NOTE UPON THE BACTERIOLOGY
OF SOME DISEASES OF SHEEP.

By James Milner Adams, M.A., M.B., and

Bertie Ronald Gordon Russell, M.B.

(From the Pathological Department, University of Aberdeen),

(401)

B B

Freliminary Note upon the Bacteriology of some
Diseases of Sheep.

The heavy losses suffered by farmers on the west coast of Scotland
from the prevalence of certain endemic diseases amongst sheep, led to
the appointment by the Board of Agriculture of a Departmental
Committee to investigate the aetiology of these diseases. Our investiga-
tion is to be regarded as an extension of the work of this Committee and
of Professor Hamilton's own researches. (Parliamentary Report of
the Departmental Committee, appointed by the Board of Agriculture,
1906.)

The material for this investigation was provided by Professor
Hamilton, to whom we are indebted for suggesting the line of the
research, as also for the very great help he has given us in the carrying
out of the same. Liquid from the peritoneal cavity of sheep which had
died from the various diseases had been preserved, and it was from this
that the various organisms were isolated and grown. The method of
isolation, in almost every case, was as follows : — The medium employed
was glucose-bouillon of varying degrees of alkalinity. The surface was
covered with olive oil to a depth of one centimetre. The oil allowed the
escape of the gases dissolved in the medium during sterilization, and
prevented later any further access of air (see Parliamentary Report,
Part H., p. 20).

A tube of bouillon heated to 80° C. was inoculated with a small
quantity of the peritoneal liquid (previously incubated for 36 hours) by
means of a sterile capillary pipette, and kept at this temperature for 20
minutes. It was then rapidly cooled and placed in an incubator at if C.

(403)

4 QUATERCENTENARY STUDIES IN PATHOLOGY

Bacillus of Braxy.

In the case of the bouillon tubes inoculated with the peritoneal liquid
obtained from a sheep which had died from this disease, the medium was
found to be muddy-looking after 12 hours, and a copious evolution of
gas was evidenced by the collection of gas-bells between the oil and the
bouillon. After 24 hours' incubation the growth was being deposited as
small floccules on the bottom and sides of the tube. At the end of 36
hours a fairly large greyish-white deposit was found at the bottom of the
tube, the supernatant fluid being clear, and now giving a marked acid
reaction to litmus.

The microscopical examination of the deposit revealed the presence
in pure culture of a bacillus possessing the following morphological
characters. In the fresh condition, the bacillus was a fairly long, thin,
delicate straight rod, with rounded ends, and averaging 4*4^1 by I'O jul.
The bacilli from the bouillon deposit were usually found to be in clumps.
Within the peritoneal liquid, and usually in the first culture of it, the
organism was only very feebly motile, if motile at all. A few sporing
forms were occasionally found in this deposit ; spore formation was best
obtained by incubating the original peritoneal liquid or growing the
bacillus on sheep's blood. The spore, which was single, was situated
at the middle or extreme end of the rod, was of a brownish colour,
produced a small swelling, and gave the rod a lanceolate or drumstick
appearance.

The bacillus reacted to all the ordinary bacillary stains, but was negative
to Gram's process. The presence of cilia was demonstrable upon culture
by the methods of staining introduced by Lcefifler and van Ermengem.
The cilia were extremely delicate, and special care had to be taken
to avoid their detachment from the body of the bacillus. The method
employed was to remove a small portion of a twelve hours' old agar
surface culture grown under hydrogen, and add it to a drop of sterile
distilled water on a perfectly clean coverglass. This was then allowed to
evaporate in the air, and was fixed, when dry, by passing it once through
the flame. A few drops of van Ermengem's mordant were poured on the
film and allowed to act for 30 minutes. After gentle lavage in distilled
water, the film was placed in a 2 ^ solution of silver nitrate for a few
seconds, and afterwards transferred direct to the sodium acetate solution,

(404)

BACTERIOLOGY OF SOME DISEASES OF SHEEP 5

where it was moved gently to and fro for 10-15 seconds. It was now finally
transferred to 2 7^ silver nitrate, and left until the silver bath began
to turn brown, when it was removed and again washed in distilled water.
To ensure a sufficiently deep staining, it was mounted in water and
examined under an oil immersion lens. If the staining was not
sufficiently deep, the latter process of washing in sodium acetate solution
and silver nitrate was repeated. By this process, very distinctive staining
of the cilia was obtained.

The cilia varied in number from i to 10, averaging five; all were of
equal diameter, but varied in length from lO/x to 20/>i. They took their
origin from any part of the periphery of the bacillus, from which they
pursued a waving tortuous course. Sporing forms of the rods also
possessed cilia, a fact which was in accordance with the slight movement
which these also showed. As stained by Loeffler's method the cilia
were hardly recognisable even with a magnification of 960 diameters
on account of their extreme delicacy.

Growth on Gelatine.

In stab cultures under oil, growth was apparent in from 3-10 days —
the longer the time the bacillus had been under cultivation the quicker
the growth. In many cases the gelatine tubes were inoculated by means
of a capillary pipette, and its track was shown by a liquefied cone of
the medium demarcated from the solid part by numerous small,
pin-point masses of a greyish-white colour. From the surface of the
cone, horizontal, thread-like processes pursued a tortuous course to the
periphery of the tube. At first, these processes were short and fluffy,
and better seen at certain points of the cone. After further incubation,
they increased in length and number. They might become so numerous
that, matted together, they almost occluded the central cone. Here and
there, throughout the growth, small, pin-point, fluffy dots were seen.
Occasionally some of these processes behaved like the first inoculation,
and repeated the original picture on a very small scale. The picture of
a growth from a bacillus which had been cultivated in the laboratory for
some time was not so complex. The liquefied cone sometimes contained
no deposit, and the hair-like processes, although numerous, long and
wavy, were still discrete. After longer incubation, the threads dissolved,

(405)

6 QUATERCENTENARY STUDIES IN PATHOLOGY

the central cone stood bare, and the fluffy dots were well picked out in
the medium, which had become dusky-looking, and slightly more opaque.
Usually only a very few gas bells developed.

Grown aerobically on surface culture, after a long incubation of
weeks, the bacillus caused a slight liquefaction of the gelatine and a
greyish filmy deposit.

Growth on Agar.

On several occasions the bacillus was isolated by plating out in the
ordinary manner, and growing under hydrogen at 37° C. After three
days, the typical colony had the following appearances : — It was about
the size of a large pin-head, rounded in shape, with sharply-defined
margin, and had produced a faint, bluish-white opalescence, of equal
density throughout the colony. Microscopically, a small colony consisted
of a large number of small dots, whilst a larger colony consisted of an
agglomeration of these smaller colonies.

In stab culture a very copious evolution of gas had torn up the
medium so that nothing characteristic except a greyish line along the
track of inoculation was observed.

A surface culture, grown anaerobically after 24 hours, showed roundish,
small, discrete colonies, which, later, ran together and spread themselves
out as a fine film over the surface.

Bacillus of Louping-lll.

Growth on Bouillon.

After 12 hours' incubation, there was marked turbidity of the medium
with copious evolution of gas collecting as a fine foam beneath the oil.
After 48 hours' incubation, the medium became clear and distinctly acid
in reaction, and the growth was deposited at the bottom of the tube in
the form of a greyish-white mass. Examination in the fresh condition at
varying periods of the incubation revealed the presence of a large rod
with blunted ends, and broad in proportion to its length. The bacillus
was only faintly motile, and no spores were visible until the incubation
had been carried out for 36 hours, at which period a single small,
centrally placed, refractile body had become apparent, which, later,
increased in size producing an oval swelling in the centre of the bacillus.

(406)

BACTERIOLOGY OF SOME DISEASES OF SHEEP 7

^ It Stained readily with the usual stains, and gave a positive reaction
with Gram's process. One peculiarity with this stain was that in some
cases a part or the whole of the rod did not give a deep blue stain, but
was coloured a more or less faint lilac. Whether this was due to a
degeneration or not could hardly be determined, since similar films
stained in a normal manner with methylene blue or fuchsin. Repeated
attempts with cilia stains gave in each case a negative result.

Growth on Gelatine.

Under aerobic conditions no surface growth was obtained.

Stab cultures under oil, after three days' incubation, began to show a
faint greyish growth along the needle track, which, after a week's
incubation, had produced a cone-shaped area of liquefaction, resembling
that of the Finkler-Prior vibrio, the growth falling as a white film to the
bottom of the cavity.

Radiating from the sides of this cavity were a few white processes,
which divided, dichotomously, each to terminate in a small, round
colony. A few gas bells developed within the liquefied area, as well as
in the neighbouring solid gelatine.

Growth on Agar.

After 24 hours' incubation at 37° C, a line of greyish growth
developed in stab culture, and a few gas bells had formed. After three
days, this line had thrown out coarse lateral projections, but the excessive
development of gas had torn up the medium.

Surface Culture grown under Hydrogen.

After two days' incubation, growth made itself apparent by the
development of small, pin-head colonies, circular in outline, with sharply-
defined margins. Later on these increased in size, fusing to form a well-
marked white line, elevated above the level of the medium. After five
days' incubation, the growth spread, laterally giving more the appearance
of a cluster of grapes. The density of colour produced by the develop-
ment of the colonies lay between that of a staphylococcus albus and a
micrococcus candicans.

(407)

8 QUATERCENTENARY STUDIES IN PATHOLOGY

Bacillus of Disease "A."

The examination of the peritoneal liquid of certain sheep said to
have died of Braxy showed the presence, not of the bacilli of Braxy or
of Louping-ill, but of bacilli which have been temporarily classified by
Professor Hamilton into two groups— " A" and "B" (cf. Report, Pt. II.,

p. 259).

The bacillus having been isolated by heating in glucose-bouillon,
growth was obtained after 24 hours' incubation, with evolution of gas
and turbidity of the medium. In three days, the medium became
clear, and of an acid reaction, with a greyish-white deposit at the bottom
of the tube. This deposit was, in great part, composed of coarse,
straight rods, with rounded ends, varying in length from 2*8 to 6/x, and
broad in proportion to their length. Movement, if any, was very slight.
They stained readily with fuchsin and methylene blue, and gave a positive
reaction to Gram's process. Spores were not developed till the second
day of incubation, when a single, small, rounded, refractile body was
present towards one end of the rod. Cilia could not be demonstrated.

Growth on Gelatine.

Streak cultures, under aerobic conditions, showed no growth, even
after weeks of incubation.

Stab cultures, grown under oil, after 7-10 days' incubation, began to
show slight liquefaction of the gelatine in the form of a cone at the top
of the tube, but, deeper down, tended to spread laterally. The quantity
of growth in each case was very slight, amounting to little more than a
faint yellow film on the sides of the cavity. Lateral projections, as in
the case of Braxy and Louping-ill, were not observed.

Growth on Agar.

Streak cultures showed no change when incubated in the presence of
oxygen, but, under hydrogen, after two days, a narrow line of growth
had developed, made up of single, rounded, faintly greyish-tinged
colonies. On the fifth day it assumed the form of a band, about \ in.
broad, whose centre was homogeneous, but whose border was crenated
from the presence of rounded colonies. The growth was very slightly

(408)

BACTERIOLOGY OF SOME DISEASES OF SHEEP 9

elevated above the surface of the agar, and had a moist, brightly
reflecting appearance.

Stab cultures under oil developed a faint greyish streak along the
line of puncture after 24 hours. This was replaced on the third day by a
coarse band, with short, coarse, spiny projections, but the development
of gas broke up the medium.

Bacillus of Disease " B."

After 24 hours' incubation, the glucose-bouillon was muddy-looking,
and there was a marked evolution of gas. In 36 hours the medium in
the upper part of the tube began to clear, while the remainder was as
milky-looking as before, if not more so. After two days, the upper half of
the medium was clear, and the whole of it became relatively transparent
in about three days. If, at any time, the tube was but slightly shaken,
the transparency of the medium was almost immediately changed to
a diffuse milky opacity. The deposit was never seen on one side of the
tube only, nor did it ever fall down in floccules as in braxy. A copious
grey homogeneous mass was ultimately found in the bottom of the tube.
The reaction of the medium became acid. The organism present was
a thin, straight, scarcely motile, rod with rounded ends. Thread forms
were common, and hence there was a great difference in the lengths,
which varied from 2'8/jt to 486 /x. The bacilli — obtained from artificial
media — were never found in clumps as was the case with braxy, but
in the oedematous fluid from a guinea-pig — dead 30 hours after
inoculation — huge aggregated clumps of bacilli were present. The
bacillus was stained by all the ordinary stains and by Gram's process.
The most favourable medium for the investigation of the sporing phase
was the original peritoneal liquid, sporing being very slight on artificial
media. The spore was single, relatively large, usually polar, and
sometimes free.

Growth on Gelatine.

Slope cultures under aerobic conditions showed slight liquefaction
and growth, only after several weeks' incubation at 21° C.

Stab cultures under oil, after two weeks, exhibited liquefaction of the
medium as a narrow cylinder, lying in which were, at several places, small,

(409) C C

lO QUATERCENTENARY STUDIES IN PATHOLOGY

roundish, sharply demarcated, greyish -white colonies. Very delicate
hair-like processes were given off from the sides of the cylinder, but in no
case were fluffy dots seen amongst these as in the case of braxy, in which
the processes were coarser. A few gas bells developed in the medium.

Growth on Agar.

Stab cultures under oil showed on the first day a greyish-white streak
along the track of inoculation. The evolution of gas seemed to be more
abundant, and to take place more rapidly than was the case with braxy,
so that the medium became torn up and the character of the growth
destroyed.

(410)

INDEX

Administrative aspects of tuberculosis, 77
Agglutination in croupous pneumonia, 393
Albumen-metaphosphate, effect on leuco-
cytes, 358
Aleurone, effect on leucocytes, 357
Aleurone, effect on opsonic index, 358
Alformant lamp, disinfecting value, 245
Alimentary canal as a source of infection,

3
Anaemia, probable intestinal origin of, 37
Appendix vermiformis, Bilharzia ova in, 48
Auricular fibrillation, 97
Autolysed staphylococci, effect on leuco-
cytes, 346
Autolysed staphylococci, effect on opsonic
index, 346

Bacillus chorese paralyticae, 13
Bacillus of braxy, characters of, 404
Bacillus of braxy, growth on agar, 406
Bacillus of braxy, growth on gelatine, 405
Bacillus of disease "A" cultures, 408
Bacillus of disease " B " cultures, 409
Bacillus of louping-ill, growth on agar,

407
Bacillus of louping-ill, growth on bouillon,

406
Bacillus of louping-ill, growth on gelatine,

407
Bacillus of louping-ill, 14
Bacillus of louping-ill, inoculation in sheep,

18
Bacillus of louping-ill, toxins of, 27
Bacillus pyocyaneus, autolysed cultures,

167
Bacillus pyocyaneus, exaltation, 168
Bacillus pyocyaneus, filtration experiments,

Bilharziosis, case of, 41
Bilharzia! lesions of intestine, 47
Bilharzial ova in appendix vermiformis, 48
Bilharzial ova in pancreas, 43
Bilharzial polyposis of large intestine, 5 1
Bouillon, effect on leucocytes, 345
Bouillon, effect on opsonic index, 345

Bulbus cordis in Xiphius, 59
Bulbus cordis malformations, 57

Carbamic acid in urine, 270
Cell necrosis produced by bacteria, 287
Chorea, intestinal origin of, 36
Cirrhosis of liver, intestinal origin of, 37
Contagious diseases of sheep, 4
Contagious diseases in sheep, immunity, 23
Contagious diseases of sheep, periodicity,

. 5
Contagious diseases of sheep, cause of

periodicity, 25
Contagious diseases of sheep, geographical

areas affected, 5
Cyllin, disinfecting value of, 255, 258

Disinfection by corrosive sublimate, 260
Disinfection by cyllin, 260
Disinfection by formaline, 240, 257
Disinfection by izal, 261
Disinfection by phenol, 254, 257
Disinfection by sanitas, 261
Disinfection by sapo-cresol, 262

Eck's fistula, 265
Eck's fistula, literature on, 267
Epidemic plague, rat theory of, 215
Epignathus, 177
Epignathus, case of, 178
Epignathus, origin of, 193
Epignathus, relations of, 190
Epignathus, structure of, 189
Epilepsy, intestinal origin of, 36
Epizootic plague in Hong-Kong, 224
Exophthalmic goitre, pathology of, 199
Exophthalmic goitre, thymus gland in,

205
Exophthalmic goitre, thyroid gland in,

207
Experiments on plague toxin, 122

Fixateur, action of heat on, 311
Fixateur, relation to opsonin, 310
Formaldehyde, disinfecting value, 240

(411)

INDEX

Guinea-pig, leucocytes of, 333
Guinea-pig, physiological variations in
blood, 333

Heart, paroxysmal irregularity, 97
Heart of Xiphius gladius, 58
Hong-Kong, plague epidemics in, 223
Hong-Kong> plague epizootics in, 223
Hong-Kong, relation of epizootic to epi-
demic plague, 224

Immunity against bacillus pyocyaneus, 157
Immunity against plague, 1 13
Immunity in pneumococcal infections, 365
Intestine, Bilharzia lesions of, 47
Intestinal origin of contagious diseases in

sheep, 23
Izal, disinfecting value of, 261

Large intestine, Bilharzial polyposis, 51
Leucocytosis and opsonic content of

serum, 321
Leucocyte in relation to opsonin, 384
Lesser curvature of heart, 69
Lingner's disinfecting apparatus, 241
Liver in Bilharziosis, 42
Louping-ill, 6
Louping-ill, Bacillus of, 13
Louping-ill, experimental, 18
Louping-ill, morbid anatomy, ii
Louping-ill, occurrence, 6
Louping-ill, peritoneal liquid, 13
Louping-ill, symptoms, 6

Maladie pyocyanique, 159
Malformations of bulbus cordis, 57
Mammalian heart, evolution of, 61
Metchnikoff on micro- and macrophages,
304

Nuclein, effect on leucocytes, 350
Nuclein, effect on opsonic index, 350

Opsonic index, fluctuations in health, 334
Opsonic index, variations in pneumococcic

infections, 375
Opsonin, action of heat upon, 313, 317,

384
Opsonin, effect of diluting, 386
Opsonin, effect of preserving, 381
Opsonin, effect of salt upon, 387
Opsonin, relation to leucocytosis, 391
Opsonin, specificity of, 380

Papayotin-albumose, effect on leucocytes,
357

Phagocytosis during hyperleucocytosis, 359
Phagocytosis of red corpuscles, 301
Pilocarpine, effect on leucocytes, 347
Pilocarpine, effect on opsonic index, 347
Plague epidemics, rats in, 215
Plague epidemics, bridging of, 232
Plague antitoxin, 130
Plague bacteriolysin, 140
Plague immunity, 113
Plague toxins, 114, 122
Pneumococcus, infections, immunity in, 365
Pneumococcus, mechanism of immunity,

370
Pneumococcus, infections, opsonic index

in, 375
Pneumonia, agglutination in, 393
Pneumonia, opsonic variation in, 375
Pulmonary artery congenital stenosis, 63
Pyocyanic immunity, 157
Pyocyanic serum, properties, 172

Rabbit, normal blood of, 329

Rabbit, leucocytes of, 330

Rats and plague, 215

Rats, epizootics in, 223

Red blood corpuscles, phagocytosis of, 301

Red blood corpuscles, hsemolosis of, 301

Right ventricle, division of, 62

Sanitas, disinfecting value of, 261
Sapo-Cresol, disinfecting value of, 262
Savtchenko's experiments on phagocytosis,

309
Sheep, contagious diseases of, 4
Sodium cinnamate, effect on leucocytes,

335
Sodium cinnamate, effect on opsonic index,

336
Stenosis, congenital, of pulmonary artery,

63

Tallianine, effect on leucocytes, 340
Tallianine, effect on opsonic index, 341
Tetanus, probable intestinal origin o.', 35
Thymus gland in exophthalmic goitre, 205
Thyroid gland in exophthalmic goitre, 207
Toxin of bacillus of louping-ill, 27
Toxins of plague bacillus, literature, 1 14
Toxins of plague bacillus, experimental,

122
Toxicity of pyocyanic filtrates, 163
Toxicity of pyocyanic cultures, 164
Tuberculosis, administrative aspects, 77
Tuberculosis, death-rate in Scotland, 79
Tuberculosis, disinfection, 85
Tuberculosis, isolation, 87

(412)

^^ L.^

PLEASE DO NOT REMOVE
CARDS OR SLIPS FROM THIS POCKET

UNIVERSITY OF TORONTO LIBRARY

RB Bulloch, Villiam (ed.)
6 Studies in pathology

B9

Biological
8c Medical