MEDICAL SCHOOL
LUBTBAKY

Medical Library Exchange

i '

t <*^s <? c <4

F

C/Ji

BACTERIOLOGY AND THE
PUBLIC HEALTH

Digitized by the Internet Archive
in 2012

http://archive.org/details/bacteriologypuble3newm

■

FORMERLY Dl ERIOLOOY IN KIN'. LONDON, Kl'C

KKDIOA I 'T HEALTH OF IHK H IUOH Of I

join !■- ,\ i THOR OF " I. MILK."

II. I i M i; \n I)

I iiii:i> EDI 1 1<»\

PHIL \I)I.I.1'HI \
P. BL VKISTON - SON AND CO
WALNUT STR1 I I

1!'

A

BACTERIOLOGY AND
THE PUBLIC HEALTH

BY GEORGE NEWMAN, M.D.. F.R.S E., D.P.H.

Printed in Great Britain

». •

PREFACE

Though nominally .1 third edition of Bacteria in Relation to the
Economy 0/ Nature, Industrial Processes, and thi Public Health, this
is, speaking generally, a new book. Several new chapters have been
added, and the whole has been enlarged and revised.

The book Lb an attempt to set forth a simple genera] statement of
our present knowledge of bacteria, especially as they are related to ihe
public health. Theoretical and practical text-books of bacteriology
abound, bat as a rule they deal largely, ind rightly so, with laboratory
methods and technique, The genera] Btudent of hygiene and the
medical officer of health require, however, an elementary book in
which, whilst ample laboratory Pacts are recorded, the subject is
viewed broadly and particularly as It concerns the practical evei
problems of health and preventive medicine. This book is aimed i«»
meet that requirement

[ am indebted to many friends and colleagues foi tionsand

criticisms, and for a number of illustrationa In addition t«> a number
of cliches used in former editions, some of which were kindly lent
by the Scient ifi< Press Limited, from 1 1.

by Slater & Spitta, I have t<> express mv ulili^atimis to the
Controller of His Majesty's Stationery Office, the Secretary of the
Royal Commission on 9 D I, and the Chairman of the

Main Drainage < Committee of the London < kranty < louncil, for permis-
sion to use several blocks illustrate derived from
oultures obtained by my friend, I>r Houston, in the course of his
sewage investigations. I am in a similar way much indebted to Mr

Ml

s 5 ?

viii PREFACE

Foulerton of the Middlesex Hospital, and Dr Harold Spitta of St
George's Hospital, for the use of some excellent photographs. My
colleague, Mr Harold Swithinbank, has kindly allowed me to use
three coloured plates of "acid-fast" cultures from our book on the
Bacteriology of Milk, and he has also supplied me with several original
plates. To each of these gentlemen I am glad to have the oppor-
tunity of expressing my sincere thanks. G. N.

London, August 1904.

INTRODUCTION

The science of biology has for its object the study of organic beings,
and for its end the knowledge of the laws of their growth,
organisation, and function From the earliest times of man that life
has been Btudied and the observations recorded Thus there has come
slowly to be a considerable accumulation of knowledge concerning the
various forms (morphology) and functions (physiology) of organised
life. Iu the midst of this gradual accumulation of facts we begin to
sec incoherence becoming coherent, chaos becoming and

apparent chance and accident becoming law.

Bacteriology rt, a chapter, of genera] biology, and is

concerned with the fi ent known, of Bomeof the lowest

forms of micro-organic lifa Owing to b variety of oircumstances, the
chief of which is the relation of these Daicro-organisma to disease, the
siuily of bacteria has assumed a place among the branches of bi<
of somewhat exceptional importance The application of bi
daily life and its problems has in recent y< < n nowhere more

marked than in the realm of bacteriology, where the great oam<
Pasteur, Koch, and Lister, represent the modern ep advance.

Turn where we will, we shall find the work of the unseen
bacteria daily nlfliming more and more attention from practical people,
and thus hioli u when concerned with the work of microscopic

cells, is coming to occupy .1 new place in the minds of men. tte
evolution begins to form part of the general Bocial evolution.

tainly the recent development 1 1 emark-

able feature in the Boientific advance of our time, Not only in the

diagnosis ami MvitiiMMit "t' .lis. Msr, n..r cvi-n in the various applications

of preventive medicine, but in every mcreasing degree and sphere
microorganisms are recognised a I or ill no longer to

1).' ignored. They occur in our drinking water, in our milk supply, in
the aiz we breathe. They ripen cream, and flavour butter. They
purifj b, and rem ic products from the land.

They are the active agents inadosen industrial fermentationa They
ix a 2

x INTRODUCTION

assist in the fixation of free nitrogen, and they build up assimilable
compounds. Their activity assumes innumerable phases and occupies
many spheres, probably more frequently proving itself beneficial
than injurious, for bacteria are both economic and industrious in the
best sense of the terms.

Yet bacteriology has its limitations. It is well to recognise this,
for the new science has in some measure suffered in fche past from
over-zealous and sanguine friends. It cannot achieve everything
demanded of it, nor can it furnish a causal agent for every disease to
which human flesh is liable. It is a science which even yet is fuller
of hope than of proved and established knowledge, for we are at
present but upon the threshold of the matter. As in tie aeighbouring
realm of chemistry, it is to be feared that bacteriology has not been
without its alchemy. The interpretations and conclusions which have
been drawn from time to time respecting bacteriological findings have
led to alarmist or optimist views which have not, by later
investigations, been fully confirmed. For the science 1ms had devotees
who have fondly believed, like the Alchemists, that the twin secret of
"transmuting the baser metals into gold," and of indefinitely
prolonging human life, was at last to be known. Neither the worst
fears of the alarmist nor the sanguine hopes of the optimist have been
verified. Science does not progress at such speed or with such kindly
accommodation. It holds many things in its hand, but Dot finally life
or death. It has not yet brought to light either "the philosopher's
stone " or " the vital essence."

What has already been said affords ample reason for a wider
dissemination of the elementary facts of bacteriological science. But
there are other reasons of a more practical nature. Municipalities
and other bodies are expending public moneys in water analysis, in
the examination of milk and the control of its supply, in the inspection
of cows and dairies, in the bacterial treatment of sewage, in pro-
tecting the oyster trade, in the ventilation of workshops and factories,
in disinfection, in the prevention of epidemic diseases, and in other
branches of public health administration. Furthermore, our increasing
colonial possessions with their tropical diseases, and the growth of
preventive medicine generally, make an increasing claim upon public
opinion and those engaged in raising the physical condition of the
people. The successful accomplishment and solution of these
questions depends in measure upon a correct appreciation of the
elements of bacteriology.

The present is a transition period in this department of knowledge.
A very large body of facts has been collected, and there has been a
natural tendency to draw somewhat sweeping deductions which
subsequent knowledge has not supported. What is now required is
that our experience in the laboratory and outside should be patiently

INTRODUCTION xi

and repeatedly checked and tested. If the science of bacteriology is
to be built solidly, the two necessities of accumulating accurate facts
and making generalisations and deductions must proceed side by side,
the former being well established before the latter are accepted. It is
the danger of a new science that too much is expected of it.
Bacteriology, except in a few well-defined spheres, cannot yet stand
alone as reliable basis for legislation. The bacteriologist must be
content at present to serve as indicator rather than as dictator. The
detection, for instance, of certain bacteria in milk or in oysters is an
indication, and not an absolute proposition, of unsatisfactory dairying or
oyster culture. Common sense and a broad view of all the ascertain-
able facts must guide those whose business it is to apply the findings
of bacteriology to preventive measures.

In the pages that follow, a large number <»t' statements occur as to
the external circumstances and conditions affecting the lit'*' of bacteria,
and to understand these rightly and hold them in righl proportion t<>
each other, it is necessary to bear in mind thai many, it' qoI most of
them, are <>t' relative importance. They are «-t' value, nol as isolated
units, but. as parts of a whola It is their co-ordination, relativity,
and correlation which must be Bought after. Again, the preseni e "t' ;i
diphtheria bacillus in the throat of a healthy man appears at first Bight
to 1"' a fact of absolute and critical importance until t be life-history of
the bacillus is inquired into and determined, and the relation of the
healthy tissues \<> the performance "t" it- function understood The
bacteriologist and worker in preventive medicine can never afford i"
neglect the inter-relationship which exists between the Beed and the
suit, h is ii.. i wholly the one or the other with which he has to deal
as .1 practical man. It i< the combination and the inter-action
between the two. [f that principle, and the relativity of our know-
Ledge "t' bacteria and t!. which they play are borne in mind,
there is litt l<- to fear from a transition period

Whilst there can then he do doubt as t" the advantage "!' a
wid»> dissemination of the ascertained concerning bacteria,

especially in relation t" water, air, milk, and other foods, it must uot
be forgotten that only patient and skilled observation, and
experimental research in well-equipped laboratories, can advance this
branch of science or indeed train bacteriologists. The li\<
Darwin and Pasteur adequately illustrate tin- truth As the world
Learns its intimate relation t<» science, and the inter-dependence
between its lite and scientific truth, Stat.-- and public authoi ities may
be expected more heartily t" rapport Bcienoe

CONTENTS

CHAPTER I

TIIK BIOLOGY OK BACTERIA

Early work — Place of Bacteria In Nature -Biology <>t' Bacteria ; Morphology,
Composition, Reproduction, Influence of External Conditions Light
— Modes of Bacteria] Action Seed and Soil Specificity of Bacteria
— Association, Antagonism, Attenuation -Bacterial Diseases of Plants 1-82

CHAFFEE II

BACTERIA IN WATER

Quantity of Bacteria In Water Quality <»t" Water Bacteria: a) Ordinary
Water Bacteria; (6) Sewage Bacteria] J> soli oommumii ; [e] Patho-
genk Bacteria In Water -Interpretation of the Findings "t Bacteri-

Ology -Natural Purification of Water Artificial Purification of Water

-Sand Filtration Domestic Purification of Water

(Haiti;!; mi

BACTERIA IN TIIK All;

Methods of Examination <>t" Kit Conditions of Bacterial Contamination of
\ir : (1) Dust and \ir Pollution ; (2) Moisture or Dampness of Surra
Bacteria in Sewer Air; (3) the Influence of GraYity; (i) Air Currents.
The Relation <>t Bacteris to CO in the Atmosphere i In Workshops, In
Bakehouses, in Railway Tubes* in the Hou» of Commons . . i

xili

xiv CONTENTS

CHAPTEE IV

BACTERIA AND FERMENTATION

TA<,E

Early Work— Kinds of Fermentation: (1) Alcoholic Fermentation, Asco-
spores, Pure Cultures, Films; (2) Acetous Fermentation; (8) Lactic
Acid Fermentation; (4) Butyric Fermentation; (.'») Ammoniacal Fer-
mentation—Diseases of Wine and Beer: Turbidity, Ropiness, Bitter-
ness, etc.— Industrial Applications of Bacterial Ferments . . 92-115

CHAPTEE V

BACTERIA IN THE SOIL

Methods of Examination— Methods of Anaerobic Cult lire Place and Function
of Micro-organisms in Soil — Denitrification, Nitrification, Nitrogen-
fixation, Bacterial Symbiosis— Saprophytic and Pathogenic Organisms
in Soil— Tetanus— Quarter-Evil— Malignant (Edema The Relation ol

Soil to Bacterial Diseases, such as Typhoid Fever . . 116-160

CHAPTER VI

THE BACTERIOLOGY OF SEWAGE AND THI BACTERIAL
TREATMENT OF SEWAGE

Composition of Sewage— Quantity and Quality of Bacteria in Sewage Treat-
ment of Sewage : (1) Disposal without Purification ; (2) Chemical Treat
ment; (3) Bacterial Treatment — Evolution of Bacterial Methods— Septic
Tank Method — Contact-Bed Method — Manchester Experiments — Effect

of Bacterial Treatment on Pathogenic Organisms . . 151-177

CHAPTEE VII

BACTERIA IN MILK AND MILK PRODI

General Principles — Sources of Pollution — Number of Bacteria in Milk —
Influence of Time and Temperature — Species of Bacteria found in Milk
— Fermentations of Milk — Pathogenic Organisms in Milk — Milk-borne
Disease: Tuberculosis, Typhoid Fever, Scarlet Fever, Sore-Throat
Illnesses, Cholera, Epidemic Diarrhoea — Preventive Measures — Pro-
tection of Milk Supply — Control of Milk Supply : Refrigeration, Strain-
ing, Sterilisation, Pasteurisation — Specialised Milk- Bacteria in Milk
Products — Cream-Ripening — Butter-Making— Cheese-Making— Abnor-
mal Cheese-Ripening — Poisonous Cheese . . . . 178-252

CONTENTS xv

CHAPTEK VIII

BACTERIA IN OTHER FOODS

PAGE

1. Shell-fish, Oysters, Cockles, Clams, and their Relation to Disease;
Symptoms of Oyster-borne Disease; Channels of Infection; Preven-
tive Methods — 2. Meat Poisoning; Tuberculous Meat — 3. Ice-cream
and Ice — 4. Bacterial Infection of Bread — 5. Miscellaneous Foods,
Watercress, etc. . . . . . . . 253-279

CHAl'TEK IX
BACTERIA AND DI8KA8S

Growth of Knowledge of Bacteria as Disease Producers Channels of [nfcc
tion — How Bacteria cause Disease Diphtherial Conditions of Infec-
tion—Scarlet Fever, Typhoid Fever, Epidemic Diarrhoea i Conditions

of Infection — Suppuration and rVbsceSS Formation -Anthrax Pneu-
monia— Influenza— Actinomj < osis Glanders . . . _-

CHAPTEB X

TUBER ii" i Tl PI "i BAI riRIAL DI8I* SX

Pathology and Bacteriology of Tul*reulosis The Baeillus of Koeh Animal
Tuberculosis, Bovine, Avian, etc Bovine and Human Tubercle Bacilli
compared Inter cofnmunicability Diagnosis of Bovine Tubercle The

Prevention of Tuberculosis Pscudo TuixTrulosi.s Vcid last Bacteria

Allied to the Tubercle Bacillus i in Man, in Animals, In Butter and
Milk* in Grass •Differential Diagnosis 9treptothrfai Group

CHAPTEB XI

THE ETIOLOGY Of TROPICAL Dl

Milan. i : Forms <>t Malarial Fever, the Mosquito Theory, Prevention "t
Malaria Cholera: Methods <>t Diagnosis Plague i Symptoms. Pats
and Plague, Bacteriology, Administrative Considerations Lepra
Yellow Fever Malta Fever Sleeping Sfckm BV beri

xvi CONTENTS

CHAPTEK XII

THE QUESTION OF IMMUNITY AND ANTITOXINS

PAGE

Bacterial Products — Toxins — Question of Immunity — Kinds of Immunity —
Theories of Immunity— Applications of Immunity — Vaccination for
Small-pox: Effect of Vaccination — Pasteur's Treatment for Rabies —
Inoculations for Cholera, Typhoid, and Plague — Antitoxin Treatment
of Diphtheria and its Effects ..... 405-431

CHAPTER XIII

DISINFECTION

General Principles— Means of Disinfection: by Heat; by Chemicals—
Practical Disinfection: Rooms, Walls, Bedding, Clothing, Excreta,
Books, Linen, Stables, etc. — Disinfection of Hands — Disinfection after
Special Diseases : Phthisis, Small-pox, Scarlet Fever, Diphtheria,
Typhoid, Plague ....... 132451

Appendix on Technique 453-488

Index 489-497

LIST OF FIGURES

I ■•!<!.

1.

9

4.
5.
6.
7.

8.
9.

10.

n.
L2.
13.
II.
15.
1G.
17.
18.
lit.
20.
21.
22,
23.
24.
26.

Various Forms of Bacteria

Diagram of Sarcma

Diagrams of Normal and Polymorphic For

Bacilli ....

Various Forms of Spore Formation and Flagella
Inoculating Needles

Media for Surface and Depth Culture .

Method of Producing Hydrogen by Kipp'f

Cultivation of Anaerobes
Koch's Steam Steriliser
Diagrams of 11. typhotut and II. colt
Pasteur-Chamberland Filter
MiqueTs Flask
Sedgwick's Sugar-tube
Diagram of Ascospore Formation
Gypsum Block ....
I Hagram of S, certvitut
Diagram of 8. eUiptouUui
Diagram of & patUnrimnu
Frttnkel's Tube

Rootlet of Pea with Nodules .
Diagram of Bacillus of Symptomatic Anthrai
A Plan of Septic lank and Filter-beds
Contact-beds ....
u I'lax " Filter ....
Diagram of Httcilltu diphtheria .
Diagram of Types of Streptococci

lis of

\|.,,

ubercle

tils \i

PACK
7

8
10

i;;

17
17

23
24

IT

71

71

~~>

lJ8

98

102

102

L02

I is

133

1 13

167

169

a 2 9

289
312

XVI 11

LIST OF FIGURES

26. Diagram of Micrococcus- tetrasonus

27. Diagram of Gonococcus ....

28. Diagram of Bacillus of Anthrax and Blood Corpuscles.

29. Quartan Malaria Parasite ....

30. Tertian Malaria Parasite ....

31. Malignant Malaria Parasite ....

32. Anopheles maculipennis .....

33. Diagram of Culex and Anopheles

34. Human and Mosquito Cycles of the Malaria Parasite .

35. Diagram of the Comma Bacillus of Cholera

36. Suspended Spinal Cord ....

37. Flask used for Preparation of the Toxin of Diphtheria

38. Petri Dish ......

39. A Diagram of Colonies of Bacteria on a Gelatine Plate

40. The Hanging Drop .....

41. Drying Stage for Fixing Films

42. Types of Liquefaction of Gelatine

43. Levelling Apparatus for Koch's Plate .

44. Moist Chamber for Koch's Plate

45. Wolfhiigel's Counter .

46. Filter-brushing Method

47. Buchner Tube ....

48. Another Form of Buchner Tube

PAGE

313
314
316
373
374
374
377
378
380
385
421
426
453
454
455
456
457
464
464
465
466
466
478 -

o face puge

18

jy

24

4G

>>

50

>>

74

7G

»>

36

LIST OF PLATES

[Note. — Photographs marked with an asterisk (*) have been kindly lent by the
Scientific Press Company ; those marked t are taken by permission from the
Report of Royal Commission on Sewage Disposal ; and those marked X from
Reports to the London County Council. ]

PLATE

1. A Form of Room Temperature Incubator .

2. Hot-air Steriliser and Blood-beat Ineubator

3. B. coll communis *; Proteus vulgaris \

4. B. coli communis ; (las Production in Gelatine t

5. Small Centrifuge; Sedgwick's Sugar-tube .

G. Air-Plate Culture from Labourer's Cot'

7. Air-Plate Cultures from Bakehouses

s. Saccharomuces cerevisia ; Ascospores; Pathogenic

Yeasi ....... 98

9. Buebner'a Tube ; kipp's Apparatus for Anaerobic

Culture ...... 110

10. A Vacuum Method of Anaerobic Culture . ,, 118

11. Nitrous Organism; Nitric Organism; Nitrogen-

fixing Organisms ....,, l_,s:

12. Nitrogen-fixing Bacteria in Nodule <>n H< >< >t 1 « t

of Pes . . . „ 134

iZ, B. tetani*; />. mucoides*; Strtpioikrix actinomuces ;

/». nidi In . . . . ,, [40

11. Sewage Proteus, Organism and Plate Culture \ . „ 1M

ID. Sewage Streptococci ! and Streptococcus puogenet ' „ 156

L6. B, mesenlericus, Organism and Plate Culture ] . ., L58

IT. />. anthracist from Septic Tank Liquor and in

Gelatine Culture (impression) t . . ., 17*.

18. B. tuberculosis (old culture); Tubercle Bacilli in

Cow's Udder . 204

xx LIST OF PLATES

PLATE

19. B. diphtherias ; B. von Hofmann . . . To face page 288

20. B. typhosus ; B. typhosus (flagella) *; Widal-Gruber

Reaction*; B. typhosus in Human Mesenteric

Gland. ...... 302

21. B. enteritidis sporogenes\; " Enteritidis change" in

Milk Cultures! 307

22. B. anthracis (Stab Culture) f; B. anthracis from

Blood*; Friinkel's Pneumococcus . . „ 318

23. B. tuberculosis, from Sputum,* Tissues, and Culture „ 328

24. Comparative Cultures of Tubercle Bacillus (Bird,

Mammal, Butter) „ 350

25. Cultures of Butter Bacillus of Rabinowitsch and

Mceller's Milk Bacillus (Chromo) . . „ 360

26. Cultures of B. jriburgensis, Nos. I. and II. (Chromo) „ 362

27. Cultures of Butter Bacilli of Binot and Grassberger

(Chromo) ...... 364

28. Comparative Cultures of Acid-fast Bacteria (Grass

and Manure) ...... 366

29. Streptothrix luteola (Foulerton) ; Streptothrix hommis

(Foulerton) .... „ 368

30. B. lepras ; B. pestis*; Staphylococcus pyogenes aureus k, 398

31. Apparatus for Filter-brushing Method in Water

Examination . . . ,, 466

BACTERIOLOGY AND PUBLIC
HEALTH

CHAPTER I

Till'. BIOLOGY 01 BAl TEBIA*

Early work Place of Bacteria In Nature Biology of Bacterial Morphology,
Composition, Reproduction, Influence of Externa] Conditions— Ughl —
Modes of Bacteria] Action Seed and Soil Specificity of Bacteria —
Association, Antagonism, Attenuation Bacterial Diseases of Plants.

The first Bciential win. demons! rated the existence of micro-organisms
was Antony von Leeuwenhoek. He was born at Delft, in Bolland,
in 1632, and enthusiastically pursued microscopy with primitive
instruments. Ee corroborated II wry of the circulation

of the l»l 1, in the web of a h * : lie defined the red blood

corpuscles of vertebrates, the fibres of tin- Lens of the human eye,
the Bcales of the skin, ami the structure of hair. He was neither
educated nor trained in science, hut in the leisure time of his
occupation as a linen-draper he Learned the ait of grinding lenses,
in which he became bo proficient that he was able to constn

microsco] f greater power than had been previously manufactured.

The compound microscope dates from 1590, and when Leeuwenhoek

We propose throughout to use the term tria) in its generic

meaning, unl< is especially stated to the contrary. It vriu also be synonymous irith
the terms mforoos, '/'/'///, and imitm. The term baeUhu will, of course, be

restricted to a rod ihaped bacterium.

A

2 THE BIOLOGY OF BACTERTA

was about forty years old, Holland had already given to the world
both microscope and telescope. Eobert Hooke did for England
what Hans Janssen had done for Holland, and established the same
conclusion that Leeuwenhoek arrived at independently, viz., that a
simple globule of glass mounted between two metal plates which
were pierced with a minute aperture to allow rays of light to pass
was a contrivance which would magnify more highly than the
recognised microscopes of that day. It was with some such instru-
ment as this that the first micro-organisms were observed in a drop
of water. It was not until more than a hundred years later that
these " animalcula," as they were termed, were thought to be anything
more than accidental to any fluid or substance containing them.
Plenciz, of Vienna, was one of the first to conceive the idea that
decomposition could only take place in the presence of some of these
"animalcula." This was in the middle of the eighteenth century.
Just about a century later, by a series of important discoveries, it
was established beyond dispute that these micro-organisms had an
intimate causal relation to fermentation, putrefaction, and disease.
Spallanzani, Pasteur, and Tyndall are the three workers who more
than others contributed to this discovery. Spallanzani was an Italian
who studied at Bologna, and was in 1754 appointed to the Chair of
Logic at Eeggio. But his inclinations led him into the realm of
natural history. Amongst other things, his attention was directed
to the doctrine of spontaneous generation, which had been propounded
by Needham a few years previously. In 1768 Spallanzani became
Professor of Natural History at Pavia, and whilst there lie demon-
strated that if infusions of vegetable matter were placed in flasks
and hermetically sealed, and then brought to the boiling point, no
living organisms could thereafter be detected, nor did the vegetable
matter decompose. When, however, the flasks were but slightly
cracked, the air gained admittance, then invariably both organisms
and decomposition appeared. Schwann, the founder of the cell-
theory, and Schultze, both showed that if the air gaining access to
the flask were either calcined or drawn through strong acid the
result was the same as if no air entered at all, namely, there were no
organisms and there was no decomposition. The result of these investi-
gations was that scientific men began to believe that no form of
life arose de novo (abiogenesis), but had its source in previous life
(biogenesis). It remained for Pasteur and Tyndall to demonstrate
this beyond dispute, and to put to rout the fresh arguments for
spontaneous generation which Pouchet had advanced as late as 1859.
Pasteur collected the floating dust of the air, and found by means
of the microscope many organised particles, which he sowed on
suitable infusions, and thus obtained rich crops of "animalcula."
He also demonstrated that these organisms existed in varying

SPONTANEOUS GENERATION 3

degrees in different atmospheres, few in the pure air of the Mer de
Glace, more in the air of the plains, most in the air of towns. He
further proved that it was not necessary to insist upon hermetic
sealing or cotton filters to keep these living organisms in the
air from gaining access to a flask of infusion. If the neck of the
flask were drawn out into a long tube and turned downwards, and
then a little upwards, even though the end be left open, no con-
tamination gained access. Hence, if the infusion were boiled, no
putrefaction would occur. The organisms which fell into the open
end of the tube were arrested in the condensation water in the
angle of the tube ; but even if that were not so, the force of gravity
acting upon them prevented them from passing up the long arm of
the tube into the neck of the flask. A few years after Pasteur's
first work on this subject, Tyndall conceived a precise method of
determining the absence or presence of dust particles in the air by
passing a beam of sunlight through a glass box before and after its
walls had been coated with glycerine. Into the floor of the box
were fixed the mouths of flasks containing an infusion. Those
were boiled, after which they were allowed to cool, and might
then be kept for weeks or months without putrefying or reveal-
ing the presence of germ life. Hero all the conditions of the in-
fusions were natural, except thai in the air above them there was do
dust.

The sum-total of result arising from these investigations was to
the effect thai no spontaneous generation was possible, that the
atmosphere contained unseen germs of life, that the smallest of
organisms responded to the law of gravitation and adhered t<» moist
surfaces, and that micro-organisms were in Borne way or other the
cause of putrefaction.

The final refutation of the hypothesis of spontaneous generation
was followed by an awakened interest in the unseen world of micro-
organic Life. [investigations into fermentation and putrefaction
followed each other rapidly, and in 1m;:; Davaine claimed that
Pollender's bacillus of anthrax, which was found in the blood and
tissues of animals which had died of anthrax, was the cause of that
disease. From tint time to this, in every department of biology,
bacteria nave been increasingly found t" play an important part.
They cause changes in milk, and flavour butter; they decompose
animal matter, yet build up the broken-down elements int<> com-
pounds Buitable for use in nature's economy; thej it in the
fixation of free nitrogen; they purify Be wage; in certain well-
established cases they arc ihc .mum- of specific disease, and in many
other oases they are the probable cause. No doubt the disposal of
spontaneous generation did much t<> arouse interest, in this branch
of science. Yet it must not ho forgotten that the advance of the

4 THE BIOLOGY OF BACTERIA

microscope and bacteriological method and technique have played a
large share in this development. The sterilisation of culture fluids
by heat, the use of aniline dyes as staining agents, the introduction
of solid culture media (such as gelatine and agar), and Koch's
" plate " method, have all contributed not a little to the enormous
advance of bacteriology.

The Place of Bacteria in Nature

As we have seen, for a considerable period of time after their first
detection these unicellular organisms were considered to be members
of the animal kingdom. As late as 1838, when Ehronberg and
Dujardin drew up their classification, bacteria were placed among the
Infusorians. This was in part due to the powers of motion which
these observers detected in bacteria. It is now, of course, recognised
that animals have no monopoly of motion. But what, after all,
are the differences between animals and vegetables so low down in
the scale of life? Chiefly two: there is a difference in life-history
(in structure and development), and there is a difference in pabulum.
A plant secures its nourishment from much simpler elements than is
the case with animals ; for example, it obtains its carbon from the
carbonic acid gas in air and water. This it is able to do, as regards
the carbon, by means of the green colouring matter known as chloro-
phyll, by the aid of which, with sunlight, carbonic acid is decomposed
in the chlorophyll corpuscles, the oxygen passing back into the
atmosphere, the carbon being stored in the plant in tin; form of
starch or other organic compound. The supply of carbon in the
chlorophyll-free plants, amongst which are the bacteria, is obtained
by breaking up different forms of carbohydrates. Beside albumen
and peptone, they use sugar and similar carbohydrates and glycerine
as a source of carbon. Many of them also have the capacity of using
organic matters of complex constitution by converting such into
water, carbonic acid gas, and ammonia. Their hydrogen comes from
water, their nitrogen from the soil, chiefly in the form of nitrates.
From the soil, too, they obtain other necessary salts. Now all these
substances are in elementary conditions, and as such plants can
absorb them. Animals, on the other hand, are only able to utilise
compound food products which have been, so to speak, prepared for
them, for example albuminoids and proteids. They cannot directly
feed upon the elementary substances forming the diet of vegetables.
This distinction, however, did not at once clear up the difficult
matter of the classification of bacteria. It is true, they possess powers
of motion, are free from chlorophyll, and even feed occasionally upon
products of decomposition — three physiological characters which

CLASSIFICATION OF BACTERIA 5

would ally them to the animal kingdom. Yet by their structure and
capsule of cellulose and by their life-history and mode of growth
they unmistakably proclaim themselves to be of the vegetable
kingdom. In 1853 Cohn arrived at a conclusion to this effect, and
since that date bacteria have become more and more limited in clas-
sification and restricted in definition.

Even yet, however, we are far from a scientific classification of
bacteria. Nor is this matter for surprise. The development in this
branch of biology has been so rapid that it has been impossible to
assimilate the facts collected. The facts themselves by their
remarkable variety have not aided classification. Names which a few
years ago were applied to individual species, like Bacillus subtilis, or
Bacterium term.0, or Bacillus coli, are now representative, not of
individuals, but of families and species. Again, isolated character-
istics of certain microbes such as motility, power of liquefying
gelatine, size, colour, and so forth, which at first sight might appear
as likely to form a basis for classification, are found to vary not only
between similar germs, but in the same germ. Different physical
conditions have so powerful an influence upon these microscopic cells
that their individual characters are constantly undergoing change.
For example, bacteria in old cultures assume a different size, and
often a different shape, from younger members of precisely the same
species ; Bacillus pyocyaneus produces a green t<> olive colour on
gelatine, but a brown colour on potato; the bacillus of Tetanus is
virulently pathogenic, and yet may not, act thus unless in com-
pany with certain other micro-organisms. Hence it will at once
appear to the Btudent of bacteriology that, though there is great
need tor classification amongst the six or seven hundred named
■' species" of microbes, our present knowledge of their life-history
is not yet advanced enough to form more than a provisional

arrangement.

We know t hat bacteria are allied to Hyphomycetes on the one
hand and Saccharomyoetes on the other, and that they have no
differentiation into root, stem, or leaf; we know that they are fungi

(having no chlorophyll), in which no sexual reproduction occurs, and

that their mode of multiplication is by division. Prom such facts as
these we may huild up a classification as follows: —

I Vegetable Kingdom.

THE BIOLOGY OF BACTERIA

VEGETABLE KINGDOM.

I

Thallophyta.

[ = The lowest forms
of vegetable life. No
differentiation into
root, stem, or leaf.]
I

Museinea?.

Pteridophyta.

I
Phanerogam ia.

Algae.

[ = Chlorophyll
present.]

Fungi.

[ = No Chlorophyll.]

Hymenomycetes.

(Mushrooms, etc.)

Hyphomycetes.

(Moulds.)

Blastomycetes.

(Yeasts, etc.) [:

Schizomycetes

multiplication by cell
division or by spores]
or
Bacteria.

( 1 ) Coccaceae * — round

cells.

(2) Bacteriaceae — rods

and threads.

(3) Leptotricheae.

(4) Cladotricheae

• hi

* Migula has suggested that the [Schizomycetes should be subdivided into CoccaeecB, Bacteriacece,
(spirilla, spirochceta), Chlamydobacteriacece (Streptothrix, Crenothrix, Cladothrix), and Beggiatoa.

Spirillaceaz

Morphology: Structure and Form

Having now located micro-organisms in the economy of nature,
we may proceed to describe their subdivisions and form. For
practical convenience rather than theoretical accuracy, we may accept
the simple division of the family of bacteria into three chief forms,
viz. : —

C (1) Bound cell form — coccus.
Lower Bacteria-] (2) Bod form — bacillus.

( (3) Thread form — spirillum.
Higher Bacteria — Leptothrix, Streptothrix, Cladothrix, etc.

A classification dependent as this is upon the form alone is not by
any means ideal, for it ignores all the complicated functions of
bacteria, but it is, as we have said, practically convenient.

1. The Coccus. — This is the group of round cells. They vary in
size as regards species, and as regards the conditions, artificial or
natural, under which they have been grown. Some are less than
2"5li(ro °f an mcn m diameter ; others are half as large again, if the
word large may be used to describe such minute objects. No regular
standard can be laid clown as reliable with regard to their size.
Hence the subdivisions of the cocci are dependent not upon the
individual elements so much as upon the relation of those elements to
each other. A simple round cell of approximately the size already
named is termed a micrococcus (/x^/cyoo?, small). Certain species of

FORMS OF BACTERIA

micrococci always or almost always occur in pairs, and such a com-
bination is termed a diplococcus. Some diplococci are united by a
thin capsule, which may be made apparent by special methods of
staining; in others no limiting or uniting membrane can be seen
with the ordinary high powers of the microscope. Again, one fre-
quently finds a species which is exactly described by saying that two
micrococci are in contact with each other, and move and act as one
individual, but otherwise show no alteration ; whilst others are seen
which show a flattening of the side of each micrococcus which is in
relation to its partner. Perhaps the diplococci in an even greater

c$xO&

degree than the micro-
cocci respond to external
conditions both as regards
size and shape. It must
further be borne in mind
that a dividing micrococcus
assumes the exact appear-
ance of a diplococcus
during the transition stage
of the fission. Hence, with
the exception of several
well - marked species of
diplococci, this form is
somewhat arbitrary. The
third kind of micrococcus is
that formed by a number
of elements in a twisted
chain, named streptococcus
(crT/)e7rTo?, twisted). This
form is produced by cells
dividing in one axis, and
remaining in contact with
each other. It occurs in a
number of different species,
or what are supposed by many authorities to be different Bpecies,
owing to their different effects. Morphologically all the streptococci
are similar, though a Bomewhal abortive attempt lias been made to

• livido them into two groups, according as i<> whether they were long
chains or short. As a matter of fact, the h-nui h of streptococci

• If ponds in some cases upon biological properties, in others upon
externa] treatment or the medium of cultivation which has boon used.
Sometimes they occur as Btraighl chains of only half a dozen
elements; at other times they may contain thirty or forty elements,
ami twist, in various ways, even forming rosaries. The elements, too,
differ not only in size, but in shape, appearing occasionally as oval

1.— Diagrams of VariO

Micrococcus.
Diplococcus.

COCCUS.

•1. Staphylococcus.

5. Leuoonostoc, show*

tag Art In •

6. Mci :

7. Sarcina.
- Bacillus.

'.'. Spirillum.

8 THE BIOLOGY OF BACTERIA

cells united to each other at their sides. The fourth form is consti-
tuted by the micrococci being arranged in masses like grapes, the
staphylococcus (<TTa</>v\is, a bunch of grapes). The elements are
often smaller than in the streptococcus, and the name itself describes
the arrangement. There is no matrix and no capsule. This is the
commonest organism found in abscesses, etc. The sarcina is best

classified amongst the cocci, for it is
composed of them, in packets of four
or multiples of four, produced by divi-
sion vertically in two planes. If the
division occurs in one plane, we have
as a result small squares of round cells
known as mcrismopedia. In both these
conditions it frequently happens that
the contiguous sides of the elements of
packets become faceted or straightened
against each other. It may happen,
too, particularly in the sarcince, that
fig. 2.-Diagram of sarcina. segmentation is not complete, and that

the elements are larger than in any
other class of cocci. They stain very readily. Nearly all the cocci
are non-motile, though Brownian movement (see p. 11) may readily
be observed.

2. The Bacillus. — This group consists of rods, having parallel
sides and being longer than they are broad. They differ in every
other respect according to species, but these two characteristics
remain to distinguish them. Many of them are motile, others not.
The ends or poles of a bacillus may be pointed, round, or almost
exactly square and blocked. They all, or nearly all, possess a
capsule. Individuals of the same species may differ greatly,
according to whether they have been naturally or artificially grown,
and pleomorphic forms are abundant.

3. The Spirillum. — This wavy-thread group is divisible into a
number of different forms, to which authorities have given special
names. It is sufficient, however, to state that the two common
forms are the non-septate spiral thread {e.g. the Spirillum Obermeier
of relapsing fever), which takes no other form but a lengthened
spirillum ; and the spirillum which breaks up into elements or units,
each of which appears comma-shaped {e.g. the cholera bacillus). The
degree of curvature in the spirilla, of course, varies. They are the
least important of the lower bacteria.

The Higher Bacteria group includes more highly organised
members of the Schizomycetes. They possess filaments, which may
be branched, and almost always have septa and a sheath. Perhaps
the most marked difference from the lower bacteria is in their

POLYMORPHISM 9

reproduction. In the higher bacteria we may have what is in fact
a flower — terminal fructification by conidia. In this group of
vegetables we have the Beggiatoa, Leptothrix, Cladothrix, and, at the
top, the Streptothrix. It has been demonstrated that Streptothrix
actinomycotica and Streptothrix madurce are the organismal cause,
respectively, of Actinomycosis and Madura-foot, two diseases which
had hitherto been obscure.

Polymorphism (or Pleomorphism). — This term is used to designate
an inconstancy of form or a tendency towards biological variation.
Vibrios may become spirilla, the ray fungus passes through a coccoid
and bacillary stage, and the diphtheria bacillus may either be long,
short, straight, or clubbed. This diversity of form appears to belong
to many species, and is transmitted from generation to generation ;
or the various forms may occur in succession, and represent different
stages in the life-history. In B. diphtheria, />'. pestis, and B. tuber-
culosis and other forms, polymorphism undoubtedly occurs. It
is particularly marked in very old cultures of the last named.
The ordinary well-known bacillus may grow out into threads witli
bulbous endings, granular filaments, "drumsticks," and diplococcal
forms, [t is now known that amongst the causes of polymorphism
are certain adverse conditions of medium <<r other physical influences
(moisture, temperature, age, etc.), and thus Borne bacteria, especially
bacilli in- vilnius, become altered in shape, Losing their ordinary form.
On transferring such aberrant and abnormal forms to fresh medium
or favourable conditions, they are generally able t<> assume their
original morphology, [ndeed the aberrant form is in all probability
only s stage in t licit- life-history. Twt forms usually imply

degenerat ion.

Biology of Bacteria

Composition. — Prom what we have -e<>n of the pabulum of
micro-organisms, wo should conclude that in Borne form or other they
contain the elements nitrogen, carbon, and hydrogen. All three
Bubstances are combined in the i <>r protoplasm of which

the body of the microbe consists. Tin- is generally homogeneous,
proteid material, and there is no \ .< nucleus. It p<

marked affinity for aniline .lyes, and by this means organisms are
Btained ha- the microscope. Besides the variable quantity of
nitrogen present, mycoprotein may also contain various mineral
Baits. The uniformity of the cell-protoplasm may he materially
affected by disintegration and segynerUatioTi due i.» degenerative
changes. Va which it. is necessary to differentiate from spores,

also may appear from a like cause. Vacuolation may also occur
as ,i result of a process of osmosis in salt solutions, the protoplasm
of the bacillus becoming contracted ami disintegrated {plasmolysis),

10 THE BIOLOGY OF BACTERIA

Two other signs of degeneration are the appearance of grannies in
the body of the cell-protoplasm known as metachromatic granules,
owing to their different staining propensities, and the polar bodies
which are seen in some species of bacteria. Surrounding the mass
of mycoprotein, we find in most organisms a capsule or membrane
composed, in part at least, of cellulose. This sheath plays a protective
part in several ways. During the adult stage of life it protects the
mycoprotein, and holds it together. At the time of reproduction or
degeneration it not infrequently swells up, and forms a viscous hilum
or matrix, inside which are formed the new sheaths of the younger
generation. It may be rigid, and so maintain the normal shape of
the species, or, on the other hand, flexible, and so adapted to rapid
movement of the individual.

Here, then, we have the major parts in the constitution of a
bacillus — its body, mycoprotein; its capsule, cellulose. But, further

f

/

f s

7 V

Fig. 3.— Diagrams of Normal and Polymorphic Forms of Tubercle Bacilli.

thau this, there are a number of additional distinctive characteristics
as regards the contents inside the capsule which call for mention.
Sulphur occurs in the Beggiatoa which thrive in sulphur springs.
Starch is commoner still. Iron as oxide or other combination is
found in several species. Many contain pigments, though these
are generally the "innocent" bacteria, in contradistinction to
the disease-producing. A pigment has been found which is
designated bacterio-purpurin. According to Zopf, the colouring
agents of bacteria are the same as, or closely allied to, the
colouring matters occurring widely in nature. Migula holds that
most of the bacterial pigments are non-nitrogenous bodies. There
are a very large number of chromogenic bacteria, some of which
produce exceedingly brilliant colours. Among some of the commoner
forms possessing this character are Bacillus et micrococcus violaceus,
B. et M. aurantiacus (orange) ; B. et M. luteus ; M. roseus (pink) ;
many of the Sarcincu ; B. aureus; B. fluorescens liquefaciens et

BACTERIAL POWERS OF MOTION 11

non-liquefaciens (green) ; B. pyocyaneus (green) ; B. prodigiosus
(blood-red).

Motility. — When a drop of water containing bacteria is placed
upon a slide, a clean cover-glass superimposed, and the specimen
examined under an oil immersion lens, -various rapid movements will
generally be observed in the micro-organisms. These are of four
chief kinds : (1) A dancing, stationary motion known as Brownian
movement. This is molecular, and depends in some degree upon heat
and the medium of the moving particles. It is non-progressive, and
is well seen in gamboge particles. (2) An undulatory, serpentine
movement, with apparently little advance being made. (3) A
rotatory movement, which in some water bacilli is very marked, and
consists of spinning round, sometimes with considerable velocity, and
maintained for some seconds or even minutes. (4) A progressive,
darting movement, by which the bacillus passes over some con-
siderable distance.

The conditions affecting the motility of bacteria are but partly
understood. Heating the slide or medium accelerates all movement.
A fresh supply of oxygen, or indeed the addition of some nutrient
substance, like broth, will have the same effect. There are also the
somewhat mysterious powers by which cells possess inherent
attraction or repulsion for other cells, known as positive and negative
chemiotaxis. These powers have been observed in bacteria by
Pfeiffer and Ali-Cohen.

The essential condition in the motile bacilli is the presence of
flagella* Theso cilia, or hairy processes, project from the sides or
from the ends of the rod, and are freely motile and elastic. Some-
times only one or two terminal flagella are present; in other cases,
like the bacillus of typhoid fever, five to twenty may occur all round
the body of the bacillus, varying in length and size, sometimes
being of greater length even than the bacillus itself. It is not yet
established as to whether these cilia arc prolongations of capsule
only, or whether they contain something of the body protoplasm.
Migula holds the former view, and states thai the position of
flagella is constant enough for diagnostic purposes. They are hut
rarely recognisable except by means of special staining methods.
Micrococcus agilis (Ali-Cohen) is one of the rare oases of a coccus
which has flagella and powers of active motion.

Modes of Reproduction. — Budding, division, and spore formation
are the t hi'cc chief wnys in which Schizomycetes and Saccharomycetes

(yeasts) reproduce their kind. /!//</, li/i;/ occurs in many kinds of

yeast-cells, and generally takes place when the nutrimenl and

* A f!(t(/i//.ii)n is a hair-like process arising from the poles or sides of the bacillus.
It must not be confused with a /''/a///, ///, which is a thread-like growth of the bacillus
itself.

12 THE BIOLOGY OF BACTERIA

environment are favourable. The capsule of a large, or "mother"
cell, shows a slight protrusion outwards, which is gradually enlarged
into a " daughter " yeast, and later on becomes constricted at the neck.
Eventually it separates as an individual. The protoplasm of the
spores of yeasts differs, as Hansen has pointed out, according to the
conditions of culture.

Division, or fission, is the commonest method of reproduction.
It occurs transversely. A small indentation occurs in the capsule,
which appears to make its way slowly through the whole body of the
bacillus or micrococcus until the two parts are separate, and each
contained in its own capsule. It has been pointed out already that
in the incomplete division of micrococci we observe a stage precisely
similar to a diplococcus. So also in the division of bacilli an appear-
ance occurs described as a diplobacillus.

Simple fission requires but a short period of time to be complete.
Hence multiplication is very rapid, for within half an hour a new
adult individual can be produced. It has been estimated that at this
rate one bacillus will in twenty-four hours produce millions of similar
individuals; or, expressed otherwise, Cohn calculated that in three
days, under favourable circumstances, the rate of increase would be
such as to form a mass of living organisms weighing many tons, and
numbering billions of individuals. Favourable conditions do not
occur, fortunately, to allow of such increase, which, it is evident, can
only be roughly estimated. But the above facts illustrate the
enormous fertility of micro-organic life. When we remember that in
some species it requires 10,000 or 15,000 fully-grown bacilli placed
end to end to stretch the length of an inch, we see also how exceed-
ingly minute are the individuals composing these unseen hosts.

Spore formation may result in the production of germinating cells
inside the capsule of the bacillus, endospores, or as modified individuals,
arthrospores. The body of a bacillus, in which sporulation is about
to occur, loses its homogeneous character and becomes granular, owing
to the appearance of globules in the protoplasm. In the course of
three or four hours the globule enlarges to fill the diameter of the
rod, and assumes a more concentrated condition than the parent cell.
At its maturity, and before its rupture of the bacillary capsule, a
spore is observed to be bright and shining, oval and regular in shape,
with concentrated contents, and frequently causing a local expansion
of the bacillus. In a number of rods lying endwise, these local
swellings produce a beaded or varicose appearance, even simulating
a streptococcus. In the meantime the rod itself has become slightly
broader and pale. Eventually it breaks down by segmentation or
by swelling up into a gelatinous mass. The spore now escapes and
commences its individual existence. Under favourable circumstances
it will germinate. The tough capsule gives way at one point,

MODES OF REPRODUCTION

13

J&

* If

v°<>o

generally at one of the poles, and the spore sprouts like a seed.
In the space of about one hour's time the oval refractile cell has
become a new bacillus. One spore produces by germination one
bacillus. Spores never multiply by fission, nor reproduce themselves.

Hueppe has stated that there are certain organisms (like
Leuconostoc, and some streptococci) which reproduce by the method
of arthrospores. Defined shortly, this is simply an enlargement of
one or more cell elements in the chain which thus takes on the
function of maternity. On either side of the large coccus may be
seen the smaller ones,
which it is supposed have (o)

contributed of their proto-
plasm to form a mother
cell. An arthrospore is
said to be larger, more re-
fractile, and more resistant
than an ordinary endospore.
Many bacteriologists of re-
pute have declined hitherto
definitely to accept arthro-
spore formation as a proved
fact.

Spore formation in bac-
teria is not to be considered
as a method of multipli-
cation. The genera] nib-
is undoubtedly that one
bacillus produces one sport1,
and one spore germinates
into one bacillus. It is a
reproduction, nol a mul-
tiplication, [ndeed, the
whole process is of the nature oi a resting stage, and is due (a) to the
arrival of the adull bacillus al its biological zenith, <t (fr) to the con-
ditions in which it finds Itself being unfavourable to further vegeta-
tive growth, and so it, endeavours to perpetuate Its species. Most
authorities are probably of the latter opinion, though there is nol a
little evidence for the former. Exactly what conditions are favour-
able to Bporulation is nol known. Nutriment has probably an
intimate effect upon it. The temperature must not be below 10 ('..
nor much above 40 C. Oxygen, as we have Been, is favourable, if
not necessary, to many species, which will in cultivation in broth
rise to the surface and Lodge in the pellicle to ham their seeds.
Moisture, too, is considered a necessity.

Koch found thai Bpore formation in B. anthracis occurred in six

<£=>&

"V

4.— Diagram Formai iom

I'l IOJ I LA.

A. Btaoe

in formation i Its After development.

15. Bpirlllum with terminal fla

14 THE BIOLOGY OF BACTERIA

hours. The spores may be situated in the middle of the bacillus
(as in B. anthracis, B. acidi hutyrici, etc.), towards one end (Bacillus
of Malignant (Edema), or actually terminal (B. tetani). Those spores
produced inside the capsule of the bacillus are termed endosporcs.
Hueppe has described the spores of certain streptococci as arthrospores.
The spores of yeast are termed ascospores. The spores of all
bacillary species possess, however, certain characters in common.
They are as follow. The spore is generally oval, though more
spherical in the Hyphomycetes : it is bright and glistening in aspect;
it is often greater in diameter than the bacillus giving rise to it;
its capsule is thicker and stronger than the capsule of the parent
bacillus; and it is generally held that the contained protoplasm is
more concentrated, so to speak, than that of the bacillus. These two
last characters are of chief importance to us, for it is owing to them
that spores possess such marked power of resistance. Colin lias
suggested that the capsule of a spore is in reality a double envelope,
an inner one of fatty and an outer one of gelatinous nature, and it is
owing to this that its resistance to heat and dessication is due. The
protoplasm of the spore contains, of course, the essential constituents
of the mother cell. It is the method by which " the continuity of
germ plasm " is secured in these lowly forms of life. Under favourable
circumstances this spore-protoplasm will germinate into a new bacillus.
It should be understood that whilst holding the view that
spores are a resting stage during adverse conditions,* we fully
recognise that certain favouring external conditions are essential

* Yeast can be effectually starved by cultivating on a small block of plaster-
of-Paris kept moist under a bell jar ; under these circumstances the yeast is
supplied with nothing but water. In a few days the protoplasm of yeast cells thus
circumstanced becomes filled with vacuoles and fat cells. The protoplasm has
been undergoing destructive metabolism, and, there being nothing to supply new
material, has diminished in quantity and at the same time been partly converted
into fat. Both in plants and animals fatty degeneration is a more or less constant
phenomenon of starvation, and to this bacteria are no exception. After a time the
protoplasm collects towards the centre of the cell, and divides simultaneously into
four masses arranged like a pyramid of four billiard balls, three at the base and
one above. These are the ascospores, and sooner or later they are liberated by
the rupture of the mother-cell wall. Certain of the Streptothrix family also
" sporulate " when they find themselves, like yeast upon gypsum, surrounded
by an unfavourable environment. Again, in old cultures, it will be found that
when the food supply has been exhausted the bacteria have either sporulated or
have died. For these reasons sporulation may be looked upon not as a method
of multiplication but one of reproduction, of carrying on the species under adverse
conditions. With regard to the rapid formation of spores under apparently
favourable circumstances (B. filamentosus, B. anthracis, etc.), it must be borne in
mind that the medium may not be by any means so favourable as appears to be the
case (Fliigge). It is clear that the food supply immediately around many of the
bacteria in a culture must soon be exhausted. Besides, there is the toxic influence
early at work, often as an inimical agency acting unfavourably towards the bacillus
producing it. So that the appearance of spores in such a culture may still be due
to conditions which are actually unfavourable.

INFLUENCE OF EXTERNAL CONDITIONS 15

to spore formation. Of these, there are at least three of which
bacteriologists have knowledge, namely, moisture, oxygen, and a
certain temperature. Fluid media forms an excellent nidus for
spor ulation so long as some oxygen can gain access to the sporulating
germs. But many organisms will not sporulate if lying deep in
such a medium. In moulds and yeasts oxygen is essential, and for
some spore-bearing bacilli a supply of oxygen is a sine qud non (the
exceptions are strict anaerobes like B. tetani, B. butyricus, etc.) of
sporulation. Prazmowski has pointed out that it is characteristic of
these forms that they are non-motile during sporulation. B. tetani,
B. butyricus, and other strict anaerobes continue to remain motile
during sporing. Temperature exerts a marked influence on the
process.* In the case of B. subtilis, an organism frequently present in
milk, spore formation did not occur below 6r C. ; at 18° C. it required
two days; at 22° G. one day; and at 30° C. only twelve hours.f

When free in the field of the microscope, spores must be dis-
tinguished from fat cells, micrococci, starch cells, some kinds of ova,
yeast cells, and other like objects. Spores are detected frequently
by their resistance to ordinary stains and the necessity of colouring
them by special staining methods. When, however, a spore has
taken on the desired colour, it retains ii with tenacity. In addition
to their shape, size, thickened capsule, and staining characteristics,
spores also resist desiccation and heat in a much higher degree than
bacilli not bearing spores. It has been suggested that bacteria
should be classified according to their method of spore formation.

The Influence of External Conditions on the
Growth of Bacteria

In the earliest days of the study of micro-organisms it was

observed that they mostly congregate where there is suitable food

for their nourishment. The reason why fluids such as milk, ami
dead animal matter such ;is a carcase, and living tissues Buch as a
man's body, contain many microbes, is because each of these three
media is favourable to their growth. Milk affords almost an ideal
food and environment for microbes. Its temperature and con-
stitution frequently meet their requirements. Dead animal matter,
too, yields a rich diet for certain species (saprophytes). In the
living tissues bacteria obtain not only nutriment, but a favourable

* Koch lias shown in the case of /.'. amtkrOCU that at least Hi ('. is necessary
for spore formation, and at this temperature limited formation of spores did not
OOC1U until after seven days. At 2] ( , spores had formed after seventy-two hours,
at 25° C. after thirty-live to forty hours, and between SO ('. and 40° C. in about
twentv-fonr hours; the best and strongest cultivations were obtained from 20"

t Flttgge. diierXHtrgamifnt. Translation by W. Watson Cheyne, 1890, p. 539.

16 THE BIOLOGY OF BACTERIA

temperature and moisture. Outside the human body it has been the
endeavour of bacteriologists to provide media as similar to the above
as possible, and containing many of the same elements of food, in
order that the life-history may be carried on outside the body and
under observation. By means of cover-glass preparations for the
microscope we are able to study the form, size, motility, flagella,
spore formation, and peculiarities of staining, all of which characters
aid us in determining to what species the organism under examination
belongs. By means of artificial nutrient media we may further
learn the characters of the organism in " pure culture," * its favour-
able temperature, its power or otherwise of liquefaction, of curdling
of milk, or of gas or acid production ; its behaviour towards oxygen ;
its power of producing indol, pigment, and other bodies ; as well as
its thermal death-point and resistance to light and disinfectants. It
is well known that under artificial cultivation an organism may be
greatly modified in its morphology and physiology, and yet its
conformity to type remains much more marked than any divergence
which may occur.

Nutritive Medium. 1 The basis of many of these artificial media is broth.
This is made from good lean beef, free from fat and gristle, which is finely minced
up and extracted in sterilised water (one pound of lean beef to every 1000 c.c. of
water). It is then filtered and sterilised. To provide peptone beef-broth, ten
grammes of peptone and five grammes of common salt are added to every litre of
acid beef-broth. It is rendered slightly alkaline by the addition of sodium car-
bonate or sodium hydrate, and is filtered and sterilised. In glycerine-broth G to 8
per cent, of glycerine has been added after filtration, in glucose-broth 1 or 2 per cent,
of grape-sugar. This latter is used for anaerobic organisms. The use of broth as
a culture medium is of great value. It is undoubtedly the best fluid medium, and
in it may not only be kept pure cultures of bacteria which it is desired to retain for
a length of time, but in it also emulsions and mixtures may be placed preparatory to
further examination. Gelatine consists of broth solidified by the addition of 100
grams of best French gelatine to the litre. Its advantage is twofold : it is trans-
parent, and it allows manifestation of the power of liquefaction. When we speak
of a liquefying organism we mean a germ having the power of producing a pepton-
ising ferment which can at the temperature of the room break down solid gelatine
into a liquid. Grape-sugar gelatine is made like grape-sugar broth. Agar was
introduced as a medium which would not like gelatine melt at 25° C, but remain
solid at blood-heat (37 '5° C. ; 98 "5° F.). It is a seaweed generally obtained in dried
strips from the Japanese market. Ten to fifteen grammes are added to every litre
of peptone-broth. Glycerine and grape-sugar may be added as elsewhere. Blood
agar is ordinary agar with fresh sterile blood smeared over its surface. Blood serum
is drawn from a jar of coagulated horse-blood, in which the serum has risen to the
top. This is collected in sterilised tubes and coagulated in a special apparatus (the
serum inspissator). Potato is prepared by scraping ordinary potatoes, washing in
corrosive sublimate, and sterilising. It may then be cut into various shapes con-
venient for cultivation. Upon any of these forms of solid media the characteristic

* A " pure culture " is a growth, in an artificial medium outside the body, of one
species of micro-organism only.

f The facts here given are obviously only general indications. The accurate
preparation of medium is of vital importance in Bacteriology, and for its accomplish-
ment text-books should be consulted (Eyre's Bacteriological Technique, 125-174).

TEMPERATURE

17

growth of the organism can be observed. Of the nutrient elements required,
nitrogen is obtained from albumens and proteids, carbon from milk-sugar,
cane-sugar, or the splitting up of proteids ; salts (particularly phosphates and
salts of potassium) are readily obtain-
able from those incorporated in the ^ f

media ; and the water which is required ^- .

is obtainable from the moisture of the
media.

&*

r ■ . 5. [n<x ii.aiiv

rialinuin wire fused into 'rlass handles.

There are two common forms

of test-tube culture, viz., on the

surface and in the depth of the

medium. In the former the

medium is sloped, and the inocu-
lating needle is drawn along its

surface ; in the latter the needle

is thrust vertically downwards

into the depth of the solid

medium. Plate cultures and anaerobic cultures willHbe described at

a later stage.

Temperature. — When the medium has been inoculated the

culture is placed at a temperature which will be favourable. For

every species of bacteria there is a favour-
able temperature, termed the optimum
temperature. This is usually the tempera-
ture of the natural habitat of the organism.
Two standards of temperature are in use in
bacteriological Laboratories. The one, room
temperature, varies from L8°-22 ( '. ; the
other is blooil-Jn^i, and varies from 35°-38 C.
< Plates 1 and 2). It is true some will

grow below L8 C, and others above 38 C.
The pathogenic (disease-producing) bacteria
i brive best as a rule at 3*7 ( '., and the non-
pathogenic ai the ordinary temperature of
the room, The different degrees of tempera-
ture arc obtained by means of incubators.
For the low temperatures gelatine is chosen
as a medium, for the higher temperatures
agar. Most ba< prow well at room

temperature about 60 F.), hut they will
grow more Luxuriantly and speedily at
blood-heat
Whilst these .in' the ordinary limit- of temperature affecting

bacteria, they <1<> qo! by any means include the extremes of beat and

cold which micro-organisms can withstand. The average thermal

death-point is about 55 C, but certain Bpecies, termed thermophilic,

B

.. 6. Media for H
and Depth Culture.

18 THE BIOLOGY OF BACTERIA

isolated from the intestine, horse manure, etc., grow at 60°-70° C.
On the other hand, investigations have shown that bacteria can
withstand exceedingly low temperatures. Koch showed that the
cholera vibrio was not killed by a temperature of —32° C. In 1900,
Swithinbank exposed cultures of the tubercle bacillus to the
temperature of liquid air ( — 193° C.) for continuous periods varying
from six hours to forty-two days, without their vitality being affected ;
and in the same year MacFadyen and Eowland found that Proteus
vulgaris, B. coli, .and several other species were not killed after an
exposure of ten hours to a temperature of liquid hydrogen ( — 252° C).
It will thus be seen that bacteria can withstand great alternations of
temperature. From a public health point of view, it is important
to remember that organisms can exist in freezing mixtures and ice,
retaining their vitality and virulence. For example, B. coli and the
typhoid bacillus can exist from the low temperatures above mentioned
to 80° C, although the usual thermal death-point for these species
is between 50°-60° C *

Moisture has been shown to have a favourable effect upon the
growth of microbes. Drying will of itself kill many species {e.g. the
spirillum of cholera), and other things being equal, the more moist a
medium is, the better will be the growth upon it. Thus it is that the
growth in broth is always more luxuriant than that on solid media.
Yet the growth of Bacillus suhtilis and some other species are an
exception to this rule, for they prefer a dry medium. Desiccation as
a rule diminishes virulence and lessens growth. But some species
can withstand long-continued drying without injury.

Light acts as an inhibitory, or even germicidal, agent. This
fact was first established by Downes and Blunt in a memoir, to the
Eoyal Society in 1877. They found by exposing cultures to different
degrees of sunlight that the growth of the culture was partially or
entirely prevented, being most damaged by the direct rays of the
sun, although diffuse daylight acted prejudicially. Further, these
same investigators proved that the rays of the spectrum which acted
most inimically upon bacteria were the blue and violet rays, next to
the blue being the reel and orange-red rays. The action of light,
they explain, is due to the gradual oxidation which is induced by the
sun's rays in the presence of oxygen. Duclaux, who worked at this
question at a later date, concluded that the degree of resistance to
the bactericidal influence of light, which some bacteria possess,
might be due to difference in species, difference in culture media, and
difference in the degrees of intensity of light. Tyndall tested the
growth of organisms in flasks exposed to air and light on the Alps,

* For the latest researches on this point, see Proc. Roy. Soc, 1900 and 1901 ; and
the Thirty-fourth Annual Report of the State Board of Health, Massachusetts, 1903,
pp. 269-281. Dewar commenced experiments of this character in 1892.

PLATE 1.

A FORM "i 1'vsikik's LaRQI INCUBATOR KOR CULTIVATION A I ROOM TkMPKRATURK,

[To fac< ]>"<" 18.

EFFECT OF LIGHT 19

and found that sunlight inhibited the growth temporarily. A large
number of experimenters on the Continent and in England have
worked at this fascinating subject since 1877, and though many of
their results appear contradictory, we may be satisfied in adopting
the following conclusions respecting the matter : —

(1) Sunlight has a deleterious effect upon bacteria, and to a less
extent on their spores.

(2) This inimical effect can be produced by light irrespectively of
rise in temperature.

(3) The ultra-violet rays are the most bactericidal, and the
infra-red the least so, which indicates that the phenomenon is due
to chemical action.

(4) The presence of oxygen and moisture greatly increase this
action, the process being largely an oxidation.

(5) Sunlight also acts prejudicially upon the culture medium,
and thereby exerts an injurious action on the culture.

(6) The time occupied in the bactericidal action depends
upon the intensity of the light and the inherent vitality of the
organism.

(7) Withregard to the ad inn of light upon pathogenic organisms,
some results have recently been obtained with Bacillus typhosus.
Janowski maintains that direct sunlight exerts a distinctly depressing
effect on typhoid bacilli At present more cannot be Baid than thai
sunlight and fresh air are two of the most powerful agents we p
with which to combat pathogenic germs,

A very simple method of demonstrating the influence of light
is to grow a pure culture in a favourable medium, either in a test-
tube or upon ;i glass plate, and then cover the whole with black
paper or cloth, A little window may then be cut in the protec-
tive covering, and the whole exposed to the light. Where it
reaches in direct rays, it will be found that little or no growth has
occurred; where, on the other hand, the culture lias been in the
dark, abundant growth occurs. In diffuse light the growth is
merely Bomewhat inhibited.

A Dumber of experiments in this direction wore made at Lawrence,
Massachusetts,* with cultures of typhoid and B. colt.

In two experiments, each with typhoid bacillus and />'. colt, water
dilutions were made from fresh cultures of the germs, 1 c.c. of this
water being placed in Petri dishes in the sun for definite periods.
After exposure, the water in the plates was mixed with agar, and all
plates were incubated twenty-four hours ;ii :'..s , after which the
number of colonies was counted In one experiment the water
dilution of typhoid was mixed with melted agar, and plates made as

Thirty -fourth Ann. /.'«/<. Stab /;./. of li<ilih of Mattaehuttii99 190U,

p. 2,

20

THE BIOLOGY OF BACTERIA

usual. After the agar had set, these plates were then exposed to the
sunlight. In one experiment with B. coli, the water culture was not
exposed to the sunlight in plates, but the exposure was made in a
clear, white glass bottle of the Blake pattern, holding 100 c.c,
samples being taken from this at the proper intervals, and plated
as usual.

In all cases control cultures were made under exactly the same
conditions as were the cultures exposed, these, however, being pro-
tected from the sunlight by a heavy, opaque cloth, or some similar
material. The temperature of these cultures was, of course, consider-
ably lower than was the temperature in the sun. The numbers of
bacteria in the controls showed the usual variation to be expected
under the circumstances, usually a slight reduction in numbers being
noted during two or three hours' standing, although in one instance
the numbers increased quite materially. The data of these control
cultures are not shown in the accompanying tables.

The brightness of the sun also varied considerably, and attempts
were made to measure the amount of light by photographic means,
but these measurements were unsatisfactory, and the data are not
included here.

"With typhoid, from 95 to 99 per cent, of all (lie germs were
destroyed by ten to fifteen minutes' exposure to direct sunlight. A
few germs may resist the sunlight for a somewhat longer time ;
usually, however, all the germs were destroyed by three or more
hours' exposure to bright sunlight. The results of the experiments
with typhoid are shown in the following tables : —

Table shotting Elimination of Typhoid Germs in Wait

t on

Exposure

to Sun light.

Exposure.

Experiment 24.

Experiment 25.

Bacteria.

Average.

Bacteria.

Average.

Start

15 minutes
30 minutes
45 minutes
1 hour .

1 .V hours

2 hours .
4 hours .
6 hours .

698 734

66 4

17 1

0 1

6 2

2 0
0 0

3 1
0 0

716
35
9
1
4
1
0
2
0

592 532
13 5

4 4

3 0
21 5

5 3

4 3
0 0
0 0

562
9
4
2
13
4
4
0
0

EFFECT OF LIGHT

21

Table showing Elimination of Typhoid Germs in Agar
Plates on Exposure to Sunlight.

Exposure.

Temperature.

Experiment 26.

Bacteria.

Average.

Start

608 642

625

10 minutes ....

91° F.

2 3

3

20 minutes ....

83

7 2

5

30 minutes ....

81

0 0

0

40 minutes ....

83

0 0

0

50 minutes ....

83

0 0

0

1 hour

78

1 0

1

1 hour, 10 minutes

74

0 0 0

1 hour, 20 minutes

71

0 0 0

1 hour, 30 minutes

69

0 0 0

1 hour, 40 minutes

68

(» 0 0

1 hour, 50 minutes

71

0 0 0

2 hours ....

68

0 0 0

With B. coli the results have been somewhat more variable, prob-
ably due to more chi mgeable conditions. In one experiment, some-
thing over 80 per cent, of the germs were destroyed by fifteen
minutes' exposure, all being destroyed after four hours. The results
of this experiment were undoubtedly influenced greatly by clouds
in the sky, so that at times the sunfighl was do1 very bright, after
about two and one-half hours the sun being entirely overcast. In
one experiment about 96 per cent, of the germs were eliminated at
the end of fifteen minutes, and after thirty minutes all of the germs
were destroyed. In these two experiments the water cultures were
exposed in plates, the results being shown in the following table: —

Table showing Elimination of B. coli in Water Cultures on
Exposure to Sunlight.

Exposure.

rlment 78.

■ iin. 'lit 7'.'.

Tem-
perature.

Bad
per cc.

\\< ■

Tern-

terla
per e.o.

Start .
L5 minutes
80 minutes
i."> minutes
1 hour

1 | hours .

2 hours
S hours

1 hours

78
80
80

82

si

70,000 7-J.S00
6,564 16,166

1,17.'. 2,
2,130 0
590 0
2,130

10 70

6 ii

71,4

12,

1,065
295

1,111
10

0

106

107
110

100
108
105
100
78

4 15,400 824,800
19,100 81,800

0 0
0 0
0 0
0 0
0 0
0 ii

.. (.

25,400

0
0
0
0
0
0
0

22

THE BIOLOGY OF BACTERIA

In one experiment the water was exposed in bottles. In this
case about 98 per cent, of the germs were destroyed after fifteen
minutes, the cultures varying somewhat. The germs persisted in
the water in considerable numbers for two hours and in small
numbers up to four hours, after five hours the sample being
completely sterilised. The results of this experiment arc shown
as follows : —

Table showing Change in Numbers of B. coli in Water in
Bulk on Exposure to Sunlight.

Temperature.

Experiment 81.

Exposure.

Bacteria pel
c.c.

Arerage.

Start .
15 minutes
30 minutes
45 minutes

1 hour .
1\ hours

2 hours .

3 hours .

4 hours .

5 hours .

94
94
92
92
94
96

109
95

102

78

2,360,000 1,620,000
30,000 43,200
85,300 22,000
44,000 55,000
53,700 45,300
35,800 34,100
57,400 76,400
150 1,17 2

3

0 0

1,990,000
36,600
53,650
19,500

1! 1,500

34,950

66,900

786

4
0

It has been found that the electric light has but little action upon
bacteria, though that which it has is similar to sunlight. Recent
experiments with the Bontgen rays have not given bactericidal
results.

In 1890 Koch stated that tubercle bacilli were killed after an
exposure to direct sunlight of from a few minutes to several hours.
The influence of diffuse light would obviously be much less. Professor
Marshall Ward has experimented with the resistant spores of
Bacillus anthracis by growing these on agar plates and exposing to
sunlight. From two to six hours' exposure had a germicidal effect.*

It should be remembered that several species of sea- water bacteria
themselves possess the property of phosphorescence. Pniiger was the
first to point out that it was such organisms which provided the
phosphorescence upon decomposing wood or decaying fish. To what
this light is due, whether capsule, or protoplasm, or chemical product,
is not yet known. The only facts at present established are to the
effect that certain kinds of media and pabulum favour or deter
phosphorescence.

* See Trans. .Tenner Inst, (second series), 1899, p. 81.

MEANS OF STERILISATION

23

Aerobiosis. — Pasteur was the first to lay emphasis upon the
effect which free air had upon micro-organisms. He classified them
according to whether they grew in air, aerobic, or whether they
flourished most without it, anaerobic. Some have the faculty of
growing with or without the presence of oxygen, and are designated
as facultative aerobes or anaerobes. As regards the cultivation of
anaerobic germs, it is only necessary to say that hydrogen, nitrogen,
or carbonic acid gas may be used in place of oxygen, or they may

Fn:. 7. Method oi produciirj Hydrogen by Klpp's Apparatus fur Cult i . .
Amu robot (a B p. 1 17.

be grown in a medium containing Borne Bubstance which will absorb
the oxygen (see p. 117).

Means of Sterilisation. — As this term occurs frequently even
in books of an elementary nature, and as it is expressive of an idea
which must always be present to the mind of the bacteriologist, it
may be desirable to make allusion to it here.

Chemical substances, perfect filtration, and heat are the three
means at- our command in order to secure germ-free conditions of
apparatus or medium. The first two, though t heoretically admissible,
arc practically Beldom used, the former of the two because the
addition of chemical substances annuls or modifies the operation,
the latter of the two on account of the great practical difficulties in
securing efficiency. Hence in the investigations involved in
bacteriological research heat is the common Bterilising agent. A
sustained temperature of 70 C. (158° F.) will kill all bacilli; oven
58 C. will kill most kinds. Boiling at 100 C. (212 V.) for five

minutes will kill anthrax Bpore8, and for thirty to sixty minutes

will kill all bacilli and their Bpores. This difference in the thermal

24

THE BIOLOGY OF BACTERIA

death-point between bacilli and their spores enables the operator to
obtain what are called "pure cultures" of a desired bacillus from
its spores which may be present. For example, if a culture contains

spores of anthrax and is contaminated
with micrococci, heating to 70° C.
(158° F.) will kill all the micrococci,
but will not affect the spores of an-
thrax, which can then grow into a
pure culture of anthrax bacilli. Frac-
tional or discontinuous sterilisation
depends on the principle of heating
to the sterilising point for bacilli (say
70° C.) on one day, which will kill the
bacilli, but leave the spores uninjured.
But by the following day the spores
will have germinated into bacilli, and
a second heating to 70° C. will kill
them before they in their turn have
had time to sporulate. Thus the
whole will be sterilised, though at a
temperature below boiling.

Successful sterilisation, therefore,
depends upon killing both bacteria
and their spores, and nothing short
of that can be considered as sterilisa-
tion. The following methods are
those generally used in the laboratory.
For dry heat (which is never so in-
jurious to organisms as moist heat*) :
(a) the Bunsen burner, in the flame of
which platinum needles, etc., are steril-
ised; (b) hot-air chamber, in which
flasks and test-tubes are heated to a temperature of 150°-170° C.
for an hour or more. For moist heat : (c) boiling, for knives and

Fig. 8.— Koch's Steam Steriliser.

* It will be observed that there is a marked difference between the effects of dry
heat and moist heat. Moist heat is able to kill organisms much more readily than
dry, owing to its penetrating effect on the capsule of the bacillus. Dry heat at
140° C. (284° F.), maintained for three hours, is necessary to kill the resistant spores
of Bacillus anthracis and B. subtilis, but moist heat at forty degrees less will have the
same effect. It is from data such as these that in laboratories and in disinfecting
apparatus moist heat is invariably preferred to dry heat. For with the latter such
high temperatures would be required that the articles being disinfected would be
damaged. Koch states the following figures for general guidance : Dry heat at a
temperature of 120° C. (248° F.) will destroy spores of mould fungi, micrococci, and
bacilli in the absence of their spores ; for the spores of bacilli 140° C. (284° F.),
maintained for three hours, is necessary; moist heat at 100° C. (212° F.) for fifteen
minutes will kill bacilli and their spores.

PLATE 2.

MODES OF BACTERIAL ACTION 25

instruments; (d) Koch's steam steriliser, by means of which a
crate is slung in a metal cylinder, at the bottom of which water is
boiled ; (e) the autoclave, which is the most rapid and effective
of all the methods. This is in reality a Koch steriliser, but with
apparatus for obtaining high pressure. The last two (d, e) are used
for sterilising the nutrient media upon which bacteria are culti-
vated outside the body. Blood serum would, however, coagulate
at a temperature over 60° C. (124° F.), and hence a special steriliser
has been designed to carry out fractional sterilisation daily for a
week at about 55° C.-580 C.

Modes of Bacterial Action

In considering the specific action of micro-organisms, it is desir-
able, in the first p]ace, to remember the two great functional divisions
of saprophyte and parasite. A saprophyte is an organism that
obtains its nutrition from dead organic matter. Its services, of
whatever nature, lie outside the (issues i>\' living animals. Its life
is spent apart from a " host/' A parasite, on the other hand, lives
always at the expense of some other organism which is its host, in
which it Lives or upon which it lives. There is a third or inter-
mediate group, known as " facultative," owing t«> their ability to act
as parasites or saprophytes, as tin* exigencies of their life may
demand.

The saprophytic organisms are, generally speaking, those which
contribute most to the benefit of man, and the parasitic the reverse,
though this statement is only approximately trua In their relation
to the processes of fermentation, decomposition, nitrification, etc., we
shall see how great and invaluable is tin- work which saprophytic
microbes perform Their result depends, in nearly all cases, upon the
organic chemical constitution of tin. Bubetances upon which they are
exerting their action, as well as upon tin- varieties of bacteria them-
selves. Nor must ii be understood that the action «•!' saprophytes is
wholly that <'t' breaking down and decomposition. As a matter of
tad. Borne of their work is, as we -hall Bee, "t" a constructive nature:
but, of whichever kind it is, the result depends upon the organism and
its environment,. This, too, may be said of the pathogenic species,
all of which are in a greater or less degree parasitic. It is well
known how various are the constitutions of man. how the bodies of
some persons are more resistant than those of others, and bow the
invading microbe will meet with a different reception according to the
constitution and idiosyncrasy of the body which it attacks. Indeed,
even after invasion the infectivity of the special disease, whatever it
happens to be, v. ill lie materially modified by the tissue-. When we
come i" bum to the micro-organisms which are pathogenic parasites

26 THE BIOLOGY OF BACTERIA

we shall further have to keep clear in our minds that their action is
complex, and not simple. In the first place, we have an infection of
the body due to the bacteria themselves. It may be a general and
widespread infection, as in anthrax, where the bacilli pass, in the
blood or lymph current, to each and every part of the body ; or it
may be a comparatively local one, as in diphtheria, where the invader
remains localised at the site of entrance. But, be that as it may, the
micro-organisms themselves, by their own bodily presence, set up
changes and perform functions which may have far-reaching effects,
It is obvious that the wider the distribution the wider is the
area of tissue change, and vice versa. Yet there is something of far
greater importance than the mere presence of bacteria in human or
animal tissues, for the secondary action of disease-producing germs —
and possibly it is present in other bacteria — is due to their poisonous
products, or toxines, as they have been termed. These may be of the
nature of ferments, and they become diffused throughout the body,
whether the bacteria themselves occur locally or generally. They
may bring about very slight and even imperceptible changes during
the course of the disease, or they may kill the patient in a few hours.
Latterly bacteriologists have come to understand that it is not so
much the presence of organisms which is injurious to man and other
animals as it is their products, which cause mischief ; and the
amount of toxic product bears no known proportion to the degree of
invasion by the bacteria. The various and widely differing modes of
action in bacteria are therefore dependent upon these three elements
(1) the tissues or medium, (2) the bacteria or agents, and (3) the
products of the bacteria or toxins ; and in all organisms! processes
these three elements act and react upon each other.

Seed and Soil. — It is of essential importance to the right under-
standing of the role which bacteria play in the production of disease
to give full place to the part taken by the soil on which they are
implanted. Few ideas in bacteriology are more erroneous, or likely
to lead to graver misconception, than to suppose that bacteria
produce the same effect under all conditions, and that the human
tissues play a small part. One might equally well expect seed to
behave in the same way in all kinds of soil. We know that as a
fact, seeds only flourish under certain conditions, and that the soil
is only second in importance to the seed-life itself. It is somewhat
the same in the production of disease. The early school of pre-
ventive medicine declared for the health of the individual and laid
the emphasis upon predisposition ; the modern school have declared
for the infecting agent, and have laid emphasis upon the bacillus.
The truth is to be found in a right perception of the action and
interaction of the tissues and the bacillus. B. diphtherial in one
person's throat (A) sets up diphtheria, in another person's throat

SEED AND SOIL 27

(B) lies quiescent, producing no apparent disease. The cause of this
extraordinary fact may be a question of different virulence in the
two bacilli, but is much more likely to be due to the greater vigour
and power of resistance of the mucous membrane of B's throat.
Sewer air, as we shall see subsequently, does not contain many
bacteria, and probably does not frequently convey germs of disease.
But this does not prove that the inhalation of sewer air will not
weaken the throat, and so form a favourable nidus for organisms
resting there, or organisms shortly to be inhaled from dust or mucous
particles from the throat of a diseased person. Which is the more
important preventive method, to maintain the resistance of the
individual or to waylay the infecting organism, is a nice point we need
not attempt to decide. Obviously, both objects should be kept in
view. Phthisis is another example. Thousands of persons inhale
the tubercle bacillus who are not attacked by the clinical disease of
consumption. This fortunate result is due to the resistant tissues
of the healthy lung, and the lesson to be derived is to maintain such
resistance at its maximum. This evidently is, in part, the scientific
explanation of Koch's dictum, " It is the overcrowded dwellings of
the poor that we have to regard as the real breeding places of
tuberculosis; it is out of them that the disease always crops up anew,
and it is to the abolition of these conditions that we must first and
foremost direct our attention if we wish to attack the evil at its root
and to wage war against it with effective weapons."* Part of the
explanation of these words is doubtless that it is in such places that
the tubercle bacillus breeds and passes from one person to another.
But every sanitarian knows that the effect of such environment is
to lower the natural resistance, to weaken the lung, Impoverish the
blood, and undermine the constitution, and thus a suitable nidus is
supplied to the invading bacillus. '-'A perfectly healthy lung is
seldom if ever primarily infected with the tubercle bacillus" (Wood-
head).

But the evidence of bacteriology as to the part played by the soil
is even stronger than at first sight appears. For we now know, by
experiment, that micro-organisms which in Borne animals produce
acute disease rapidly ending in death, result only in mild disease in
other animals, and in yet a third group produce no apparent disease
whatever. This is not d\w to variation in virulence but t<» variation
in soil.

Tin; advance of bacteriology has been so rapid and marked by
such striking discoveries that there has been a tendency to over-rate
altogether the potentiality of the bacillus apart from its medium.
The latest findings in the study of comparative culture work, of
immunity and of the production of antitoxins have, however, demon-
Trans. Brit, Con;/, of Tuberculosis^ 1901, vol. i., p. 31.

28 THE BIOLOGY OF BACTERIA

strated beyond all doubt the enormous part played by the medium
or soil in which the micro-organism is growing.*

Specificity of Bacteria

A species may be defined as a group of individuals which, however
many characters they share with other individuals, agree in present-
ing one or more characters of a peculiar and hereditary kind with
some certain degree of distinctness.^ There is no doubt that separate
species of bacteria occur and tend to remain as separate species.
But it must be remembered that species are merely arbitrary divisions
which present no deeper significance from a philosophical point of
view than is presented by well-marked varieties, out of which
they are in all cases believed to have arisen, and from which it is
often a matter of individual opinion whether they shall be
separated by receiving a specific label. What degree or character
of variation shall be considered as sufficient for the demarcation of
a species of bacteria ? B. coli and B. typhosus have certain distinctive
features, which are accepted as factors of provisional differentiation.
But they have many points in common, the peculiarity and heredity
of which are not as yet determined. And they have many allies,
para-typhoid and para-colon organisms, in the same way as the
tubercle bacillus possesses many allies, both bovine and human, among
acid-fast species having similar characters but differing in degree of
virulence. The fact is, that our present knowledge of these matters
is very small, and it is impossible to dogmatise. The future may
reveal some unlooked-for relationships, and organisms now classified
as morphologically separate may ultimately prove to be nearly
related. Further, it may be found that their respective action
in the human body is not greatly dissimilar (the production,
of diarrhoea, for example, by the colon group). Medium and
tissue have their effect in the production of variations of greater or
lesser mark in bacteria. B. typhosus may, in the course of sub-
culture, become morphologically indistinguishable from B. coli, and
its pathogenicity may also be reduced. The tubercle bacillus in old
culture and in saprophytic existence becomes almost indistinguish-
able from the streptothrix family. Streptococcus conglomerate on
certain media simulates in a marked degree the Klebs-Loffler diph-
theria bacillus, and by passage through a mouse loses its streptococcal

* The writer has been impressed in particular as to the truth of this view by
observation of a number of epidemics, by the study of a long series of cultures of
the same bacillus on different media, and by antitoxin production. But the same
conclusion has been reached from other premises. See a suggestive paper by Sir
W. J. Collins, M.D., in the Jour, of the Sanitary Institute 1902 (Oct.), xxiii., pt. iii.,
p. 335.

t Darwin and After Darwin, G. J. Romanes, F. R.S., vol. ii., p. 231.

ASSOCIATION OF ORGANISMS 29

form (Gordon). The Klebs-Loffler bacillus in its turn may be
greatly modified in morphology and pathogenicity by environment.
Nor is the change necessarily in descending order. Non-pathogenic
organisms may possibly become pathogenic. We do not know.
The subject is one full of difficulty in a transition period of knowledge
in any branch of science. But there is no reason to suppose that
bacteria are exceptional in nature and outside the influence of
natural selection ; and it is not improbable that the views of the early
bacteriologists will have to be very much revised, and that eventually
it will be found that many "species" of micro-organisms are in
reality varieties of a single species showing involution and pleomorphic
forms. At the same time it should be recognised that amongst the
lowliest forms of life specific distinctions are, as a rule, less definite,
and less permanent, than amongst forms of life much higher in the
organic scale.

The Association of Organisms

At a later stage we shall have an opportunity of discussing
Symbiosis and allied conditions. Here it is only necessary to draw
attention to a fact that is rapidly becoming of the first importance
in bacteriology. When species were first isolated in pure culture it
was found that they behaved very differently under varying
circumstances. This modification in function lias been attributed to
differences of environment and physical conditions. Whilst it is true
that such external conditions must have a marked effect upon such
sensitive units of protoplasm as bacteria, it has recently been
proved that one great reason why modification occurs in pure
artificial cultures is that the species has been isolated from amongst
its colleagues and doomed to a separate existence. One of the most
abstruse problems in the immediate future of the science of bacteri-
ology is to learn what intrinsic characters there are in Bpecies or
individuals which act as a basis for the association of organisms for a
speeilie purpose. Some, bacteria appear to be unable to perform their
ordinary rdle without the aid of others.* An example of such
association is well illustrated in the case of Tetanus, for it has been
shown thai if the bacilli and spores of tetanus alone obtain entrance
to a wound the disease dues not follow the same course as when with
the speeilie organism the lactic acid bacillus or the common
organisms of Buppuration or putrefaction also gain entrance. There
is here evidently something gained by association. Again the viru-

* The three different degrees of association have been expressed by the following
terms: Symbiorit, the co-operation for a mutual advantage, not obtained other-
wise; metabiosis, where one organism prepares the way for another; antibiosis
[antagonism of bacteria), where one of the two associated organisms is directly or

indirectly injuring the other.

30 THE BIOLOGY OF BACTERIA

lence of other bacteria is also increased by means of association.
The Bacillus coli is an example, for, in conjunction with other
organisms, this bacillus, although normally present in health in the
alimentary canal, is able to set up acute intestinal irritation, and
various changes in the body of an inflammatory nature. It is not
yet possible to say in what way or to what degree the association of
bacteria influences their role. That is a problem for the future. But
whilst we have examples of this association in Streptococcus and the
bacillus of diphtheria, B. coli and yeasts, Tetanus and putrefactive
bacteria, Diplococcus pneumoniae and Proteus vulgaris, and Streptococcus
erysipelatis and Proteus vulgaris, we cannot doubt that there is an
explanation to be found of many, hitherto unknown, results of
bacterial action. This is the place in which mention should also be
made of higher organisms associated for a specific purpose with
bacteria. There is some evidence to support the belief that sonic of
the LeptotricheaB (Crenothrix, Beggiatoa, Leptothrix, etc.) and the
Cladotricheae (Cladothrix) perform a preliminary disintegration of
organic matter before the decomposing bacteria commence their
labours. This occurs apparently in the self -purification of rivers, as
well as in polluted soils.

Antagonism of Bacteria (Antibiosis).— Study of the life-
history of many of the water bacteria will reveal the fact that they
can live and multiply under conditions which would at once prove
fatal to other species. Some of these water organisms can indeed
increase and multiply in distilled water, whereas it is known that
other species cannot even live in distilled water owing to the lack of
pabulum. Thus we see that what is favourable for one species may
be the reverse for another.

Further, we shall have opportunity of observing, when consider-
ing the bacteriology of water and sewage, that there is in these
media in nature a keen struggle for the survival of the fittest
bacteria for each special medium. In a carcase it is the same. If
saprophytic bacteria are present with pathogenic, there is a struggle
for the survival of the latter. Now whilst this is in part due to a
competition owing to a limited food supply and an unlimited popula-
tion, as occurs- in other spheres, it is also due in part to the inimical
influence of the chemical products of the one species upon the life of
the bacteria of the other species. Moreover, in one culture medium,
as Cast has pointed out, two species will often not grow. When
Pasteur found that exposure to air attenuated his cultures, he
pointed out that it was not the aiv per se that hindered growth, but it
was the introduction of other species which competed with the
original. The growth of the spirillum of cholera is opposed by
Bacillus pyogenes fmtidus. B. anthracis is, in the body of animals,
opposed by either B. pyocyancus or Streptococcus erysipelatis, and yet

ANTAGONISM AND ATTENUATION 31

it is aided in its growth by B. prodigiosus. B. accti is under certain
circumstances antagonistic to B. coli.

In several of the reports of the late Sir Kichard Thorne issued
from the Medical Department of the Local Government Board, we
have the record of a series of experiments performed by Dr Klein
upon the subject of the antagonisms of microbes. From this work it
is clearly demonstrated that whatever opposition one species affords
to another it is able to exercise by means of its poisonous properties.
These are of two kinds. There is, as is now widely known, the
poisonous product named the toxin, into which we shall have to
inquire in more detail at a later stage. There is also in many
species, as several workers have pointed out, a poisonous constituent,
or constituents, included in the body protoplasm of the bacillus, and
which he therefore terms the intracellular poison. Now, whilst the
former is different in every species, the latter may be a property
common to several species. Hence those having a similar intracellu-
lar poison are antagonistic to each other, each member of such a
group being unable to live in an environment of its own intracellular
poison. Further, it has been suggested that there are organisms
possessing only one poisonous property, namely, their toxin — for
example, the bacilli of Tetanus and Diphtheria — whilst there are
other species, as above, possessing .1 double poisonous property, an
intracellular poison and a toxin. In this latter class would be
included the bacilli of Anthrax and Tubercle.

There can be no doubt that these complex biological properties
of association and antagonism, as well as the parasitic growth of
bacteria upon higher vegetables, arc ;is yet little understood, and wo
may be glad thai any light is being Bhed upon them. In the
biological study of soil bacteria in particular may we expect in the
future to find examples of associal ion, even as already there are signs
that this is so in certain pathogenic conditiona In the alimentary
canal, on the <>thor hand, and in conditions where organic matter is
greatly predominating, wo may expect to see further light on the
subjecl of antagonism.

Attenuation of Virulence or Function. — It wis pointed out,
by some of the pioneer bacteriologists that the function of bacteria
Buffered under certain circumstances a marked diminution in power.
Later workers found that such a change might be artificially pro-
duced. Pasteur introduced the firs! method, which was the simple
one of allowing cultures to grow old before sub^ulturing. Obviously
a pure culture cannot last for ever. To maintain the species in
characteristic condition it is accessary frequently t<» sub-culture upon
fresh media. If this Bimple operation be postponed as long as
possible consistent with vitality and then performed, it will be found
that the snb-culture is nit, nuaU <l , i.<\, weakened. Another mode is

32 THE BIOLOGY OF BACTERIA

to raise the pure culture to a temperature approaching its thermal
death-point. A third way of securing the same end is to place it
under disadvantageous external circumstances, for example in a too
alkaline or too acid medium. A fourth method is to pass it through
the tissues of an insusceptible animal. Thus we see that, whilst
the favourable conditions which we have considered afford full
scope for the growth and performance of functions of bacteria, we are
able by a partial withdrawal of these, short of that ending fatally,
to modify the character and strength of bacteria. In future chapters
we shall have opportunity of observing what can be done in this
direction.

Bacterial Diseases of Plants

Eeference has been made to the associated work of higher
vegetable life and bacteria. The converse is also true. Just as
we have bacterial diseases affecting man and animals, so also plant
life has its bacterial diseases. Wakker, Prillieux, Erwin Smith,
and others have investigated the pathogenic conditions of plants
due to bacteria, and though this branch of the science is in its
very early stages, many facts have been learned. Hyacinth disease
is due to a flagellated bacillus. The wilt of cucumbers and pumpJdns
is a common disease in some districts of the world, and may
cause widespread injury. It is caused by a micro-organism
which fills the water-ducts. Wilting vines are full of the same
sticky germs. Desiccation and sunlight have a strong prejudicial
effect upon these organisms. Melon blight must not be confused
with the bacterial wilt of cucumbers and melons. The blight disease
is caused by Plasmopara cubensis, a sporulating fungus. Bacterial
brown-rot of potatoes and tomatoes is another plant disease probably
due to a bacillus. The bacillus passes down the interior of the
stem into the tubers, and brown-rots them from within. There is
another form of brown-rot which affects cabbages. It blackens the
veins of the leaves, and a woody ring which is formed in the stem
causes the leaves to fall off. This also is due to a micro-organism,
which gains entrance through the water -pores of the leaf, and
subsequently passes into the vessels of the plants. It multiplies
by simple fission, and possesses a flagellum. Certain diseases of
Sweet Corn have been investigated by Stewart, and traced to a
causal bacillus possessing marked characters. Professor Potter
believes that white-rot of the turnip is produced by Pseudomonas
destructans, a liquefying, motile, aerobic bacillus.

CHAPTEE II

BACTERIA IN WATER

Quantity of Bacteria in Water — Quality of Water Bacteria : (a) Ordinary Water
Bacteria ; (/>) Sewage Bacteria ; B. coli communis ; (c) Pathogenic Bacteria in
Water — Interpretation of the Findings of Bacteriology — Natural Purification
of Water — Artificial Purification of Water — Sand Filtration — Domestic Puri-
fication of Water.

The collection of samples, though it appeals simple enough, is
sometimes a difficult and responsible undertaking. Complicated
apparatus is rarely necessary, and fallacies will generally be avoided
by observing two directions. In the first place, the sample should
be chosen as representative as possible of the real water or conditions
we wish to examine. Some authorities advise that it is necessary
to allow the tap to run for some minutes previously to collecting
the sample; but if we desire to examine chemically for lead or
biologically for micro-organisms in the pipes, then Buch a proceeding
would be injudicious.* If it is well water thai is to be examined,
the well should be pumped for some minutes before taking the
sample. If it is river water which is to be examined, it is important
to collect the sample without incorporating any deposit. In short,
we must use common sense in the selection and obtaining of a
sample, following this one guide, namely, to collect as nearly as
possible a sample of the exact water, the quality of which it is
desired to learn. In the second place, we must observe strict

"Water from a house cistern is rarely a fair sample of a town supply. It
should be taken from the main. If taken from a stream or still water, the collect-
ing bottle should be held about a foot below the surface before the stopper is
removed.

33 c

34 BACTERIA IN WATER

bacteriological cleanliness in all our manipulations. This means that
we must use only sterilised vessels or flasks for collecting the sample,
and in the manipulation required we must be extremely careful to
avoid any pollution of air or any addition to the organisms of the
water from unsterilised apparatus. A flask polluted in only the
most infinitesimal degree will entirely vitiate all results. Vessels
may be sterilised by heating at 150° C. for two or three hours. If
this is impracticable the vessel may be washed with pure sulphuric
acid, and then thoroughly rinsed out in the water which is to be
examined.

Accompanying the sample should be a more or less full statement
of its source. There can be no doubt that, in addition to a chemical
and bacteriological report of a water, there should also be made a
careful examination of its source. This may appear to take the
bacteriologist far afield, but until he has seen for himself what " the
gathering ground " is like, and from what sources come the feeding
streams, he cannot judge the water as fairly as he should be able to
do. The configuration of the gathering ground, its subsoil, its
geology, its rainfall, its relation to the slopes which it drains, the
nature of its surface, the course of its feeders, and the absence or
presence of cultivated areas, of roads, of houses, of farms, of human
traffic, of cattle and sheep— all these points should be noted, and
their influence, direct or indirect upon the water, carefully borne in
mind.

When the sample has been duly collected, sealed, and a label
affixed bearing the date, time, and conditions of collection and full
address, it should be transmitted with the least possible delay to the
laboratory. Frequently it is desirable to pack the bottles in a small
ice -case for transit. Miquel, Pakes, and others have constructed
special forms of packing-cases, and these have their advantages.
But the ordinary bottle of water may be quite satisfactorily con-
veyed, as a rule, packed in sawdust and ice. On receipt of such a
sample of water the examination must be immediately proceeded
with, in order to avoid, as far as possible, the fallacies arising from
the rapid multiplication of germs.

Multiplication of Bacteria in Water. — In almost pure water, at the
ordinary temperature of a room, Frankland found that organisms
multiplied as follows :

No. of Germs.
Hours. per c.c.

0 1,073

6 6,028

24 7,262

48 48,100

Another series of observations revealed the same sort of rapid

MULTIPLICATION OF WATER BACTERIA 35

increase of bacteria. On the date of collection the micro-organisms
per c.c. in a deep-well water (in April) were seven. After one day's
standing at room temperature the number had reached twenty-one
per c.c. After three days under the same conditions it was 495,000
per c.c. At blood-heat the increase would, of course, be much greater,
as a higher temperature is more favourable to multiplication. But
this would depend in part also upon the degree of impurity in the
water, a pure water decreasing in number of germs on account of the
exhaustion of the pabulum, whereas, for the first few days at all
events, an organically polluted water would show an enormous
increase in bacteria.

It is desirable to remember that organisms, in an ordinary water,
do not continue to increase indefinitely. Cramer, of Zurich, examined
the water of the Lake after it had been standing in a vessel for
different periods, with the following results: —

No. of Micro-organisms
per cc

143

12,457
328,543
283,452

17,486

"00

In a general way it may be said that foul waters, rich in putrescible
animal matter, show a rapid increase of bacteria; surface waters, such
as river water, show a slow and persistent multiplication of organisms ;
and deep-well waters and spring water show comparatively little*
increase in contained bacteria. [ndeed it may be said thai the
condition of a water is partly indicated by the rapidity or Blowness
with which its bacteria increase, A low temperature (5° C.)
undoubtedly diminishes the multiplication, and there arc other
conditions such as exposure to air, movement, and antagonism of
organisms which exerl an indirect effect As will be inferred from
what has been Baid, the most important condition affecting the
number of bacteria in a water is the organic matter contained in it.*

The Bacteriology of Water

In many natural waters there will be found varied contents even
in regard to flora alone: algas, diatoms, sjpirogyrcB, desmids, and all

* For suggestions and hints on points of technique in the systematic examination
of a water, see 1 )elepine*s Bacteriological Survey ofSurfaci Water Supplies : .lour, of
Stale Medicine* 1898, vol \i., pp 145,198,241,289? and Bacteriological Examina-
tion of Water, by W. II. Eiorrocks. (See also presenl volume, pp. 463-473.)

Hours

and Days

of

Exa

minatfon.

0

hours

'24

»*

3

days

8

,.

17

,,

70

»,

36 BACTERIA IN WATER

sorts of vegetable detritus. Many of these organisms are held
responsible for certain disagreeable tastes and odours. The colour
of a water may also be due to similar causes. Dr Garrett, of
Cheltenham, has recorded the occurrence of redness of water owing
to a growth of Crenothrix polyspora, and many other similar cases
make it evident that not unfrequently great changes may be produced
in water by contained microscopic vegetation.

With the exception of water from springs and deep wells, all
unfiltered natural waters contain numbers of bacteria.* The actual
number roughly depends, as we have seen, upon the amount of
organic pabulum present, and upon certain physical conditions of
the water. In some species multiplication does not appear to
depend on the presence of much organic matter, and, indeed, some
bacteria can live and multiply in almost pure water ; e.g., Micrococcus
aquatilis and Bacillus crythrosporus. Again, others depend not upon
the quantity of organic matter, but upon its quality. And frequently
in a water of a high degree of organic pollution it will be found that
bacteria have been restrained in their development by the competi-
tion of other species monopolising the pabulum. It will be necessary
to deal with the subject under the two subdivisions of (1) quantity
and (2) quality of bacteria found in water.

Quantity of Bacteria in Water

Percy Frankland has quoted in his book*)* a number of records
of the quantity of organisms found in various waters. These tables
give the returns for the rivers Seine (Miquel), Rhone, Saone (Eoux),
Spree (Frank), Isar (Prausnitz), Limmat (Schlatter), lihine (Moers),
etc. Here it is unnecessary to do more than give typical illustra-
tions, and for comparative purposes English rivers may be taken.
Prof. Frankland himself collected water from the river Thames
at various times and seasons, and some of his results were as
follow : —

* Bacteria, of course, exist in the water of the sea. Near land, as might be
expected, the number is greatest, and diminishes rapidly further out to sea.
Currents sometimes bring them to the surface from a depth of 596 fathoms
(Fischer). At a depth of 100-200 fathoms bacteria have been found in large
numbers. The comparative paucity at the surface is due to the germicidal effect
of sunlight. Ocean bacteria vary widely in size and shape. Apparently, typical
cocci and bacilli are never met with on the high seas. Spirilla and zooglcca
masses are common. Most sea bacteria are motile and furnished with flagella ;
some are anaerobes.

t Micro-organisms in Water (1894), pp. 89-116.

NUMBER OF BACTERIA IN WATER

37

River Thames Water Collected at Hampk

Number of Micro-organisms obtained from 1 c.c. of Water

Month.

1886.

1887.

1888.

January

45,000

30,800

92,000

February

15,800

6,700

40,000

March

11,415

30,900

66,000

April .
May .

12,250

52,100

13,000

4,800

2,100

1,900

June .

8,300

2,200

3,500

July .

3,000

2,500

1,070

August

6,100

7,200

3,000

September

8,400

16,700

1,740

October

8,600

6,700

1,130

November

56,000

81,000

11,700

December

63,000

19,000

10,600

Another example from the river Lea was as follow a :-

River Lea Water Collected at Chitlgford.
Number of Micro-organisms obtained from i c.c ol Water.

Month.

1886.

1887.

1888.

January

39,300

37,700

31,000

February

20,600

7,900

March

9,025

2 l ,000

£rpril '

7,800

1,880

84,000

May .

2,050

2,200

1,124

June .

1,700

12,200

7,000

July .

.

12,800

2,100

Augusl

1,300

,800

2,000

September

3,700

0,200

1,670

October

6,400

2,810

November

12,700

27,000

December

] • 1 ,000

11,000

1,400

" During the summer months these waters are purest as regards
micro-organisms, this being due bo the facl that during dry weather
these rivers are mainly composed of spring water, whilst ;ii other
seasons they receive tin- washings of much cultivated land " ( Frank-
land). ^ Prausnitz lias shown that water differs, as would be expected,
according fc0 the locality in the stream at which examinatioD is made
His investigations were made from tin- river [sar before and after it
receives the drainage of Munich: —

Above Munich

Near entrance of principal sewer

! 8 kilometres from Munich .

22 „ ...

N". 'if ( felonies
per c.c.

227,81
9,111

I.,

38

BACTERIA IN WATER

Frankland has shown that the river Dee affords another example,
even' more perfect, of pollution and restoration repeated several times
until the river becomes almost bacterially pure.

Professor Boyce and his colleagues have recently made an
examination of the river Severn before and after its waters pass the
town of Shrewsbury.* Their findings may be represented briefly as
follows : —

Average Total

Average

Position of Examination.

No. of Bacteria

No. of B. coli

per c.c.

per c.c.

At Asylum, 2 miles above Shrewsbury

7,000

13

,, Waterworks, opposite Shrewsbury .

13,000

46

,, Ferry i., 0'6 of a mile lower down .

20,000

177

,, English Bridge, 1*6 of a mile lower down

23,000

321

,, Ferry iii., 2*5 miles lower down

19,000

600

,, Uffington, 4*7 miles lower down

17,000

142

,, Alcham, 9 miles lower down .

13,000

48

,, Cressage, 16 miles lower down

5,000

36

This table and that of Prausnitz — and many other workers have
produced similar records — illustrate the effect of (a) local pollution,
and (b) river purification, upon the bacterial content of water, to
which subsequent reference will be made. The record respecting the
Severn includes also the indication of sewage pollution by the
presence of B. coli. An elaborate examination has also been made
of. the water of the river Thames and the Thames estuary, by
Houston, and the report dealing with it is full of information on
the subject, to which reference should be made.f

Lastly, the accompanying table (pp. 39 and 40), for 1902 and 1903,
deals with the London water supply as examined by Crookes and
Dewar. \ It is concerned, it should be added, only with numerical
results.

This record, compiled from the monthly reports respecting the three
waters supplied to the metropolis, illustrates many interesting points
upon which we have not space to dwell fully. A few notes, however,
upon an actual example are more useful than much theoretical
information, and therefore a brief study of these figures may be made.
In the main the table illustrates two points more clearly than the
preceding tables. The first is the effect of filtration, and the second
is the effect of season, upon the number of bacteria in water. In
respect to the former, comment is needless. It is only necessary to

* Royal Commission on Sewage Disposal, Second Report, 1902, p. 99.
t Ibid, Fourth Report, 1904, vol. iii., pp. 1-75.
t Metropolitan Water Supply, 1902 and 1903.

BACTERIA IN LONDON WATER

39

!>.

\a>

iO

Oi

CO

o

CD

o

U

i^

CM

OO

co

o

CM

*o

03

CO

GO

CO

CO

ft

IO

i-~

CD

CM

lO

CM

o

CO

CO

>

lr^

i-H

CD

CO

»o

o

rH

o

CM

CM

^

CM

fc

CM

■>#

i-H

CO

o

IO

o

CO

t^

4J

i—i

T-H

«#

CM

CD

"cH

l-H

O

CO

cq

CM

CM

o

00

a>

H

CO

«*H

rH

0O

X.^

o

*a

O

ft

«D

i-H

Oi

(M

Oi

o

CM

CO

J-H

-«*

i-H

CM

03

M

<m*

•4J

o3

tH

CO

-*

rH

CO

o

CO

CO

Pi

to

o

rH

CO

CM

Oi

CO

CO

O

■s

o

<1

CO

T-H

1—1

CM

J

fi

CM

^

CM

-*

CM

o

o

"*

•r-l

>»

OS

l^

rH

l-H

>~

tK

03

3

*3

(N

T-H

CM

CM

CM

|

+3

03

•

CM

CO

CO

CM

OO

o

-^

»o

s

i—i

T— 1

CO

CO

CO

CD

CO

cj

2

co

CO

IO

Tfl

o

1-5

"*

2

o

CO

CM

CD

I-H

CO

o

rH

x^

Oi

i-H

Oi

CO

Oi

l^

CO

03

ft

ri

00

CO

CO

03

f*<

oi

£

w

"B

•

OI

Oi

M

l-H

CD

o

1 -

CD

&

•rj

GO

1 ~

CO

Oi

■*

CM

&

pj

CM

CD

th

'-.

o

«4

O

s

l-H

Jg

CO

CO

CO

CO

CM

o

o

CM

<t

3

CO

CO

©

•M

Oi
M

I ~

w

Oi

Oi

I-

O

o

CO

rQ

at

N

M

o

l-H

CO

CO

n

ri

T— 1

' ;

CD

O

co

o

CO

i -

fl

CO

i— i

Oi

co

CO

3

IN

CO

n

71

P

3

B

/-

C ^

^X
I- u
qj «

til

► w

W CM

> o

CU

«4H .

o

S?

u c

S 2

w n

M

«C CO

Oh"
. '--
> 1 -
1

C*H .

O
Sic

§ c

v a

-t-> **

13 CO

= 1-

SB*'

B o a«»
o pjj
£ =o w ex

h T s §

*£~-

~ S « O

W3 ^ CO

c ^ •— ■— ■

2 7 ~

. | 8 = •

ri D o .

- r ^ co

■2 ^

~ | = 0

> 1 ~ 5 "

1 jj v— ' 03

ft % CO r-

c ^
P c
o

•~T

■*->

ri

s «-
-- ^.

- s
IS
1j

Y «
to

= -5

1

X 'H-l

o o
a a

- BQ ^

T;.- ^

I?

^ s

1 °

to •. O
- "•

W E =

C- cu •

11-5

O g •

£ c?
£■§ -^

0j - S ~

§•§ f a

a 53 3

»3^s

V.

S5

H

H

H

H

^v->

a

40

BACTERIA IN WATER

r-H

o

CD

t^

co

o

CO

C30

o

<X)

"<*

rH

CO

i>-

CM

iS

OO

CM
CM

co

CM

VO

lO

o

CO

CM

o

o

oo

t*

■>#

CM

Oi

T*

J>.

1 — 1

OJ

o

oo

co

CO

CO

(23

CO

co

O

1— <

co

CO

o

o

iro

<*

CO

t^

CO

>o

w

CN

O

oo

»

O

**

CO

CO

o

r-i

B

CO

Oi

CM

Oi

CO

o

o

Oi

CO

r-i

o

CO

Oi

CO

01

co

CM

-*■

co

9

CO

CO

+3

c3

?

o

o

o

co

01

o

09

DQ

pj

tc

1-

T-H

-#

1^.

oo

■:

O

(N

Oi

CO

t3

■q

a

©

o

►5

oo

,_,

KJ

CO

o

o

,»,

CO

i^

i rs

rH

o

<N

IO

CO

CO

0)

3

CN

oo

CM

oq

u

a

t-3

00

-*-=

a

d

r— I

1^

i.^

OI

oo

3

o*

O0

CO

Ttl

0O

l>.

2

§

t— 1

CO

>.~

^<

o

1-3

©

IS

1—1

2

O

t— 1

iH

UO

Oi

GO

o

I— 1

CO

u

J?

CO

F— 1

01

1 ".

CM

5v

CM

CO

»— 1

::

1

s

■*r

*3

^4

OS

CO

o

O

CO

o

EC

o

c3

"2

Tf<

01

CO

-.

OI

tc

00

o

«J

of

6

Ifl

, — 1

!>.

l-

CM

o

CM

o.

£

CO

r-H

CO

CM

T— 1

oo

oq

co

co

CN

■ -.

S

oT

KO

00

CO

co

o

o

-r

o

rQ

O

o

-H

i.'O

I— 1

£

-^

CO
O

0^

CM

Oi

CO

Oi

CM

o

CO

CO

CO

CO

CO

rH

CO

aj

CO

o

CO

ha

CO
I— 1

S o

«+H .

=4H .

So ft^

£ co « a

- V- i ' .

- '-+- 1 - .

13

II

JJ tfl^

o

o

fflfT

g 3 cu

* « M ft

>- 5i £ s

►MS

^Sc '

.S Jj €3 ee
*- co ^3 cu

,a'cjH£*

1 w j_j CU

co "S co r-

§ £ "cu ft^:

«*H CO J Q.

s * .

ft cu •

£ £ S

!■§!

s cu cu

►^ 4-> JS

CO c ^ ft

^^3 i co

C co co

S3 S - '^

<

e

C
c

c
y

.

i

5

i

2

si

CU CO

±i ^

cS CO

3 a

fS

<U CO

-§o

£ o

fig

CU to
CX2 co

^"ft

£s

CO -M

CU c
CU y

IS

cu CO
go

-It §

CU !? > CO

rO £ CU O

.5 5- ca co

co Th co

=3T3 S •

u Sf <u 5

i ^> n

•S Js cc co
fa B n (i)
cu 1>EJ S

co ■£ co cu

§ « "Sb.22 ^

§53-5 ft^

CU-3
cu co

^ CU
cS CO

c «

3 CU

co co

CU CO

Jo

CM

> u

cu

cu

J3

, t-j

l~l

^c

fc

£

H

H

H

H

Qj

£

BACTERIA IN LONDON WATER 41

examine the returns to recognise the marked reduction in the number
of bacteria, in some cases amounting to 98 per cent., brought about
by filtration. In respect to the latter, the effect of season, some note
is required. It will be seen that during 1902 the figures are fairly
uniform throughout the year, showing, on the whole, a rise in winter
and spring, and a fall during summer. But in 1903 the returns show
wide variation which calls for explanation, which is as follows : —

The water supply in December 1902, on the whole, maintained an equal microbic
purity to that of November. This exceptional condition of the supply, the com-
paratively small number of bacteria, for the winter months is no doubt due to the
absence of floods in the Thames Valley, and to the unusually mild character of the
season. "It* the large deficit in the rainfall is made up," wrote Crookes and
Dewar, " no doubt there will be in the near future a period when the filtration of
the London waters will require more than usual care. As the general filter-beds
are, however, in good working order, we believe these difficulties, should they
arise, will be overcome satisfactorily.1'

The standard of general organic purity during 1902, as defined by chemical
methods, was maintained. As regards the month of December, the Thames-derived
Companies showed decided differences among themselves, which, as the supply
comes from the same source, were essentially connected with differences in storage
capacity and variations in the structure of the filter-beds, although the latter is of
less importance in removing soluble organic matter.

Crookes and Dewar add: "The longer our experience of the bacteriological
method, as applied to the analysis of the filtered supply, and the wider its application,
the more we are convinced of its primary importance as a safeguard to the public.
It enables us to define in a much more delicate way than is possible by chemical
analysis what is an efficiently filtered water, and thereby enables the chemist to
warn the engineer the moment any one of his filters show signs of defective
working. Whether the supply as regards the organic matter in solution varies
more or less according to the season of the year, is of relatively small moment
as compared with the knowledge th.it the microbic impurity is reduced to a
minimum.11

That was the position at the end of 1902. But at the turn of the year, owing to
the great increase in the rainfall, the microbes in the unfiltered Thames water rose
from about tiOOO to 13,000, that is, the bacteriological impurity about doubled,
whereas the unfiltered New River water underwent little or no alteration. The
result of this increase was that the filters of the Thames-derived Companies, which
were not working at their best, furnished a larger number of samples from the filter
wells, showing an increase in the number of bacteria which was the inevitable result
of an increased rainfall.

Things remained thus until April, when in comparison with the month of March
the bacteriological quality showed considerable improvement, a result which might
have been anticipated from the advent of summer, and the improved natural
conditions associated with vegetable growth; a state of things which generally
improves the quality of the water obtained from such collecting areas as the \ alleys
of the Thames and Lea. I Jut in June, when the number of bacteria ought to have
been low, as ordinarily there would be a small rainfall, an exception il condition of
things arose. The total excess of rainfall amounted to 19*9 per cent on the thirty
years' average, so that during the month of June actually 22*5 per cent., or an
amount approaching one-half oi' the previous excess, of rain fell in the valley of the
Thames. Such an amount of rain is altogether exception il in twenty -two years'
experience. 'The result was that the proportion of vegetable matter in solution,

and therefore the colour of the water, were both quite exceptional for the summer
months. Nevertheless, the general filtration was adequately and effectively
performed, as is shown by the bacteriological results. Similar conditions prevailed
in \ugust and October. The exceptional rainfall, which amounted to sixty per

42 BACTERIA IN WATER

cent, in excess of the average, kept the colour and amount of soluble vegetable
matter in solution abnormally high. In December, owing to the continued rains, the
New River and Thames unfiltered waters contained a maximum number of bacteria.
In dry weather the number per c.c. had been as low as 149 and 2013 respectively,
but owing to seasonal changes they had risen to 861 and 27,216 bacteria per c.c.
respectively.

From these various records we find that in the result the number
of bacteria in river water depends upon a variety of circumstances,
amongst which the most important direct conditions are four, namely,
(1) local pollution, (2) natural purification (to which subsequent
reference will be made), (3) season and rainfall, and (4) sedimentation
and filtration. Behind these direct conditions we have also seen that
time, temperature, light, exposure to air, and the presence of organic
matter play an essential part.

Bacteriological Examination of Water. — [See Appendix, p. 463.]

Quantitative Standard. — In arriving at a conclusion respecting the
number of organisms in a water and their bearing upon its suitability
for use, it should be remembered that a chemical report and a
bacteriological report are desirable before forming an opinion. The
former is able to tell us the quantity of salts and condition of the
organic matter present : the latter the number and quality of micro-
organisms. Neither can take the place of the other, and, generally
speaking, both are more or less useless until we can learn, by inspec-
tion and investigation of the source of the water, the origin of the
organic matter or contamination. Hence a water report should con-
tain not only a record of physical and microscopical characters, of
chemical constituents, and of the presence or absence of micro-
organisms, injurious and otherwise, but it should also contain infor-
mation obtained by personal investigation of the source. Only thus
can a reasonable opinion be expected. Moreover, it is generally only
possible to form an accurate judgment of a water by watching its
history ; that is to say, not from one examination only, but from a
series of observations. The writer has examined a certain water
supply for thirty-six consecutive months. In 1901 the average
number of bacteria per c.c. was 93, in 1902, 136, and in 1903, 57.
This shows a stable bacterial content which in itself is favourable.
A water yielding a steady standard of bacterial content is a much
more satisfactory water, from every point of view, than one which
is unstable, one month possessing 50 bacteria per c.c. and another
month 5000. It is obvious that rainfall and drought, soil and trade
effluents, time and temperature, will have their influence in materially
affecting the bacterial condition of a water.

Miquel and others have suggested standards which allow " very
pure water " to contain up to 100 micro-organisms per c.c. Pure
water must not contain more than 1000, and water containing up to

NUMERICAL STANDARDS 43

100,000 bacteria per c.c. is contaminated with surface water or
sewage. Mace gives the following table: —

Ba

cteria per c.c.

Very pure water

0

to 50

Good water

50

500

Passable (mediocre) water

500

3,000

Bad water .

3,000

,, 10,000

Very bad water

10,000

„ 100,000 and over.

Koch first laid emphasis on the quantity of bacteria present as an
index of pollution, and whilst different authorities have all agreed
that there is a necessary quantitative limit, it has been impossible to
arrive at a settled standard of permissible impurity. Besson adopts
the standard suggested by Miquel, and on the whole French
bacteriologists follow suit. They also agree with him, generally
speaking, in not placing much emphasis upon the numerical estima-
tion of bacteria in water. In Germany and England it is the custom
to adopt a stricter limit. Koch in 1893 suggested 100 bacteria
per c.c. as the maximum number of bacteria which should be present
in a properly filtered water. Miquel holds that not more than ten
different species of bacteria should be present in a drinking water,
and such is a useful standard. The presence of many rapidly
liquefying bacteria or organisms associated with sewage or surface
pollution would, even though present in fewer numbers than a
standard, condemn a water. From a consideration of all the facts
it will be seen that it is impossible to judge alone by the numbers.
As the science of bacteriology advances less emphasis is laid
upon quantitative estimation, for the reason that it is impossible
to gauge the quality of a water only by such estimation. The
character of the organisms present and the relative abundance of
each species is of more importance than quantitative estimations.
Such estimations of water bacteria, based upon the counting of
colonies in plate cultures, are of little value, and are in no sense
an adequate bacteriologica] examination of a water. It is such
"examinations" which have brought bacteriology into disrepute,
for it is certain that estimations of this kind are frequently not ever
approximately correct, aor do they furnish any final indication as to
safety or otherwise of a water Bupply. At the Bame time it should
not be forgotten that, other things being equal "nil constant, a low
number of organisms bends t<> indicate thai a water has uot been
contaminated with organic matter or the addition of foreign bacteria,
.•mil has nnt been in a condition to favour multiplication of bacteria,
and vice versd. Broadly Bpeaking, it must he tine that a water
containing' a large degree of organic matter, the pabulum of bacteria,
will contain a higher number of bacteria than a water containing a

u

BACTERIA IN WATER

low degree, and this, of course, is the reason for quantitative
estimations.

Quality of Water Bacteria

The species of bacteria found in water vary widely. Many of
them are common in pure water, and may be strictly termed
" water bacteria " ; others are as clearly " sewage bacteria," with
an allied group belonging to the soil and washed into rivers, or
wells, by rain, and which may be described as " surface bacteria " ;
and a third group are the pathogenic bacteria, which have under
exceptional conditions been isolated from water. Prof. Marshall
Ward, in his fifth report to the "Water Kesearch Committee of the
Eoyal Society, drew up a classification of water bacteria,* which was
adopted two years later by Boyce and Hill.f In 1899 Johnson
and Fuller made other groups,* and many other workers have sug-
gested classifications. The two most recent have been constructed
by Horrocks of Netley§ and Jordan of Chicago.||

Both authorities recognise that provisional classification is all that
is at present possible. Their groups are as follows : —

CLASSIFICATION OF HOUROCKS.

CLASSIFICATION OF JORDAN.

i. Fluorescent bacilli.

ii. B. aquatiiis sulcalns.

iii. B. suhtilis and " Potato bacilli."

iv. B. liquefaciens.

v. Chromogenic (red) bacilli.

vi. Chromogenic (yellow) bacilli.

vii. Chromogenic (blue) bacilli.

viii. Chromogenic (milk-white) bacilli.

ix. Chromogenic (brown) bacilli.

x. Micrococci.

xi. Sarcinse.

xii. Spirilla.

xiii. Denitrifying and nitrifying bac-
teria,

xiv. B. coli communis.

xv. B. enteritidis sporogenes.

xvi. Staphylococci,

xvii. Streptococci,

xviii. The Proteus group.

xix. Sewage bacteria.

xx. B. typhosus.

n.
iii.
iv.

v.
vi.

II. roll communis.
11. Ixrtis aerogenrs.
Proteus.
II. i nteritidu.
B. fluorescent liquefaciens.
/L Jluorescens rwn-Uquefaciens.
vii. B. suhtilis.

viii. Non-gas forming, non-fluorescent,
non-sporulating, liquefy gela-
tine and acidify milk,
ix. Similar to Group viii., but milk

rendered alkaline.
x. Similar to Group viii. , but gelatine

not liquefied,
xi. Similar to Group ix., but gelatine
not liquefied.
Similar to Group xi. , but the

reaction of milk not altered.
Chromogenic bacilli, not included

in above groups.
Chromogenic Staphylococci.
Non-chromogenic Staphylococci.
Sarcinse.
Streptococci.

xn.

xiv.

xv.

xvi.

xvii.

":: Proc. Rot/. Soc, 1897, lxi., p. 415.

t Jour, of Path, and B act., 1899, vi., p. 32.

t Jour, of Exp. Med., 1899, iv., p. 609.

§ An Introduction to the Bacteriological Examination of Water, 1901, p. 12 el seq.

II Jour, of Hygiene, 1903, vol. hi., No. 1, p. 5.

QUALITY OF WATER BACTERIA 45

Both the above quoted authorities furnish a large body of facts
illustrative of the characteristics of the various groups suggested, to
which the reader is referred for further particulars. Broadly it may
be said that the organisms classified in twenty groups by Horrocks
are divisible into a few general divisions. Groups i.-xii. are the
ordinary water bacteria ; Group xiii. is the denitrifying and nitrify-
ing organisms found in soil, water, etc. ; Groups xiv.-xix. are the
sewage bacteria ; and Group xx. represents the pathogenic group of
organisms occurring occasionally in water. Brief reference will
now be made to these four groups, with the exception of the second,
which will be dealt with subsequently.

(a) Ordinary Water Bacteria.— These are organisms usually
found in pure or approximately pure waters. They are common in
well waters and unpolluted river water. They include the common
fluorescent bacilli, liquefying and non-liquefying, and which create
an iridescent green colour in the nutrient media. In this class also
are B. aquatilis sulcatics, the "potato bacilli " (JB. mesentericus,
vulgatus, fuscus, et ruber), the " hay bacilli " (B. subtilis, B. mycoides,
B. megatherium), the liquefying bacilli common in un filtered waters,
the chromogenic organisms {B. prodigiosus, B. lactis erythrogenes,
B. rubescens, B. arborescens, B. aqtcatilis, B. aurantiacus, B. violaceus,
etc.), and the micrococci, sarcinse, and ordinary water spirilla.* The
presence of these species of bacteria in water, unless in very
exceptional numbers, indicates little of importance. They vary
according to season, geological formation over or through which the
water passes, surface washings, aeration of the water, forma of
vegetation existing in the water, and many other similar natural
conditions. Tin4 fluorescent and oon-gas-producing and non-liquefy-
ing bacilli are generally Less abundant in recently polluted waters
than in purer waters, and aon-chromogenic staphylococci more
abundant.

(b) Sewage Bacteria. This group includes B. colt communis and
its allies, the Proteus family, B. enteritidis sporogenes of Klein, and
certain streptococci and staphylococci They will be treated of
subsequently in a chapter devoted to the bacteriology of sewage (see
pp. 152-157). Exception will, however, be made in the case of R colt
communis, as this organism is perhaps the most important in relation
to water. It will, therefore, be considered here. In the firsl place
the chief biological and cultural facts may be stated, and in the
second place a general note may be added.

The biological characters of these various groups of Mater bacteria will be
found in Frankland's Micro-organumt in WaUr% pp. 399-508; Lehmano and
Neumann's Bacteriology* vol. ii., pp. 133-381; Crookshank's Bacteriology and
Infective IHsecues ; Horrocks1 Baetsnotogicai Examination of WaUrt pp. 42-80; and
in tin- systematic works of Sternberg, Plugge, Besson, Mace, etc

46 BACTERIA IN WATER

BACILLUS COLI COMMUNIS (Escherich)

Source and habitat — An organism of wide distribution, normally present in the
excreta of man and animals. Abundant in crude sewage (100,000 per c.c. in
London Sewage, Houston). In polluted water, milk, soil, etc.

Morphology — A short rod with round ends ; size and shape may vary in same
colony ; polymorphism, depending upon age of culture, products of culture, com-
position of medium, etc., 2 to 3 n long, 0*5 to 0*6 /n broad ; sometimes oval, hardly
longer than broad. Usually single, but occasionally in pairs, bundles, or even
chains and threads (Plate 3).

Staining reaction — Ordinary aniline dyes. Decolorised by Gram. (Schmidt
states that B. coli from fatty stools of infants holds the Gram.)

Capsule — Present.

Flagella — 3 or 4 in number, fragile, short, and not wavy. Sometimes only a
terminal one ; sometimes several long ones ; but polar staining and vacuolation
frequently present in old cultures, or cultures grown under unfavourable con-
ditions.

Motility — Present, especially in young cultures, but not, as a rule, so active as
B. typhosus ; oscillatory rather than progressive. Sometimes apparently absent.

Spore formation — None.

Biology : cultural characters (including biochemical features) — Grows best at 37° C. ,
but will also grow at room temperature. Gordon showed that many varieties of
B. coli exist with many minor modifications (Jour, of Path, and Bact., 1897).

In gelatine plate cultures the colonies appear generally within 24 hours at 20° C.
The deep colonies appear as small white dots, the surface colonies as delicate,
slightly granular films of an irregularly circular shape. They are bluish-white by
reflected, and amber colour by transmitted light. The diameter of the colony is
1 to 2 mm. The colonies are transparent, and sometimes iridescent, especially
towards the periphery, but at the centre and over the entire surface in old cultures
an opacity due to a greater thickness of the bacterial growth is observed (Plate 4).

It has been observed that species derived from water grow in transparent
colonies, whereas those from the alimentary canal or excreta may show opacity of
the colony, which characteristic disappears if the culture be passed through milk.
About the second or third day the surface colonies attain a diameter of 5 to 6 mm.,
and become marked by concentric, or radiating, or irregular markings. The
surrounding gelatine very frequently acquires a dull, cloudy, faded appearance,
and the edges of the colony become more crenated and thinner. The whiteness of
the colony turns to yellow. There is no liquefaction of the gelatine.

In gelatine stab-cultures the organism grows rapidly. On the surface, in twenty-
four hours, the growth is often 2 to 3 mm. in diameter, and closely resembles a
surface colony in a plate culture, though more luxuriant. A thick white growth
extends along the whole length of the track of the needle, and not infrequently gas
bubbles or fissures appear. The gelatine is not liquefied, even in old cultures.

In gelatine streak cultures growth is also abundant. In twenty-four hours the
elongated milky surface colony may be 5 mm. in diameter. It consists of a delicate
faintly-granular film with transparent and irregular margins. Down the centre
longitudinally the growth is thicker and therefore more opaque. Irregular thick-
enings, foldings, and corrugations may occur in old cultures. Sometimes the film
shows iridescence, and the medium, though not liquefied, becomes clouded. The
growth, as on the plate cultures, is bluish- white by reflected, and yellowish-amber
in colour by transmitted light.

In 25 per cent, gelatine at 37° C — In 48 hours the melted gelatine remains clear,
but a thick pellicle forms on the surface (Klein).

Gelatine shake cultures become turbid, and within twenty-four hours at 20° C. are
riddled with bubbles of gas, which are generally more numerous and larger towards
the foot of the tube. They increase in size by the second day, sometimes even
forming fissures. The gas is mainly carbonic acid. The presence of a small per
cent, of fermentable sugar in the medium increases the gas production (Plate 4.).

On potato-gelatine the colonies of B. coli are similar in appearance to those

PLATli 3.

Bacillus coli command*. From agar culture, 4S hours at >6T

X 1000.

Proteus vulgaris, tmp

ression preparation from "swarming islands" on gelatine
20 hours at 20 c. 8000.

[To face page 46.

BACILLUS COLI 47

occurring on ordinary gelatine, except that they grow more slowly, are more
circumscribed, and of a characteristic brown colour (Houston).

On carhol-gelatine (*05 per cent, of phenol), the growth does not differ from
ordinary gelatine cultures except that it is delayed.

Broth — In less than twelve hours at 37° C. the medium becomes uniformly
turbid. It may be very pronounced. Frequently there is also at a later
stage a marked amorphous flocculent sediment consisting of bacteria. Only a
faint film forms on the surface, which rarely becomes a pellicle. There is a foetid
odour, and sometimes gas formation. In glucose, lactose, saccharose broth
(2 per cent), and glucose-formate broth (Pakes), and bile-salt broth (M'Conkey),
the growth is abundant, and gas is produced. In phenolated broth (*05 per
cent, of phenol), and in broth containing formalin (1 to 7000), there is
also growth.

On agar at 37° C. the organism grows rapidly, producing thin, moist, translucent
creamy greyish-white colonics of irregular shape and size. The colonies grow more
rapidly on the surface than in the depth of the medium. The same appearances
occur on agar at 20° C, except that the growth is delayed. Gas bubbles frequently
occur in the condensation fluid.

Litmus lactose agar (2 per cent.)- The medium is turned red in twenty-four hours,

B. Typhosus. li. coli.

Pio, 9. i 'i.-i I'.t'i'- of uid /■'. coli.

and the surface growl h becomes tinged slightly with the reddened litmus. Numer-
ous gas bubbles are produced in the medium.

On potato at 87 C. there is produced in twenty-four hours a thick, moist,
yellowish-grey growth, becoming brown in old cultures. The colour varies widely
in degree, sometimes being richer than at other times. The potato becomes changed
in colour near the growth. If the potato is not fresh, or its reaction has been made
alkaline, the growth of />'. coli may be almost colourless. There are, of course, a
very large number of bacteria which produce a growth on potato not readily
distinguishable from II. coli.

Litmus milk — Usually an acid curdling of the milk occurs in twenty-four hours at
37° C, though sometimes slightly delayed. The bluish-purple colour changes to
pink, then the whole of the milk is turned into a solid compact coagulum, the milk
itself becoming white. Liter the redness extends from the surface downwards
until the whole contents of the tube are bright red in colour.

On blood si rum at 37 ('. an abundant white glistening layer is rapidly developed,
somewhat similar to the growth on agar. There is no liquefaction.

Indol is produced in bouillon fluid cultures (<.'/. peptone water). The reaction
is frequently obtainable in forty-eight hours at 37 C., but in any ease is generally
well-marked in bouillon cultures kept at "»7 C. for five days. The "red reaction"
may be obtained by adding to such a culture 1 c.c. of a 0'02 per cent solution of
potassium nitrite, and 0*5 c.c. of strong sulphuric acid. If the colour (due to

48

BACraiUA IN WATER

nitroso-indol) does not appear at once, the culture may be incubated for a brief
period.

Reduction of nitrate. — B. coli is a vigorous denitrifying organism. In twenty-
four hours at 37° C. the reduction of nitrates to nitrites is well marked. (Bouillon
5 per cent, KNO;J 0"1 per cent, water 94*9 per cent).

Aerobic or facultative anaerobic.

Vitality and -powers of resistance, not considerable, but more than the typhoid
bacillus.

The following table of comparative features of B. coli and B. typhosus is a
provisional scheme of some of the differences between a typical B. coli and a
typical typhoid bacillus. As is pointed out elsewhere, the Coli group is large and
its characteristics vary according to origin, race, cultivation, and many other
conditions. In some ways the table is misleading, as it is exceptional to find a
bacillus which gives all these features, but the table is inserted for reference,
because in a general way it states the broad differences between the types : —

Comparative Features of B. coli and B. typhosus

B. typhosus.
Morphology — Bacillus of unequal lengths;

some filaments.
Flagella — Long, wavy, spiral, numerous

(9 to 18) ; movement very active.

On gelatine and agar — Angular, irregular,

slightly raised colonies ; slow growth ;

medium remains clear.
In gelatine — In ordinary gelatine and in

lactose gelatine no gas is produced (at

20° C). No liquefaction.
Milk— Not curdled by the bacillus (at

37° C). No acid production.
Indol — In bouillon and Witte's peptone

water, no production of indol.
Bouillon containing 0*3 per cent. Phenol,

or Formalin (1 : 7000) — No growth.
Lactose — bouillon at 37° C. — No gas pro-
duction.
Neutral-red glucose-agar — No change.
Glucose or lactose media, shake cultures —

No gas production.
Potato — An " invisible growth " if the

potato is acid in reaction.
25 per cent, gelatine at 37° C— Strongly

and uniformly turbid (Klein). No

pellicle.
Eisner s iodised potato-gelatine — Slow

growth ; small transparent colonies.
Proskauer and Capaldi's Medium, No. 1

— No growth ; no change in reaction.
WidaVs reaction — Bacilli became motion-
less and agglutinated when suspended

in blood serum from a typhoid patient.

{See Appendix. )
M'Conkeys lactose agar — Surface colonies

transparent ; medium clear.
Vitality in water or sewage — B. typhosus

soon ceases to multiply and more or

less readily dies.
Pfeijfer's inoculation test with anti-typhoid

serum — Negative result.

B. coli.

Bacillus shorter and thicker; filaments
rare.

Shorter, stiffer, few (average 3), move-
ment less active, and sometimes almost
absent.

Colonies with even margin, homogenous,
much larger and quicker growth,
medium becomes turbid or coloured.

Under the same circumstances abundant
gas is produced. No liquefaction.

Milk is curdled, within 24 to 48 hours at

37' C. Abundant acid production.
Indol is present as a rule.

Grows well and uniformly throughout

medium.
Gas production occurs.

Marked green fluorescence.
Marked gas production.

Thick, yellowish-white growth, later be-
coming brown in colour.

Gelatine remains clear within 48 hours,
but a thick pellicle forms on the
surface.

Rapid growth ; large brown colonies.

Growth ; acid reaction.

B. coli remains motile and not aggluti-
nated.

Surface colonies white with yellow

centre ; haze on medium.
B. coli retains vitality and power of

self-multiplication.

Positive result, variable symptoms ac-
cording to virulence of bacillus.

BACILLUS COLI 49

General Note. — Whilst the above description applies to the
normal type of B. coli, it should be clearly understood that a large
number of bacilli have been described which possess some, but not
all, of the above characters. Eefik has described (Ann. de Vlnst.
Pasteur, x., 1896, 242), five varying types very similar to the normal
B. coli, but differing in one or more characters. Almost all forms,
however, have some features in common, e.g., motility, few flagella,
and characteristic growth on potato. Moreover, there are a group
of organisms allied to B. coli, and often associated with it. Like it
also, they are related, etiologically or otherwise, to similar pathological
processes. Befik's types are briefly as follows : —

A. Ferments lactose, coagulates milk, but gives no indol reaction.

B. Ferments lactose, does not coagulate milk, gives indol reaction.

C. Ferments lactose, does not coagulate milk, does not give indol
reaction.

D. Does not ferment lactose, coagulates milk, does not give indol
reaction.

E. Does not ferment lactose, does not coagulate milk, does not
give indol reaction.

Mervyn Gordon lias made a careful study of the B. coli and its
allies which he classified according bo their read ions and their flagella.
He differentiated 16 varieties.* Horrocks studied i he cultural char-
acters of 150 "varieties" of B. roll isolated partly from normal and
partly from typhoid stools.f Other workers have observed an
enormous variety of minor differences. The important point ifl the
diagnosis of B. colit&nd the following characters are now chiefly relied
upon (see also p. 472). 1. The B. coli group is non-sporing and non-
liquefying; 2. The members of the group rarely Btain by (nam's
method; 3. They produce acid and gas with both glucose and lactose;
4. They produce acid in milk and they usually also coagulate it; 5.
They produce acid and gaa in bile-salt-glucose broth; 6. They grow
well at a temperature of 42 C.J Other fairly reliable features are
motility, a small number of flagella, a fairly typical growth on potato,
and more rapid development on all media than the typhoid bacillus.
But there is not at the present time a complete unanimity of opinion
as to the mosi reliable characters for diagnostic purposes. §

Jour, of Path, <>»</ Bact., 1897, vol. hr., p,
I Bacteriological Examination of Wat,,-, L901, p. '.'I ; Jour, of Hyg., 1901, p.
; Roy. Cm, i. mi 8«wagt Disposal, Second Report, L902, p. 101. See also Brit.
Med. Jour., 1903, i. 11^ (Klein), for summary of characters or B. coU.

§ Houston considers the following the mosi useful tests for /;. coli: (1) Gas
formation in ordinary gelatine "shake" cultures; (•_') indol in broth cultures;
(3) acid and clot in litmus milk -cultures; (i) greenish-yellow fluorescence in

neutral-red broth cultures; (5) gas and acid in lactose pep'tone cultures; (6) gas,
acid, and clot in peptone l;i< tose milk cultures : (~ gas and acid in glucose peptone
cultures; (8J reduction of nitrate to nitrite in nitrate broth cultures; (9) Strong
acid in Proskaucr and Capaldi'fl medium Xo. 1, and no definite production ot acid

D

50 BACTERIA IN WATER

The significance of B. coli is of course its potential pathogenicity,
and its similarity to the typhoid bacillus, but above all its relation
to sewage. Eoux, Koclet, and others have stated that B. coli, under
certain circumstances, may assume a character not distinguishable
from B. typhosus, both in its biological and cultural characteristics
and in its pathogenic properties. Chantemesse, Widal, and others
have held that polluted waters owe their power to produce typhoid
fever to the presence of B. coli, and that possibly the organisms
are transformable the one into the other. Klein and many other
bacteriologists, as the result of very numerous experiments, have
been unable to effect any transformation of one form into the
other. Each organism has retained unimpaired its differential
characters.

Certain strains of B. coli are distinctly pathogenic for lower
animals, and there is some ground for considering the organism a
cause of disease (epidemic diarrhoea and other conditions) in man,
either by itself or in association with other organisms (Deh'pine).
In the third place, as is pointed out elsewhere, B. coli is a sewage
organism, and the chief importance of its detection in water is an
indication of sewage pollution and therefore of possible contamination
of the water with specific bacteria. It is therefore a most reliable
test of pollution. Klein and Houston have emphasised the importance
of the presence of B. coli and the B. at celt ides sporogencs in water as
indication of sewage pollution, and by this means a demonstration of
the presence of sewage in water can be carried to an incomparably
higher degree than by chemical examination. Chemistry is powerless
to detect pollution by pathogenic germs or the small amount of
organic pollution which can be detected by bacteriology, which is ten

in Proskauer and Capaldi's medium No. 2; (10) presence of motility; (11)
non-liquefaction of gelatine ; and (12) acidity in litmus whey cultures, varying from

about 20-40 c.c. — Na2C03 per 100 c.c. of culture. In dealing with sewage,

effluents, and non-drinking-water streams, Houston employs the first three tests,
but in dealing with drinking-water, the first five tests {Fourth Report of Royal
Commission Sewage Disposal, 1904, p. 106).

McWeeney relies chiefly upon (a) the character of gelatin colony and non-
liquefaction of that medium, even after a long time ; (b) non-retention of Gram's
stain ; (c) fermentation of lactose with gas and acid formation ; {d) coagulation of
milk within four days at 37° C. ; (e) production of yellowish-green fluorescence in
neutral-red-agar-shake culture ; and (/) production of indol in liquid peptone media.
(Report of Local Government Board for Ireland, 1904). Klein describes B. coli
as a motile, non-spore-bearing bacillus, possessing a limited number of flagella,
capable of fermenting glucose and lactose, of curdling milk with the production of
acid, of forming indol in broth culture, reducing neutral red with the production of
a green fluorescence, producing gas-bubbles in nutrient jelly, of forming a more or
less brownish growth on steamed potato, and of producing on the surface of gelatin
a dry, translucent growth which does not liquefy the gelatin. The bacilli, under the
microscope, appear as cylindrical rods, showing more or less pronounced motility,
and they do not stain by the method of Gram (see also Appendix, pp. 466 and 472).

PLATE 4.

Bacillus call communis. Surface gelatine plate culture, 0-1 c.c. ol ,,„' c.c. of Rugby Bew<

Gas in Gklatuti Bhaki I . houra at 20 C.

K,'"m '"n '" ri*ht ,h" bubM »P* • ., ,...' , ,„.,' <-, . \ tttngham crudi

[To face j>"'j< 60

BACILLUS COLI 51

to one hundred times less than that detectable by chemistry.* It is,
however, important to bear in mind that something more than the
mere presence of B. coli must be ascertained. The comparative
numbers present, the relative abundance, and the general character
and source of the water must be considered. Waters containing no
B. coli in 100 c.c. are of course of a high degree of purity.f In
upland surface waters the presence of B. coli in such a small amount
as 1 c.c, may be sufficient to condemn the waters. Certainly drinking-
water from a deep well should contain no B. coli. The presence in
a water of B. coli in conjunction with streptococci or even the spores
of B. cnlcritidis sporogenes, or both, would of course indicate serious
pollution.

The differential diagnosis of B. coli from its allies or other
organisms is not always a simple matter. An adherence to the
characteristics set out above will generally prove safe guidance, but
reliance should not be placed upon any single character or test.
The tendency to adopt some rapid and easily-applied fcegl for this
organism is strongly to be deprecated, as likely to lead bo error.
Nothing can take the place of the careful study and Bub-culture of
the suspected organism in this and in all other species. At the same
time, it has been found thai diagnostic aid is obtained by a
comparison of some of the biological characters of the colon and
allied groups of bacteria. They may be divided into four divisions: —
(1) The proteus group, the members of which are motile, liquefy
gelatine, produce gas in glucose and Bucrose hut ool in lactose, curdle
and acidulate mills very slowly, and usually produce indol ; (2) the coli
group include motile bacilli, producing gas in glucose and lactose,
curdle milk rapidly, nearly always produce indol, bu1 do not liquefy
gelatine, and do not retain Gram's stain; (3) the group including
B. lactis cerogenes are non-motile bacilli, which do not liquefy gelatine
but which curdle and acidulate milk and fermenl sugars other than
glucose; and (4) the enteritidis -roup contain bacilli which are
motile, which only fermenl glucose, and which do not liquefy gelatine
or curdle milk, which is ultimately rendered alkaline. Tins group
includes />'. enteritidis of Gaertner, the para-colon and the para-
typhoid bacilli.

Streptococci in Water. — Houston considers the presence of strepto-
cocci in water as indication n\' recent and dangerous pollution of
water. They are absent even in large quantities of pure water and
in virgin soils.^; Streptococci, as a class, are delicate germs thai
readily lose their vitality and die when the physical conditions are
unfavourable, and they comprise species highly pathogenic i<> human

* Medical Supplement to Report of Local Government Board, 1898-99, p. 19S.

i Set- .Us.) Jour. of Hug., 1902, p. 889 S.i\

; Report of Local Government Board, I 399 1900, p. 183.

52 BACTERIA IN WATER

beings. They are present in human faeces and in crude sewage in
considerable number; and as we have said, they are absent from
relatively large amounts of pure waters and virgin soils, but present
in abundance in water and soil recently polluted with animal
dejecta. It is not claimed that all streptococci are necessarily delicate
germs, or pathogenic, or of recent animal outcome. It may be that
certain streptococci are comparatively hardy germs, and that others
may be capable of multiplying in Nature outside the animal body.
Again, there may be streptococci in Nature which do not owe their
origin to excremental matter, and doubtless many of them may be
non-pathogenic, although this latter circumstance is no proof that
at a stage prior to their isolation they were non-virulent, nor does
it impair the value of the test as an indication of recent fouling with
objectionable matters.

Houston found streptococci habitually present in crude sewage
in ToVo c-c-> present in human faeces in one milligramme, and present
in minimal quantities of soils and water recently polluted with
matters of animal outcome. These results encourage the belief that
the streptococcus test is one of the most delicate yet suggested for
detecting recent, and therefore, presumably, specially dangerous,
pollution.

The question of relative abundance in connection with the strepto-
coccus test also deserves consideration. For if streptococci are absent
from 10 c.c. or more of pure waters and present in y^Vo c-c- °f
crude sewage the distinctions as regards streptococci between water
and sewage is sufficiently great to allow of considerable latitude being
observed in framing a standard without seriously impairing the value
of the test. What standard should be adopted is a matter of opinion,
but as a rule it may be said that the presence of streptococci are to
be thought of as indicating extremely recent, and B. coli less recent,
but still not remote, pollution of animal sort (Houston). The
presence of B. cntcritidis sporogenes, however, cannot be considered to
afford evidence of pollution bearing a necessary relation to the recent
evacuations of animals. Streptococci and B. coli are either altogether
absent or present in sparse amount in virgin soils, and may be
absent even from polluted soils, unless the contamination is of
comparatively recent sort. In soils recently polluted with animal
matters streptococci and B. coli ar*e of course present in abundance.
B. cntcritidis sporogencs may be present even in seemingly virgin soils,
but in sparse proportion compared with the large number found in
cultivated and polluted soils. Lastly, the presence of streptococci
in any number in a water supply points not only to recent animal
pollution, but also implies that the antecedent conditions — condi-
tions intervening between the period of pollution of the water and the
time of collection of the sample — could hardly have been of so un-

PATHOGENIC BACTERIA IN WATER 53

favourable a character as to destroy the vitality of seemingly more
hardy microbes — for example, the typhoid bacillus. The same
cannot be said for the B. coli test, since B. coli is a more hardy germ
than B. typhosus.

Broadly, therefore, it will be seen that the presence of B. coli or
B. enteritidis sporogencs or Streptococci in a water is presumptive
evidence of sewage pollution. But that in forming an opinion it is
essential to bear in mind the relative abundance of organisms per
c.c. and the relative abundance of certain species.

(c) Pathogenic Bacteria in Water. — The two chief types of
disease-producing organisms found in water are the bacillus of
typhoid fever and the bacillus of cholera. These diseases and their
causal organisms are dealt with subsequently (see pp. 298 and 384).
Here it will only be necessary to note one or two general facts as to
the relation of pathogenic organisms to water supplies.

In sterilised water, and in very highly polluted water or sewage,
pathogenic bacteria do not flourish. In the former case they die of
starvation, although there are experiments on record which appear
not to support this view ; in the latter case they are killed by the
enormous competition of common bacteria. Even in ordinary water
there is a wide divergence of behaviour. Some bacteria are destroyed
in a few hours ; others appear to flourish for weeks. In all cases the
spores are able to resist whatever injurious properties the water may
have much more persistently than the bacilli themselves. These
changes in the vitality of bacteria in water, partly due to the water
and partly to the other micro-organisms, bring about two character-
istics which it is important to remember, viz., that pathogenic germs
in water are, as a rule, scanty and intermittent. It is these features
in conjunction with the enormous quantities of common water bacteria
which make the search for the bacillus of typhoid fever whal Klein has
called "searching foraneedle in a rick of hay." Not that it cannot
be detected, but its detection is cue of the most difficult of investiga-
tions. In recent years the typhoid bacillus has been isolated from
water which had given rise to of typhoid fever at Pierrefonds

(Widal & Chantemesse), Dijon (Vaillard), Chateaudun, Cuxhaven
(Dunbar), and possibly one or two Other instances.* Undoubtedly
a large number of epidemics have been due to typhoid infected water,
hut for obvious reasons (long incubation of typhoid, the tact that
the bacillus only lives in water for a few day-, etc.), the cases where

the bacillus has been actually isolated are very few. In the Milroy

Lectures for 1902, Professor Corfield gives records of between 50 and
GO typhoid epidemic, since 1864. We -hall refer to this matter
subsequently when Bacillus typhosus is under consideration.

Tn artificial cultivation water bacteria respond very readily to
/;/•;/. M,<l. Jour., 1900, ii. p. 11'.'-.

54 BACTERIA IN WATER

external conditions. Increase of alkalinity ('01 grams of sodium
carbonate added to 10 c.c. of ordinary gelatine) causes the number
of colonies to be five or six times greater than that revealed by using
ordinary gelatine ; on the other hand, very slightly increasing the
acidity of a medium as markedly diminishes the number of bacteria.
Advantage is taken of this in culturing the bacillus of typhoid, which
is not inhibited by an acid medium.

Water may become contaminated with pathogenic bacteria in a
variety of ways, as pollutions at the source, in the course, and at the
periphery. Gathering grounds are frequently the source of the
pollution. The Maidstone typhoid epidemic was an example. Here
some of the springs supplying the town with water were con-
taminated by several typhoid patients. Frequently on the gathering
ground one may find a number of houses the waste and refuse of
which will furnish ample surface pollution, which in its turn may
readily pass into a collecting reservoir or a well. On one occa-
sion the writer investigated the cause of typhoid fever in a large
country house in Oxfordshire, and traced it to pollution of the
private well by surface washings from the stable quarters. Link-
age of house drains into wells is not an infrequent source of
contamination.

The same cause is generally operative in cases of pollution of a
water supply in its course from the source to the cisterns or taps
at the 2)eriphery, viz., a sewer or drain leaking into the water supply.
Water companies and those responsible for water supply appear
frequently to hold the opinion that so long as there is sand filtration
or subsidence reservoirs, it is unnecessary to consider the gathering
ground or possible contamination during transit. But it happens
that a frost may completely dislocate the efficient action of a filter,
and times of flood may prevent proper sedimentation ; then our
dependence for pure water is wholly upon the gathering ground and
source. Hence we find water contaminated at its source by polluted
wells, by sewage-infected rivers and streams, by drainage of manured
fields, by innumerable excremental pollutions over the areas of the
gathering grounds, and in transit by careless laying, bad construction
and jointing of pipes, and close proximity of such drain pipes to the
water supply.

In the third place, we may get a water infected at the periphery,
in the house itself. Such cases are generally due to two causes:
filthy cisterns and pipes or suction. Cisterns per se are more or less
indispensable where a constant service does not exist, but they should
be inspected from time to time and maintained in a cleanly condition.
Suction into the tap has been emphasised by Dr Vivian Poore as
a cause of pollution. It is liable to occur whenever a tap is left
turned on, and a vacuum is produced in the supply pipe by inter-

INTERPRETATIONS OF BACTERIOLOGY 55

mission of the water supply, so that foul gas or liquid is sucked back
into the house-pipe.

A further point has relation to bacterially polluted water when it
has gained entrance to the body. It has been known for some time
past that not all waters polluted with disease germs produce disease.
As we have before said, this depends upon the infective agent, its
quantity and quality, and upon the human body. The body is able
in many cases to resist a small dose of poison. It is, however,
necessary to infection, especially in water-borne disease, that the
tissues shall be in some degree disordered, weakened, or injured. For
instance, the perverted action of the stomach influences the acid
secretion of the gastric juice, through which bacilli might then pass
uninjured. Particularly must this be so in the bacillus of cholera,
which is readily killed by the normal acid reaction of the stomach.
Hence, in this disease at least, it is the opinion of bacteriologists that
the condition of the mucous membrane of the stomach is of primary
importance. Metchnikoff has indeed demonstrated the presence of
the bacillus of cholera in the intestinal excretion of apparently healthy
persons, which shows that they were protected by the resistance of
their tissues to the bacilli. Further light has been thrown on this
question by the researches of MacFadyen, who has pointed out: thai
suspensions of cholera bacilli in water passed through the stomach
untouched, and were thus able to exert their evil influence in other
parts of the alimentary canal. When, however, cholera bacilli were
suspended in milk, none appeared to escape tin' germicidal action
of the gastric juice. The explanation of this is probably the simple
one that the stomach reacted with its secretion of gastric juice only
to food (milk), but passed the water on into the lower and more
absorptive parts of the alimentary canal. Such a condition of affairs
dearly increases the danger due to water-borne germs.

The Interpretation of the Finding's of Bacteriology

Bacteriology is the mosl direct and delicate test of the safety of a
water for drinking purposes. By it we obtain exact information not
alone as to the constitution of a water, but as to its potentiality to
cause disease. It is also a more delicate test than a chemical
examination.* Klein and others have shown that by bacteriological
methods it is possible to detect smaller degrees of sewage pollution
than by chemistry. On the other hand, it is useless to expect to
Learn of the exact chemical constitution of a water by bacteriological
methods. Bacteriology must he interpreted by what it can

(lark and Gage state that polluted waters which might become unfit for
drinking purposes arc more plainly indicated by a single chemical anal) bis than by
a tingle determination of B. eolL

56 BACTERIA IN WATER

do and not by what it cannot; and in a general way it may
be said that there are three groups of facts contained in a
systematic bacteriological report of water. These findings are
concerned with the number of bacteria per c.c, the presence of
any organisms of contamination, and the presence of any specific
organisms of disease.

(1) Number of Bacteria per c.c. — It would appear that in the past
a great deal too much weight has been attributed to the number of
bacteria per c.c. This fact is not of the first importance for two
obvious reasons. In the first place there is no standard as to how
many bacteria should be present in 1 c.c. of a potable water, and in
the second place there is no known means by which this number
can be accurately measured. In this country any number of bacteria
under one hundred per c.c. is generally considered low. The
metropolitan water supply, as consumed, usually contains less than
twenty bacteria per c.c. Deep - well waters and spring waters
frequently contain very few bacteria. Polluted or surface waters
contain thousands of organisms per c.c. More than this, no
standard exists. Nor would any numerical standard taken alone
be of much value, for the reason that the number of bacteria in
water is of comparatively little value apart from a knowledge of the
species, and moreover a really accurate record of the number of
bacteria per c.c. is not obtainable. Whether the organisms detected
be many or few depends upon a variety of external circumstances,
such as medium used for cultivation, temperature and period of
incubation, length of time of cultivation before counting, or the
use or not of a lens when counting. For these reasons it is
evident that great reliance cannot be placed upon the number
of bacteria per c.c. returned in bacteriological reports, and it is
well that should be understood. The only circumstances under
which such returns are valuable are (a) when used in a series of
examinations of the same water supply, when such returns, if always
obtained under the same conditions, are of great comparable value,
and (b) when used in the examination of water before and after
filtration. In these two circumstances the number of organisms per
c.c. is of great value in forming an opinion as to pollution or as to
failure of filtration.

(2) Presence of Organisms of Contamination. — In the general
bacteriological examination of water this point is perhaps the most
important. Judgment must be formed on two facts, namely, the
presence of any of the "bacteria of indication," such as B. coli,
B. enteritidis sporogenes, streptococci, and the para-colon types
(enteritidis, G-aertner, and the chologenes type), and the relative
abundance of these species. The latter point is one of importance.
The chief organism of indication is B. coli, including under that

ORGANISMS OF CONTAMINATION 57

term the typical bacillus and closely allied organisms. When
this bacillus can be detected in a small measured quantity of
water, that is to say, in 1, 2 or 3 c.c, it is assumed (a) that the
organism has gained access to the water from sewage, and (b) that
recently, (c) It is further assumed that certain disease-producing
bacteria which occur frequently in sewage may also be present in
the water, though if present at all in the water, in considerably
smaller numbers than B. coli. (d) Further, judging the matter
broadly, the higher the number of B. coli the heavier will have been
the recent sewage pollution, and the greater the probability of the
presence of disease-producing bacteria. Conversely, if B. coli is
not present, one may assume with some probability of being correct,
that such disease-producing bacteria as the bacillus of typhoid fever
will also be absent, and that the particular sample of water under
examination might safely be used for drinking purposes.

There is difference of opinion as to the exact quantity of a water
which must be free from a single specimen of B coli in order that it
may be said that the sample is a "safe" one; but many would in
practice accept the standard 1 or 2 c.c.

It has already been stated that the presence of B. coli in a water is
not of importance, because this organism itself, under the ordinary
conditions, is likely to be harmful, but rather because it serves as an
index of sewage or surface pollution. In this connection it may be
said that a single examination of a water is of practically no value
when the results of the bacteriological examination are favourable ;
it is only after repeated examination has shown that B. coli is absent
from the water for a prolonged period, and after local inspection has
shown that there are no possible sources of dangerous sewage con-
tamination, that one is justified in giving a positive opinion as to
the safety of a water. On the other hand, a single bacteriological
examination with an unfavourable result will prove the actual
occurrence, and suggest the possible recurrence, of sewage contamina-
tion, and will necessitate renewed inspection if no obvious source <>f
contamination is known to exist.

B. coli is commonly considered as evidence of contamination by
sewage, but it is possible for the bacillus to gain access to the water
from other sources also. The bacillus is present in the excreta of
mammals generally, and lias been found in the excreta of birds, and
in surface waters there will undoubtedly be a certain amount of
contamination caused in this way. The question as to whether any
contamination of this kind can be caused by various fishes, and other
forms of aquatic life, is not fully established, though Eyre has
recently found the B. coli iii the excreta of fishes, as well as mammals
and birds.*

lancet, 1904, L, p. 648.

58 BACTERIA IN WATER

Most bacteriologists would condemn a water containing the
typical B. coli in 1 c.c. as showing signs of sewage pollution. In the
case of a recent pollution the presence of B. coli affords therefore a
much more delicate test of pollution than any chemical examination
which can be made.*

B. enteritidis sporogenes is another organism of indication as to
sewage pollution, and its presence in bacillary form or as spores is
now accepted as showing recent or remote contamination.

The presence of streptococcus is held by many bacteriologists to
be a sign of sewage contamination, although some contend that the
presence of streptococci does not indicate dangerous contamination
unless accompanied by B. coli. The following table (p. 59), from the
Thirty-fourth Annual Eeport of the Lawrence Sta., 1903, sets forth,
in less space and with more accuracy than could be recorded in many
words, the relative presence of the chief organisms of contamination,
and it is therefore inserted.

Lastly, there are a number of organisms which appear to be fre-
quently present in waters contaminated with sewage, and are rarely
if ever found in pure supplies. The occurrence of such bacteria in a
water should arouse suspicion as to its origin or contamination.
Among this group of bacteria are B. Jluorcscens putridus, B. erythro-
spores, B. ct M. urea, B. pyocyanevs, B. lactis cyanogenus, and B.
megaterium.

(3) The presence of pathogenic species. — The presence of any
pathogenic organisms, in however few numbers, is of course sufficient
for the condemnation of a water. For instance, the presence of the
bacillus of typhoid fever or the bacillus of cholera at once condemns
a water. There are very few authentic records of such organisms
being found, and it is therefore necessary to judge of waters by the
presence of organisms of contamination.

Note. — A water may be considered safe and potable (a) if it
contains comparatively few organisms; (b) an absence of organisms
capable of fermenting glucose or lactose media; (c) an absence of
B. enteritidis sporogenes; and (d) an absence of any pathogenic
species, and especially if these conditions are found to exist as a
result of several examinations or of periodic examinations. A water
should be condemned, as a rule, (a) if it contains a very large
number of bacteria per c.c. of whatever kind; (b) if it contains
B. coli communis, or B. enteritidis sporogenes or streptococci in 1 c.c. or
any such small quantity; (c) if it gives the enteritidis change in milk
cultures , or ferments glucose or lactose media. It should be con-
demned without hesitation if it contains B. coli and B. enteritidis sporo-
genes (or spores), and streptococci, or if it contains any pathogenic
organism, in however small a quantity. But in condemning or
* See also Fourth Report Roy. Com. Sewage Disposal, 1904, pp. 106-109.

ORGANISMS OF CONTAMINATION

59

oa

"HH

CN

0

O

O

■**

O

O

0

CO

CO

■o*o 001

O

O

0

O

O

CN

O

O

0

co

0

O) <U

CN

5 >»

oH

kO

O

0

0

O

0

O

-*

CO

0

0

0

OS

0

O

0

O

0

O

O

0

O

CO

co

0

l-H

OS

CN

I— I

OS

1>-

0

O

OO

*tf

co

t^

0

CN

»c

xn

•o-o 001

l-H

O

0

O

O

CN

lO

rH

0

CO

rH

§ s

CN

o >»

CD

.53 Eh

OS

t>-

0

O

O

OS

O

-*

0

0

O

>o

>

<j

•o-o 1

,_,

O

0

0

O

-f

t-

CO

i-H

,_!

O

OS

CN

f

1— 1

"-H

CN

o

6

*

(M

<N

l>-

O

CO

eo

■>*

O

.

GN

CO

Ol

<2

&P<

•o-o 001

0

O

■*

l-H

CO

w

10

CO

O

co

rH

21

03

g>5

CN

7*

©

O

O

O

OS

0

I-

O

O

O

OS

kO

O

£ 0

■Q'O I

0

O

O

O

O

T*

0

I— 1

kfl

O

CO

OS

kO

^

£ 0

m

CN

J-
©

eo

•0*0 001

0

CO

l-H

CO

tN

CO

t*.

O

OS

OS

p

p

O
CI

•

CO
CO

p

Ph

Eh

rH

rH

l-H

•«#

CN

T— 1

O

O

CN

O

O

0

-f

O

0

0

O

0

OS

Th

D

•o-o x

«o

CN

O

O

0

CN

O

-H

OS

,_,

H<

OS

OO

CO

M

O)

+3

0

><1

O

O

0

O

O

0

O

"N

O

•o-o 001

-h

t«.

,_,

t—

OO

OS

OT>

-H

:~

CO

CO

CN

<N

pH

CO

CM

CO

to

CN

CN

3 gl

O

O

^

t^

0

O

0

0

0

O

0

OS

C

X

'0*0 I

H<

OO

CN

H

01

r- •

r->

BC

r-(

\Q

OS

O

a

|H

1-1

CD

^~

'*

OO

1-1

N

"N

-a J

I

<M

<co

CO

b-

00

rH

K

O)

-f

*

CO

OS

10

-S c'+;

CI

1 *<

t

Ifl

CN

-r

r— 1

>-.

OO

0

l^

OS

n

t— 1

OS

Eh 3

fc

/rt

r— '

Tl

"8

x

F

-

c3

r.
1

3

W

CU

^

en

T3
C

tn

O

T!

S

-

h

a

T3
3
O

en

O
O

CU

■
x

CM

O

CU

£

CU

en

en

s

b

8

«*

X
*

13

13

13

O

3
O

CU

c

_u

C

*
CU

44

CU

5

X

O

E

L

*

3

u

u

CU

■
E

CU

C

a

CU

3
CU

3

-Q

s

3
CU

c

i-

£

r.

3

H

S

s

g

pg

C3

CU

O

H

cu

_y

CU

X

It

ctf

^

J

•sj.i

i''.\\

60 BACTERIA IN WATER

approving a water supply it is important to take all the findings of
chemistry, bacteriology, and topography into consideration. The whole
history of the sample must be considered, and too much reliance
must not be placed upon the mere presence or absence of B. coli, or
any single phenomenon or reaction. No ultimate reliance should, as
a rule, be placed upon any single test.

Natural Purification of Water

We have already noticed that rivers purify themselves as they
proceed. There are many excellent examples of such self-
purification. The Seine as it runs through Paris becomes highly
polluted with every sort of ^ filthy contamination. It receives daily
about 250,000 cm. of sewage. But 20 or 30 miles below the city it
is found to be even purer than above the city before it received the
sewage. In small rivers it is the same, provided the pollution is less
in amount. The Thames and the Severn are excellent examples.
Whilst authorities differ with regard to the means of self-purification
which operate most effectually, all agree that in some way rivers
receiving crude sewage are able in a marvellous degree to become
pure again.

The chief conditions influencing this phenomenon are as follow : —

(a) The movement of the water. — It is probable that any beneficial
result accruing from this cause is due not to any mechanical factor in
the movement, but to the extra surface of water available for oxida-
tion processes. Dek'pinc has shown that the effect of agitation is an
increase in the number of suspended bacteria which he attributes to
the dislodgment of deposit and side adhesions. The greatest amount
of purification in his experiments occurred when the rate of flow was
about 8 cm. per hour.*

(b) The pressure of the water. — It is believed that the volume of
water pressing down upon any given area beneath it weakens the
vitality of certain microbes. In support of this theory, it is urged
that the number of bacteria capable of developing is less the greater
the depth from the surface. Yet it must be remembered that mud at
the bottom of a river, or at the bottom of shallow sea, is teeming with
living organisms, and there is no evidence to show that pressure in
river water ever reaches a degree capable of affecting the life of
bacteria. Delepine found that in the Manchester mains increase of
pressure did not reduce the number of bacteria. f

(c) Light. — We have seen how prejudicial is light to the growth
of organisms in culture media. This is so, though to a less extent,
in water (see p. 18). Arloing held that sunlight could not pierce

* The Natural Purification of Running Water, Jour, of State Med., 1901, p. 517.
f Report to the Manchester Water Works Committee, 1894.

NATURAL PURIFICATION OF WATER 61

a layer of water an inch in thickness and still act inimically on
micro-organisms. But Buchner found that the sun's rays could
pass through 15 or 20 inches and yet be bactericidal. This evidence
appears contradictory. On the whole, however, authorities agree
that the influence of the sun's rays upon water is in some degree
bactericidal and causes a diminution in the quantities of organisms
after acting for some hours. Especially will this be so when the
water is spread out over a wide area and is therefore shallow and
stationary, or moving -but slowly.* But taken as a whole it may
be said that light does not exert a marked influence in water puri-
fication. There is, on the other hand, evidence to prove that water
in its passage through dark mains of various sizes gradually becomes
deprived of a great part of its bacterial contents.

(d) Vegetation in water. — Pettenkoter, in his observations upon
the Iser below Munich, has .shown how algoe bring about a marked
reduction in the organic matters present in water. Boyce has
pointed out that in the river Severn, in addition to the temperature
and movement being unfavourable to B. coli and presumably patho-
genic bacteria, that (a) lack of pabulum, and (b) antagonism due bo
the fauna and flora of the river exert an unfavourable influence upon
those bacteria. The organic matter so abundant when the river
becomes polluted at Shrewsbury is diluted and destroyed lower down
stream, and therefore the water becomes purified of bacteria living
on the organic matter. Fish, birds, rats, protozoa, and forms of river
life generally contribute their share to the consumption of organic
pabulum. The water Ranunculus, Sphcerotilus, Leptomitus lacteus,
sewage fungi, chlorophyll containing protophytes, and river plants
generally assist, in the destruction oi organic matter and bacteria.f

(e) Dilution. — The pollutions passing into a flowing river are very
soon diluted with the large quantities of comparatively pure water
always forthcoming. And this, whilsl ii lowers the percentage of
impurity, also raises the percentage of oxygenated water; Delepine
has pointed out as a result of artificial experiments thai dilution
exerts a double effect on the bacterial content of water. In the firsl
place it has the mechanical effect of increasing the space occupied by
a definite uumber of bacteria, .and in the second place it causes a
diminution in the amount of pabulum present in a given bulk of the
impure fluid. Dilution and deposition acting together exert a power-
ful influence as purifiers. Clark and Gage of the Lawrence station
pointed out in L903 that the Dumber of ]L<<>!i in a polluted river
varies in inverse ratio with the dilution of the entering Bewage by
the river water, and is affected by the temperature, the uumber of
B. coli being larger during the warm weather than in the cold. In

Sec also Spina's work on the Spree .it Berlin, Archwfiir IIii<j.. vol. xxxviii.
i Roy. emu. on Sewage l>i>i«>s<il, Second Report, 1902, pp. 10-1-109.

G2 BACTERIA IN WATER

effluents from water filters the effect upon filter efficiency of dilution
of the water in winter is less marked than the effect of high tem-
perature in summer, the work of a filter in warm weather being, of
course, more satisfactory than in cold.

(f) Sedimentation. — Whilst Pettenkofer attributes self-purifica-
tion to oxygenation and vegetation, most authorities are now agreed
that it is largely brought about by the subsidence of impure matters,
and by their subsequent disintegration at the bottom of the river.
Sedimentation and side-adhesion to the banks in rivers and streams
of solids in suspension removes a large number of bacteria in the
Severn (Boyce). Sedimentation obviously is greatest in still waters.
Hence lake water contains as a rule very few bacteria. "The
improvement in water during subsidence is the more rapid and pro-
nounced the greater the amount of suspended matter initially
present" (Frankland). Tils has pointed out that the number of
micro-organisms was invariably smaller in the water collected from
the reservoir than in that taken from the source supplying the latter.
Percy Frankland has demonstrated the same effect of sedimentation
by storage as follows : —

No. of Colonies in
1 c.c. of Water.

1. Intake from Thames, 25th June 1892 . 1991

2. First small storage reservoir . . 1703

3. Second small storage reservoir . . 1156

4. Large storage reservoir . . . 464

The large reservoir would of course necessitate a prolonged sub-
sidence, and hence a greater diminution than in the small reservoirs.
Karlinski gives the following distribution of bacteria in the Borka
Lake (Herzegovina) : —

Bacteria per c.c.
Surface water ..... 4000
Five inches below surface . . . 1000

Ten inches below surface . . . 600

Twelve to sixteen inches below surface . 200

Bottom when mud was stirred up . . 6000

Delepine. considers that bacteria die rapidly in the deposit,
although their large numbers are evidence of the effect of sedimenta-
tion. He examined some water mains after the sediment had been
disturbed and also with the sediment undisturbed. The results
were as follows : —

Sediment undisturbed. Sediment disturbed.

1. 51 living bacteria per c.c. 334 living bacteria per c.c.

2. 356 ,, „ 3164

3. 10 „ „ 852

He concludes (1) that sedimentation is a very important factor
of bacterial purification in flowing water, and (2) that the effects of

NATURAL PURIFICATION OF WATER 63

sedimentation are most manifest when the flow of water is rapid
enough to prevent the accumulation at any point of the products of
bacterial multiplication, but not so rapid as to interfere with a
comparatively rapid action of gravity.*

In the case of a tidal stream the conditions are different, as
recently pointed out by Foulerton.f In such rivers the disease-
producing bacteria are deposited not only on the bed of the stream,
but also on the mud, or sludge, on the banks, and are uncovered by
water at low tide. It now requires only the agency of a fly, feeding-
first on the organic matter in the sewage-contaminated mud and then
on some human food, milk for instance, to convey the bacillus of
typhoid fever from the river to some human being. An additional
way by which a bacillus of this kind may survive after it has been
discharged into a river is by its being deposited on the bed of the
stream where there are shell-fish layings. It has been proved that
the typhoid bacillus can survive for a considerable time in the liquor
contained in the shell of the oyster or the mussel, and in this way it
may escape destruction by finding itself once more inside the con-
sumer of the shell-fish. Therefore, in the case of sewage discharged
into a tidal river, owing to lack of dilution and sedimentation, it is a
menace to the inhabitants on the banks in one or both of these ways.
The exact degree of danger depends first upon the extent to which
the sewage is purified before its discharge into the stream, and
secondly upon the distance from the source of pollution at which con-
tamination of the water by special sewage bacteria is still appreciable.

This principle of sedimentation operates upon all bacteria, which
are often carried down on gross particulate matter. The number
of B. coli is reduced quite appreciably by storage of water (Clark
and Gage). Many species remain in the mud, sand, or other deposit
at the bottom of the stream or reservoir. The parasitic organisms
die on account of the unfavourable environment.

(g) Oxidation. — Many experiments and observations have been
made to prove that large quantities of oxygen are used up daily in
oxygenating the Thames water. Oxygenated water will come up
with the tide and down with the fresh water from above London.
There will also be oxygen absorption going on upon the surface of
the water, and from these three sources enough oxygen is obtained
to oxidise impurities and produce what is really an "effluent." In
many smaller streams the opportunity for oxidation is afforded by
weirs and falls.

Probably all these factors play a part in the self-purification of
rivers, but we may take it that oxidation, dilution, and sedimentation
are three of the principal agencies. The test of purification is in the

Jour, of Stat. Med., 1901, p. 518.

Report on Pollution of Tidal (Jus?, 1903, p. 11.

64 BACTERIA IN WATER

number and character of the bacteria at different stages of the
river (e.g., see Table of Bacteria in Severn, p. 38). Jordan has
pointed out the peculiar value of the reduction of B. coli*

We may here refer in passing to the facts obtainable from the
late Sir Edward Frankland's report on Metropolitan water supply in
1894, as they will afford a connecting link between natural purifica-
tion and artificial purification. First, judged by the relatively low
proportion of carbon to nitrogen, the organic matter present in the
water was, as usual, found to be chiefly of vegetable origin. Secondly,
an immense destruction of bacteria was effected by storage in
subsidence reservoirs. Thirdly, the bacterial quality of the water
might differ widely from its chemical qualities. It is, of course, a
much liner index of pollution. These three facts are of primary
importance in the interpretation of water reports, and it will be well
to bear them in mind. Sir E. Frankland also referred to the physical
conditions affecting microbial life in river waters, and, as in previous
reports, to the importance of changes of temperature, the effect of
sunlight, and rate of flow. Eespecting the relative proportion of
these factors, he wrote : " The number of microbes in Thames water is
determined mainly by the flow of the river, or, in other words, by
the rainfall, and but slightly, if at all, by either the presence or
absence of sunshine, or a high or low temperature. With regard to
the effect of sunshine, the interesting researches of Dr Marshall
Ward leave no doubt that this agent is a powerful germicide, but it
is probable that the germicidal effect is greatly diminished, if not
entirely prevented, when the solar rays have to pass through even
a comparatively thin stratum of water before they reach the living
organisms." Subsequent investigations have confirmed the im-
portance of these broad principles, and from which it is clear that
evidence favours the effect of sedimentation and dilution. These two
factors in conjunction with filtration are, practically speaking, the
methods of artificial water purification, to which reference will now
be made.

Artificial Purification of Water

Sedimentation and Precipitation. — In nature we see this
factor in operation in lakes and reservoirs. For example, the water
supply of Glasgow is the untreated overflow from Loch Katrine.
Purification has been brought about by means of subsidence of
impurities. Nothing further is needed. Much of the purification
obtained in reservoirs supplying large towns is due to the same
factor. Artificially we find it is this factor which is the mechanical
purifier of biological impurity in such methods as Claries process.
By this mode "temporary hardness," or that due to soluble

* Jour, of Hyc/., 1901, p. 293.

ARTIFICIAL PURIFICATION OF WATER 65

bicarbonate of lime, is converted into insoluble normal carbonate of
lime by the addition of a suitable quantity of limewater. Carbonates
of lime and magnesia are soluble in water containing free carbonic
acid, but when fresh lime is added to such water it combines with the
free C02 to form the insoluble carbonate, which falls as a sediment : —

CaC03 + C02 + CaH202 (limewater) = 2 CaC03 + H20.

As the carbonate falls to the bottom of the tank it carries down
with it the organic particles. Hence sedimentation is brought about
by means of chemical precipitation. It is obviously a mechanical
process as regards its action upon bacteria. Nevertheless its action
is well-nigh perfect, and 400 bacteria per c.c. may be reduced to 4
or 5 per c.c. We shall refer to this same action when we come to
speak of bacterial purification of sewage. Alum has been frequently
used to purify water which contain much suspended matter. Five
or six grains of alum are added to each gallon of water, plus some
calcium carbonate by preference. Precipitation occurs, and with it
sedimentation of the bacteria, as before. But, as Babes has pointed
out, alum itself acts inimically on germs ; in such treatment, there-
fore, we get sedimentation and germicidal action combined.

As a matter of actual practice, however, sedimentation alone is
rarely sufficient to purify water. It is true that the collection of
water in large reservoirs permits subsidence of suspended matters,
affords time for the action of light, and the suicidal competition
among the common water bacteria. But in small collections of
water it is otherwise. Here filtration is the most important and
most reliable method.

Sand filtration, as a means of purifying water, has been practised
since the early part of last century. But it was not till 1885 lli.it
Percy Frankland first demonstrated the great difference in bacterial
content between a water unfiltered and a water which had passed
through a sand filter (only about 3 per cent, of the bacteria originally
present being left in the water). Previous to this time the criterion
of efficiency in water purification had been a chemical one only,
and the presence or absence of bacteria in any appreciable quantity
was described not in mathematical terms, but in indefinite descriptive
words, such as " turbid," " cloudy," etc. It is needless to say that this
difference in estimation was largely duo i<» the introduction by Koch
of the gelatine-plate method of examination. As a result of investi-
gation Percy Prankland formulated the following conclusions as
regards the chief factors i nil uencing the number of microbes passing
through the filter. The efficiency of filtration, he held, depended
upon (a) the storage capacity for unfiltered water, by which it was
possible to obtain the preliminary advantage of subsidence; (b) the
thickness of fine sand through which the filtration is carried on;

E

66 BACTERIA IN WATER

(c) rate of filtration ; (d) the renewal of the filter-beds. After a
certain time the filter-bed becomes worn out and inefficient, and at
such times renewal is necessary. Not only may the age of the
fdter act prejudicially, but the extra j^ressure required will tend to
force through it bacteria which ought to have remained in the filter.

In 1890 a special study of filtration was made by the Massa-
chusetts State Board of Health, and in annual reports published
from 1890, a number of experiments are recorded which have proved
of classic importance, and which should be consulted by the student
or practical worker desiring to acquire a thorough grasp of the
principles of biological filtration. There it is shown that water can
be filtered through sand filters at the rate of 3,000,000 gallons per
acre daily and 99'95 per cent, of the bacteria removed. In actual
practice it was found that the finer sands were more effective than
the coarser, and under moderate pressure 1 foot of sand was as effec-
tive as 5 feet. Over 80 per cent, of the bacteria removed were
found in the upper inch of sand and 55 per cent, in the upper
quarter inch. If the surface of the filter was scraped, it was shown
that an increased number of bacteria passed through the filter, which
was therefore much less effectual. Subsequently, Koch emphasised
the importance of this vital layer. But it was the Massachusetts
Board that first proved by experiment that the oxidation which occurs
in a filter-bed was due to the nitrifying organisms in the surface
or scum layer. When nitrification is established in a filter, the rate
of filtration within certain limits was found to exert comparatively
little effect upon the removal of the organic matter.

In 1893 Koch brought out his monograph upon Water Filtration
and Cholera, and his work had a deservedly great influence upon the
whole question. He showed how the careful filtration of water
supplied to Aitona from the Elbe saved the town from the epidemic
of cholera which came upon Hamburg as a result of drinking
unfiltered water, although Aitona is situated several miles below
Hamburg, and its drinking water is taken from the river after it has
received the sewage of the latter.

Now, from his experience of water filtration, Koch arrived at
several important conclusions. In the first place, he maintained
that the portion of the filter-bed which really removed micro-organisms
effectively was the slimy membranous organic layer upon the surface of
the sand. This layer is produced by a deposit from the still unpurified
water lying immediately above it. The most vital part of the filter-
bed is this organic layer, which, after formation, should not be dis-
turbed until it requires removal owing to its impermeability. A
filter-bed, as is well-known, consists of, say, 3 feet of sand and 1
foot of coarse gravel. The water to be filtered is collected into large
reservoirs, where subsidence by gravitation occurs. From thence it

FILTRATION OF WATER 67

is led by suitable channels to the surface of the filter-bed. Having
passed through the 3 or 4 feet of the bed, it is collected in a storage
reservoir and awaits distribution. Such being the principles of
construction, it will be apparent that the action of the whole process
is both mechanical and chemical. Mechanically by subsidence,
much suspended matter is left behind in the reservoir. Again,
mechanically, much of that which remained suspended in the water
when it reached the filter-bed is waylaid in the substance of the
sand and gravel of the filter-bed. The next change is a chemical
one. Oxidation of the organic matter occurs to some extent as the
water passes through the sand. Until recently this chemical action
and the double mechanical action (sedimentation and straining) was
believed to be the complete process, and its efficiency was tested by
chemical oxidation and alteration, and absence of the suspended
matter. Now, however, it is recognised that the second portion of
the chemical action is vastly the more important, indeed, the only
vital part of the process. This is the chemical effect of the layer
of scum and mud on the surface of the sand at the top of the filter-
bed. The mechanical part of this layer is, of course, the holding
back of the particulate matter which has not subsided in the reservoir ;
the vital action consists in what is termed nitrification of unoxidised
substance, which is accomplished in this layer of organic matter.
We shall deal at some length with the principles of nitrification
when we come to speak of soil. But we may say here that by
nitrification is understood a process of oxidation of elementary
compounds of nitrogen, by which these latter arc built up into stable
bodies which can do little or no harm in drinking-water. The action
of a filter-bed may, therefore, be summarised as follows: — There is
(1) subsidence of the grosser particles of impurity in the settling
tank; (2) mechanical obstruction to impurities in the interstices of
the scum, sand, and gravel in the filter; (3) oxidation <>t' organic
matter by the oxygon lid. I in the pores of the sand and gravel;
(4) nitrification in the vital scum layer, which is accomplished by
micro-organisms themselves. This hitter is now considered to be
incomparably the mosl Important part of the filter. That being so,
its removal, except when absolutely necessary, is t<> l»e avoided as
detrimental to the efficiency of the filter. New filters have obviously
but little of this action. Kummel found that when a filter had new
sand placed upon it the number of bacteria in the filtered water was
as follows : —

Per cc.

Before cleaning ... 42

One day after cleaning . . . 1880

Two days after cleaning . . . 752

Three days after cleaning . . 208

Four days after cleaning . . 156

Five days after cleaning . . 10'2

Six days after cleaning ... 84

68 BACTERIA IN WATER

Hence it is necessary to allow a new filter-bed to act for a short
period (say four clays) before the filtered water is used for domestic
purposes, in order to allow a fresh film, the organic layer, to be formed.
This must also be borne in mind after a filter-bed has been cleaned.*

To maintain this nitrifying action of a filter in efficiency, Koch
suggested, in the second place, that the rate of filtration must not
exceed four inches per hour. At the Altona water-works this rate of
filtration was maintained, and the number of organisms always
remained below 100 per c.c, which, as we have seen, is the standard.
Thirdly, it is important that periodic bacteriological examinations
should be made. Koch's emphasis upon this point is well known, and
the cumulative experience of bacteriologists only further supports such
a course being taken. Clark and Gage of the Lawrence Experimental
Station, claim that the test for the presence of B. coli is a more delicate
indication of filter efficiency when filtering polluted water than tests for
the total number of bacteria present. Fourthly, Koch maintained that
the thickness of the sand of the filter-bed should never be less than one
foot. Fifthly, if it be true that efficient sand filtration is a safeguard
against putrefactive and disease-producing germs, then there can be
but one criterion of efficiency, viz., their absence in the filtered water,
which can only be ascertained by regular examination. But it is
not alone for pathogenic germs that filtration is proposed. Hence
Koch laid down that filtered water containing more than 100 micro-
organisms of any kind per c.c. is below the standard of purity, and
should not, if possible, be distributed for drinking purposes. In this
country chemical analysis, with a more or less cursory microscopic
examination, has been almost invariably accepted as reliable indication
of the condition of the water. But such an examination is not really
any more a fair test of the working of the filter than it is of the
actual condition of the water. It is true, the quantity of organic
matter can be estimated and the condition in which it exists in
combination obtained ; but it cannot tell us what a bacteriological
examination can tell us, viz., the quantity and quality of living
micro-organisms present in the water. Upon this fact, after all, an
accurate conclusion depends. There is abundant evidence to show
that no valuable opinion can be passed upon a water except by both
a chemical and a bacteriological examination, and further by a
personal investigation, outside the laboratory, of the origin of the
water and its liabilities to pollution.

So convinced was Koch of the efficiency of sand filtration as
protection against disease-producing germs, that he advocated an
adaptation of this plan in cases where it was found that a well
yielded infected water. Such pollution in a well may be due to

* See also Thirty-fourth Ann. Hep. State Bd. of Health, Massachusetts, 1903,
p. 228.

FILTRATION OF WATER

69

various causes ; surface-polluted water oozing into the well is probably
the commonest, but decaying animal or vegetable matter might also
raise the number of micro-organisms present almost indefinitely.
Koch's proposal for such a polluted well was to fill it up to its
highest water level with gravel, and above that, up to the surface
of the ground, with fine sand. Before the well is filled up in this
manner it must, of course, be fitted with a pipe passing to the
bottom and connected with a pump. This simple procedure of filling
up a well with gravel and sand interposes an effectual filter-bed
between the subsoil-water and any foul surface-water percolating
downwards. Such an arrangement yields as good, if not better,
results than an ordinary filter-bed, on account of there being
practically no disturbance of the bed nor injury done to it by frost.
The evidence that filter-beds remove pathogenic bacteria has not only
been demonstrated by experiment but by actual experience. At
Lawrence, Hamburg, Mount Vernon, and other towns, a marked
decline in water-borne typhoid fever has occurred as a result of
filtration.

The effect of filtration upon the number of bacteria was demon-
strated in the results which Sir Edward Frankland arrived at in his
investigation of London waters so long ago as 1887.*

Mean of Monthly Examinations for the )

car.

Name of Company.

Source of
Supply.

Micro-organisms per c.c.

Average % of

Micro-
organisms
removed l>y
Filtration.

At

Source.

After

After
Filtration.

The Chelsea Co.

West Middlesex Co. .
Southwark & Vauxhall Co.

Grand Junction Co. .

Lambeth Co.

I Thames at '\
\ Hampton /

16,138

16,138
16,138

16,138

16,138

1,067
1,788

2,500
7,820

34

58
80

f ,;-:: !
10° f

I 96 J
75

98-96

99-40
97-72

98-46

99-50

In 1899 the Massachusetts Board of Health found that by
continuous filtration through 45 inches of sand (size 0*23 nun.)
99-49 per cent, of the bacteria were removed ; and by intermittent
filtration 99-08 per cent, of the organisms were removed. In 1902
the intermittent filter removed 987 per cent of the total bacteria,
99-9 per cent. B. coli, and 100 per cent. B. typhosus. The continuous

Report on tfo Metropolitan Wain- Supply, 1887.

70 BACTERIA IN WATER

filter removed 98*7 per cent, of the total bacteria, 99'8 per cent, of
B. coli, and 99*9 of the typhoid bacillus.*

The teaching of these figures could, with great ease, be emphasised
again and again if such was necessary ; but sufficient has been said
to show that sand filtration, when carefully carried out, offers a more
or less absolute barrier to the passage of bacteria, whether non-
pathogenic or pathogenic.

Domestic Purification of Water

Something may, however, be added, from a bacteriological point
of view, relative to what is called domestic purification. There is but
one perfectly reliable method of sterilising water for household use,
viz., hoiliiu). As we have seen, moist heat at the boiling-point main-
tained for a few minutes will kill all bacteria and their spores. The
only disadvantages to this process are the labour entailed and the
"flat" taste of the water. Nevertheless, in epidemics due to bad
water, it is desirable to revert to this simple and effectual purification.

There are a large number of domestic filters on the market with,
in many cases, but little difference between them. The materials
out of which they are made are chiefly the following: carbon and
charcoal, iron (spongy iron or magnetic oxide), asbestos, porcelain
and other clays, natural porous stone, and compressed siliceous and
diatomaceous earths. From an extended research in 1894 by Prof.
Sims Woodhead and Dr Cartwright Wood, who repeated and
extended experiments by Freudenreich, Schofer, and others, our
knowledge of the quality of these substances as protectives against
bacteria has been largely increased.-)- They concluded that a filter
failed to act in one of two ways. It was either pervious to micro-
organisms, or its power of filtering became modified owing to (a)
structural alteration of its composition, or (b) to the growing through
of the micro-organisms, which had been demonstrated by previous
workers. The conditions which chiefly influence the growth of
bacteria through a filter appear to be the temperature, the inter-
mittent use of the filter, and the species of bacteria. The higher the
temperature and the longer the organisms are retained in the filter
the more likely is it that they will grow through, and in the next
usage of the filter appear in the filtrate. As to the species, those
multiplying rapidly and possessing the power of free motility will
naturally appear earlier in a filtrate than others. Woodhead and
Wood concluded that out of 18 different kinds of domestic filter, each
of which had its supporter, the Pasteur-Chamberland candle filters

* Thirty-fourth Ann. Rep. State, Bd. of Health, Massachusetts, 1903, pp. 224
and 269.

+ Brit. Med. Jour., 1894, i., pp. 1053, 1118, 1182, 1375, 1486.

DOMESTIC FILTRATION OF WATER

71

(composed of porcelain formed by a mixture of kaolin and other
clays) were the only filters out of the substances named above which
were reliable and protective against bacteria. They tested over three
dozen of the Pasteur filters, and " in every case these gave a sterile
filtrate/' Pure cholera bacillus in suspension (5000 bacilli to every
c.c.) and typhoid bacillus in suspension (8000 per c.c.) were passed
through these filters, and not a single bacillus was detectable in the
filtrate. The Berkefeld filter (siliceous earth) came second on the
list as an effective filter, and had but the one fault of not being a
"continuous" steriliser. A certain Parisian filter ("Porcelaine
d'Amiante "), made of unglazed porcelain, ren-
dered water absolutely free from bacteria. Its
action was, however, very slow. Setting aside
these three efficient filters, we are face to face
with the fact that most filters do not produce
germ-free filtrates, even though they are nomin-
ally guaranteed to do so. It is professed for
animal charcoal, which is widely used, that it
absorbs oxygen, and so fully oxidises whatever
passes through it. This may be so at first, but
after a little use it does more harm than good.
It appears to add nitrogen and phosphates to
water, which are both nutritive substances on
which bacteria grow, and it readily absorbs im-
purities from the air. As a matter of experiment
and practice, it has been found by Frankland,
Woodhead, and others, that charcoal actually
adds to the number of germs after it has beeu
in use for some time.

Subsequent experiments were made in this
country by Lunt and Horrocks. Lunt working
in 1897 investigated the power of the Berkefeld
filter to intercept pathogenic bacteria, especially
the typhoid bacillus. He concluded as a result
of his inquiry that on the first day an efficiently sterilised Berkefeld
filter gave an absolutely sterile filtrate, but that on the second or
third day of using some water bacteria passed through. For thirty-
nine days the B. coli did not pass through, though the organism could
be detected on the outside of the filtering candle. Lunt found that
the action of the filter depended very much upon its method of
use: forcing or intermittently pumping water through the filter
resulted in a filtrate containing bacterin, whilst if the same filter was
steadily used a germ-free filtrate was obtained. In short, the result
of Lunt's work was to show the necessity for a frequent sterilisa-
tion of the, filter, for though it allows ordinary water bacteria to

Fig. 10. — Partki b-

Ch W l:l.KI. \N1> Fll.l BR.

Attached to Water

Supply.

72 BACTERIA IN WATER

pass on the second or third day, B. coli and the typhoid bacillus
do not appear with them in the filtrate until a subsequent date.
Probably, if reliance is to be placed upon such a filter from a
bacteriological point of view, daily sterilisation is advisable.

Experiments of a similar nature have been done by Horrocks,*
who arrives at the following conclusions. First, the B. typhosus is not
able to grow through the walls of a Pasteur-Chamberland candle, and
if proper care be taken to prevent the direct passage of organisms
through flaws in the material and imperfections in the fittings, the
Pasteur-Chamberland filter ought to give complete protection from
water-borne disease. Secondly, typhoid bacilli can grow through the
walls of Berkefeld candles, the time required for the passage being
largely dependent on the nutriment supplied to the organisms by the
filtering fluid. Possibly the weakness of the candle from a bacteri-
ological point of view is due to the large size of the lacunar spaces,
which cannot be avoided if a fair delivery is to be obtained, but
which " appears to militate against the immobilising and devitalising
influences which operate so strongly in filters made with very narrow
lacunar spaces." Thirdly, Horrocks concluded that when a highly
polluted liquid containing typhoid bacilli is filtered through a
Berkefeld candle the bacilli may appear in the filtrate in four days.
Consequently, it is necessary to sterilise these candles every third day.
The method of sterilisation of filters is not washing or brushing or
any other kind of cleansing or soaking in water, but by exposing
them to steam or boiling them.

* Bacteriological Examination of Water, 1901, pp. 273-280.

CHAPTER III

BACTERIA IN THE AIR

Methods of Examination of Air — Conditions of Bacterial Contamination of Air :
(1) Dust and Air Pollution ; (2) Moisture or Dampness of Surfaces : Bacteria
in Sewer Air ; (3) the Influence of Gravity ; (4) Air Currents. The Relation of
Bacteria to C0.2 in the Atmosphere : in Workshops, in Bakehouses, in Rail-
way Tubes, in the House of Commons.

The basis of the usual methods in practice for bacterially examining
air is to pass the air over or through some nutrient medium. By
this means the contained organisms are waylaid, and finding them-
selves under favourable conditions of pabulum, temperature, and
moisture, commence active growth, and thus reveal themselves in
characteristic colonies. These are examined by the microscope and
sub-culture. Returns of the number of bacteria in the sample taken
may be made for the sake of information, but little or no conclusion
of value can be drawn from such data. The standard recognised in
Europe is the cubic metre or litre, and one may report, for example,
of the air of a room containing 500 or more germs per cubic metre.

Methods of Examination of Air

1. The Plate, Method. — Koch adopted the simplest of all the culture methods,
viz., exposing a plate of gelatine or agar for a longer or shorter time to the air of
which examination is desired. By gravity the suspended bacteria fall on the plate
and start growth. As a matter of quantitative exactitude, this method is not to be
recommended, but it frequently proves an excellent method for qualitative
estimation. It will be found in practice that nutrient agar is better for the purpose
than nutrient gelatine. Greater latitude is obtained both in point of temperature
and length of incubation, and the result is uncomplicated by the, at times, very
rapid liquefaction of the gelatine by liquefying organisms. Care should be taken
in preparing the plates to allow them to cool on a level surface, and at least 15 c.c.
73

74

BACTERIA IN THE AIR

of the medium should be employed for each Petri dish in order to ensure an even
surface and sufficient depth of medium all over the plate. After exposure the
plates are, under ordinary circumstances, best left at room temperature during the
development of the colonies, but if it is desired to examine the bacteria alone it will
be found well to favour the growth of these at the expense of the moulds, by first
incubating the dish at a temperature of 37° C. for, say, eighteen hours. In any
case the plate should be shielded from light, or otherwise many of the chromogenic
organisms will not assume their typical coloration. Should it be desired to photo-
graph the plates in order to obtain a permanent record, the growth should be
arrested, and the organisms killed about the third or fourth day of incubation. The
best method of doing this is to reverse the dish, and to pour upon a piece of blotting
paper placed on the inner surface of the lid, which will now be undermost, a
sufficient quantity of Formalin to saturate it. The results of this method of
examination may be expressed per square foot per minute, the area of the Petri dish

being calculated j = (radius)- x —

2. The Flask Method of Miquel. — Pasteur was the first to analyse air by the culture

method, and he adopted a plan which, in
principle, is washing the air in some fluid
culture medium which will retain all the
particulate matter, which may then be
cultured directly or sub-cultured into any
favourable medium. Miquel has con-
trived a simple piece of apparatus for the
carrying out of this principle. It consists
of a flask with a central tube through its
own neck for the entrance of the air. On
one side of the flask is a tube to be con-
nected with the aspirator, on the other
side of the flask a tube through which to
pour off the contained fluid at the end of
the process. In the flask are placed 30 c.c.
of sterilised water (or, indeed, if it be pre-
ferred, sterilised broth). The entrance
tube is now unplugged, and the aspirator
draws through a fair sample of the air in
the room (say ten litres). This air perforce
passes through the water, and by the exit
tube to the aspirator, and is thereby
washed, leaving behind in the water its
bacteria. The aspiration is then stopped,
and the entrance tube closed. The water
(plus bacteria) is now poured out into
test-tubes of media or plated out on
Petri's dishes. Provided that the appar-
atus has been absolutely sterilised, and

that only sterilised water is used, any colonies developing upon the Petri dish are

composed of micro-organisms from the air examined.

3. The Method of Hesse.— This method is somewhat akin to Pouchet's aeroscope,
but is in addition a culture method. Hesse's tube is 50-70 cm. long and 3-5 cm.
bore throughout. At one end is an indiarubber stopper bored for a glass tube to
the aspirator. The other end is open. Before using, the tube is sterilised, and 40
or 50 c.c. of sterilised gelatine are placed in it. The tube is now rapidly rotated
in a groove on a block of ice or under a cold-water tap, and by this simple means
the gelatine becomes fixed and forms a layer inside the tube throughout. We have
therefore, so to speak, a tube of glass with a tube of gelatine inside it. The
apparatus is now ready for use. It is fixed on the tripod, and 10-20 litres of air
are drawn through, and the tube is properly plugged and incubated at room
temperature. In two or three days the colonies appear upon the gelatine. They
are most numerous generally in the first part of the tube. The disadvantages of

Fig. 11.— Miquel's Flask

PLATE 5.

$*3HMirtifSt"~

Sedgwick's Sugar Turk, in position on tripod,
with siphon.

Small II \m> Ceh hum oe.

[To face page 74.

BACTERIOLOGICAL EXAMINATION OF AIR 75

this process are that dried gelatine does not catch germs like the broth cultures
of Pasteur or Miquel, and that many organisms are carried straight through the
tube, and failing to be deposited, pass out at the aspirator exit, and thus are
neither caught nor counted. The Hesse tube is generally used in practice with a
pump consisting of two flasks and a double-way indiarubber tube. The flasks have
a capacity for one litre of water. By a simple arrangement it is possible to secure
syphon action, and hence measure with considerable exactitude the amount of air
passing through the tube (Plate 5).

4. Methods of Filtration. — Frankland, Petri, Pasteur, Sedgwick, and others have
suggested the adoption of methods of filtration. These depend upon catching the
organisms contained in the air by filtering them through sterilised sand or sugar,
and then examining these media in the ordinary way. Many different kinds of
apparatus have been invented. Petri aspirates through a glass tube containing
sterilised sand, which after use is distributed in Petri dishes and covered with
gelatine. The principal objection to this method is the presence of the opaque
particles of sand in and under the gelatine. Probably it was this which suggested
the use of soluble filters like sugar. Pasteur introduced the principle, and Frank-
land and others have followed it out. Sedgwick's Tube consists of a comparatively
small glass tube, about a foot long. Half of it has a bore of 2*5 cm., and the
other half a bore of '5 cm. It is sterilised at 150° C, after which the dry, finely
granulated cane-sugar is inserted in such a way as to occupy an inch or more of
the narrow part of the tube next the wide part. Next to it is placed a wool plug,
and the whole is again sterilised. After sterilisation an indiarubber tube is fixed
to the end of the narrow portion, and thus it is attached to the aspirator. The
measured quantity (5-20 litres) of air is drawn through, and any particulate matter

K

J

I,,.. 12.— Sedgwick's Bugar-tube.

is caught in the sugar. Warm, nutrient gelatine (10-15 C.C.) is now poured into the
broad end of the tube, and by means of a sterilised stilette the sugar is pushed down
into the gelatine, where it quickly dissolves. We have- now in the gelatine all the
micro-organisms in the air which has been drawn through the tube. After plugging
with wool at both ends, the tube is rolled on ice, or under a cold-water tap, in order
to fix the gelatine all round the inner wall of the tube, which is incubated at room
temperature. In a day or two the colonies appear, and may be examined.

Frankland used finely powdered sugar and glass wool as filtering-medium, and
a tube with two constrictions. After passing sufficient air through, the tube is
broken in halves and the wool and sugar are pushed by means of a sterile needle
Into liquefied gelatine. The sugar dissolves and the organisms arc distributed in
the medium. Andrewes has used a modification of this method, and the aspiration
was carried out with a large brass syringe of known capacity, fitted with a two-way
DOZzle and cock, so that the requisite number of syringefuls could be aspirated
without disturbance.*

Various other methods, including Miquel's filtration method, and the methods
of Laveran, and Wurtz and Strauss, have been used, but the principal are those
mentioned above.

In respect of the results obtained in the examination of an- bacteriologically, it
may be said that they are twofold. First, a quantitative result is obtained" by
which we may arrive at the approximate number of bacteria and moulds. Secondly,
the quality or species of organisms is determined. Reference will be made to both
these points in the pages which follow.

■ Brit. Med. .Innr., 1902, ii., p. 1534; and Report to London County Council
1902.

76 BACTERIA IN THE AIR

Conditions of Bacterial Contamination of Air

There are, speaking in a general way, four chief external conditions
affecting the occurrence of bacteria in air. They are as follow : —

1. The presence of dust and air pollution.

2. Dampness of surfaces.

3. Gravity.

4. Air currents.

1. Dust and Air Pollution. — Schwann was one of the first to
point out that when a decoction of meat is effectually screened from
the air, or supplied solely with calcined air, putrefaction does not
set in. It is true that Helmholtz and Pasteur confirmed this, and
greatly added to our knowledge of the subject, but on the whole it
may be said that Schwann originated the germ -theory of air, and
Lister applied it in the treatment of wounds. Lister believed that
if he could surround wounds with filtered air free from dust and
particulate or germ matter, the result would be as good as if the
wounds were shut off from the air altogether.

It was Tyndall* who first laid down the general principles upon
which our knowledge of organisms in the air is based. That the
dust in the air was mainly organic matter, living or dead, was a
comparatively new truth ; that epidemic disease was not due to
" bad air " and " foul drains," but to germs conveyed in the air, was
a prophecy as daring as it was novel. From these and other like
investigations it came to be recognised that putrefaction begins as
soon as bacteria from the air gain an entrance to the putrefiable
substance, that it progresses in direct proportion to the multiplication
of these bacteria, and that it is retarded when they diminish or lose
vitality.

Tyndall made it clear that, both as regards quantity and quality
of micro-organisms in the air there neither is nor can be any
uniformity. The degree in either case will depend on air pollution
and on dust particles. Bacteria may be conducted on particles of
dust — "the raft theory" — but being themselves endowed with a
power of flotation commensurate with their extreme smallness and
the specific lightness of their composition, dust as a vehicle is not
really requisite. Nevertheless the estimation of the amount of dust
present in a sample of air may be a very good index of danger. It
is to Dr Aitken that we are indebted for devising a method by which
we can measure dust particles in the air, even though they be
invisible. His ingenious experiments, reported in the Transactions
of the Royal Society of Edinburgh (vol. xxxv.), have demonstrated
that by supersaturation of air the invisible dust particles may become
visible. As is now well known, Dr Aitken believes that fogs, mists,
* John Tyndall, F.R.S., Floating Matter of the Air, 1878.

PLATE 6.

Air-Plate exposed in Laboi rbr's Cottage in Buckinohamse
Agar culture, :: 'lays at 22 C.

(Grown and photographed by Bwithlnbank).

i ri (80 minutes).

[To fao

DUST AND AIR POLLUTION 77

and the like do not occur in dust-free air, and are due to condensation
of moisture upon dust particles. And much the same has been
found to be true in respect to dust and bacterial pollution. As a
rule, when the former is abundant the latter is considerable. Haldane
and Osborn (vide infra) found bacteria most numerous in a workplace
where dust was most abundant, and their finding was merely con-
firmatory of many other previous researches. On the other hand
it should be remembered that, though dust forms a vehicle for
bacteria, dusty air is sometimes comparatively free from bacteria.

For the conditions which affect the number of bacteria in the
air are various. In open fields, free from habitations, there are
fewer, as would be expected, than in the vicinity of manufactures,
houses, or towns. A dry, sandy soil or a dry surface of any kind
will obviously favour the presence of organisms in the air. Frank-
land found that fewer germs were present in the air in winter than in
summer, and that when the earth was covered with snow the number
was greatly reduced. Miquel and Freudenreich have declared that
the number of atmospheric bacteria is greater in the morning and
evening between the hours of six and eight than during the rest of
the day.

There is no numerical standard for bacteria in the air as there is
in water. In houses and towns it would rise according to circum-
stances, and frequently in dry weather reach thousands per cubic
metre. When it is remembered that air possesses no pahulum for
bacteria, as do water and milk, it will be understood I hat bacteria
do not live in the air. The quality and quantity of air organisms
depend entirely upon environment and physical conditions. In some
researches which the writer made into the air <»f workshops in Soho
in 1896, it was instructive to observe that fewer bacteria were isolated
by Sedgwick's sugar-tube in premises which appeared to the naked
eye polluted in a larger degree than in other premises apparently less
contaminated. In the workroom of a certain skin-curer the air was
densely impregnated with dust particles from the skin, yet scarcely
a single bacterium was isolated. Macfadyen and Lunt have also
found that the number of dust particles does not bear any relatioD l<>
the number of bacteria, They found thai air containing oven
millions of dust particles might be almosl germ-free. In the polish-
ing room of a well-known hat firm, in which the air appeared to the
naked eye to be pure, and in which there was ample ventilation,
there were found by the writer numerous bacteria belonging to four
or five species of saprophytes. The public analyst for the city of
Nottingham, estimated the bacterial quality of the air of the streets
of that town during "the goose fair" held in the autumn. He
used a modification of Hesse's apparatus, in which the gelatine is
replaced by glycerine. The air was slowly drawn through and

78 BACTERIA IN THE AIR

measured in the usual way. Sterilised water was then added to bring
the glycerine to a known volume, the liquid thoroughly mixed, and a
series of gelatine and agar plates made with quantities varying from
01 to 2 c.c. By this method a large number of bacteria were
detected in this particular investigation, including Staphylocorcus
pyogenes aureus and albus, the common Bacillus sublilis, and, appar-
ently, B. coli communis* Carnelly, Haldane, and Anderson found
11 bacteria per litre of air in a classroom of the High School at
Dundee with the boys at rest. But when the boys were instructed
to stamp on the floor, and thus raise the dust, the number rose to 150
bacteria per litre.

During a six years' investigation the air of the Mont Souris
Park yielded, according to Miquel, an average of 455 bacteria per
cubic metre. In the middle of Paris the average per cubic metre
was nearly 4000. Fliigge accepts 100 bacteria per cubic metre as a
fair average. From this fact he estimates that " a man during a life-
time of seventy years inspires about 25,000,000 bacteria, the number
contained in a quarter of a litre of fresh milk."f Many authorities
would place the average much below 100 per cubic metre, but even
if we accept that figure it is at once clear how relatively small it is.
This comparative freedom from bacteria is due to sunlight, rain,
desiccation, dilution of air, moist surfaces, etc. So essentially does
the bacterial content of air depend upon the facility with which
certain bacteria withstand drying that Dr Eduardo Germano has
addressed himself first to drying various pathogenic species and then
to mixing the dried residue with sterilised dust and observing to
what degree the air becomes infected.^: The typhoid bacillus appears
to withstand comparatively little desiccation, without losing its viru-
lence. Never theless it is able to retain vitality in a semi -dried con-
dition, and it is owing to this circumstance, in all probability, that it
possesses such power of infection. The bacillus of diphtheria, on the
other hand, is capable of lengthened survival outside the body,
particularly when surrounded by dust. The question of its power of
resistance to long drying is an unsettled point. The power of
surviving a drying process is, according to Germano, possessed by the
St rcijtococcus pyogenes. This is not the case with the organisms of
cholera or plague. Dr Germano classifies bacteria, as a result of his
researches, into three groups : first, those like the bacilli of plague,
typhoid, and cholera, which cannot survive drying for more than a
few hours ; second, those like the bacilli of diphtheria and strepto-
cocci, which can withstand it for a longer period ; thirdly, those like
the tubercle bacillus, which can very readily resist drying for months

* Public Health, vol. x., No. 4, p. 130 (1898).

t Fliigge, Grmulrixs der Hygime, 1897, pp. 161, lb2.

X Zeitschrift far Hygiene, vols, xxiv.-xxvi.

DAMP SURFACES 79

and yet retain their virulence. It will be obvious that from these
data it is inferred that Groups 1 and 2 are rarely conveyed by the air,
whereas Group 3 is frequently so conveyed. Miquel has recently
demonstrated that certain soil bacteria or their spores can remain
alive in dried dust in hermetically sealed tubes for as long a time as
sixteen years. Even at the end of that period such soil inoculated
into a guinea-pig produced tetanus.

The presence of pathogenic bacteria in the air is, of course, a
much rarer contamination than the ordinary saprophytes. The
tubercle bacillus has been not infrequently isolated from dry dust
in consumption hospitals, and in exit ventilating shafts at Brompton
the bacillus has been found. From dried sputum, it has, of course,
been many times isolated, even after months of desiccation. Indeed,
a very large mass of experimental evidence attests the fact that the
air in proximity to dried tubercular sputum or discharges may
contain the specific bacillus of the disease. The bacillus of diphtheria
in the same way, but in a lesser degree, may be isolated from the
air, and from the nasal mucous membrane of nurses, attendants, and
patients in a ward set apart for the treatment of the disease, and
from the throats and nasal mucous membrane of persons who have
been in contact with cases of the disease. Delalivcsse, examining
the air of wards at Lille, found that the contained bacteria varied
more or less directly with the amount of floating matter, and
depended also upon the vibration set up by persons passing through
the ward and the heavy traffic in granite-paved streets adjoining.
B. coli, staphylococci, and streptococci, as well as B. tuberculosis,
were isolated by this observer. Other observers have found B. <<>/{
very rarely present in air (Chick, Andrewes, etc.).

2. Moisture or Dampness of Surfaces. — It is an interesting
and important fact that except under special circumstances micro-
organisms do not leave moist surfaces, but remain adhering to them.
A clear recognition of this fact is essential to a right understanding
of the pollution of air by bacteria. They cannot Leave the moist
surface of fluids either under evaporation or by means of air currents.*
Only when there is considerable molecular disturbance, such as
splashing, can microbes be transmitted to the surrounding air.
This is the reason why sewer gas and ;ill an- contained within moist
perimeters is almost germ-free^ whereas from dry surfaces the Least
air current is able to raise countless numbers of organisms.

This principle has been admirably illustrated in investigations
made upon expired and inspired air. In a report to the Smithsonian
Institute of Washington (1895) upon the composition of expired air,

Fliigge has lately attempted to demonstrate that an air current having a
velocity of four metres per second can remove bacteria from surfaces of liquids by
detaching drops of the liquid itself.

80 BACTERIA IN THE AIR

it is concluded that "in ordinary quiet respiration no bacteria
epithelial scabs or particles of dead tissue are contained in the
expired air. In the act of coughing or sneezing such organisms or
particles may probably be thrown out." The mucous membrane
lining the cavity of the mouth and respiratory tract is a moist
perimeter, from the walls of which no organisms can rise except
under molecular disturbance. The popular idea that bacteria can
be "given off by the breath" is therefore contrary to the laws of
organismal pollution of air. The required conditions are not
fulfilled, and such breath infection must be of extremely rare
occurrence except in speaking, spitting, sneezing or coughing
(Fliigge). Air can only become infective when impregnated with
organisms arising from dried surfaces.

Another series of investigations were conducted by Drs Hewlett
and St Clair Thomson, and dealt with the fate of micro-organisms
in inspired air and micro-organisms in the healthy nose. They
estimated that from 1500 to 14,000 bacteria were inspired every
hour. Yet, as we have seen, expired air contains practically none
at all. It is clear, then, that the inspired bacteria are detained some-
where. Lister has pointed out, from observation on a pneumo-thorax
caused by a wound of the lung by a fractured rib, that bacteria may
be arrested before they reach the air cells of the lung, and other
observations confirm this fact, although of course there are several
well-known exceptions {e.g. tubercle of the lung). Hence it is at
some intermediate stage that they are detained. Hewlett and
Thomson examined the mucus from the wall of the trachea, and
found it germ-free. It was only when they examined the mucous mem-
brane and moist vestibules and vibrissse of the nose that |they found
bacteria. Here they were present in abundance. The ciliated
epithelium, the mucus, and the bactericidal influence of the wandering
or " phagocyte " cells, probably all contribute to their final removal.*

There can be no doubt that the large number of bacteria present
in the moist surfaces of the mouth is the cause of a variety of
ailments, and under certain conditions of ill-health organisms may
through this channel infect the whole body. Dental caries will occur
to everyone's mind as a disease probably due in part to bacteria. As
a matter of fact, acids (due to acid secretion and acid fermentation)
and micro-organisms are two of the chief causes of decay of teeth.
Defects in the enamel, inherent or due to injury, retention of debris
on and around the teeth, and certain pathological conditions of the
secretion of the mouth, are predisposing causes, which afford a

* Hewlett and Thomson graphically demonstrated the bactericidal power of the
nasal mucous membrane by noting the early removal of Bacillus prodigiosus,
which had been purposely placed on the healthy Schneiderian membrane of the

DUST AND AIR POLLUTION 81

suitable nidus for putrefactive bacteria. The large quantities of
bacteria which a decayed tooth contains are easily demonstrated.

From the two series of experiments which we have now con-
sidered we may gather the following facts: —

(a) That air may contain great numbers of bacteria which may
be readily inspired.

(b) That in health those inspired do not, as a rule, pass beyond the
moist surface of the nasal and buccal cavities, except in persons who
practice oral instead of nasal respiration.

(c) That in the nose and mouth there are various influences of a
bactericidal nature at work in defence of the individual.

(d) That expired air in normal quiet breathing contains, as a
rule, no bacteria whatever.

The practical application of these things is a simple one. To
keep air free from bacteria, the surroundings must be moist. Strong
acids and disinfectants are not required. Moisture alone will be
effectual. Two or three examples at once occur to the mind.
Anthrax spores are conveyed from time to time from dried infected
hides and skins to the hands or bodies of workers in warehouses in
Bradford, Bermondsey, Finsbury, and other places. If the surround-
ings are moist and the hides moist, anthrax spores and other bacteria
do not remain free in the air. As a matter of actual experience, it has
been found that handling dried hair or dried skins leads to more anthrax
infection than handling the same articles in a moist condition.*

Again, the bacilli (or "spores") of tuberculosis present in sputum
in great abundance cannot infect the air until and unless the
sputum dries. So long as the expectorated matter remains on the
pavement or handkerchief wet, the surrounding air will derive from
it no bacilli of tubercle. But when in the course of time the sputum
dries, then the least current of air will at once in tret itself with the
dried spores or bacilli. It should, however, be remembered thai the
"cough-spray" and microscopic particles of saliva emitted in shout-
ing, heavy breathing through the mouth, etc., have been shown by
Fliigge and others to carry the bacilli of tubercle. Such conveyance
may, of course, prove a channel of infection betweeu diseased and
healthy persons. The typhoid bacillus, too, occupies the same position.
Only when the excrement dries can the contained bacteria infect
the air. It is of course well known that the common channel of
infection in typhoid fever is, not the air, whereas the reverse holds
true of tuberculosis. But if it happens thai the excrement of
patients suffering from typhoid dries, the air may become infected;
if, on the other hand, it passes in a moist state into the sewer, even
though untreated with disinfectants, all will ho well as regards the
surrounding air.

Annual Reports of Medical Inspector of Factories and Workshops, 1902 and 1903.

F

82 BACTERIA IN THE AIR

A still more remarkable illustration of the effect of a moist
perimeter upon the contained bacteria is to be found in sewer air.
For long it has been known that air polluted by sewage emanations
is capable of giving rise to various degrees of ill-health. These
chiefly affect two parts of the body; one is the throat and the other
the intestine. Irritation and inflammation may be set up in both or
either by sewer air. Such conditions are in all probability produced
by a lowering of the resistance and vitality of the tissues, and not by
a conveyance of bacteria in sewer air or by any stimulating effect
upon bacteria exercised by sewer air. What evidence we have is
against such factors. Several series of investigations have been
made into the bacteriology of sewer air, amongst others by Uffel-
mann, Carnelly and Haldane, and Laws and Andrewes. From their
labours we may formulate four simple conclusions : —

1. The air of sewers contains very few micro-organisms indeed,
sometimes not more than two organisms per litre (Haldane), and
generally fewer than the outside air (Laws and Andrewes).

2. There is not, as a rule, intimate relationship between the
microbes contained in sewer air and those contained in sewage.
Indeed, there is a marked difference which forms a contrast as
striking as it is at first sight unexpected. The organisms isolated from
sewer air are those commonly present in the open air. Micrococci
and moulds predominate, whereas in sewage moulds and micrococci
are rare, and bacilli are most numerous. Liquefying bacteria, too,
which are common in sewage, are extremely rare in sewer air.
Bacillus coli communis, which occurs in sewage from 20,000 to
200,000 per c.c, is altogether absent from sewer air.

3. As a rule it may be said that only when there is splashing in
the sewage, or when bubbles are bursting (Carnelly and Haldane), is
it possible for sewage to part with its contained bacteria to the air
of the sewer. But under these conditions it may part with a
considerable number.

4. Pathogenic organisms and those nearly allied to them are
found in sewage, but are absent in sewer air. Uffelmann isolated
the Staphylococcus pyogenes aureus (one of the organisms of sup-
puration), but such a species is exceptional in sewer air. Hence,
though sewer air is popularly held responsible for directly conveying
virulent micro-organisms of various diseases, there is up to the
present no evidence of a substantial nature in support of such views.
In 1894, Laws and Andrewes found an average of 2,781,650 bacteria
per c.c. in fresh sewage, and in older sewage from 3,400,000 per c.c,
to 11,216,000, and they pointed out that temperature and dilution
of sewage were determining factors in the number of bacteria present.
They consider that sewage may become a medium for the dissemina-
tion of the typhoid bacillus, and that sewage-polluted soil may possibly

BACTERIA IN SEWER AIR 83

give up germs to the subsoil air, but they are satisfied that the air
of sewers themselves does not play any part in the conveyance of
the typhoid bacillus.*

In passing, mention may be made of some interesting observations
recorded by Mr S. G. Shattock on the effect of sewer air upon the
toxicity of lowly virulent bacilli of diphtheria. Some direct relation-
ship, it has been surmised, exists between breathing sewer air and
"catching" diphtheria. Clearly, it cannot be that the sewer air
contains the bacillus. But some have supposed that the sewer air
has had a detrimental effect by increasing the virulent properties of
bacilli already in the human tissues. Two cultivations of lowly
virulent B. diphtheria? were therefore grown by Mr Shattock in flasks
upon a favourable medium over which was drawn sewer air. This
was continued for two months in the one case, and five weeks in the
other. Yet no increased virulence was secured.f Such experiments
require ample confirmation, but even now it may be said that sewer
air does not necessarily have a favouring influence upon the virulence
of the bacilli of diphtheria. Such experiments do not affect the
contrary question of the possibility of sewer air depressing the vitality
of the individual, and so allowing even lowly virulent bacilli to do
mischief. Of such depression caused by breathing sewer air there
is clinical proof, and although sewer-men do not appear to be affected,
persons freshly breathing sewer air may be.

It should be noted that the bacilli of diphtheria are capable of
lengthened survival outside the body, and are readily disseminated
by very feeble air - currents. The condition necessary for their
existence outside the body for any period above two or three days
is moisture. Dried diptheria bacilli soon lose their vitality. It is
possible, owing to this fact, that the disease is not as commonly con-
veyed by air as, for example, tubercle.

3. The Influence of Gravity upon bacteria in the air may be
observed in various ways, in addition to its action within a limited
area like a sewer or a room. Miquel found in some investigations
in Paris that, whereas on the Rue de Rivoli 750 germs were present
in a cubic metre, yet at the summit of the Pantheon only 28 were
found in the same quantity of air. Frankland found that air at the
top of Primrose Hill contained 9 organisms per ten litres, and air
at the bottom 24. On the spire of Norwich Cathedral (310 feet),
ten litres of air yielded 7 organisms, on the tower (180 feet) 9, and
on the ground 18. At the level of the golden gallery of St Paul's
Cathedral he found in every ten litres 11 bacteria, at the stone gallery
34, and in St Paul's Churchyard 70. As Tyndall has pointed out,

Report to the London County Council on the Result of Investigation on the Micro-
organisms of Sewage, by J. Parry Laws and F. W. Andrewes, 1894, p. 14.
r Pathological Society of London, Transactions* 1897.

84

BACTERIA IN THE AIR

even ultra-microscopic cells obey the law of gravitation. This is equally
true in the limited areas of a laboratory or warehouse, and in the open
air. At high altitudes, the air may be looked upon as practically
germ-free, although here again the lighter spores of the mould fungi
may cause them to be carried by air currents to a very great height.
In the recent researches of Dr Jean Binot of the Pasteur Institute,*
100 litres of air taken at the summit of Mont Blanc did not contain
a single microbe, and the total number of organisms varied between
4 and 11 per metre cube (1000 litres). An examination of the air of
the interior of M. Janssen's Observatory, situated on the highest
point of Mont Blanc, and taken in two different rooms, gave, on the
other hand, 540 and 260 organisms per metre cube. The gradual
increase of the number of organisms as descent to lower level takes
place is of interest. Thus 6 per metre cube were found in the Grand
Plateau, 8 at the Grand Mulet, and 14 at the Plon de PAiguille.
Upon the Mer de Glace 23 organisms were found, and 49 at Montan-
vert. Graham Smith found that at the top of the Clock Tower of
the Houses of Parliament in London there was only about one-third
of the number of bacteria found at the ground level.f

4. Air Currents. — Miquel, Pasteur, Cornet, and other workers
have shown that the presence of micro-organisms in air depends in
part upon air currents, winds, etc. In the month of August, with
the wind from the south, i.e. blowing from the country citywards, the
number of organisms was found by Miquel to be 40 in the Mont Souris
Pare around the Observatory, while at the same moment a record of
14,800 was obtained in the 4th Arrondissement, which may be taken
as the centre of Paris, and comprises the surroundings of Notre Dame
and of the Hotel de Ville. In the month of June, on the other hand,
with the wind blowing from the N.E., i.e. across the city towards
Mont Souris, the numbers were, in the 4th Arrondissement, 10,000 per
metre cube, and in the Park of Mont Souris itself, 1180 per metre cube.

The seasonal variations of the organisms present in the air are
also worthy of note, and depend chiefly upon dust and air currents.
The following table shows the mean over a period of ten years in the
air taken at Mont Souris : —

Average per metre cube.

Season.

Bacteria.

Moulds.

Winter .
Spring .
Summer
Autumn

170
327
480
195

175
145
210
235

* Communication a V AcacUmie des Sciences de Paris, 17 Mars 1902.
t Jour. ofHyg., 1903, p. 513.

INFLUENCE OF CARBONIC ACID GAS 85

Similar experiments have been carried out by Frankland, Fliigge,
Delalivesse, Neisser, Chick, Andrewes, and others.* The last named
conducted some experiments in London streets in 1902, and reported
his results to the Pathological Society. He found the number of
organisms varied greatly, but no pathogenic species were detected.
The four species he isolated were staphylococci, sarcinae, strepto-
thricese, and moulds.

Carnelly, Halclane, and Anderson found the ratio of organisms
in the air increased according to whether the air was examined on
still damp days, windy damp days, still dry days, and windy dry
days, and in brief this expresses the findings of most investigators.

Some new light has been thrown upon the subject of pathogenic
organisms in air by Neisser in his investigations concerning the
amount and rate of air currents necessary to convey certain species
through the atmosphere. He states that the bacteria causing
diphtheria, typhoid fever, plague, cholera, and pneumonia, and
possibly the common Streptococcus pyogenes, are incapable of being
carried by the molecules of atmospheric dust which the ordinary
insensible currents of air can support, whilst Bacillus anthracis, B.
pyocyaneus, and the bacillus of tubercle are capable of being aerially
conveyed. This work will require further confirmation before being
entirely accepted.

Finally, some mention may be made of the relationship alleged
to exist between the presence of a considerable degree of carbonic
acid gas in an atmosphere and the number of bacteria contained in
the same atmosphere. As far as may be judged, it would appear
that the relationship is but slight. But to illustrate the subject as
well as other points of importance in the bacteriology of air, four
separate investigations may be mentioned.

(i.) Haldane and Osborn, in their inquiry into the ventilation of
factories and workshops, made a number of bacteriological examina-
tions.! The determinations of bacteria wore made by a slightly modified
form of Frankland's method. The air was drawn through a sterilised
plug of glass wool by means of a brass syringe of known capacity.
The glass tubes containing the glass wool plugs were each enclosed
in a separate outside sterilised glass tube, with an asbestos plug.
In taking the sample of air the inside tube was attached directly to
the pump by means of a short piece of stout rubber tubing. The
plug was afterwards transferred with the necessary precautions to a
shallow, flat-bottomed flask, containing a small quantity of liquefied
gelatine, which was shaken so as to disintegrate and spread the glass

* See also Jour, of Sanitary Institute (Oct 1002), vol. xxiii., pt iii., p. 209-236.
The Dust Problem, by Sir J. Crichton-Browne, P.R.S.

I First Report of the Departmental Committee appointed to inquire into the
Ventilation of Factories and Workshops, 1002. J. S. Haldane, M.D., F.R.S., and
E. II. Osborn.

86

BACTERIA IN THE AIR

wool. The gelatine having set, the flask was incubated at 20c C. till no
further colonies of bacteria or moulds developed. Some of the chief
results were as follows : —

CO.,

Bacteria and Moulds

Cub.

per 10,000" parts.

per Litre of Air.

Content.

Inside.

Outside Air.

Bacteria.

Moulds.

Tailor, Whitechapel .

67,500

35-8

3-5

17

T2

»» »> • •

21,953

9'2

3-5

8

1

»» i» • • •

2,750

4'6

3-5

16

2

»» >» • • •

18,636

10-0

3-5

9

8

?» »» • • •

9,800

7'4

3-5

9

0

,, London, E. .

27,265

14-6

3-5

10

2

,, London, E.C.

26,460

14-6

3-5

25

3

Capmaker, London, E.

4,296

23*0

3-5

9

2

Dressmaker, London, W.

21,600

13-2

3-5

8

0

Boot Workshop, London, E. .

8,688

8'8

3-5

25

6

Railway Works, Wilts. .

93,786

4'6

3*5

20

2

Chocolate Factory, Bermondsey

12,000

6'2

3-5

8

0

Newspaper Printer, Lond. , E.C.

24,098

16-5

3-5

9

0

»» »» »»

45,259

15-2

3*5

6

6

»» »» »>

23,562

25'4

3-5

10

2

Ropemaker, Chatham * .

20

6

»» »» •

82

8

»» »» • •

850

18

* The ventilation of this large room was considerable, but having the effect of keeping dust in
suspension rather than expelling it from the room. Three tests made here were all in the same work-
place, differing only in degree of dust present.

(ii.) In 1902 the writer made some observations in Finsbury on the
number of bacteria to be found in the air of underground bakehouses.
Four were selected, and the degree of carbonic acid gas was estimated
by Pettenkofer's method, and examinations were made as follow of
the bacteria pollution. In each of these bakehouses, whilst work
was going on, three agar-plates (of 9 '6 inches area each) were exposed
for thirty minutes. One plate was placed on the floor, one on the
table or trough where the bread was being made, and one on a shelf
near the ceiling. After exposure for thirty minutes the plates were
re-covered and incubated at blood-heat (37° C), for exactly twenty-
two hours. All the plates then showed abundant growth. Doubtless
if the plates had been incubated for forty-eight hours, or three or
four days, there would have been a greater growth of colonies, and
it is probable also that if some of the plates had been placed at room
temperature certain bacteria would have grown which did not appear
at blood-heat in twenty -two hours. It is not suggested that these
plates provide an adequate record of the bacteria present in the air
of these bakehouses. The object was merely to obtain a comparative

AIR-PLATES EXPOSED IN BAKEHOUSES (30 minutes).

PLATE 7.

(i.) Above-Ground Bakehouse (Z.) Agar culture, 2-2 hours at 37 C.

(ii.) Under-ground Bakehouse (0.) Agar culture, 22 hours at 37" C

[ To fact page *<'>.

BACTERIA IN BAKEHOUSE AIR

87

idea of the air of underground bakehouses and above-ground bake-
houses in Finsbury. Accordingly, the whole of the 30 plates used
in this examination were treated exactly the same in every way, the
medium, exposure, and temperature and period of incubation being
precisely similar. The results, therefore, whilst of little value as a
complete examination of the air, are useful and reliable for comparison
with each other.

The results were as follow : —

Carbonic Acid Gas

Average No. of

in parts

Bacteria

per 10,000 (Colwell).

on each Plate.

Underground Bakehouse

B

12-0

800

55 »»

C

17-5

680

'5 »»

D

16-9

600

E

13-6

600

Above-ground Bakehouse

Z

4-9

200

Outside Air in street

(C) .

4-5

160

Inside the bakehouses there was also an interesting distribution
of bacteria as follows : —

No. of Bacteria

per Plate on
Shelf.

No. of Bacteria

per Plate on

Table.

No. of Bacteria

per Plate on

Floor.

Underground Bakehouse C
Above-ground Shop of C
Above-ground Bakehouse Z

490
130
150

150
170*

850
720
300

1 Illustrations of these two plates arc attached (\

From these figures it will be seen (a) thai underground bakehouse
air contained at least four times more bacteria than sired air around
it; (ft) at least three times more bacteria than the air of* the Bhop
over it; and (c) at least three times more bacteria than the above-
ground bakehouse. The general result of the investigations was that
the air of the typical underground bakehouses examined (1) contained
14-8 volumes per 10,000 of carbonic acid gas, CO, (as compared with
49 in above-ground bakehouses and 4*3 in the streets of Finsbury);
(2) that it contained between 10 and 24 per cent, less moisture than
outside air surrounding the bakehouses; and (3) that it contained at
least four times more bacteria than surrounding street air, and three

88

BACTERIA IN THE AIR

times more bacteria than the air of a typical above-ground bake-
house.*

Dr Scott Tebb has also made a somewhat parallel examination
of the air of London streets as compared with the railway tube of
the City and South London railway.^ As the result of a large
number of investigations carried out in a similar way to the writer's
examinations in bakehouses, the following figures were arrived at : —

Air of

co2

per 10,000 parts.

No. of

Micro-organisms

per Plate.

The open streets ....
Platforms in Railway Tube
Railway Carriages in Tube

3'8

7-9

11-6

459
114
218

(hi.) Thirdly, some of the results of the investigations of Graham
Smith into the condition of the atmosphere of the House of Commons
may be mentioned.^ He used a modification of Frankland's method of
filtering the air to be examined (45 litres in each case) through glass
wool. An air-pump and a rubber tube of 10 feet in length were
used for drawing the air through, and gelatine was used as the
medium, the cultures being incubated at 20° C. for five days or
longer. The results may be expressed in tabular form in three
series : —

Experiments on Outside Air, ISth July.

Position.

No. of

Bacteria and Moulds

per litre.

No. of Moulds

only

per litre.

1. Terrace (ground level)

(10 feet from ground)
(20 feet from ground)
Clock Tower (half-way up)

Peers' Inner
Star Court

(top)

ourt

4*2
2'9
3'3
1-5
1-3
4'2
6*0

1-1

l'l
2-0
0-2
0*6
0*6
1-3

Similar experiments were performed in the House itself during a

* Report on Bakehouses in Finsbury (Newman), 1902, p. 51.

t Report of Public Analyst of Southwark on Condition of Air on City and South
London Raihoai/, 1903. W. Scott Tebb, M.D.
t Jour. ofHyg., 1903, pp. 498-513.

OF HOUSE OF COMMONS

89

sitting. The air as it entered the House contained 2 '6 bacteria and
moulds per litre.

Experiments in Debating Chamber, 21st July. Bacteria and Moulds per litre.

Position of Examination.

1. Government Side (third seat) .

2. ,, ,, (back seat) .

3. Opposition Side (third seat) .

4. Equalising Chamber

(Air before it passes into Debating
Chamber.)

5. Roof

Early

Early

Late

Late

Series.

Series.

Series.

Series.

7 P.M.

7.15 p.m.

10.30 p.m.

10.45 p.m.

10*6

5'2

7-0

6-0

5-1

4'4

4-4

5*2

6'2

5*2

5-7

8-0

9'2

8-4

7-7

8-8

8*2

7-0

6-4

Average

of all

Experiments.

7'2
4-8
5'4
8'3

7*6

A third series of examinations was made by Graham Smith of
the air in certain committee rooms, etc., as follows : —

Experiments in Committee, Dining, and Smoking Rooms.

Position of Examination.

No. of

Bacteria and Moulds

par litre.

No. of Moulds

only

per litre.

1. Committee Room 9, fans working, 1">0

persons present, 1.45 P.w

2. Committee Room 9, fans working, 150

persons present, 1.45 p.m. .

3. Committee Room 1, fans not working, 11

persons present, 1.30 p.m. ....

4. Committee Room 1, fans not working, 11

persons present, 1.30 p.m

5. Central Dining-room, 36 persons present,

8 p.m.

6. Central Dining-room, 86 persons present,

8 P.M

7. Members' Smoking-room, 2 1 persons present,

9 p.m.

8. Members* Smoking-room, 24 persons present,

9 P.M

13*3

20-9
35 '5
33 '7

n -a

1 1-2

30*6

3-6

1-0
I-.;
.v:;
l-j
3-4
12*0
10*6

Separate investigation as to the degree of C02 present in llic
Debating Chamber of the House of Commons revealed between 3-4
pints per 10,000.

Dr Graham Smith, as a result of bis investigations, arrived at
the following conclusions: —

1. The number of micro-organisms in the open space surrounding
the Eouse of Parliamenl is comparatively small (4'2 per litre).

90

BACTERIA IN THE Altl

2. The air in the debating chamber is from a bacteriological
point of view remarkably pure (5*8 per litre as average of eleven
experiments).

3. The number of bacteria found in the committee, dining, and
smoking rooms was several times greater than in the chamber (32'3
per litre as average of six experiments).

4. No organisms pathogenic to man were isolated, and only a few
which were pathogenic to animals.

(iv.) Fourthly, in 1902 Andrewes furnished a report to the London
County Council on the micro-organisms present in the air of the tube
of the Central London Kail way.* The method he employed was
in principle that of Frankland, viz., the aspiration by means of a brass
syringe (capacity 425 c.c.) of a known volume of air (5 litres), through
a plug of glass wool and finely-powdered cane-sugar. The latter
retains the micro-organisms, which can be subsequently distributed
through a suitable cultivating medium (such as gelatine) in a Petri
dish. The gelatine plate-cultures were incubated at 20° C. for four
days, when the colonies were counted, examined, and sub-cultured.
Special control experiments were made, and search was also made
for the presence of anaerobic organisms. The twelve series of
experiments yielded results which may be abstracted and tabulated
as follow : —

u
a

03

e

03

O

03

tD 33 o

* o »

O
Q

2

a

to

03

Q
A

a

a

2

ci

«

U

<D

a
O

o

U 03 o

£ *g

O o rH

3 £
to
u
O

d

o

d

o

O

d

'A

1. Platform .

11.30 a.m.

66

30*3

•09

•0019

14

13

2. Lift .

11.45 a.m.

61

30*5

•109

•0028

20

19

3. Carriage (smoking)

11.45 a.m.

70

29'5

•108

•0026

51

10

4. Tunnel . .

11.45 a.m.

67

30'2

•082

•0010

10

8

5. Carriage (non-smoking)

2.45 p.m.

68

30*5

•111

•0027

13

14

6. Platform .

5.10 p.m.

67

30-1

•103

•0012

30

11

7. Platform .

11.40 a.m.

68

30*4

•094

•0010

106

6

8. Carriage (non-smoking)

11.20 a.m.

72

30-2

•134

•0016

90

13

9. Tunnel

11.15 a.m.

70

30-2

•104

•0018

10

4

10. Lift ....

11.0 a.m.

66

29-9

•152

•0042

64

9

11. Staircase, Passage

11.0 a.m.

64

29*8

•078

•0013

13

3

12. Staircase, Passage

11.0 a.m.

68

30-0

•102

•0023

17

7

* Examination of the Atmosphere of the Central London Railway, London County
Council, 1902. No. 615. F. W. Andrewes, M.D., F.R.C.P.

OF CENTRAL LONDON RAILWAY 91

By way of summary, it may be said that Andre wes found that
micro-organisms were present in the air of the Central London
Eailway in a somewhat greater proportion (as 13 to 10) than in the
fresh air outside. The number was high in proportion to the
concentration of human traffic, being highest in carriages, platforms,
and lifts. The tube air does not compare unfavourably with that
known to exist in ordinary dwelling-rooms. No pathogenic germs
were discovered, though the number of organisms capable of growing
at body temperature was greater in the tube air than in the outside
air, and the number of organisms in the tube air was found generally
proportional to the degree of chemical contamination, but this rule
was subject to striking exceptions. It is evident that bacterial
contamination of air, though, as a rule, jmrallel to chemical con-
tamination, may yet vary quite independently as the result of special
conditions, such as air currents, which indicates that chemical
examination alone cannot always be taken as a trustworthy guide to
the contamination of air.

The species of bacteria which Andre wes found in the railway
air were in the main identical with those occurring in fresh air,
and included Staphylococcus cereusfiavus ct alius, Micrococcus candicans,
M. flavus, M. citrcus, M. lactis, M. albicans tardissimus, Sarcina lutea,
S. flava, S. alba, Bacillus luteus, B. lactis innocuus, Strcptothrix
Fbrsteri, S. chromogencs, S. albido-flava, Torula alba, and Saccharomyces
ccrevisio3.

Interpretation of Reports on Bacterial Content of Air

In the present position of our knowledge of the bacteriology of
air, reports are only of comparative value. Mere numbers of sapro-
phytic bacteria in air are not of great service. Up to the present it
has not been possible to isolate pathogenic organisms, though such
must inevitably occur in air under certain circumstances, though eveu
then probably only in very small numbers. To detecl pathogenic
organisms, it will probably be necessary to examine large volumes of
air, and by methods which will eliminate the common saprophytes.
The truth is that the foundations of our knowledge concerning the
bacteria] flora of the air are* only beginning to be laid, ami until we
can detect, by bacteriological examination, organisms of disease, the
bacteriology of air can only be a subject of relative importance.

CHAPTER IV

BACTERIA AND FERMENTATION

Early Work — Kinds of Fermentation: (1) Alcoholic Fermentation, Ascospores,
Pure Cultures, Films ; (2) Acetous Fermentation ; (3) Lactic Acid Fermenta-
tion ; (4) Butyric Fermentation ; (5) Ammoniacal Fermentation — Diseases of
Wine and Beer: Turbidity, Ropiness, Bitterness, etc. — Industrial Applica-
tions of Bacterial Ferments.

It was Pasteur who, in 1857, first propounded the true cause and
process of fermentation. The breaking down of sugar into alcohol
and carbonic acid gas had been known, of course, for a long period.
Since the time of Spallanzani (1776) the putrefactive changes in
liquids and organic matter had been prevented by boiling and subse-
quently sealing the flask or vessel containing the fluid. Moreover,
this successful preventive practice had been in some measure
correctly interpreted as due to the exclusion of the atmosphere, but
wrongly credited to the exclusion of the oxygen of the air. It was
not until the beginning of the present century that authorities modi-
fied their view and declared in favour of yeast cells as the agents in
the production of fermentation. That this process was due to
oxygen per se was disproved by Schwann, who showed that so long
as the oxygen admitted to the flask of fermentable fluid was sterilised
no fermentation occurred. It was thus obvious that it was not the
atmosphere or the oxygen of the atmosphere, but some fermenting
agent borne into the flask by the admission of unsterilised air. It
was but a step further to establish this hypothesis by adding
unsterilised air plus some antiseptic substance which would destroy
the fermenting agent. Arsenic was found by Schwann to have this
germicidal property. Hence Schwann supported Latour's theory

5)2

WORK OF PASTEUR 93

that fermentation was due to something borne in by the air, and that
this something was yeast.

Passing over a number of counter-experiments of Helmholtz and
others, we come to the work of Liebig. He viewed the transforma-
tion of sugar into alcohol and carbonic acid gas simply and solely as
a non-vital chemical process, depending upon the dead yeast com-
municating its own decomposition to surrounding elements in contact
with it. Liebig insisted that all albuminoid bodies were unstable,
and if left to themselves would fall to pieces — i.e. ferment — without
the aid of living organisms, or any initiative force greater than
dead yeast cells. It was at this juncture that Pasteur intervened to
dispel the obscurities and contradictory theories which had been
propounded.

As in all the conclusions arrived at by Pasteur, so in those relat-
ing to fermentation, there were a number of different experiments
which were performed by him to elucidate the same point. We will
choose one of many in relation to fermentation. If a sugary solution
of carbonate of lime is left to itself, it begins after a time to effervesce,
carbonic acid is evolved, and lactic acid is formed ; and this latter
decomposes the carbonate of lime to form lactate of lime. This lactic
acid is formed, so to speak, at the expense of the sugar, which little
by little disappears. Pasteur demonstrated the cause of this trans-
formation of sugar into lactic acid to be a thin layer of organic
matter consisting of extremely small moving organisms. If these be
withheld or destroyed in the fermenting fluid, fermentation will
cease. If a trace of this grey material be introduced into sterile milk
or sterile solution of sugar, the same process is sol up, and lactic
acid fermentation occurs.

Pasteur examined the elements of this organic layer by aid of the
microscope, and found it to consist of small short rods of protoplasm
quite distinct from the yeast cells which previous investigators had
detected in alcoholic fermentation. One series of experiments was
accomplished with yeast cells and these bacteria, a second series with
living yeast cells only, a third series with bacteria only, and the con-
clusions at which Pasteur arrived as the result of these labours he
expressed in the following words: — *

"As for the interpretation of the group of new facts which I have
met with in the course of these researches, I am confident that who-
ever shall judge them with impartiality will recognise that the
alcoholic fermentation is an act correlated to the life and to the
organisation of these corpuscles, and not to their death or their putre-
faction, any more than it will appear as a case of contact action in
winch tht' transformation of the sugar is accomplished in the
pivsiMiee of tin1 ferment without the latter giving or taking anything
from it,"

94 BACTERIA AND FERMENTATION

Pasteur occupied six years (1857-1863) in the further elucidation
of his discovery of the potency of these hitherto unrecognised agents,
and the establishment of the fact that "organic liquids do not alter
until a living germ is introduced into them, and living germs exist
everywhere." It must not be supposed that to Pasteur is due the
whole credit of the knowledge acquired respecting the cause of
fermentation. He did not first discover these living organisms ; he
did not first study them and describe them ; he was not even the
first to suggest that they were the cause of the processes of fermenta-
tion or disease. But nevertheless it was Pasteur who " first placed
the subject upon a firm foundation by proving with rigid experiment
some of the suggestions made by others."

Kinds of Fermentation

Although fermentation is nearly always due to a living agent, as
proved by Pasteur, the process is conveniently divided into two
kinds.* (1) When the action is direct, and the chemical changes
involved in the process occur only in the presence of the cell, the
latter is spoken of as an organised ferment ; (2) when the action is
indirect, and the changes are the result of the presence of a soluble
material secreted by the cell, acting apart from the cell, this soluble
substance is termed an unorganised soluble ferment, or enzyme. The
organised ferments are bacteria or vegetable cells allied to the
bacteria ; the unorganised ferments, or enzymes, are ferments found
in the secretions of specialised cells of the higher plants and animals.
It will be sufficient to illustrate the enzymes by a few of the more
familiar examples, such as the digestive agents in human assimilation.
This function is performed, in some cases, by the enzyme combining
with the substance on which it is acting and then by decomposition
yielding the new " digested " substance and regenerating the enzyme ;
in other cases, the enzyme, by its molecular movement, sets up
molecular movement in the substance it is digesting, and thus changes
its condition. These digestive enzymes are as follow : in the saliva,
ptyalin, which changes starch into sugar; in the gastric juice of the
stomach, pepsin, which digests the proteids of the food and changes
them into more soluble forms ; the pancreatic ferments, amylopsin,
trypsin, and steapsin, capable of attacking all classes of food stuffs ;
and the intestinal ferments, which have not yet been separated in
pure condition. In addition to these, there are ferments in
bitter almonds, mustard, etc. Concerning these unorganised ferments
we have little further to say. Perhaps the commonest of them all
is diastase, which occurs in malt, and to which some reference will be
made later. Its function is to convert the starch, which occurs in

* E. A, Sehafer, F.R.S., Text-book of Physiology, vol. i., p. 312.

CONDITIONS OF FERMENTATION 95

barley, into sugar. These unorganised ferments act most rapidly at a
high temperature.*

We may preface our consideration of the organised ferments by
an axiom by which Professor Frankland sums up the vitalistic theory
of fermentation, which was supported by the researches of Pasteur :
" No fermentation without organisms, in every fermentation a particular
organism" From these words it is to be inferred that there is no one
particular organism or vegetable cell to be designated the micro-
organism of fermentation, but that there are a number of fermenta-
tions each started by some specific form of agent. It is true that the
chemical changes, induced by organised ferments, depend on the life
processes of micro-organisms which feed upon the sugar or other
substance in solution, and excrete the product of the fermentation.
Fermentation always consists of a process of breaking down of
complex bodies, like sugar, into simpler ones, like alcohol and
carbonic acid. Of such fermentations we may mention at least five :
the alcoholic, by which alcohol is produced ; the acetous, by which
wine absorbs oxygen from the air and becomes vinegar ; the lactic,
which sours milk ; the butyric, which out of various sugars and
organic acids produces butyric acid; and the ammoniacal, which
is the putrefactive breaking down of compounds of nitrogen into
ammonia. We shall have occasion to refer at some length to this
process when considering denitrifying organisms in the soil.

There are four chief conditions common to all these five kinds of
organised fermentation. They are as follow : —

1. The presence of the special living agent or organism of the
particular fermentation under consideration. This, as Pasteur
pointed out, differs in each case.

2. A sufficiency of pabulum (nutriment) and moisture to favour
the growth of the micro-organism.

3. A temperature at or about blood-heat (35-38 C, 985 F.).

4. The absence from the solution or substance of any obnoxious
or inimical substances which would destroy of retard the action of
the living organism and agent. Many of the products of fermenta-
tion are themselves antiseptics, as in the case of alcohol; hence
alcoholic fermentation always arrests itself at a certain point.

The causal micro-organisms of particular fermentations are of
various kinds, belonging, according to botanical classification, to

" The unorganised ferments are frequently otherwise classified than as above,
according to function. The chief are these : — <(»>i//o/j/fii\ those which change starch
and glycogen (amyloses) into sugars, <\;/., ptyalin, diastase, amylopsin (organisms
of the subtilis group and the micrococcus of mastitis are said to produce amylolytic
ferments); proteolytic, those which change proteids into proteoses and peptones,
«.//., trypsin, pepsin; mvertWS, those which change maltose, sucrose, and lactose
into glucose, e.g., invertin (various species of bacteria produce inversive ferments);
eoagulative, those which change soluble proteids into insoluble, r </., rennet ; steato-
lytic, those which split up fats into fatly acids and glycerine, e.g., steapsin.

9G BACTERTA AND FERMENTATION

various different subdivisions of the non-flowering portion of the
vegetable kingdom. A large part of fermentation is based upon the
growth of a class of microscopic plants termed yeasts. These differ
from the bacteria in but few particulars, mainly in their method of
reproduction by budding (instead of dividing or sporulating, like the
bacteria). Their chemical action is closely allied to that of the
bacteria. Secondly, there are special fermentations and modifications
of yeast fermentation due to bacteria. Thirdly, a group of somewhat
more highly specialised vegetable cells, known as moulds, make a
perceptible contribution in this direction. According to Hansen,
these latter, so far as they are really alcoholic ferments, induce
fermentation, that is, inversion of sugar, not only in solutions of
dextrose, but also in maltose. Mucor raccmosus is the only member
that is capable of inverting a cane-sugar solution ; Mucor erectus is
the most active fermenter, yielding 8 per cent, by volume of alcohol
in ordinary beer wort. Both of these will be referred to as they
occur in considering the five important fermentations already
mentioned.

The general microscopic appearance of yeast cells may be shortly
stated as follows : They are round or oval cells, and by budding
become " daughter " yeast cells. Each consists of a cellulose
membrane and clear homogeneous contents. As they perform their
function of fermentation, vacuoles, fat-globules, and granules make
their appearance in the enclosed plasma. Just as in many vegetable
cells a nucleus was detected by Schmitz by means of special methods
of staining, so Hansen has found the nucleus in old-yeast cells from
" films " without any special staining.

1. Alcoholic Fermentation

Cause, yeast ; medium, sugar solutions ; result, alcohol and carbonic acid.

It was Caignard-Latour who first demonstrated that yeast cells, by
their growth and multiplication, set up a chemical change in sugar
solutions which resulted in the transference of the oxygen in the
sugar compound from the hydrogen to the carbon atoms, that
is to say, in the evolution of carbonic acid gas and the production,
as a result, of alcohol. Expressed in chemical formula, the change
is as follows : —

•C^H^O^ (plus the fermenting agent) = 2 C2H(iO + 2 C02.

A natural sugar, like grape-sugar, present in the fruit of the vine,
is thus fermented. The alcohol remains in the liquid ; the carbonic
acid escapes as bubbles of gas into the surrounding air. If we go
a step further back, to cane-sugar (which possesses the same elements
as grape-sugar, but in different proportions), dissolve it in water,

ALCOHOLIC FERMENTATION 97

and mix it with yeast, we get exactly the same result, except that
the first stage of the fermentation would be the changing of the cane-
sugar into grape-sugar, which is accomplished by a soluble ferment
secreted by the yeast cells themselves. If now we go yet one step
further back, to starch, the same sort of action occurs. When starch
is boiled with a dilute acid it is changed into a gum-like substance,
dextrin, and subsequently into maltose, which latter, when mixed
with these living yeast cells, is fermented, and results in the evolution
of carbonic acid gas and the production of alcohol. In the manu-
facture of fermented drinks from cereal grains containing starch there
is, therefore, a double chemical process : first, the change of starch
into sugar by means of conversion, a chemical change obtained by
the action of sulphuric or some other acid, or by the influence of
diastase; and secondly, the change of the sugar into alcohol and
carbonic acid gas by the process of fermentation , an organic change
brought about by the living yeast cells.

In all these three forms of alcoholic fermentation the principal
features are the same, viz., the sugar disappears ; the carbonic acid
gas escapes into the air ; the alcohol remains behind. Though it is
true that the sugar disappears, it would be truer still to say that it
reappears as alcohol. Sugar and alcohol are built up of precisely the
same elements: carbon, hydrogen, and oxygen. They differ from
each other in the proportion of these elements. It is obvious, there-
fore, that fermentation is really only a change of position, a breaking-
down of one compound into two simpler compounds. And this
redistribution of the molecules of the compound results in the
production of some heat. Hence, we must add heat to the results
of the work of the yeasts.*

It will be necessary subsequently to consider a remarkable faculty
which bacteria possess of producing products inimical to their own
growth. In some degree this is true of the yeasts, for when they
have set up fermentation in a saccharine fluid there (Mines a time
when the presence of the resulting alcohol is injurious to further
action on their part. It has become indeed a poison, and, as we
have already mentioned, a necessary condition for the action of a
ferment is the absence of poisonous substances. This limit of
fermentation is reached when the fermenting fluid contains 13 or
14 per cent, of alcohol.

The Biology of Yeast. — Having briefly discussed the "medium"
and the results, we may now turn to the other side of the matter,
and enumerate some of the chief forms of the yeast plant. Jorgensen

■ When alcohol is pure and contains no water it is termed absolute alcohol. If,
however, it is mixed with 16 per cent, of water, it is called rectified spirit, and when
mixed with more than half its volume of water (66'8 per cent.) it is known as proof

spirit .

G

98

BACTERTA AND FERMENTATION

gives more than a score of different members of this family of
Saccharomycetes* But before mentioning some of the chief of
these, it will be desirable to consider a number of properties common
to the genus. The yeast cell is a round or oval body of the nature
of a fungus, composed of granular protoplasm surrounded by a
definite envelope, or capsule. It reproduces itself, as a general
rule, by budding, or gemmation. At one end of the cell a slight
swelling or protuberance appears, which slowly enlarges. Ulti-
mately there is a constriction, and the bud becomes partly and
at last completely separated from the parent cell. In many cases
the capsules of the daughter cell and the parent cell adhere, thus
forming a chain of budding cells. The character of the cell and its
method of reproduction do not depend merely upon the particular
species alone, but are also dependent upon external circumstances.
There are differences in the behaviour of species towards different
media at various temperatures, towards the carbohydrates (especially

Fio. 13.— "Diagram of Ascospore Formation.

Fig. 14.— Gypsum Block.

maltose), and in the chemical changes which they bring about in
nutrient liquids. In connection with these variations Professor
Hansen has pointed out that, whilst some species can be made use
of in fermentation industries, others cannot, and some even produce
" diseases " in beer.t

One of the most remarkable evidences of the adaptability of the
yeasts to their surroundings, and a specific characteristic, occurs in
what is termed ascospore formation. If a yeast cell finds itself
lacking nourishment or in an unfavourable medium, it reproduces
itself not by budding, but by forming spores out of its own intrinsic
substance, and within its own capsule. To obtain this kind of
spore formation Hansen used small gypsum blocks as the medium on
which to grow his yeast cells. Well-baked plaster-of -Paris is mixed
with distilled water, and made into a liquid paste. The moulds are
made by pouring this paste into cardboard dishes, where it hardens

* Micro-organisms and Fermentation.

f E. C. Hansen, Studies in Fermentation (Copenhagen), p. 98.

PLATE 8.

•w

%

Saccharomyces cerevisice.

Film preparation, x 1000.;

Ascospoee Formation in Yeast. The capsule
of th.- parent cell around the spores is
invisible, x 1000.

o
O

c

(POO ^

tto

Pathoobnk Vkast. (Poulerton). x 1000.

[To face page 98.

ALCOHOLIC FERMENTATION 99

again. The mould is then sterilised by heat, a few cells of yeast are
placed on its upper surface, and the whole is floated in a small
vessel of water and covered with a bell-jar. Under these conditions
of limited pabulum the cell undergoes the following changes: it
increases in size, loses much of its granularity, and becomes homo-
geneous, and about thirty hours after being sown on the gypsum
there appear several refractile cells inside the parent cell. These
are the ascospores. In addition to the gypsum, it is necessary to
have a plentiful supply of oxygen, some moisture (gained from the
vessel of water in which the gypsum stands), a certain temperature,
and a young condition of the protoplasm of the parent yeast cells.
Hansen found that the lowest temperature at which these ascospores
were produced was '5 — 3° C, and at the other extreme up to 37° C.
The rapidity of formation also varies with the temperature, the
favourable degree of warmth being about 22 — 25° C. (Plate 8).

Hansen pointed out that it was possible by means of sporulation
to differentiate species of yeasts. For it happens that different
species show slight differences in spore formation, e.g. : —

(a) The spores of Saccharomyces cerevisiai expand during the first
stage of germination, and produce partition walls, making a com-
pound cell with several chambers. Budding can occur at any point
on the surface of the swollen spores. To this group belong S.
pastorianus and S. ellipsoideus.

(b) The spores of Saccharomyces Ludwigii fuse in the first stage, and
afterwards grow out into a promycelium, which produces yeast cells.

(c) The spores of Saccharomyces anomalus are different in shape
from the others in that they possess a projecting rim round
the base.

Another point in the cultivation of yeasts has been elucidated by
a number of workers, among whom is Hansen, namely, the methods
of obtaining pure cultures. Only by starting with one individual
cell can it be hoped to secure a pure culture of yeasts. For the
study of the morphology of yeasts under the microscope the problem
was not a difficult one. It was comparatively easy to keep out
foreign germs from a cover-glass preparation sufficiently to perceive
^v rii dilation of spores and the growth of yeasts. But when pure
cultures are required for various physiological purposes then a
different standard and method is necessary.

Hansen employed dilution with water in the following manner : —
Yeast is diluted with a certain amount of sterilised water. A drop
is carefully examined under the microscope, a single cell of yeast is
taken, and a cultivation made upon wort. When it has grown
abundantly a quantity of sterilised water is added. From this,
again, a single <ln>]> is taken and a<l<lc<| in say 20 c.c. of sterilised
water in a fresh flask. This flask will contain, let us suppose, ten

100 BACTERIA AND FERMENTATION

colls. It is now vigorously shaken, and the contents arc divided
into twenty portions of 1 c.c. each, and added to twenty tubes of
sterilised water. It is highly probable that half, of these tubes
have received one cell each. In the course of a few days it can be
seen how far the culture is pure. If only one colony is present,
the culture is a pure one, and as this grows we obtain an absolutely
pure culture in necessary quantity. Even when the gelatine plate
method is used, it is desirable to start with a single cell (Hansen).
The advantage of Hansen's yeast method over Koch's bacterial plate
method is that it has a certain definite starting-point. This is
obviously impossible when dealing with such microscopic particles
as the bacteria proper.

A third point in the differentiation of yeast species is the question
of films. Hansen set to work, after having obtained pure cultures
and ascospores, to examine fdms appearing on the surface of liquids
undergoing fermentation. The object of this was to ascertain
whether all yeasts produced the same mycelial growth on the surface
of the fermenting fluid. To produce these films the process is as
follows : Drop on to the surface of sterilised wort in a flask a very
small quantity of a pure culture of yeast ; secure the flask from
movement, and protect it, not from air, which is necessary, but from
falling particles in the air. In a short time small colonies appear,
which coalesce and form patches, and finally a film or membrane
which covers the liquid and attaches itself to the sides of the flask.
By the differences in the films and the temperatures at which they
form it is possible to obtain something of a basis for classification.
The further advances in yeast culture and in our knowledge of the
agencies of fermentation have, however, tended to show that no
strict dividing lines can be drawn. Hansen's researches have, not-
withstanding, been of the greatest moment to the whole industry of
fermentation. What has been found true in bacteriology has also
boen demonstrated in fermentation, namely, that, though many
yeasts differ but little in structure and behaviour, they may produce
very different products and possess very different properties.
Industrial cultivation of these finer differences in fermentative action
has to a large extent revolutionised the brewing industry. The
formation of films is not a peculiarity of certain species, but must
be regarded as a phenomenon occurring somewhat commonly amongst
yeasts. The requisites are a suitable medium, a yeast cell, a free,
still surface, direct access of air, and a favourable temperature. The
wort loses colour, and becomes pale yellow. Microscopic differences
soon appear between the sedimentary yeast and the film yeast of the
same species, the latter growing out into long mycelial forms, the
character of which depends in part upon the temperature. This
often varies from 3° to 38° C.

ALCOHOLIC FERMENTATION 101

A fourth point helpful in diagnosis is the temperature which
proves to be the thermal death-point. Saccharomyccs ccrevisice is
killed by an exposure to 54° C. for five minutes, and 62° C. kills the
spores. As a rule, yeasts can resist a considerably higher tempera-
ture when in a dry state than in the presence of moisture.

It should be noted that yeasts may be cultivated on solid
media. Hansen employed wort-gelatine (5 per cent, gelatine), and
found that at 25° C. in a fortnight the growths which develop show
such microscopic differences as to aid materially in diagnosis.
Saccharomyces ellipsoideus I. exhibits a characteristic network which
readily distinguishes it.

There is one other feature to which reference must be made.
The process of fermentation may be set up by a "high" or a
" low " yeast. These terms apply to the temperature at which the
process commences. " High " yeasts rise to the surface as the action
proceeds, accomplish their work rapidly, and at a comparatively high
temperature, say about 16° C. ; " low " yeasts, on the contrary, sink-
in the fermenting fluid, act slowly, and only at the low temperature
of 4° or 5° C. This is maintainable by floating ice in the fluid.
Formerly all beer was made by the "high" mode, but on the con-
tinent of Europe "low" yeast is mostly used, whilst the ' high " is in
vogue in England. This latter method is move conducive to the
development of extraneous organisms, and therefore risky in all but
well-ordered brewing establishments.

Before proceeding to mention shortly sonic of the commoner
forms of yeast we must again emphasise Hansen's method of analysis
in separating a species. The shape, size, and appearance of colls are
not sufficient for differentiation, because it is found that the same
species, when exposed to different external conditions, can occur in
very different forms. Hence Hansen established the analytical
method of observing (1) the microscopic appearance, (2) the forma-
tion of ascospores, and (3) the production of films. In addition, the
temperature limits, cultivation on solid media, and behaviour towards
carbohydrates, are characters which aid in the separation of yeasts.
In well-grown cultures on wort-gelatine, -i broad division can be
made of yeasts according as they produce (") a dry, hard, cohesive
growth; (b) a soft, moist growtb with liquefaction of gelatine; and
(c) those producing pigment. By basing differentiation of species
upon these features, the following can be distinguished : —

Saccharomyces Cerevisiof.— Oval or ellipsoidal cells; reproduction by budding;
ascospores, rapidly at 30 C, , slowly at 12 C, nol formed at all at lower tempera-
tures; film formation, seven to ten days at 22 C. ; an active alcoholic ferment,
producing in a fortnighl in beer wort from I to (i per cent, by volume of alcohol
(Jorgensen> This species is a typical English high yeast, possessing the power of
"inverting" cane sugar prei IOUS to producing alcohol' and carbonic acid. It is said
to have no action on milk-sugar. It is the M true brewing jreasl " (Plate 8).

102 BACTERIA AND FERMENTATION

Saccharomyces Ellipsoidus I. — Round, oval, or sausage-shaped cells, single or in
chains; ascospores in twenty-four hours at 25° C. (not above 30° C, not below
4° C). Grown on the surface of wort gelatine, a network is produced by which
they can be recognised (in eight to twelve days at 33° C). At 13-15" C. a
characteristic branching mass is produced. It is an alcoholic ferment as active as

o ,00 ^ do *o Q y

03 §§^

/

I

Fig. 15. — Diagram of S. cerevisice. Fig. 16. — Diagram of S. ellipsoid*

S. cerevisice. S. ellipsoideus II. — Round and oval, rarely elongated, a widely dis-
tributed yeast, causing " muddiness " in beer and a bitter taste. It is essentially a
"low " yeast, and one of the so-called " wild yeasts " injurious to beer.

Saccharomyces Conglomeratus is a round cell, often united in clusters, and
occurring in rotting grapes, and at the commencement of fermentation.

Saccharomyces Pastorianus I. — Oval or club-shaped cells, occurring in after-
fermentation of wine, etc., and producing a bitter taste, unpleasant odour, and
_ turbidity. The spores frequently occur in the air of

£Z--)r-^ ^-p breweries.

" /^^^ & Pastor. II. — Elongated cells, possessing an in-

Q vertose ferment. They do not, like 8. pastor. 7.,
A 0 [\ produce disease in beer.
/) f\ II S. Pastor. 111. — Oval or elongated cells, producing

II l\ v/ \ I turbidity in beer. Grown on yeast-water gelatine ;

(/ *S II the colonies show after sixteen days crenated hairy

Ss ffi— ■ \ // \J Saccharomyces Apknlatus. — Lemon-shaped cells.

£/ " "" a/ They give rise to a feeble alcoholic fermentation, and

^^^ // produce two kinds of spores— round and oval ; they

^^^P (/ appear at the onset of vinous fermentation, but give

G^^ way later on to S. cerevisice.

Fig. 17.— Diagram of S. Saccharomyces Mycoderma. — Oval or elliptical cells,

pastorianus. often in branching chains. They form the so-called

"mould" on fermented liquids, and develop on the

surface without exciting fermentation. When forced to grow submerged, they

produce a little alcohol.

Saccharomyces Exiguus. — Conical cells, appearing in the after-fermentation of
beer.

Saccharomyces Pyriformis. — Oval cells, converting sugary solutions containing
ginger into ginger-beer.

Saccharomyces lllicis, Hansenii, and Aquifolii produce a small percentage of
alcohol.

2. Acetous Fermentation

Cause," Mycoderma aceti ; medium, wine and other alcoholic liquids ; result, the

formation of vinegar.

If alcohol be diluted with water, and the specific ferment mixed
with it and exposed to the air at 22° C, it is rapidly converted into

ACETOUS FERMENTATION 103

vinegar. The change is accompanied by the absorption of oxygen,
one atom of which combines with two of hydrogen to form water, and
a substance remains termed aldehyde, further oxidation of which pro-
duces the acetic acid. We may express it chemically thus : —

C2H0O ( + oxygen and the ferment) = C2H40 + H20.

Alcohol. Aldehyde. Welter.

The aldehyde becomes further oxidised : —

C,H40 + O = C2H402 (acetic acid).

This method of simply oxidising alcohol to obtain acetic acid may
be carried out chemically without any ferment. If slightly diluted
alcohol be dropped upon platinum black, the oxygen condensed in
that substance acts with energy upon the spirit, and union readily
occurring, acetic acicl results. Here the whole business of the plati-
num sponge is to persuade the oxygen of the air and the hydrogen of
the alcohol to unite. In the ordinary manufacture this is accom-
plished by the vegetable cells of Mycoderma aceti.

There are two chief methods adopted in the commercial manu-
facture of vinegar, both of which depend upon the presence of the
mycoderma. The method in vogue at Orleans when Pasteur (about
1862) commenced his studies of the vinegar organism, was to fill vats
nearly to the brim with a weak mixture of vinegar and wine. Where
the process is proceeding the surface is covered with a fragile pellicle,
"the mother of vinegar," which is produced by, and consists of,
certain micro-organisms whose function is to convey the oxygen of
the air to the liquor in the vats, thus oxidising the alcohol into
vinegar. This oxidation may be carried on even beyond the Btage of
acetic acid (when no more alcohol remains to be oxidised), resulting
in carbonic acid gas, which escapes into the air. Bui as in the
alcoholic, so in the acetic, fermentation there comes a time when the
presence of an excess of the acid inhibits the further growth of the
organism. This point is approximately when the acetic acid has
reached a percentage as high as 14. But if the acid be removed, and
fresh alcohol added, the process recommences.

The second method, sometimes called by the Germans the "quick
vinegar process," is to pour the weakened alcohol through a tall
cylinder filled with wood-shavings, having first added some warm
vinegar to the shavings. After a number of hours the resulting fluid
is charged with acetic acid. What 1ms occurred ? [iebig maintained
that a chemical and mechanical change had brought about the change
from the alcohol put into the cylinder and the vinegar drawn offal
the exit tube. It was reserved for Pasteur to demonstrate by experi-
ment that the addition of the warm vinegar to the shavings was in
reality an addition of a living micro-organism, which, forming a film

104 BACTERIA AND FERMENTATION

upon the shavings, became " the mother of vinegar," and oxidised the
alcohol which passed over it, inducing it to become aldehyde and then
acetic acid.

Mycoderma aceti (described by Persoon 1822, Klitzing 1837, and
Pasteur 1864), is the name rather of a family than an individual.
Pasteur believed it to be a specific individual, but Hansen pointed
out that it was composed of two distinctly different species (Bacterium
aceti and B. Pastcurianum), and subsequently other investigators
have added members to the acetic fermentation group of which M.
aceti is the type. This bacterium is made up of small, slightly
elongated cells, with a transverse diameter of 2 or 3 /ul, sometimes
united in short chains of curved rods. They frequently show a
central constriction, are motile, and produce in old cultures involution
forms. The way in which the cells act and are made to perforin their
function is as follows: A small quantity, taken from a previous
pellicle, is sown on the surface of an aqueous liquid, containing 2 per
cent, of alcohol, 1 per cent, of vinegar, and traces of alkaline phos-
phates. Very rapidly indeed the little isolated colonies spread, and
becoming confluent, form a membrane or pellicle over the whole area
of the fluid. When the surface is covered the alcohol is converted to
acid. After this it is necessary to add, each day, small quantities of
alcohol. When the oxidation is completed the vinegar is drawn off,
and the membrane is collected and washed, and is then again ready
for use. It ought not to remain long out of fermenting liquid, nor
ought it to be allowed to over-perform its function, for thus having
oxidised all the alcohol it will commence oxidation of the vinegar.

In wort-gelatine Bacterium Pasteurianum develops as round
colonies with a smooth or wavy border, whilst B. aceti has a tendency
towards stellate arrangement. Spores have not been observed, and
from a morphological point of view the two species behave alike.
Neither produces any turbidity in the liquid containing them. In
order to flourish, B. aceti requires a temperature of about 33° C. and
a plentiful supply of oxygen. In a cool store or cellar there is,
therefore, nothing to fear from B. aceti. Frankland has isolated a
Bacillus cthaceticus, which is a fermentative organism producing
ethyl-alcohol and acetic acid. By oxidation the ethyl-alcohol may be
converted into acetic acid.

3.— Lactic Acid Fermentation

Cause, Bacillus acidi lactici; medium, milk-sugar, cane-sugar, glucose, dextrose,

etc. ; result, lactic acid.

The process set up by the lactic ferment is simply a decomposition,
an exact division of one molecule of sugar into two molecules of
lactic acid, there being neither oxidation nor hydration. The con-

LACTIC ACID FERMENTATION 105

ditions under which the ferment acts are very similar to those we
have already considered (see also p. 196). There is frequently car-
bonic acid gas formed ; there is a cessation of fermentation when the
medium becomes too acid ; there is the same method of starting the
process by inoculation of milk or cheese or any such substance with
the specific bacillus. It is probable that such inoculated matter will
contain a mixture of micro-organisms, but if the lactic bacillus is
present, it will grow so vigorously and abundantly that the fermen-
tation will be readily set up.*

In 1877 Lister was able, by means of the " dilution method," to
isolate from sour milk, in a form of pure culture, an organism to
which he gave the name B. lactis, and which he believed gained
access to milk from the air of dairies and similar places.f For some
time this organism was held to be causally related to lactic
fermentation. But in 1884, by means of culture on solid media, as
introduced by Koch, Hueppe was able to isolate a bacillus which he
named the Bacillus acidi lactici. This was probably identical with
Lister's bacillus, and is now a term used to cover a whole family of
organisms having somewhat similar characters, and possessing the
property of sotting up lactic fermentation. J In 1894 Gunther and
Thierf elder published the result of their work on lactic acid
fermentation, from which they concluded that Lister and Hueppe
had discovered one and the same species, and thai it was the causal
agent of lactic acid production in Europe. Esteu found a similar
organism to be the cause of lactic acid fermentation in America, and
Conn holds that three organisms, or rather types of species, are the
chief agents in the production of Lactic fermentation, namely J>\ acidi
lactici, Nos. i. and ii., and ./>'. lactis cerogenes. The firsl named forms
between 75 to 90 per cent, of the bacteria present. No. ii. is also
very abundant. B. lactis cerogenes is found almost universally,
although never in large numbers. It is a type of ;i species which
produces intense acid on Litmus gelatine cultures, produces much
gas in milk or milk-sugar broth, curdles milk at high temperatures,
and produces a distinctive odour in the milk, which it ferments. §
According to Escherich, the formation of lactic acid by this
organism prevents fermentation in the stomach and intestines.

It was Hueppe who made the importanl discovery thai many

* For full discussion of the subject of lactic fermentation of milk, sec- Bacteriology
of Milk, 1903 (Swithinbank and Newman), pp. 149-159.

I Trans, of Path. S<>c, 1878, p. 187.

X Hueppe isolated five forms of his lactic acid bacillus, and Fliigge described
eleven forms. Maddox, Beyer, Fokker, Krucger, Grotenfeld, and other workers
isolated lactic acid organisms.

§ Storr's Agricultural Expt. Sta. Rap., L899, p. 22. Others than those named
are II. acidi lactici of Giintner, />. actdi /<><-/iri of Leichmann, Bacillus XIX. of
Adametz, Bacillus a of Freudenreich, B. and .1/. acidi l&volactici of Leichmann,
Grotenfeld's B. acidi lactici (Nos. i. and ii.), X<>. 8 of Eckles, and B. casei.

106 BACTERIA AND FERMENTATION

different species of bacteria are capable of setting up lactic fer-
mentation, and what we have now said amply supports that view.
Indeed, it has been estimated that upwards of 100 different bacteria
possess this property.*

Bacillus acidi lactici (Hiippe) consists of rods about 2 ij. long and *4 /jl wide,
occurring singly or in pairs, chains or threads. Its habitat is sour milk.

It grows best at blood-heat, but much above that it fails to produce its fermenta-
tion. It ceases to grow under 10° C. It inverts milk-sugar and changes it to
dextrose, from which it then produces lactic acid. Sugars do, however, differ
considerably in the degrees to which they respond to the influence of the lactic;
ferment, and some which are readily changed by the alcoholic ferment are un-
touched by Bacillus acidi lactici.

Staining reaction— Ordinary stains and slightly by Gram's method.

Motility — No flagella ; non-motile.

Spore formation — Absent.

Biology: cultural characters (including biochemical features) — Good growth at
room temperature and blood-heat.

Bouillon — Diffuse turbidity ; abundant sediment.

Gelatine plates and tubes— Colonies similar to B. coli; small, smooth, round,
white.

Agar plates and tubes — Colonies similar to B. coli ; small, smooth, white growths,
moist and shiny.

Potato — A wavy, smooth-edged growth, elevated ; greyish-white or yellowish-
white in colour, sometimes turning brown.

Milk — Solid coagulation, leaving clear fluid ; occasionally some gas-bubbles.
Lactic and acetic acids are produced. Powers of acid coagulation of milk are
gradually lost after long cultivation upon gelatine or agar.

Anaerobic or aerobic — Grows well aerobically, and if sugar present in medium
anaerobically also.

Non- pathogenic.

The lactic fermentation bacteria are short rods, do not liquefy
gelatine, nor do they form spores. They grow readily on gelatine at
room temperature, forming as a rule small circular colonies, white or
grey in colour, with sometimes a tinge of yellow, and the surface of
the colony is smooth and glistening. The lactic acid organisms pro-
duce appreciable amounts of lactic acid only at somewhat elevated
temperatures. If the amount of acid rises much above 2 }Der cent.,
the growth of the lactic acid bacteria is inhibited. Many other sub-
stances, as we have seen, are produced in addition to lactic acid {e.g.
acetic and ferric acids, alcohol, methane, C02 .H.N"., etc.). Lactic acid
organisms (as non-spore bearers) are readily killed by pasteurisation.

* Delbriick, Zopf, Krause, Peters, Lindner, Weigmann, Storch, and Marpmann,
are amongst those who, in addition to workers we have named, have described
bacteria possessing the power of setting up lactic fermentation. Only provisional
classifications are possible at present, as, owing to variations in biology and
terminology, it is probable that certain lactic organisms are described under several
different terms. Generally, it may be said that some grow well in the presence of
oxygen, and others do not. The latter group, facultative anaerobes, are perhaps
the most common. They sour milk best in deep vessels, and produce a right-
handed lactic acid. They are widely distributed in nature, and may form 90 per
cent, of the total bacteria in milk. Some produce gas, others liquefy gelatine, and
yet others produce spores.

BUTYRIC ACID FERMENTATION 107

The economic function of the lactic ferments concerns, of course, the
manufacture of butter and cheese.

Van Laer has described a saccharo-bacillus which produces lactic
acid amongst other products, and brings about a characteristic
disease in beer, named tourne. The liquid gradually loses its bright-
ness and assumes a bad odour and disagreeable taste. The bacillus
is a facultative anaerobe. A number of workers have separated
organisms having a lactic acid effect, which diverge considerably from
the ordinary type of lactic acid bacillus.

4. Butyric Acid Fermentation

Cause, Bacillus butyricus and other forms ; medium, milk, butter, sugar and
starch solutions, glycerine ; result, butyric acid.

When sugars are broken down by Bacillus acidi lactici the lactic
acid resulting may, under the influence of the butyric ferment
become converted into butyric acid, carbonic acid, and hydrogen.
Neither butyric acid nor lactic acid is as commonly used as alcohol
or vinegar. Both, like vinegar, can be manufactured chemically,
but this is rarely practised. Butyric acid is a common ingredient
in stale milk and butter, and its production by bacteria was
historically one of the first bacterial fermentations understood.
Moreover, in its investigation Pasteur first brought to light the
fact that certain organisms acted only in the absence of oxygen.
In studying a drop of butyric fermenting fluid, it was observed
that the organisms at the edge of the drop were motionless and
apparently dead, whilst in the central portion of the drop the
bacilli were executing those active movements which are character-
istic of their vitality. To Pasteur's mind this at once suggested
what he was able later to demonstrate, namely, that these bacilli
were paralysed by contact with oxygen. When he passed a stream
of air through a flask containing a liquid in butyric fermentation,
he observed the process slacken and eventually cease. So were
discovered the anaerobic micro-organisms. The aerobic ferments
give rise to oxidation of certain products of decomposition; the
anaerobic organisms, on the other hand, only commence to grow
when the aerobic have used up all the available oxygen. Thus in
such fermentations certain bodies (carbohydrates, fatty acids, etc.)
undergo decomposition, and by oxidation become carbonic acid gas,
and the remainder is left as a "reduced" product of the whole
process. Hence sometimes this is termed fermentation by reduction.
The chemical formula, of this butyric reaction may be expressed
thus : —

CtfH1206(by simple decomposition) = 2C8H6Og,

Glucose. Lactic acid.

108 BACTERIA AND FERMENTATION

which is followed by the fermentation of the lactic acid:-

2C3H6Os

C,H802

+

2C02

+ 2 H2.

Lactic acid.

Butyric acid.

Carbonic
acid gas.

Free hydrogen,

Previously to 1880, the only work which had been done in the
elucidation of the bacterial origin of butyric fermentation had been
accomplished by Prazmowski and Pasteur; the former designating
the organism he found Clostridium butyrieum, and the latter naming
his " infusoires " Vibrion butyrique. Prazmowski emphasised the
motility and resistance of the bacillus, and found that the latter was
due to the spores produced by the organism. These spores were
able to withstand boiling for several minutes. Fitz went so far as
to say that butyric spores could resist 1 toiling for twenty minutes.
Prazmowski was unable to obtain pure cultures. Clostridium buty-
rieum grows most readily at a temperature of about 40° C, and is
very widely distributed in nature. It is capable of dissolving cellu-
lose, and therefore plays a part in the cellulose fermentation, which
is employed in various maceration industries. It is generally held
that in such fermentations there is symbiotic action between the
butyric bacillus and an organism incapable of causing "retting" by
itself. The organisms discovered by Prazmowski and Pasteur were
anaerobic. But Fitz and Hucppe isolated an aerobic butyric bacillus.
This fact was confirmed by Gruber, working with a pure culture in
1887, and it wras at the same time demonstrated that the Clostridium
butyrieum of Prazmowski consists of a number of closely allied, but
distinct, species. Lafar states that nearly related to this is a ferment
isolated by Liborius from old cheese, and introduced into literature
under the name of Clostridium fcetidum. This organism liberates
foul-smelling gases, in addition to producing butyric acid, and forms
one of the many connecting links between the butyric acid bacteria
and the so-called " potato " bacilli. No sharply defined limit can be
drawn between these two groups.

The following are the three chief organisms of butyric acid : —

1. Bacillus Butyricus (Botkin)

Source and habitat — Widely distributed in milk, water, soil, dust.

Morphology — Rods, 1 to 3 /j, long, 0*5 n thick. Sometimes in threads, sometimes
in chains.

Staining reaction — Stains by Gram's method.

Motility— Motile.

Spore formation — Spores in middle or at end of bacillus ; about 1 n thick
(provisional) ; sporulation not proved (Botkin).

Biology : cultural characters {including biochemical features) — Favourable tempera-
ture 37° C. ; organism contains starch granules.

Bouillon — Slight growth; at 18° C. involution forms may occur. Vigorous
growth if glucose present, with opaque turbidity.

Gelatine plates and tubes — Round or oval colonies with sinuous edges ; medium

BUTYRIC ACID FERMENTATION 109

is rapidly liquefied ; gas development ; slight undulating colonies, as if consisting of
mass of felted threads. No odour.

Agar plates and tubes — A luxuriant growth with gas development and ramifica-
tions in medium. Odour present.

Potato— Growth extends into potato substance; smell of alcohol.

Milk— At-37° C. , after fifteen hours casein precipitated and butyric acid is formed
without intermediate formation of lactic acid. Coagulum eventually dissolves ;
before that stage the appearance is very characteristic ; there is a spongy fatty
layer on surface, then clear fluid, and then a white deposit. The presence of this
organism is readily proved in almost all milk. Fill a half litre flask with milk, and
steam at 100° C. for half an hour. Incubate at 37° C. In less than twenty-four
hours the characteristic changes will occur, with strong odour of butyric acid.
Care must be taken that the gas pressure does not burst the flask. There is a
marked odour of butyric acid. Other acids present are acetic, formic, and lactic.

Anaerobic or aerobic — Facultative anaerobe.

Non-pathogenic — It has been suggested that the B. enteritidis sporogenes of Klein
is a pathogenic form of this bacillus.

2. Bacillus Butyricus (Hiippe)

Source and habitat — Milk.

Morphology — Slender rods ; 1 "2 to 4 /x long, 05 /x thick ; round ends. May
grow into filaments ; rods slightly bent ; 2 1 /x long, 0*3 /x broad.

Staining reaction — Stains by Gram's method.

Flagella ; motility— Many flagella ; actively motile.

Spore formation— Oval spores at 37° C. ; mesially situated.

Biology : cultural characters (including biochemical features').

Bouillon — A pellicle is formed ; bouillon remains clear. No indol.

Gelatine plates arid tubes — Small whitish-yellow colonies with crater-shaped
depression ; liquefaction ; whitish-grey wrinkled pellicle produced in liquid
cultures in tubes ; liquefied medium cloudy and yellowish in colour.

Agar plates and tubes — A thin yellow layer, similar to B. mesentericue.

Potato — A fawn-coloured transparent layer, sometimes wrinkled. Somewhat
similar to B. megatherium.

Milk — Is coagulated. Precipitated casein subsequently dissolved. Bitter taste.
Butyric acid produced from salts of lactic acid ; also from milk-sugar when it is
previously hydrated.

Facultative anaerobe.

Non-pathogen ic.

3. Bacillus Butyricus (Pasteur). (Vibrion Butyrique)

Source and habitat — Air, and thence to milk.

Morphology — Cylindrical rods with rounded extremities; 3 u to 5 fx long by
6 ix to *8 /x broad. Isolated or in chains ; at times in long filaments indistinctly
articulated.

Staining reaction— Ordinary aniline stains.

Motility — Feebly motile ; motility ceases at once in presence of free oxygen.

Spore forma/ion — Ovoid sports.

biology : cultural characters (including biochemical features').

Bouillon — Grows freely under strictly anaerobic conditions in bouillon contain-
ing lactate of lime.

Agar plates and tidies — In agar " shake M cultures free from oxygen the medium
becomes clouded in the lower portion, and is soon broken up, with copious gas
formation accompanied by strong smell of butyric acid.

(ielatiue plates and tubes — As upon agar, but in a less degree, the medium
liquefying in the neighbourhood of the forming colonics

Anaerobic.

Non-pathogenic.

Several other butyric acid organisms have been isolated, of which a few notes
may be added : —

110 BACTERIA AND FERMENTATION

Bacillus acidi butyrici — (Kedrowski's Butyric acid bacillus). Anaerobic.
Kedrowski (Z. 16. 3) has isolated from mixtures of sugar solution with bad cheese
or rancid cream-butter which has been placed in the incubator, two organisms
which only show small deviations from one another. (Cf. the B. saccharo-butyricus
of Klencki from cheese). Kedrowski's B. acidi butyrici is a motile bacillus, which
towards one end produces ellipsoidal spores. The staining of the spores is readily
accomplished. The colonies in gelatine show rays — those in agar partly reticulated —
and interlaced spurs. Liquefaction of gelatine is more or less marked. Milk is
coagulated with separation of serum on the surface (acid reaction). There is
gradual peptonisation and simultaneous gas development.

Although, according to Pasteur's researches, the butyric acid
ferment performs its function anaerobically, many butyric organisms
can act in the presence of oxygen, and yield somewhat different
products.

All of them, however, ferment most actively at a temperature at
or about blood-heat, and the spores are able to withstand boiling for
from three to twenty minutes (Fitz). It will be observed that as in
lactic acid fermentation so in butyric, the results are not due to one
species only.

5. Ammoniacal Fermentation (see under Soil).

From what has now been said, it is obvious that although we learn
many important facts by a study of these different forms of fermen-
tation, we may also learn on the one hand how to prevent or correct
those conditions constantly occurring in fermented beverages, which
are known as " diseases," and on the other, the opportunities which
occur in industrial processes for the application of fermentation.
We will first deal with the former.

Diseases in Beer and Wines

We have seen how the knowledge of fermentation has been com-
piled by a large number of workers. Spallanzani, Schwann, Pasteur,
and Hansen all contributed epoch-making researches. In the same
way the investigations of diseases in beers and wines were carried on
by many observers, and were, at all events in the early stage, closely
connected with those relating to spontaneous generation and mixed
cultures of bacteria, or of yeasts occurring in fermentation. These
so-called "diseases" are analogous to the taints occurring in milk.
It was in 1883 that Hansen demonstrated that the universally
dreaded yeast turbidity and the disagreeable changes in taste, odour,
or colour of beer were caused not by the water or malt or particular
method of brewing, as was commonly believed, but that these
diseases had their origin in micro-organisms or in the yeast itself.*
A clue had been given by Scheele and Appert, who had prevented
these diseased conditions by physical agents which had destroyed the

* Practical Studies in Fermentation, E. C. Hansen, pp. 156-231.

DISEASES OF BEER 111

able to produce the diseases. The demonstration by
experiment of the cause of these diseases was worked out by Pasteur,
who, as we have seen, established the fact that there are different
micro-organisms inducing different kinds of fermentation, and there-
fore if it be desired to procure a pure fermentation, a pure and
not a mixed ferment must be used at the commencement; and
immediately after the primary fermentation the wine must be
" pasteurised " to destroy the disease-producing organisms. In short,
disease-producing organisms must either be excluded from, or killed
in, the wine.

By carrying out a large number of experiments, partly with
single species of yeast, and partly with mixtures, Hansen was able to
declare that many of these diseases were due to particular yeasts.
The number of such yeasts is by no means small. Hence we have
two groups of yeasts, namely, " culture " or " brewery yeasts," those
that are employed in brewing ; and " wild yeasts," occurring widely
distributed in nature, and which on gaining entrance to breweries set
up diseases in the fermentations. The development of wild yeasts is
promoted by vigorous aeration of the beer whilst it is being drawn
off, and also through the bottles being badly corked. Beer which has
undergone a feeble fermentation, and which has a high extract, is
more subject to contamination than a beer which has not. When
beer which has remained sound in the larger casks is attacked after
it has been drawn off, it is clear that the agent of the disease
obtained entrance into the beer from the surrounding air or from
unclean vessels. If the infection is not great in amount, it may,
particularly in a good beer, have practically no effect. There can be
no doubt that some of the Saccharomycetes can live for months in soil
and dust, even atmospheric dust, and amongsf these may be various
disease-yeasts.

The diseases of wines and beers are various. Generally speaking,
the chief forms are comprised in the following simple classifica-
tion : —

1. Turbidities. — (a) Gluten turbidities, or albuminous scud, due
to precipitation of albuminoids.

(b) Chemical suspension and deposits, e.g. calcium tartrate,
reduced sulphur scud, resins, essential oils, etc.

(a) Starch turbidity, due to the presence of unsacchariiied starch,

(d) Yeast turbidity, due to a high content of yeast cells.

(e) Barter i<i I turbidity, brought about by fission fungi.

2. Ropiness, which may be thus classified separately, although
doubtless frequently due to a high degree of turbidity. This con-
dition of ropiness in wine, formerly attributed to a coagulation of the
album inoids, was traced by Pasteur to a number of organisms
of which he described two chief forms, namely, a streptococcus and

112 BACTEIUA AND FERMENTATION

Bacillus viscosus vini. This latter organism occurs in the form of
small rods, frequently united in pairs, and capable of producing
ropiness in white wines in the absence of air. The presence of sugar
is a sine qua non for the occurrence of the malady, since it forms the
material from which the strings of mucus are produced. Nessler
maintains that wines containing over 10 per cent, of alcohol are proof
against ropiness.

Pasteur also investigated ropiness in beer, and traced it to
Micrococcus viscosus. But in all probability there are a number of the
Schizomycctes possessing the power of rendering beer and wine viscid.
The so-called Sarcina turbidity of beer has been traced to the Pedio-
coccus cerevisice. But it should be borne in mind that such conditions
may be easily mistaken for turbidities set up in other ways.

3. Changes in Colour.— The browning of wines — changing of
colour with turbidity and unpleasant flavour, sometimes occurring in
white wines — is said to be due to oxydaxes, enzymes produced by
some of the yeasts and setting up an oxidation.

4. Alteration of Flavour, Bitterness, Acidity, etc.— Bitterness
of wine almost exclusively affects red wines. The wine decolorises
and develops a strange odour and a bitter after-taste. Pasteur
attributed the disease to bacteria, but up to the present no species
has been isolated able to bring about this condition upon inoculation
in healthy wines. Bittering of beer may be occasioned by a disease-
yeast (Saccharomyces pastorianus I.) introduced at the commencement
of the primary fermentation, even in such small quantity as one-fifth
of the pitching yeast. This organism, according to Hansen, not only
injuriously affects the taste and odour of the beer, but also its
stability. It is of very frequent occurrence in breweries.

The turning (tourne) of wines is by no means a clearly-defined or
uniform phenomenon. The most frequent form, perhaps, is that due
to the vinegar taint (caused by Mycoderma accti). But the condition
may be set up by the lactic acid bacteria. It mostly attacks young
vintages. The wine becomes turbid, eventually having an appearance
of diluted milk, and even later it may assume a condition of brown
or inky-black liquid.

The turning of beer, on the other hand, is a simple malady due
to lactic acid fermentation, set up by the Saccharo-bacillus pastorianus
III. The beer at first loses its brightness, then becomes turbid, and
ultimately, according to some authorities, of unpleasant smell and
taste. If the sample be shaken delicate waves or films of the organ-
ism are apparent to the naked eye, and eventually the beer becomes
muddy. Hansen has shown that there are two species of yeast, S.
pastor. III. and S. ellipsoideus II, which produce the disease wThen
they are present in the pitching yeast, and are, therefore, introduced at
the commencement of the primary fermentation. Both species are

INDUSTRIAL APPLICATIONS 113

injurious when present at this stage, and indeed only at this stage.
S. ellipsoideus II is the stronger of the two species. Whilst upon this
particular subject, we may add that in 1883 Hansen demonstrated
that these much-dreaded turbidities and other beer diseases may be
due to mixtures of two yeasts, even though each of them by itself
gives a faultless product.

The Industrial Application of Bacterial Ferments

We may commence our brief category of the industrial application
of bacteria by referring the reader to fermentations, like the acetous
(which results in the manufacture of vinegar), the alcoholic {alcoholic
beverages), the lactic acid {souring of milk for dairying purposes,
cheese, etc.), the butyric (resulting in butyric acid), and those fer-
mentations occurring in the soil and improving the fertility of land
for farming purposes. With the principal facts concerning each of
these applications of bacteria to industrial processes we have dealt
elsewhere. It remains for us to mention other spheres of industry
where bacteria are, whether we recognise it or not, playing a leading
role. Their industrial effects are often secondary to vital processes.
For instance, in securing their food bacteria break down organic
material and bring about chemical and physical change. Now this
power which organisms have of chemically destroying compounds
may, or may not, be of primary importance, but there can be no doubt
that many of the products which arise as a result are of an importance
in the world which it is difficult to over-estimate. Perhaps the most
remarkable examples occur in soil and in milk. But other illustra-
tions which will at once occur to the reader are the maceration
industries. For example, linen, as is well known, is produced from
flax. The flax stem is made up of cellular substance, flax fibres and
wood fibres; the latter are of no service in the making of linen, but
the whole is bound together by a gummy, resinous Bubstance termed
"the central lamellae" (an intermediate inter-cellular Bubstance
consisting probably not of pectose, but of calcium pectate). The
solution of this cementing Bubstance can be brought about by
chemical means by treating the plant with very dilute sulphuric
acid and then neutralising the adherent acid by a weak alkali bath
(Bura). But it can also be solved by the process known as retting.
There is dew-retting and water-retting. The former is practised in
Eussia, and consists in spreading the flax ou the grass and exposing
it to the influence of dew, air, rain, and light. The result is a soft
and silky fibre. Water-retting is the method more commonly
adopted, and is accomplished by means of Bteeping the flax in bundles,
roots downwards, in innks or ponds, with appliances bo arranged as
to keep the flax below water. In ten to fourteen days, according to

li

114 BACTERIA AND FERMENTATION

the warmth of the weather, fermentation is completed hy the break-
ing away of the " shore " or " shive " (the woody core) from the flax
fibres. This decomposition and eventual breaking-down is due to
bacteria, which, under favourable circumstances, multiply rapidly
and set up the decomposition of the pectin resinous substance.
Winogradsky, in 1895, proved that the process was due to a large
bacillus (10-15 jul long, 1 jm broad). It is an anaerobe, growing not
in gelatine, but in the presence of nitrogenous food will ferment
saccharose, lactose, and starch.

A precisely similar process is used in the preparation of jute and
hemp. The former is of course used in various fabric industries, the
chief centre of such manufactures being at Dundee. Jute fibre is
obtained from the bark of at least two species of plants allied to the
lime-tree order. The fibre, which is the inner bark, is separated from
the stem by retting, either in rivers or tanks. The retting lasts for
different periods, from two days to three weeks, and when the
cementing substance between the fibres and the stem is sufficiently
decomposed to allow of it, the jute fibre is separated, and may be
woven into sacking, carpets, curtains, etc. It is said that many of
the brightly-dyed prayer-carpets used in the East by Moslems are
made of this material in Dundee, and exported. Hemp also is
cultivated in Poland and European Paissia for the sake of its fibre,
which is used for sail-cloth and other coarse material. This fibre is
also separated by retting. Another example of the same putrefactive
process is the preparation of cocoanut fibre for matting, etc. Some-
times retting for as long as twelve months is necessary to separate
the fibres from the unripe husk of the cocoanut. Sponges are cleared
in much the same manner by the putrefaction and softening of the
organic matter in their interstices, set up by micro-organisms. The
preparation of indigo from the indigo plant is brought about by a
special bacterium found on the leaves. If the leaves are sterilised
no fermentation occurs, and no indigo is formed. If, however, some
of the specific bacteria are added to the mass, the fermentation soon
begins, and the blue colour of the indigo makes its appearance. In
the treatment of ox-hides for the production of certain kinds of
leather the first object of the tanner is to clean and soften the hide,
which is accomplished by washing. The unhairing and removal of
the scarf-skin is the next operation, and this is achieved in America
by "sweating" the hides, or artificially heating them till incipient
putrefactive fermentation is set up by means of bacteria. Even in
the subsequent tanning bacteria probably play an important part.
But little is known at present of their work in this respect.

In the production of tobacco, the leaves, when gathered, are allowed
to become somewhat withered, and are then arranged in moderate-
sized heaps, where they undergo a so-called " sweating," after which

INDUSTRIAL APPLICATIONS 115

they are tied in bundles and arranged in huge heaps, containing

sometimes 50 tons of tobacco. Hereupon active decomposition
rapidly ensues, and the temperature rises to 50° or 60° C. This
fermentation is clue to bacteria, and was studied by Schloesing and
Suchslan, who used pure cultures of bacteria for the purpose of
favourably influencing the fermentation of tobacco, and producing
a definite aroma. There is some evidence to show that certain of
the family of Aspergillus co-operate with the bacteria in this process.
Throughout the needful operations in tobacco-curing the producer
has to contend with a number of micro-organisms which may produce
disease in the tobacco.

The fermentation of cellulose is an example of bacterial action
which has been more or less widely applied to industry. The
process is due to Bacillus amylobactcr, which acts, it is supposed, in
symbiotic relationship with some other organism incapable of
fermenting cellulose by itself. In relation to these so-called
industrial symbioses it will be remembered by some that Calmette
drew attention at the British Association Meeting at Dover (1899)
to the application of bacteria to various processes carried out in the
East. For example, the Japanese manufacture their sale with a
form of aspergillus described by Ahlburg in 1879, and the eau dc vie
and vins de riz of the Chinese and Javanese have their source in
symbiotic fermentations. Thus, in many cases, without the manu-
facturer even knowing it, micro-oro;anic ferments are utilised in
industrial operations.

In all these applications it is obvious we have advanced only the
first stage of the journey. Nevertheless, here, as in u.it urc on a
large scale in the formation of fertile soils and coal measures, we find
bacteria or their allies silently at work achieving great ends by
co-operating in countless hordes.

CHAPTER V

BACTERIA IN THE SOIL

Methods of Examination — Methods of Anaerobic Culture — Place and Function of
Micro-organisms in Soil — Denitrification, Nitrification, Nitrogen-fixation,
Bacterial Symbiosis — Saprophytic and Pathogenic Organisms in Soil —
Tetanus — Quarter -Evil — Malignant CEdema — The Relation of Soil to
Bacterial Diseases, such as Typhoid Fever.

Surface soils and those rich in organic matter supply a varied field
for the bacteriologist. Indeed, it may be said that the introduction
of the plate method of culture and the improved facilities for
growing anaerobic micro-organisms have opened up possibilities of
research into soil micro-biology unknown to previous generations of
workers.

From the nature of bacteria it will be readily understood that
their presence is affected by physical conditions of the soil, and in
all soils they occur only within a few feet of the surface. As we go
down below 2 feet, bacteria become less, and below a depth of
5 or 6 feet we only find a few anaerobes. At a depth of 10 feet,
and in the "ground water region," bacteria are scarce or absent.
This is held to be due to the porosity of the soil acting as a filtering
medium. Regarding the numbers of micro-organisms present in soil,
no very accurate standard can be obtained. Ordinary earth may
yield anything from 10,000 to 5,000,000 per gram, whilst from
polluted soil even 100,000,000 per gram have been estimated.
These figures are obviously only approximate, nor is an exact
standard of any great value. Nevertheless Frankel, Beumer, Miquel,
and Maggiora have, as the result of experiments, arrived at a number
of conclusions respecting bacteria in soil which are of practical use.
From these results it appears that, in addition to the " ground water
region " being free, or nearly so, virgin soils contain much fewer than
cultivated lands, and these latter, again, fewer than made soils and

116

PLATE 9.

METHODS OF EXAMINATION 117

inhabited localities. In cultivated lands the number of organisms
augments with the activity of cultivation and the strength of the
fertilisers used. In all soils the maximum occurs in July and
August.

But the condition which more than all others controls the quantity
and quality of the contained bacteria is the degree and quality of
the organic matter in the soil. The quantity of organic matter
present in soil having a direct effect upon bacteria will be materially
increased by placing in soil the bodies of men and animals after
death. Dr Buchanan Young two or three years ago performed some
experiments to discover to what degree the soil bacteria were affected
by these means. "The number of micro-organisms present in soil
which has been used for burial purposes," he concludes, "exceeds
that present in undisturbed soil at similar level, and that this excess,
though apparent at all depths, is most marked in the lower reaches
of the soil/'* The numbers were as follows : —

Virgin soil, 4 ft. 6 in. = 53,436 m.o. per gram of soil.
Burial soil (8 years), 4 ft. 6 in. = 363,411 m.o. per gram of soil.
(3 „ ), 6 ft. 6 in. = 722,751

Methods of Examination of Soil. — Two simple methods are generally adopted.
The first is to obtain a qualitative estimation of the organisms contained in the soil.
It consists simply in adding to test-tubes of liquefied gelatine or broth a small
quantity of the sample, finely broken up with a sterile rod. The test-tubes are now
incubated at 37 ° C. and 22° C, and the growth of the contained bacteria observed
in the test-tube, or after a plate culture has been made on gelatine, agar, or glucose-
litmus agar. The second plan is adopted in order to secure more accurate quanti-
tative results. One gram or half-gram of the sample is weighed on the balance,
and then added to 100 c.c. or 1000 c.c. of distilled sterilised water in a sterilised
flask, in which it is thoroughly mixed and washed. From either of these two
different sources it is now possible to make sub-cultures and plate cultures. The
procedure is, of course, that described under the examination of water (p. 463 el seq.),
and Petri's dishes, Koch's plates, or Esmarch's roll cultures are used. I Many of
the commoner bacteria in soil will thus be detected and cultivated. Spores may
be isolated, as is described under Examination of Sewage. Hut it is obvious that this
by no means covers the required ground. It will be necessary for us here to con-
sider the methods generally adopted for growing anaerobic bacteria, that is to say,
those species which will not grow in the presence of oxygen. This anaerobic
difficulty may be overcome in a variety of ways.

Methods of Anaerobic Cultivation

1. The oxygen may be displaced by some other gas, and though coal-gas,
nitrogen, and carbon dioxide may all be used for this purpose, it has become the
almost universal practice to grow anaerobes in hydrogen. The hydrogen is readily
obtained by Kipp's or some other suitable apparatus for the generation of hydrogen

I* roc Royal Sue. of Edin., xxxvii., pt. iv. , p. 759.
I See also Report <>/'//,,■ Medical Officer t<> the Local Government B<><ir<l (1SP7-9S]
A. C. Houston, pp. l'.:>1-:507.

118

BACTERIA IN THE* SOIL

H

from zinc and dilute sulphuric acid, or it may be provided in a cylinder. The free
gas is passed through various washbottles to purify it of any contaminations ; e.g.
lead acetate (1-10 of water) removes any traces of sulphuretted hydrogen, silver
nitrate (1-10) doing the same for arseniated hydrogen ; whilst a flask of pyrogallate
of potash will remove any oxygen. It is not necessary to have these three purifiers
if the zinc used in the Kipp's apparatus is pure. Occasionally a fourth flask is added
of distilled water, and this, or a dry cotton-wool stopper in the exit tube, will ensure
germ-free gas. From the further end of the exit tube of the Kipp's apparatus an
indiarubber tube will carry the hydrogen to its desired destination. With some it is
the custom to place anaerobic cultures in test-tubes, and the test-tubes in a large
flask, tube, or desiccator, having a two-way tube for entrance and exit of the
hydrogen, or Petri dishes may be used and placed in well-sealed jar or desiccator ;
others prefer to pass the hydrogen immediately into a large test-tube containing the
culture (Friinkel s method). Either method, if properly carried out, will be found
effectual, and the growth of the culture in hydrogen is readily observed. Another
plan is to use a yeast flask, and after having passed the
hydrogen through for about half an hour, the lateral exit
tube is dipped into a small capsule containing mercury
(as in Plate 9). The entrance tube is now sealed, and
the whole apparatus placed in the incubator. The
interior of the flask containing the culture is filled with
an atmosphere of hydrogen. No oxygen can obtain
entrance through the sealed entrance tube, or through
the exit tube immersed in mercury Yet through this
latter channel any gases produced by the culture may
escape.

2. The Absorption Method. — Instead of adding hydro-
gen to the tube or flask containing the anaerobic culture,
it is feasible to add to the medium substances, such as
glucose or pyrogallic acid, which will absorb the oxygen
which is present, and thus enable the anaerobic require-
ment to be fulfilled. To various media — gelatine, agar,
or broth (the latter used for obtaining the toxins of
anaerobes) — 2 per cent, of glucose may be added.
Pyrogallic acid, or pyrogallic acid one part and 20 per
cent, caustic potash one part, is also readily used for
absorptive purposes. A large glass tube of 25 c.c. height,
termed a Buchner's cylinder, having a constriction near
the bottom, is taken ; and about two drachms of the
pyrogallic solution are placed in the bulb. A test-tube
containing the culture is now lodged in the upper part
above the constriction, and the mouth of the Buchner
tube is carefully sealed. The apparatus is then placed in the incubator at the desired
temperature, and the contained culture grows under anaerobic conditions. As the
pyrogallic solution absorbs the oxygen it assumes a darker tint.

3. Mechanical Methods. — These include various ingenious methods for preventing
an admittance of oxygen to the culture. An old-fashioned one was to plate out the
culture and protect it from the air by covering it with a plate of mica. A more
serviceable mode is to inoculate, say, a tube of agar with the anaerobic organism,
and then pour over the culture a small quantity of melted agar, which will readily
set, and so protect the culture itself from the air. Oil or vaseline may be used
instead of melted agar. Another mechanical method is to make a deep inoculation,
and then melt the top of the medium over a Bunsen burner, and thus close the
entrance puncture and seal it from the air.

4. Absorption of Oxygen by an Aerobic Culture. — This method takes advantage
of the power of absorption of certain aerobic bacteria, which are planted over the
culture of the anaerobic species. It is not practically satisfactory, though occasionally
good results have been obtained.

5. Lastly, there is the Vacuum Method. — By this means it is obviously intended
to extract air from the culture and seal it in vacuo. The culture tubes are connected

<J

Fig. IS. — Frankkf/s Tube.
For Cultivation of Anaerobes.

PLATE 10.

A V\m M BlE I A\ tEROBIl ('mm i:k.

[ To face page 1 1 8.

QUALITATIVE EXAMINATION 119

with the air-pump, and exhausted as much as possible. The method can be applied
in many different ways (for example, with pyrogallic solution, as in Bulloch's
apparatus).

Of these various methods it is on the whole best lo choose either the hydrogen
method, the vacuum, or the plan of absorption by grape-sugar or pyrogallic. In
anaerobic plate cultures grape-sugar agar plus 0*5 per cent, of formate of soda may
be used. The poured inoculated plate should be placed over pyrogallic solution
under a sealed bell-glass and incubated at 37° C. Pasteur, Roux, Joubert, Chamber-
land, Esmarch, Kitasato, and others have introduced special apparatus to facilitate
anaerobic cultivation, but the principles adopted are those which have been
mentioned.

The Qualitative Examination of Bacteria in the Soil. — We

may now turn to consider the species of bacteria found in the soil.
They may be classified in five main groups ; the division is somewhat
artificial, but convenient : —

1. The Denitrifying Bacteria. — This group, whose function has
been elucidated largely by the investigations of Professor Warington,
is held responsible for the breaking clown of nitrates. With its
members may be associated the Dccomjiosition or Putrefactive
Bacteria, which break down complex organic products other than
nitrates into simpler bodies.

2. The Organisms of Nitrification. — To this group belong the two
chief types of nitrifying bacteria, viz., those which oxidise ammonia
into nitrites, and those which change nitrites into nitrates.

3. The Nitrogen-fixing Bacteria, found mainly in the nodules on
the rootlets of certain plants.

4. The Common Saprophytic Bacteria, whose function is at present
but imperfectly known. Many are putrefactive germs.

5. The Pathogenic Bacteria. — This division includes three types,
the bacilli of tetanus, malignant oedema, and quarter evil. Under this
heading we shall also have to consider in some detail the intimate
relation between the soil and such important bacterial diseases as
tubercle and typhoid.

To enable us to appreciate the work which the "economic
bacteria" perform, it will be necessary to consider shortly the plate
they occupy in the economy of nature. This may be perhaps most
readily accomplished by studying the scheme shown on p. 120.

The threefold function of ordinary plant life is nutrition, assimi-
lation, and reproduction, i.e., the food of plants, the digestive and
storage power of plants, and the various means they adopt for multi-
plying and increasing their species. With the two latter we have
little concern in this place. Eespecting the nutrition of plant life,
it is obvious that, like animals, plants must feed and breathe to
maintain life. Plant food is of three chief kinds, viz., water, inorganic
salts, and gases. Water is an actual necessity to the plant as a direct
food and as a food-solvent, i.e. as the vehicle of important inorganic
materials. The hydrogen, too, of the organic compounds is obtained

120

BACTERIA IN THE SOIL

from the decomposition of the water which permeates every part of
the plant, and is derived by it from the soil and from the aqueous
vapour in the atmosphere. The chief inorganic salts of which proto-
plasm is constituted are composed in part of potassium, magnesium,
calcium, iron, phosphorus, or sulphur. These inorganic elements do
not enter the plant as such, but combined with other substances or
dissolved in water. Potassium, calcium, and magnesium are absorbed
chiefly as nitrates, phosphates, and carbonates. Iron contributes to
the formation of the green colouring-matter of plants, indeed, is
essential to it, and is also derived from the soil. Phosphorus, one of
the chief constituents of seeds, generally occurs as nucleo-albumin.

A SCHEME SHOWING THE PLACE AND FUNCTION OF THE
ECONOMIC MICRO-ORGANISMS FOUND IN SOIL

Water.

\

Inorganic Sails.
[Nitrates, etc.]

I

Gases.
[C02,H,N,0].

PLANT LIFE.

I

Carbohydrates

[albumoses, sugar

starch, etc.].

Fats. Proteids

[bodies containing
Nitrogen].

ANIMAL LIFE.
I

Vegetable Mineral Water,
Acids. Salts.

I
Gases [CO,,, etc.].

Water.

I

Urea, Albuminoids,

Ammonia compounds,

etc.

Nitrogen in many

forms locked up

in the body.

Decomposition and Denitrifying Bacteria.
I

Free Nitrogen. Gases [CO.,].

Nitrooen-Fix i no
Bacteria.

[In soil and in the nodules
on the rootlets of Legu-
minosce. ]

I I I

Water. Ammonia, [Nitrites].

and other elements
of broken-down
complex bodies.

Y

Nitrifying Bacteria.

Nitrites [ = Nitrous organism
(Nitrosomonas)] .
Nitrates [ = Nitric organism
(Nitromonas, or
Nitrobacter)].
[In soil and
available for
plant life. ]

CONDITIONS OF PLANT LIFE 121

Sutyhur, which is an important constituent of albumen, is derived
from the sulphates of the soil. In addition to the above, there are
other elements, sometimes described as non-essential constituents of
plants. Amongst these are silicon (to give stiffness), sodium, chlorine,
iodine, bromine, etc. All these elements contribute to the formation
or quality of the protoplasm of plants.

The gases essential to plants, and absorbed as such, are two :
Carbon dioxide (carbonic acid) and Oxygen ; the necessary hydrogen
and nitrogen being absorbed in the form of salts. By the aid of the
green chlorophyll corpuscles, and under the influence of sunlight, we
know that leaves absorb the carbon dioxide of the atmosphere, and
effect certain changes in it. The hydrogen, as we have seen, is
obtained from the water. Oxygen is absorbed through the leaves and
through the root from the interstices of the soil. Each of these gases
contributes vitally to the existence of the plant. The fourth gas, nitro-
gen, which constitutes more than two-thirds of the air we breathe (79
per cent, of the total volume and 77 per cent, of the total weight of the
atmosphere), is also an absolutely necessary food required by plants.
Yet, although this is sot the plant cannot absorb or obtain its nitrogen
in the same manner in which it acquires its carbon — viz., by absorption
through the leaves — nor can the plant take nitrogen into its own
substance by any means as nitrogen. Hence, although this gas is
present in the atmosphere surrounding the plant, the plant will
perish if nitrogen does not exist in some combined form in the soil.
Nitrates and compounds of ammonia are widely distributed in nature,
and it is from those bodies that the plant obtains, by means of its
roots, the necessary nitrogen.

Until comparatively recently it was held that plant life could not
be maintained in a soil devoid of nitrogen or compounds thereof.
But it has been found that certain classes of plants (the Lcguminosw
for example), when they are grown in a soil which is practically free
from nitrogen at the commencement, do take up this gas into their
tissues. One explanation of this fact is that free nitrogen becomes
converted into nitrogen compounds in the soil through the influence
of micro-organisms present there. Another explanation attributes
this fixation of free nitrogen to micro-organ isms existing in the
rootlets of the plant. These two classes of organisms, known as the
nitrogen-fixing organisms, will require our consideration at a later
stage. Here we merely desire to make it clear that the main supply
of this gas, absolutely necessary to the existence of vegetable life
upon the earth, is drawn not from the nitrogen of the atmosphere,
but from that contained in nitrogen compounds in t lie soil. The
most important of these are the nitrates. Here then we have the
necessary food of plants expressed in a sentence : water, inorganic suits,
gases; some of the salts containing nitrogen in the form of nitrates.

122 BACTERIA IN THE SOIL

Plant life seizes upon its required constituents, and by means of
the energy furnished by the sun's rays builds these materials up into
its own complex forms. Its many and varied forms fulfil a place in
beautifying the world. But their contribution to the economy of
nature is, by means of their products, to supply food for animal life.
These products of plant life are chiefly sugar, starch, fat, and proteids.
Animal life is not capable of extracting its nutriment from soil, but
it must take the more complex foods which have already been built
up by vegetable life. Again, the complementary functions of animal
and vegetable life are seen in the absorption by plants of one of the
waste materials of animals, viz., carbonic acid gas. Plants abstract
from this gas carbon for their own use, and return the oxygen to the
air, which in its turn is of service to animal life.

By animal activity some of these foods supplied by the vegetable
kingdom are at once decomposed into carbonic acid gas and water,
which goes back to nature. Much, however, is built up still further
into higher and higher compounds. The proteids are converted by
digestion into more soluble forms, such as albumoses and peptones ;
these in their turn are reconverted into less soluble proteids, and
become assimilated as part of the living organism. In time they
become further changed into carbonic acid, sulphuric acid, water, and
certain not fully oxidised products,* which contain the nitrogen of
the original proteid. In the table these bodies have been represented
by one of their chief members, viz., urea.

It is clear that there is in all animal life a double process
continually going on; there is a building up (anabolism, assimilation),
and there is a breaking down (katabolism). These processes will not
balance each other throughout the whole period of animal life. We
have, as possibilities, elaboration, balance, degeneration ; and the
products of animal life will differ in degree and in substance accord-
ing to which period is in the predominance. These products we
may subdivide simply into excretions during life and final materials
of dissolution after death, both of which may be used more or less
immediately by other forms of animal or vegetable life, or immediately
after having passed to the soil. We may shortly summarise the
final products of animal life as carbonic acid, water, and nitrogenous
remnants. These latter will occur as urea, new albumens, compounds
of ammonia, and nitrogen compounds of great complexity stored up
in the tissues and body of the animal. The carbonic acid, water, and
other simple substances like them will return to nature and be of
immediate use to vegetable life. But otherwise the cycle cannot be
completed, for the more complex bodies are of no service as such
to plants or animals.

* E. A. Schafer, Text-book of Physiology, vol. i., p. 25 (W. D. Halliburton).

CONCERNED WITH PUTREFACTION 123

1. Decomposition and Denitrification

In order that this complex material should be of service in the
economy of nature, and its constituents not lost, it is necessary that
it should be broken down again into simpler conditions. This
prodigious task is accomplished by the agency of two groups of
organisms, the decomposition and denitrifying * bacteria. The organ-
isms associated with decomposition processes are numerous ; some
denitrify as well as break down organic compounds. This group
will be referred to under " Saprophytic Bacteria." The reduction by
the denitrifying bacteria may* be simply from nitrate to nitrite, or
from nitrate to nitric or nitrous oxide gas, or indeed to nitrogen
itself. In all these processes of reduction the rule is that a loss of
nitrogen is involved. How that free nitrogen is brought back again
and made subservient to plants and animals we shall understand at
a later stage.

Professor Warington has set forth the chief facts known of
this decomposition process.f That the action in question only
occurs in the presence of living organisms was first established
by Mensel in 1875 in natural waters, and by Macquenne in 1882
in soils. If all living organisms are destroyed by sterilisation
of the soil, denitrification cannot take place, nor can vegetable
life exist. "Bacteria reduce nitrates," says Professor Warington,
" by bringing about the combustion of organic matter by the oxygen
of the nitrate, the temperature distinctly rising during the operation."
The reduction to a nitrite is a common property of bacteria. But
only a few species have the power of reducing a nitrate to gas.
These few species are, however, widely distributed. In 1886 Gayon
and Dupetit first isolated the bacteria capable of reducing nitrates
to the simplest element, nitrogen. They obtained their species from
sewage, but ten years later denitrifying bacteria were isolated from
manure. That soil contains a number of these reducing organisms
is proved by introducing a particle of surface soil into some broth,
to which has been added 1 per cent, of nitre. During incubation of
such a, tube gas is produced, and the nitrate entirely disappears.

Whenever decomposition occurs in organic substances there is a
reduction of compound bodies, and in such cases the putrefying
substances obtain their decomposing and denitrifying bacteria from
the air. The chief conditions requisite for bringing about a loss of
nitrogen by denitrification are enumerated by Professor Warington
as follows: — (1) the specific micro-organism; (2) the presence of a
nitrate and suitable organic matter; (3) such a condition as to

* " Denitrifying" means reducing nitrates.

t R. Warington, M.A., F.R.S., Jour. Roy. Agric. Soc. Eng.t series iii., vol. viii/J
part. i\\, p. .".77 et seq. See also Trans. Chem, Si>c, 1884, 1SSS, ete.

124 BACTERIA IN THE SOIL

aeration that the supply of atmospheric oxygen shall not be in excess
relatively to the supply of organic matter; (4) the usual essential
conditions of bacterial growth. "Of these," he says, "the supply of
organic matter is by far the most important in determining the
extent to which denitrification will take place." The necessarily
somewhat unstable condition facilitates its being split up by means
of bacteria. The bacteria in their turn are ready to seize upon any
products of animal life which will serve as their food. Thus, by
reducing complex bodies to simple ones, these denitrifying organ-
isms act as the necessary link to connect again the excretions of
the animal body, or after death the animal body itself, with the
soil.

In a book of this nature it has been deemed advisable not to
enter into minute description of all the species of bacteria mentioned.
Some of the chief are described more or less fully. We cannot,
however, do more than name several of the chief organisms
concerned in reducing and breaking down compounds. As we shall
find in the bacteria of nitrification, so also here, the entire process
is rarely, if ever, performed by one species. There is indeed a
remarkable division of labour, not only between decomposition
bacteria and denitrification bacteria, but between different species
of the same group. Bacillus fluorescens non-liquefaciens, Myco-
derma urece, and some of the staphylococci break down nitrates
(denitrification), and also decompose other compound bodies. Amongst
the group of putrefactive bacteria found in soil may be named B.
coli, B. mycoides, B. mcsentericus, B. liquidus, B. prodigiosus, B.
ramosus, B. vermicularis, B. liquefaciens, and many members in the
great family of Proteus. Some perform their function in soil, others
in water, and others, again, in dead animal bodies. Dr Buchanan
Young, to whose researches in soil we have referred, has pointed out
that in the upper reaches of burial soil, where these bacteria are
most largely present, there is as a result no excess of organic carbon
and nitrogen. Even in the lower layers of such soil it is rapidly
broken down.

It will be observed, from a glance at the table (p. 120), that the
chief results of decomposition and denitrification are as follow : free
nitrogen, carbonic acid gas, and water, ammonia bodies, and some-
times nitrites. The nitrogen passes into the atmosphere, and is
" lost " ; the carbonic acid and water return to nature, and are at
once used by vegetation. The ammonia and nitrites await further
changes. These further changes become necessary on account of the
fact, already discussed, that plants require their nitrogen to be in the
form of nitrates in order to use it. Nitrates obviously contain a
considerable amount of oxygen, but ammonia contains no oxygen, and
nitrites very much less than nitrates. Hence a process of oxidation

CONCERNED WITH NITRIFICATION 125

is required to change the ammonia into nitrites and the nitrites into
nitrates.

2. Nitrification

This oxidation is performed by the nitrifying micro-organisms,
and the process is known as nitrification. It should be clearly
understood that the process of nitrification may, so to speak, dovetail
with the process of denitrification. No exact dividing line can be
drawn between the two, although they are definite and different
processes. In a carcase, for example, both processes may be going
on concomitantly, so also in manure. There is no hard-and-fast line
to be drawn in the present state of our knowledge. Other organisms
beside the true nitrification bacteria may be playing a part, and it is
impossible exactly to measure the action of the latter, where they
began and where the preliminary attack upon the nitrogenous com-
pounds terminated. In all cases, however, according to Professor
Warington, the formation of ammonia has been found to precede the
formation of nitrous or nitric acid.

It was Pasteur who (in 1862) first suggested that the production
of nitric acid in soil might be due to the agency of germs, and it is
to Schlosiiiff and Muntz that the credit belongs for first demonstrat-
ing (in 1877) that the true nature of nitrification, the conversion of
ammonia into nitric acid, depended upon the activity of a living
micro-organism.* Partly by Schlosing and Miintz and partly by
Warington (who was then engaged in similar work at Eothamsted), it
was later established (1) that the power of nitrification could be com-
municated to substances which had not hitherto nitrified by simply
seeding them with a nitrified substance, and (2) that the process of
nitrification in garden soil was entirely suspended by the vapour of
chloroform or carbon disulphide. The conditions for nitrification,
the limit of temperature, and the necessity of plant food, have
furnished additional proof that the process is due bo a living organism.
These conditions, according to Warington, are as follows: —

1. Food (of which phosphates are essential constituents). "The
nitrifying organism can apparently feed upon organic matter, but it
can also, apparently with equal ease, develop and exercise all its
functions upon purely inorganic food" (J. M. H. Munro).f Wino-
gradsky prepared vessels and solutions carefully purified from
organic matter, and these solutions lie sowed with the nitrifying
organism, and found that they flourished. Professor Warington lias
employed the acid carbonates of sodium and calcium with distinct
success as ingredients of an ammoniacal solution undergoing nitrifi-
cation.

2. The next condition of nitrification is the presence of oxygen.

Compt. Rend., ls77, pp. si, 301. i Trans. Chem. Sue, lbstj, etc.

126 BACTERIA IN THE SOIL

Without it the reverse process, denitiification, occurs, and instead
of a building up we get a breaking down, with an evolution of
nitrogen gas. The amount of oxygen present has an intimate pro-
portion to the amount of nitrification, and with 16 to 21 per cent, of
oxygen present the nitrates are more than four times as much as
when the smallest quantity of oxygen is supplied. The use of tillage
in promoting nitrification is doubtless in part due to the aeration of
the soil thus obtained.

3. A third condition is the presence of a base with which nitric
acid when formed may combine. Nitrification can only take place
in a feebly alkaline medium, but an excess of alkalinity will retard
the process.

4. The last requirement is a favourable temperature. As low as
37° or 39° F. (3-4° C.) will suffice, but at a higher temperature it
becomes much more active. According to Schlosing and Muntz, at
54° F. (12° C.) nitrification becomes really active, and it increases
as the temperature rises to 99° F. (37° C), after which it falls. A
high temperature or a strong light are prejudicial to the process.

We are now in a position to consider shortly some of the char-
acters of these nitrification bacteria. They may readily be divided
into two chief groups, not in consideration of their form or biological
characteristics, but on account of the duties which they perform.
Just as we observed that there were few denitrifying organisms
which could break down ammonia compounds to nitrogen gas, so is it
also true that there are few nitrifying bacteria which can build up
from ammonia to the nitrates. Nature has provided that this shall
be accomplished in two stages, viz., a first stage from ammonia bodies
to nitrites (nitrosificalion), and a second stage from nitrites to
nitrates. The agent of the former is termed the nitrous organism,
the latter the nitric organism. Both are contributing to the final
production of nitrates which can be used by plant life.*

The Nitrous Organism (Mtrosomonas). Prior to Koch's gelatine
method the isolation of this bacterium proved an exceedingly difficult
task. But even the adoption of this isolating method seemed to give
no better results, and for an excellent reason : the nitrifying
organisms will not grow on gelatine. To Winogradsky f and Percy
FranklandJ belongs the credit of separately isolating the nitrous
organism on the surface of gelatinous silica containing the necessary

* The saltpetre beds of Chili and Peru are an excellent example of the applica-
tion of these facts. Nitrates are there produced from the faecal evacuations of sea-
fowl in such quantities as to form an article of commerce. A like form of utilisation
of the action of these bacteria was once practised on the continent of Europe.
Considerable nitrate deposits have recently been discovered in California. Economic
application is also seen in the treatment of sewage referred to elsewhere.

t Ann. de VInst. Pasteur, 1890, p. 213.

XPhil Trans. Roy. Soc, 1890, B. 107.

THE NITROUS ORGANISM 127

inorganic food. Professor Warington, in his lectures under the Lawes
Agricultural Trust, has described this organism as follows : —

" The organism as found in suspension in a freshly nitrified solu-
tion consists largely of nearly spherical corpuscles, varying extremely
in size. The largest of these corpuscles barely reaches a diameter of
one-thousandth of a millimetre, and some are so minute as to be
hardly discernible in photographs. The larger ones are frequently
not strictly circular, and are sometimes seen in the act of dividing.

"Besides the form just described, there is another, not universally
present in solutions, in which the length is considerably greater than
its breadth. The shape varies, being occasionally a regular oval, but
sometim.es largest at one end, and sometimes with the ends truncated.
The circular organisms are probably the youngest.

" This organism grows in broth, diluted milk, and other solutions
without producing turbidity. When acting on ammonia it produces
only nitrites. It is without action on potassium nitrite. It is, in
fact, the nitrous organism which, as we have previously seen, may be
separated from soil by successive cultivations in ammonium carbonate
solution."*

The elongated forms appear to be a sign of arrested growth.
Normally, the organism is about 1"8 /ul long, or nearly three times as
long as the nitric organism. It possesses a gelatinous capsule. " The
motile cells, stained by Lolller's method, are seen to have a flagelliun
in the form of a spiral." When grown on silica jelly the nitrous
organism appears in the same two forms — zooglea and free cells — as
when cultivated in a fluid. It commences to show growth in about
four days, and is at its maximum on about the tenth day. Upon
gypsum, to which 1 per cent, of magnesium carbonate has been added,
the organism grows in the same form of small brown colonies, but
more rapidly. Winogradsky found that there were considerable
differences in the morphology of the organism according to the soil
from which it was taken. The solution used by him consisted of
water containing 1 per 1000 ammonium sulphate, 1 per 1000 potas-
sium phosphate, and 1 per 100 magnesium carbonate.

As we have already seen, an astonishing property of this organism
is its ability to grow and perform its specific function in solutions
absolutely devoid of organic matter (Munro). Some authorities hold
that it acquires its necessary carbon from carbonic acid. The mode
of culturing it was as follows: — To sterilise. I Bat-bottomed flasks add
100 c.c. of a solution made of two grams of ammonium sulphate, one
gram of potassium phosphate, and L000 c.c. of distilled water. To
this was added half a gram of magnesium sulphate, two grams of
common salt, and 0 '4 of a gram of ferrous sulphate. Now the flask

U.S.A. Dept of Agriculture: Lectures under the Lawes Agricultural Trust.

By Robert Warington, F.K.S., 1801, pp. 58, 59.

128 BACTERIA IN THE SOIL

was inoculated with a small portion of the soil under investigation,
and after four or five days sub -cultured on the same medium in fresh
flasks, and repeated half a dozen times. Now, as this inorganic
medium was unfavourable to the ordinary bacteria of soil, it was
supposed that after several sub-cultures the nitrous organism was
isolated in pure culture. Winogradsky employed for sub-culturing
upon a solid medium a mineral gelatine, silica jelly.* Upon this
medium it is possible to sub-culture a pure growth from the film at
the bottom of the flasks in which the nitrous organism is first
isolated. In 1899 Winogradsky showed that the nitrous organism
(nitroso-bacterium) was able to grow in the presence of large amounts
of organic matter, and since that date Fremlin has carried this branch
of work to a further stage of advancement. He has shown that
cultures developed in inorganic solutions become eventually pure
cultures of this species of nitrifying organism, and when inoculated
into solutions containing small quantities of organic matter they were
able to oxidise the ammonia present. Fremlin has also demonstrated
that the nitrous organism grows well on silica jelly and ammonia agar,
and colonies from these media transferred to beef-broth agar and
gelatine also grew well. From these experiments he concluded " that
the nitroso-bacterium grows well on any ordinary medium" but " that
in the presence of large percentages of organic matter the nitroso-
bacterium, although growing very profusely, loses for a time the power
of converting ammonia into nitrites."f

The Nitric Organism. — It was soon learned that the nitrous
organism, even when obtainable in large quantities and in pure
culture, was not able entirely to complete the nitrifying process. As
early as 1881 Professor Warington had observed that some of his
cultures, though capable of changing nitrites into nitrates, had no
power of oxidising ammonia. These he had obtained from advanced
sub-cultures of the nitrous organism, and somewhat later Wino-
gradsky isolated and described this companion of the nitrous
organism. It develops freely in solutions to which no organic matter
has been added ; indeed, much organic matter will prevent it growing.^

The temperature for incubation is 30° C. Winogradsky con-

* Two per cent, of dialysed silicic acid mixed with neutral salts and magnesium
carbonate in order to solidify it.

t Jour, of Hyg., 1903, pp. 378, 379.

X Compt. Rend., 113 (1891) p. 89— Winogradsky isolated it from soils from various
parts of the world on the following medium : — Water, 1000*0 ; potassium phosphate,
1*0; magnesium sulphate, 0'5; calcium chloride, a trace; sodium chloride, 2*0.
About 20 c.c. of this solution was placed in a flat-bottom flask, and a little freshly
washed magnesium carbonate was added. The flask was closed with cotton-wool,
and the whole sterilised. To each flask 2 c.c. of a 2 per cent, solution of ammonium
sulphate was subsequently added. Recently, the following medium has been used
for cultivation of the nitric organism : — Sodium nitrite, 1 "0 ; sodium carbonate, 1*0 ;
sodium chloride, 0*5 ; potassium phosphate, 0*5 ; magnesium sulphate, 0*3 ; ferrous
sulphate, 0*4, in 1000 parts of distilled water.

PLATE It.

Wit '^

Nitrous Organism.

x 1000.

Nitric Organism.
x 1000.

NlTROOEN-FlXING I >R<

danisms from Secretions of
Film preparations.

ROOT-NODULES TAKES PROM LEG! MINOR*.

x 1000.

[To face page 128.

THE NITRIC ORGANISM 129

eluded that the oxidation of nitrites to nitrates was brought about by
a specific organism independently of the nitrous organism. He suc-
cessfully sub-cultured it from his inorganic medium on to silica jelly
and also on to purified agar. He believes the organism, like its com-
panion, derives its nutriment solely from inorganic matter, but this
is not finally established.

The form of the nitric organism (or nitromonas, as it was once
termed) is allied to the nitrous organism. The cells are elongated,
rarely oval, but sometimes pear-shaped. They are more than half a
micromillimetre in length, and somewhat less in thickness. The
cells have a gelatinous membrane. Like the other nitrifying bacteria,
its development and action are favoured by the presence of the acid
carbonates of calcium and sodium. Of the latter, six grams per litre
or even a smaller quantity gives good results. The sulphate of
calcium can be used, but the organism prefers the carbonates. The
differences between these two bacteria are small, with the exception
of their chemical action. The nitric organism has no action uj)on
ammonia, and its presence in very small amount (five parts per
million) hinders its development, and in sixty-four parts per million
prevents its action on a nitrite.*

We may here summarise the general facts respecting nitrification.
Winogradsky proposes to term the group nitroso-baderia, and to
classify thus: —

(Nitrosomonas, containing at least two
species, viz., the European and the
Java.
Nitrosococcus.
Nitric organism Nitrobacter.

Nitrification occurs in two stages, each stage performed by a
distinct organism. By one (nUrosomonas) ammonia is converted into
nitrite; by the other (nitrobacter) the nitrite is converted into
aitrate.f Both organisms are widely and abundantly distributed in

* The course of nitrification maybe followed by means of chemical tests. 1.
The disappearance of ammonia. 2. The appearance of nitrite. 3. Its disappear-
ance. 4. Appearance ot nitrate.

! Professor Warington, in Report IV. (p. 526) of his admirable series of papers
on the subject, draws attention to Miintz's criticism that the nitrifying organisms
only oxidise from nitrogenous matter to nitrites, and not from nitrites to nitrates.
Muntz held that the conversion of nitrite into nitrate is brought about by the joint
action of carbonic acid and oxygen. Professor Warington's experiments, however,
clearly illustrate that the production of nitrates from nitrites in an ammoniacal solu-
tion can be determined by the character of the bacterial culture with which the
solution is seeded, and that in a solution of potassium nitrite conversion into
nitrate can be determined by the introduction of the nitric organism. Professor
Warington still adheres to the opinion, in favour of which he has produced so much
evidence, that the formation of nitrates in the soil is due to the nitric organism
which soil always contains.

I

130 BACTERIA IN THE SOIL

the superficial soils. They act together and in conjunction, and for
one common purpose. They are separable by employing favourable
media. " If we employ a suitable inorganic solution containing potas-
sium nitrite, but no ammonia, we shall presently obtain the nitric
organism alone, the nitrous organism feeding on ammonia being
excluded. If, on the other hand, we employ an ammonium carbonate
solution of sufficient strength, we have selected conditions very un-
favourable to the growth of the nitric organism, and a few cultiva-
tions leave the nitrous organism alone in possession of the field " *
(Warington). Adeney has summarised conclusions respecting
nitrification as follows: (1) In organic solutions containing ammonia
nitrous organisms thrive, but nitric organisms gradually lose their
vitality ; (2) nitrous organisms cannot oxidise nitrites to nitrates in
inorganic solutions ; (3) nitric organisms thrive in inorganic solutions
containing nitrites; (4) the }3resence of peaty or humous matter
appears to preserve the vitality of nitric organisms during the
fermentation of ammonia, and establishes conditions whereby it is
possible for the nitric organisms to thrive simultaneously in the
same solution as the nitrous organisms. Other conditions necessary
for nitrification are, of course, the presence of ammonia preceding
the appearance of nitrous or nitric acid, the presence of a fixed base,
not too high a degree of alkalinity, and darkness and free admission
of air.

A word may be said upon the natural distribution of these nitrify-
ing bacteria before we leave them. They belong to the soil, river-
water, and sewage. They are also said to be frequently present in
well-water. From experiments at Eothamsted it appears that the
organisms occur mostly in the first 12 inches, in subsoils of clay down
to 3 or 4 feet, and in sandy soils probably at even a greater depth.
These facts are of the first importance in relation to the biological
treatment of sewage.

We have now given some consideration to the chief events in the
life-cycle of nature depicted in the table (p. 120). There is but one
further process in which bacteria play a part, and which requires some
mention. It will have been noticed that at certain stages in the
cycle there is a more or less appreciable " loss " of free nitrogen. In
the process of decomposition brought about by the denitrifying
bacteria, a very considerable portion of the nitrogen is dissipated
into the air in the form of a free gas. This is the last stage of all
proteid decomposition, so that wherever putrefaction is going on
there is a continual "loss" of an element essential to life. Thus it
would appear at first sight that the sum-total of nitrogen food must
be diminishing.

But there are other ways also in which nitrogen is being set free.
* Lawes Agricultural Trust Lectures^ 1891, p. 63.

NITROGEN-FIXING BACTERIA 131

In the ordinary processes of vegetation there is a gradual draining of
the soil and a passing of nitrogen into the sea ; the products of
decomposition pass from the soil by this drainage, and are "lost" as
far as the soil is concerned. Many of the methods of sewage dis-
posal are in reality depriving the land of the return of nitrogen,
which is its necessity. Again, nitrogen is freed in explosions of
gunpowder, nitroglycerine, and dynamite, for whatever purpose they
are used. Hence the great putrefactive "loss" of nitrogen, with its
subsidiary losses, contributes to reduce this essential element of all
life, and if there were no method of bringing it back again to the
soil, it would seem that plant life, and therefore animal life, would
speedily terminate.

3. Nitrogen-Fixing' Bacteria

It is at this juncture, and to perform this vital function,
that the nitrogen-fixing bacteria play their wonderful part: they
help to recover the free nitrogen and fix it in the soil. Excepting a
small quantity of combined nitrogen coming down in rain and in
minor aqueous deposits from the atmosphere, the great source of the
nitrogen of vegetation is the store in the soil and subsoil, whether
derived from previous accumulations or from recent supplies by
manure.

Sir William Crookes has pointed out the vast importance of
using all the available nitrogen in the service of wheat production.*
The distillation of coal in the process of gas-making yields a certain
amount of its nitrogen in the form of sulphate of ammonia, and this,
like other nitrogenous manures, might be used to give back to the
soil some of the nitrogen drained from it. But such manuring
cannot keep pace, according to Sir W. Crookes, with the present
loss of fixed nitrogen from the soil. We have already referred to
several ways in which "loss" of nitrogen occurs. To these may well
be added the enormous loss occurring in the waste of sewage when it
is passed into the sea. As the President of the British Association
pointed out, the more widely this wasteful system is extended,
recklessly returning to the sea what we have taken from the land,
the more surely and quickly will the finite stocks of nitrogen, locked
up in the soils of the world, become exhausted. Let us remember
that the plant creates nothing in this direction ; there is no com-
bined nitrogen in wheat which is not absorbed from the soil, and
unless the abstracted nitrogen is returned to the soil, its fertility
must be ultimately exhausted. When we apply to the land sodium
nitrate, sulphate of ammonia, guano, and similar manurial substances,
we are drawing on the earth's capital, and our drafts will no! bo

The Wheat Problem. 1S99.

132 BACTERIA IN THE SOIL

perpetually responded to.* We know that a virgin soil cropped for
several years loses its productive powers, and without artificial aid
becomes unfertile. For example, through this exhaustion forty
bushels of wheat per acre have dwindled to seven. Eotation of crops
is an attempt to meet the problem, and the four-course rotation of
turnips, barley, clover, and wheat witnesses to the fact that practice
has been ahead of science in this matter. It is unnecessary to add
that rotation of crops and the use of the Leguminosse does not absolve
the agriculturist from maintaining the land in ripe condition by
manuring and ordinary tillage.

The store of nitrogen in the atmosphere is practically unlimited,
but it is fixed and rendered assimilable only by organic processes of
extreme slowness. We may shortly glance at these, for it is upon
these processes, plus a return to the soil of sewage, that we must
depend in the future for storing nitrogen as nitrates.

1. Some combined nitrogen is absorbed by the soil or plant from
the air, for example, fungi, lichens, and some algse, and the absorption
is in the form of ammonia and nitric acid. This is admittedly a
small quantity.

2. Some free nitrogen is fixed within the soil by the agency of
porous and alkaline bodies.

3. Some, again, may be assimilated by the higher chlorophyllous
plants themselves, independently of bacteria (Frank).

4. Electricity fixes, and may in the future be made to fix more,
nitrogen. If a strong inducive current be passed between terminals,
the nitrogen from the air enters into combination with the oxygen,
producing nitrous and nitric acids.

5. Abundant evidence has now been produced in support of the
fact that there is considerable fixation by means of bacteria.

Bacterial life in several ways is able to reclaim from the atmo-
sphere this free nitrogen, which would otherwise be lost. The first
method to which reference may be made is that involving
symbiosis. This term signifies " a living together " of two different
forms of life, generally for a specific purpose. Marshall Ward has
recently defined it as the co-operation of two associated organisms to
their mutual advantage, each symbiont being incapable of carrying
on alone the work which the symbiotic association is able to per-
form.']" It is convenient to restrict the term symbiosis to comple-
mentary partnerships such as exist between algoid and fungoid
elements in lichens, or between unicellular algae and Radiolarians,J

* Sir John Lawes and Sir Henry Gilbert {Times, 2nd December 1898) have
pointed out that the addition of nitrates only would be of no permanent use to the
wheat crop. They rely upon thorough tillage and proper rotation of crops as the
means of improving the nitrogen value of the soil.

f British Association for the Advancement of Science Repm^t, 1899, p. 693.*

X Geddes, Nature, xxv., 1882.

SYMBIOSIS

133

or between bacteria and higher plants. The partnerships between
hermit crabs and sea-anemones and the like are sometimes defined
by the term commensalism (joint diet), which is applied to such
associations having negative results, neither partner gaining much
advantage from the association. Symbiosis and commensalism must
be distinguished from parasitism, which indicates that all the
advantage is on the side of the parasite, and nothing but loss on
the side of the host. Association
of organisms together for increase
of virulence and function should
be distinguished from symbiosis,
and mere existence of two or more
species of bacteria in one medium
is not, of course, symbiosis. Most
frequently such a condition would
result in injury and the subsequent
death of the weaker partner, an
effect precisely opposite to that de-
fined by this term.

The example of bacteriological
symbiosis with which we are con-
cerned here is that partnership be-
tween bacteria and some of the
higher plants (Leguminosse) for the
purpose of fixing nitrogen in the
plant and in the surrounding soil.*

The nitrogen-fixing1 bacteria,
the third group of micro-organisms
connected with the soil, exist in
groups and colonies situated inside
the nodules, appearing, under cer-
tain circumstances, on the rootlets
of the pea, bean, and other Legu-
minosae. It was Hellriegel and
Wilfarth who first pointed out
that, although the higher chloro-
phyllous plants could not directly
obtain or utilise free nitrogen,
some of them at any rate could acquire nitrogen brought into com-
bination under the inlluence of bacteria. Hellriegel found that the
gramineous, polygonaceous, cruciferous, and other orders depended

r\r * Examples of bacteria symbionts arc numerous ; e.g. the dissolution of cellulose
(Van benus); the decomposition of sound potato in water exhausted of air (Ward) ;
the reduction of sulphates ; the oxidation of sulphuretted hydrogen ; the iron
bacteria, etc.

l'i'. L$>.— -Rootlet of Pea with Nodules.

134 BACTERIA IN THE SOIL

upon combined nitrogen supplied within the soil, but that the
Leguminosee did not depend entirely upon such supplies.

It was observed that in a series of pots of peas to which no
nitrogen was added most of the plants were apparently limited in
their growth by the amount of nitrogen locked up in the seed.
Here and there, however, a plant, under apparently the same cir-
cumstances, grew luxuriantly, and possessed on its rootlets abundant
nodules. The experiments of Sir John Lawes and Sir Henry Gilbert
at Eothamsted * demonstrated further that under the influence of
suitable microbe-seeding of the soil in which Leguminosae were
planted there is nodule formation on the roots, and coinciden tally
increased growth and gain of nitrogen beyond that supplied either
in the soil or in the seed as combined nitrogen. Presumably this is
due to the fixation, in some way, of free nitrogen. Nobbe proved
the gain of nitrogen by non-leguminous plants (Elceagnus, etc.) when
these grow root nodules containing bacteria, but to all appearances
bacteria differing morphologically from the Bacillus radicicola of the
leguminous plants.

These facts being established, the question naturally arises, How
is the fixation of nitrogen to be explained, and by what species of
bacteria is it performed ? In the first place, these matters are
simplified by the fact that there is very little fixation indeed by
bacteria in the soil apart from symbiosis with higher plants. Hence
we have to deal mainly with the work of bacteria in the higher
plant. Sir Henry Gilbert concludes f that the alternative explana-
tions of the fixation of free nitrogen in the growth of Leguminosse
seem to be :

" 1. That under the conditions of symbiosis the plant is enabled
to fix the free nitrogen of the atmosphere by its leaves ;

" 2. That the nodule organisms become distributed within the
soil and there fix free nitrogen, the resulting nitrogenous compounds
becoming valuable as a source of nitrogen to the roots of the higher
plant ;

" 3. That free nitrogen is fixed in the course of the development
of the organisms within the nodules, and that the resulting nitrogenous
compounds are absorbed and utilised by the host. " Certainly," he
adds, "the balance of evidence at present at command is much in
favour of the third mode of explanation." If this is finally proved
to be the case, it will furnish another excellent example of the power
existing in bacteria of assimilating an elementary substance.

Experiments at Eothamsted have confirmed those of others, in
showing that, by adding to a sterilised sandy-soil growing leguminous

* Sir Henry Gilbert, F.R.S., The Lawes Agricultural Trust Lectures, 1893,
p. 129.

t Ibid., p. lib.

PLATE 12.

y Ml ', • , :-o

Nitrogen-Fixing Bacteria in situ in Nodule on Rootlet of Pea.
X 400.

NllROGEN-FlXING BACTERIA in SitU IN ROOT-NODULE

of Pea. (Section of Nodule), x 500.

1 ^v

Nitrogen-Fixing Bacteria in situ in Root-Nodule
of Pea. (/Section of Nodule), x 000.

[To face page 134.

NITROGEN-FIXING BACTERIA 135

plants, a small quantity of the watery extract of a soil containing
the appropriate organisms, a marked development of the so-called
leguminous nodules on the roots is induced, and that there is confi-
dently increased growth, and gain of nitrogen. There is no evidence
that the leguminous plant itself assimilates free nitrogen ; the supposi-
tion is, that the gain is due to the fixation of nitrogen in the course of
development of the lower organisms within the root-nodules, the nitro-
genous compounds so produced being taken up and utilised by the
higher plant.

It would seem, therefore, that in the growth of leguminous crops,
such as clover, vetches, peas, beans, sainfoin, lucerne, etc., at any
rate some of the large amount of nitrogen which they contain, and of
the large amount which they frequently leave as nitrogenous residue
in the soil for future crops, may be due to atmospheric nitrogen
brought into combination by the agency of lower organisms. It has
yet to be ascertained, however, under what conditions a greater or
less proportion of the total nitrogen of the crop will be derived — on
the one hand from nitrogen-compounds within the soil, and on the
other from such fixation. It might be supposed, that the amount
due to fixation would be the less in the richer soils, and the greater
in soils that are poor in combined nitrogen, and which are open and
porous. On the other hand, recent results obtained at Kothamsted
indicate that, at any rate with some leguminous plants, there may be
more nodules produced, and presumably more fixation, with a soil
rich in combined nitrogen, than in one poor in that respect.

Most authorities would agree that all absorption of free nitrogen,
if by means of bacteria, must be through the roots. As a matter of
fact, legumes, especially when young, use nitrogen, like all other
plants, derived from the soil. It has been pointed out that, unless
the soil is somewhat poor in nitrogen, there appears to be but little
assimilation of free nitrogen and but a poor development of root
nodules.* The free nitrogen made use of by the micro-organism is
in the air contained in the interstices of the soil. For in all soils,
but especially in well-drained and light soils, there is a large quantity
of air. Although it is not known how the micro-organisms in legumes
utilise free nitrogen and convert it into organic compounds in the
tissues of the rootlet or plant, it is known that such nitrogen com-
pounds pass into the stem and leaves, and so make the roots really
poorer in nitrogen that the foliage. But the ratio is a fluctuating
one, depending chiefly on the stage of growth or maturity of the
plant.

If the nodules from the rootlets of Leguminosee be examined, the
nitrogen-fixing bacteria can be readily seen. They may be isolated

■ This has been denied in the official report by the chemist of the Experi-
mental Farm to the Minister of Agriculture at Ottawa (Report, 1896, p. 200).

136 BACTERIA IN THE SOIL

and grown in pure culture as follows : — The nodules are removed, if
possible, at an early stage in their growth, and placed for a few
minutes in a steam steriliser. This is advisable in order to remove
the various extraneous organisms attached to the outer covering of
the nodule. The latter may then be washed in antiseptic solution,
and their capsules softened by soaking. When opened with a
sterilised knife, thick creamy matter exudes. On microscopic
examination this is found to be densely crowded with small round-
ended bacilli or oval bodies, known as bacteroids. By a simple
process of hardening and using the microtome, excellent sections of
the nodules can be obtained which show these bacteria in situ. In
the central parts of the section may be seen densely crowded colonies
of the bacteria, which in some cases invade the cellular capsule of
the nodule derived from the rootlet.

The organisms are of various shapes, sometimes rod-like', and at
other times assuming a V or Y shape. Probably these latter forms
are due either to circumflex arrangement, branching or pleomorphism.
At the end of the summer most nodule-bearing roots, being annuals,
perish, and the nitrogen-fixing bacteria are liberated in the surround-
ing soil. Probably they are able to exist for long periods in the soil,
and re-infect other rootlets.

Other Bacterial Symbioses. — As we have already pointed out,
incidental association of organisms must not be mistaken for sym-
bioses. The decomposition organism, B. ramosus, may be found
associated with Nitrosomonas and Nitrobacter in the processes of
denitrification and nitrification, but this does not necessarily fulfil
the conditions of symbiosis, even though each of the three produces
substances which provide pabulum for the other two. True symbiosis
involves a much closer relationship than this, namely, the inability
of each symbiont to produce its effect apart from its partner.

Now, in addition to the case of the nitrogen-fixing bacteria we
have other bacterial examples, and brief reference must be made to
them. Van Senus, for instance, found an anaerobic bacillus capable of
dissolving cellulose if associated with another organism, also incapable
by itself of attacking cellulose. Winogradsky, too, found that a
certain anaerobic organism (Clostridium Pasteurianum), if supplied
with abundance of dextrose but no oxygen, could fix atmospheric
nitrogen. This capacity was found to be due to the organism being
surrounded with aerobic bacteria acting in partnership with it.
Probably, also, the bacteria concerned in the reduction of sulphates,
and the oxidation of sulphuretted hydrogen, as also the iron bacteria,
are further examples of symbiosis. Kephir and the so-called ginger-
beer plant must also be named in the same category. Kephir is a
common beverage amongst the Caucasians. The " Kephir grains " are
in reality composed of three separate organisms. The first is a fila-

OTHER FORMS OF SYMBIOSIS 137

mentous bacterium forming " zooglea." The second is a lactic-acid-
producing bacillus, and the third is a yeast. By these agents a
fermentation is set up in the milk of cows, goats, or sheep. " The
yeast and the bacteria, either jointly or separately, split up the
lactose or milk-s'ugar into two other sugars, galactose and glucose.
The yeast then forms alcohol from the latter, and the bacterium
lactic acid from the former" (Green). This filamentous bacillus
probably affects the casein. The outline is, it is true, only the
probable course of action, as full details as to the whole function of
the separate factors are not yet known.

The gingerbeer plant is the agent of fermentation in the so-called
"stone gingerbeer," and is composed essentially of two organisms,
one a yeast, Saccharomyces pyriformis, the other a bacillus, Bacterium
vermiforme. It rarely happens that these two forms are found pure,
there being as a rule an admixture of other organisms with them.
Professor Green describes B. vermiforme as growing in two different
ways, namely, as long rods or convoluted threads, invested by a
translucent wrinkled sheath, and as constituent microbes contained
within the sheath, yet able to escape from it. The sheathing form
of the organism can only be produced when oxygen is replaced by
carbon dioxide. In the symbiotic association the yeast absorbs the
oxygen, and during its fermentative activity produces carbon dioxide,
thus providing the necessary conditions for the formation of the
sheath. The bacterium benefits by substances excreted by the yeast,
and the latter profits in its turn by the removal of these matters
through the agency of the former. The yeast sets up the usual
fermentation of cane-sugar.

A third organism manifesting Bymbiosia occurs in Madagascar
as a curious gelatinous substance found attacking the sugar-cane,
and consisting again of a yeast and a bacterium associated together
in very much the same way as are the organisms in the gingerbeer
ferment.

Before leaving this subject of symbiosis as illustrated in the
lichens, in Winogradsky's Clostridium, in the nodule-bacteria, in the
gingerbeer plant, and in Kepliir, we may suitably inquire whether
anything is at present known as to how the symbionts are related to
each other. Obviously the matter presents many difficulties, and
the problem is by no means solved. There arc, however, three chief
hypotheses. First, the provision of definite food materials by the
one Bymbiont for the other may be an Important factor; e.g. an alga
supplies a fungus with carbo-hydrates, or a fungus converts Btarch
into the fermentable sugars which the associated yeast needs. In
other cases the advantage derived is one of protection from some
injurious agent; e.g., the aerobic bacterium prevents the access of
oxygen to the anaerobic one. Thirdly, there is some evidence,

138 BACTERIA IN THE SOIL

according to Professor Marshall Ward, in support of the hypothesis
that one symbiont may stimulate another by exciting some body
which acts as an exciting drug to the latter — just as truly as certain
drugs act as stimulants to some cell or organ of a higher animal, and
probably in a fundamentally similar manner.

Before we leave the subject of the economic bacteria present in
the soil, it may be well to refer briefly to the application of the new
knowledge to agriculture. Whilst many of the details of our know-
ledge concerning "the living earth" have not passed beyond the
experimental stage, it is not to be wondered at that the New Soil
Science has been received with some caution, and possibly in some
quarters even scepticism. This is neither surprising, nor, as regards
the details, altogether undesirable. A number of the cardinal
principles, however, are now obtaining very general acceptance
amongst practical agriculturists. Briefly, these may bo stated as
follows. That a soil which has been sterilised, or is otherwise not
occupied by soil bacteria, is necessarily an unfertile soil ; that the
disintegration and oxidation of organic matter in the soil are the
result of bacterial life and work; that the sowing, growing, and
feeding of the desirable soil germs are of as much importance to the
agriculturist to-day as is the sowing of seeds, or the growing and
feeding — by manuring — of plants; that the physical and chemical
conditions of soil favourable to these bacteria are of as much value
to the agriculturist as the requisite physical and chemical conditions
for the growth of the yeast cell are to the brewer ; and indeed that
one of the essential functions of manure in the soil is to provide
favourable pabulum and conditions for the operation of these soil
ferments.

In the further elucidation of these principles various series of
experiments, in addition to those at Eothamsted (under Sir J. B.
Lawes and Sir J. H. Gilbert) and at Woburn (under Dr Voelcker),
have been designed and carried out. Of such a nature are the well-
known Dalmeny Experiments originated by Lord Eosebery some
years ago. The chemists engaged in this series were aware that
though large doses of caustic lime would kill outright certain of the
economic soil bacteria, annual or biennial dressings of mild lime
added to the culture media, that is the soil, would materially assist
the process of nitrification. Five acres of land have been worked as
a miniature farm, each division being divided into sixteen plots. The
soil is of very uniform character, and is of the usual loamy kind met
with in the low grounds of the Lothians. Each plot has been
manured, or left unmanured as the case may be, on a regular system,
so that the residual values of the different manures, as well as the
yield of crop, may be accurately dealt with. The crops grown are
regularly analysed in order to determine the feeding value. Concern-

SAPROPHYTES 139

ing results, it may be said that the wheat plot points to the fact that
where the soil is in good condition, through the application of farm-
yard manure, the artificials that may be most profitably applied are
lime (4 parts), superphosphate (3 parts), and sulphate of ammonia
(1 part). On the other hand, where the land is not in such high
condition, this dressing should be supplemented by a dressing of
potash salt. The analyses show that by the application of these
dressings the value and quality of the crop are increased because
the operations of the nitrifying organisms have been thus favoured.

4. The Saprophytic Bacteria in Soil

This group of micro-organisms is by far the most abundant
as regards number. They live on the dead organic matter of the
soil, and their function appears to be to break it down into simpler
constitution. Specialisation is probably progressing among them,
for their name is legion, and the struggle for existence keen. After
we have eliminated the economic bacteria, most of which are obviously
saprophytes, the group is greatly reduced. It is also needless to add
that of the remnant little beyond morphology is known, for as their
function is learned they are classified otherwise. It is probable, as
suggested, that many of the species of common saprophytes normally
existent in the soil act as auxiliary agents to denitrification and
putrefaction. At present we fear they are disregarded in equal
measure, and for the same reasons, as the common water bacteria.
An excess of either, in soil or water, is not of itself injurious as far
as we know; indeed, it is probably just the reverse. It is, however,
frequently an index of value as to the amount and sometimes con-
dition of the contained organic matter. The remarks made when
considering water bacteria apply here also, viz., that an excess of
saprophytes acts not only as index of increase of organic matter,
but as at first auxiliary, and then detrimental, bo pal bogenic organisms.
It will require accurate knowledge of soil bacteria generally to be
able to say which saprophytic germs, if any, have no definite function
beyond their own existence. It maybe doubted whether the stern
behests of nature permit of such organisms. However t hat may be,
we may feel confident, though at present there are many common
bacteria in soil, as also in water, the life object of which is not
ascertained, that as knowledge increases and becomes more accurate,
this special provisional group will become gradually absorbed into
other groups having a part in the economy of nature, or in the
production of diseasa At presenl the decomposition, denitrifying,
nitrifying,* and nitrogen-fixing organisms are the only saprophytes

;: It has already been pointed out that the nitrifying baeteria, though able to
live on organic matter, do not require such either for existence or for the performance
of their function.

140 BACTERIA IN THE SOIL

which have been rescued from the oblivion of ages, and brought more
or less into daylight. It is but our lack of knowledge which requires
the present division of saprophytes, whose business and place in
the world is unknown.

5. The Pathogenic Organisms found in Soil

In addition to these saprophytes and the economic bacteria,
there are, as is now well known, some disease-producing bacteria
finding their nidus in ordinary soil. The three chief members of this
group are the bacillus of Tetanus, the bacillus of Quarter-Evil, and
the bacillus of Malignant (Edema.

Tetanus

The pathology of this disease has, during recent years, been
considerably elucidated. It was the custom to look upon it as
"spontaneous," and arising no one knew how; now, however, after
the experiments of Sternberg and Nicolaier, the disease is known to
be due to a micro-organism common in the soil of certain localities,
existing there either as a bacillus or in a resting stage of spores.
Fortunately, Tetanus is comparatively rare, and one of the peculiar
biological characters of the bacillus is that it only grows in the
absence of oxygen. This fact contributed not a little to the difficulties
which were met with in securing its isolation.

Tetanus occurs in man and horses most commonly, though it may
affect other animals. There is usually a wound, often an insignificant
one, which may occur in any part of the body. The popular idea
that a severe cut between the thumb and the index linger leads to
tetanus is without scientific foundation. As a matter of fact, the
wound is nearly always on one or other of the limbs, and becomes
infected simply because the limbs come more into contact with soil
and dust than does the trunk. It is not the locality of the wound
nor its size that affects the disease. A cut with a dirty knife, a gash
in the foot from the prong of a gardener's fork, the bite of an insect,
or even the prick of a thorn, have before now set up tetanus.
Wounds which are jagged, and occurring in absorptive tissues, are
those most fitted to allow the entrance of the bacillus. The wound
forms a local factory, so to speak, of the bacillus and its secreted
poisons ; the bacillus always remains in the wound, but the toxins
may pass throughout the body, and are especially absorbed by the
cells of the central nervous system, and thus give rise to the spasms
which characterise the disease. Suppuration generally occurs in the
wound, and in the pus thus produced may be found a great variety
of bacteria, as well as the specific agent itself. After a few days or,

PLATE 13.

"A

\

1 r

«

«-'

•♦

r

Bacillus tetani.

Film preparation, from broth culture, showing^spore
formation, x 1000.

Streptoth r-i '.c actinomyces.

Ray fungus in tissue. Stained by Gram's method,
x 700.

Bacillus mycoides.

Film preparation, from agar culture, 37° C. Spore
formation, x 1000.

Bacillus mallei (Glanders).

Film preparation, agar culture, 37* C.
x 1000.

[To face page 140.

TETANUS 141

it may be, as much as a fortnight, when the primary wound may be
almost forgotten, general symptoms occur. Their appearance is often
the first sign of the disease. Stiffness of the neck and facial muscles,
including the muscles of the jaw, is the most prominent sign. This
is rapidly followed by spasms and local convulsions, which, when
affecting the respiratory or alimentary tract, may cause a fatal result.
Fever and increased rate of pulse and respiration are further signs
of the disease becoming general. After death, which results in the
majority of cases, there is very little to show the cause of fatality.
The wound is observable, and patches of congestion may be found
on different parts of the nervous system, particularly the medulla
(grey matter), pons, and even cerebellum. Evidence has recently
been forthcoming at the Pasteur Institute to support the theory
that tetanus is a " nervous " disease, more or less allied to rabies,
and is best treated by intra-cerebral injection of antitoxin, which
then has an opportunity of opposing the toxins at their favourite
site. The toxins diffuse throughout the tissues of the body, but
particularly affect the spinal cord. The long incubation period
indicates that the toxins arc probably produced by a ferment of
some kind. Whatever its exact nature, it is undoubtedly a most
powerful poison.

Tetanus bacilli spores have been found in considerable quantities
in the dust of dry jute fibre (Andrewes), and various cases are on
record where the disease has been contracted by workers in jute
mills in Dundee and elsewhere. Legge attributes the presence of
the bacilli in the jute {Corchorus) to the soil in which it is grown in
Bengal.

The Bacillus of Tetanus. — In the wound the bacillus is present
in large numbers, but mixed up with a great variety of suppurative
bacteria and extraneous organisms. It is in the form of a straight
short rod with rounded ends, occurring singly or in pairs or threads,
and slightly motile. It has been pointed out that by special methods
of staining nagella may be demonstrated. These are both lateral
and terminal, thin and thick, and arc shed previously to sporulation.
Branching also has been described. Indeed, it would appear that,
like the bacillus of tubercle, this organism has various polymorphic
forms. Next to the ordinary bacillus, filamentous forms predominate,
particularly so in old cultures. Clubbed forms, not unlike the
bacillus of diphtheria, may often be obtained from agar cultures.
Without doubt the most peculiar characteristic of this bacillus is its
sporulation. The well-formed round spores occur readily at incuba-
tion temperature. They occupy a position at one or other pole of
the bacillus, and have a diameter considerably greater than the
organism itself. Thus the well-known "drumstick" form is produced.
In practice the spores frequently occur free in the medium and in

142 BACTERIA IN THE SOIL

microscopical preparation. Like other spores, they are extremely
resistant to heat, desiccation, and antiseptics.*

As we have seen, this bacillus is a strict anaerobe, growing only
in the absence of oxygen. The favourable temperature is 37 ° C., and
it will only grow very slowly at or below room temperature. The
organism is readily stained by the ordinary stains and by Gram's
method.

An excellent culture is generally obtainable in glucose gelatine.
The growth occurs only in the depth of the medium, and appears as
fine threads passing horizontally outwards from the track of the
needle. At the top and bottom of the growth these fibrils are
shorter than at the middle or somewhat below the middle. For
extraction of the soluble products of the bacillus, glucose broth may
be used. (For isolation and detection of the B. tetani, see Appendix,
p. 481.)

In some countries, and in certain localities, the bacillus of tetanus
is a very common habitant of the soil, and when one thinks how
frequently wounds must be more or less contaminated with such
soil, the question naturally arises, How is it that the disease is,
fortunately, so rare ? Probably we must look to the advance of
bacteriological science to answer this and similar questions at all
adequately. Much has recently been done in Paris and elsewhere to
emphasise the relation which other organisms have to such bacteria
as those of typhoid and tetanus. In tetanus, Kitasato, Vaillard, and
others have pointed out that the presence of certain other bacteria,
or of some foreign body, is necessary to the production of the disease.
The common organisms of suppuration in particular appear to
increase the virulence of the bacillus of tetanus. How these
auxiliary organisms perform this function has not been fully
elucidated. Probably, however, it is by damaging the tissues and
weakening their resistance to such a degree as to afford a favourable
multiplying ground for the tetanus bacillus. Some authorities hold
that they act by using up the surrounding oxygen, and so favouring
the growth of the germ of tetanus. In any case it is now generally
held that in natural infection the presence of some foreign body or
suppurative bacteria is necessary to produce the disease.

Quarter-Evil and Malignant (Edema

Quarter-Evil (or symptomatic anthrax) is a disease of animals,
produced in a manner analogous to tetanus. It is characterised by
a rapidly-increasing swelling of the upper parts of the thighs, sacrum,
etc., which, beginning locally, may attain to extraordinary size and

* Atlas and Principles of Bacteriology, by Lehmann and Neumann, part ii.,
pp. 330-337.

QUARTER-EVIL 143

extent. The swelling may assume a dark colour, and crackles on
being touched. There is high temperature, and secondary motor and
functional disturbances. The disease ends fatally in two or three
days.

Slight injuries to the surface of the skin or mucous membrane
are sufficient for the introduction of the
causal bacillus. This organism is, like the ^ i 0
bacillus of tetanus, an anaerobe, existing in j 0

the superficial layers of the soil. From its . I ^ A / I

habitat it readily gains entrance to animal | / ' I

tissues. It has spores, but though they
are of greater diameter than the bacillus
itself, they are not absolutely terminal. " | 9 q

are of greater diameter than the bacillus i 0 / \ \

itself, they are not absolutely terminal. | 0

Hence they merely swell out the cap-

sule of the bacillus, and produce a club- S I

shaped rod. The bacillus forms gas while

growing in the tissues and in artificial cul- _ OA _. __ ...

& o -i 1 • i -i • • FlG- 20.— Diagram of Bacillus

ture. External physical conditions have of symptomatic Anthrax.

little effect upon this organism, and dried

and even buried flesh retains infection for a long period of

time.

Quarter-Ill, Quarter-Evil, or Black-Leg

Quarter-ill may be said to lack much of the importance and interest which is
attached to anthrax, inasmuch as it is confined to two domestic animals — sheep and
cattle — and is not communicable to man. It, however, resembles anthrax, in so
far as they are both caused by the introduction into the blood of the healthy
animal of specific bacilli. Both diseases have a tendency to recur on farms or
premises on, or in, which animals affected with these diseases have been previously
kept. On the other hand neither anthrax nor quarter-ill is communicable by
association of the affected with the healthy animal, and in that respect they differ
from most of the contagious diseases which are legislated for in this country.
Another peculiar feature of quarter-ill is that while it is very fatal to sheep at any
age, cattle over two years may be said to have an immunity against the disease.

The symptoms of quarter-ill in young cattle are so strikingly different from any
other disease that an error in diagnosis is almost impossible. The first indication of
an animal being affected with quarter-ill is a marked stiffness or lameness of one of
the limbs ; it is exceedingly dull, and presents a most anxious and dejected
appearance, does not feed, and it is with extreme difficulty that it can be forced to
move. Very soon after the limb is attacked a swelling appears beneath the skin,
usually upon one of the hind quarters, which is extremely hot, increases in size
rapidly, and is most painful to the animal when touched. This swelling has a
disposition to extend down the leg, or perhaps along the loins and back, and when
pressed gives a peculiar crackling sensation to the fingers. In almost every instance
death supervenes within a few hours after the swelling has appeared.

In the case of sheep the symptoms are not of so marked a character. The first
indication is lameness, but -the swelling is not so observable in sheep as in cattle,
being hidden to a great extent in the case of the former by the fleece.

There is no doubt that the disease exists to a greater extent among the sheep in
certain counties in England than has been generally known, and from the rapidity
with which sheep frequently die it is often locally called " strike."

Should any doubt exist as to whether a sheep has died from quarter-ill, the
difficulty can easily be solved by making an incision through the skin of the dead

144 BACTERIA IN THE SOIL

animal into the tumour or swelling, which contains a large quantity of dark
coloured fluid, which emits a very strong and peculiarly offensive odour. Any
fluid that may thus escape should be carefully collected and destroyed.

The carcases of animals which have died of quarter-ill should be buried as in
anthrax, or, still better, cremated on or in the place where the animal died. All
dung, fodder, litter, or other materials of a like character which may have been
on or about places or sheds where animals have died should be burnt, or thoroughly
mixed with some powerful disinfectant, and buried in a part of the premises to
which cattle and sheep do not have access. The sheds, particularly the flooring
and mangers, should be thoroughly washed and scrubbed with a 5 per cent, solution
of carbolic acid, and it would be prudent to repeat the process before they are
again used for cattle or sheep.

A third disease-producing microbe found naturally in soil is that
which produces the disease known as Malignant (Edema. Pasteur
called this disease gangrenous septicaemia, and the bacillus vibrion
septique. Unlike quarter-evil, malignant oedema may occur in man
in cases where wounds have become septic. It is usually described
as a spreading inflammatory oedema, with emphysema, and followed
by gangrene. Man and animals become inoculated with this bacillus
from the surface of soil, straw-dust, upper layers of garden-earth, or
decomposing animal and vegetable matter.

The bacillus occurs in the blood and tissues as a long thread
(3 /a to 10 /a in length), composed of slender segments of irregular
length. It is motile and anaerobic, and readily stained by aniline
stains but not by Gram's method, in this way differing from the
anthrax bacillus. The spores are larger than the diameter of the
bacillus, and usually centrally placed. The organism produces gas,
and so much is this the case in artificial culture, that the medium
itself is frequently split up. The bacillus liquefies gelatine. The
most suitable medium for cultivation is glucose agar at 37° C.

Both malignant oedema and symptomatic anthrax are similar in
some respects to anthrax itself. There are, however, a number of
points for differential diagnosis. The enlargement of the spleen, the
enormous numbers of bacilli throughout the body, the square ends
of the bacillus, its non-mo tility, its equal inter- bacillary spaces, its
aerobic growth, and its characteristic staining, afford ample evidence
of the anthrax bacillus.

Frankel and Pasteur have both demonstrated the possible presence
in soil of the bacillus of anthrax itself. Frankel maintained that it
could not live there long, and at 10 feet below the surface no growth
occurred. This may have been clue to the low temperature of such
a depth. Pasteur held that earthworms are responsible for convey-
ing the spores of anthrax from buried carcases to the surface, and
thus bringing about re-infection. In any case it is well-known that
the spores of anthrax may infect a soil for months. The bacillus of
cholera, too, has been successfully grown in soil, except during
winter. The presence of common saprophytes in the soil is prejudicial

RELATION TO DISEASE 145

to the development of the cholera spirillum, and under ordinary
circumstances it succumbs in the struggle for existence. Other
species of bacteria have also been isolated from soil from time to
time.

Now whilst since the early days of bacteriology the three organ-
isms we have described have been looked upon as the typical bacteria
of soil, modern research has brought to light a new relationship
between soil and disease, which has greatly enhanced the importance
of our knowledge of the subject. Directly, it has been shown that
soil may harbour germs of disease, acting sometimes as a favourable
and at other times as an unfavourable nidus. Indirectly, it has been
shown that a right understanding of the bacteriology of water and
its potentiality of disease production, depends upon a knowledge of
bacteria in the soil over, or through which, the water has passed.
The matter must, therefore, be briefly considered here.

The Relation of Soil generally to certain Bacterial Diseases

It is now some years since Sir George Buchanan, for the English
Local Government Board, and Dr Bowditch, for the United States,
formulated the view that there is an intimate relationship between
dampness of soil and the bacterial disease of Consumption (tuber-
culosis of the lungs). The matter was left at that time sub judice,
but the conclusion has since been drawn, and it is surely a legitimate
one, that the dampness of the soil acted injuriously in one of two
ways. It either lowered the vitality of the tissues of the individual,
and so increased his susceptibility to the disease, or in some unknown'
way favoured the life and virulence of the bacillus. That is one fact.
Secondly, Pettenkofer traced a definite relationship between the rise
and fall of the ground water with pollution of the soil and enteric
(typhoid) fever.* A third series of investigations concluded in the
same direction, viz., the researches by Dr Ballard respecting summer
diarrhoea. This, it is generally held, is a bacterial disease, although
no single specific germ has been isolated as its cause. Ballard
demonstrated that the summer rise of diarrhoea mortality does not
commence until the mean temperature of the soil, recorded by the
4-foot thermometer, lias attained 564° F., and the decline of such
diarrhoea coincides more or less precisely with the fall in soil
temperature. This temperature (56*4° F.) is, therefore, considered

p ,! The conditions requisite for an outbreak of enteric fever were, according to
Fettenkofer, (aU rapid tall (after a rise) in the ground water, (/,) pollution of the
soil with animal impurities, (c) a certain earth temperature, and lastly (d) a specific
micro-organism in the soil. These four conditions have not, particularly in England,
always been fulfilled preparatory to an epidemic of typhoid. Yet the observations
necessary for these deductions were a definite step in advance of the idea of the
significance of mere dampness of soil.

K

146 BACTERIA IN THE SOIL

as the "critical" 4-foot earth temperature, that is to say, the
temperature at which certain changes (putrefactive, bacterial, etc.)
take place either, primarily, in the earth, or secondarily, in the atmo-
sphere, with the consequent development of the diarrhoeal poison.

After prolonged investigation on behalf of the Local Govern-
ment Board, Dr Ballard formulated the causes of diarrhoea in the
following conclusions : — *

(a) The essential cause of diarrhoea resides ordinarily in the super-
ficial layers of the earth, where it is intimately associated with the
life processes of some micro-organism not yet detected or isolated.

(b) That the vital manifestations of such organism are dependent,
among other things, perhaps principally upon conditions of season
and the presence of dead organic matter, which is its pabulum.

(c) That on occasion such micro-organism is capable of getting
abroad from its primary habitat, the earth, and having become air-
borne, obtains opportunity for fastening on non-living organic
material, and of using such organic matter both as nidus and as
pabulum in undergoing various phases of its life-history.

(d) That from food, as also from contained organic matter of
particular soils, such micro-organism can manufacture, by the
chemical changes wrought therein through certain of its life
processes, a substance which is a virulent chemical poison.

Here, then, we have a large mass of evidence from the data
collected by Buchanan, Bowditch, Pettenkofer, and Ballard. But
much of this work was done anterior to the time of the application
of bacteriology to soil constitution. Eecently the matter has
received increased attention from various workers abroad, and in this
country from Eobertson, Martin, Houston, and others, and we must
now consider the new facts brought forward by these investigators.

From the first, experiments on this subject have been more or
less confined to the behaviour of the typhoid bacillus than any other
pathogenic organism. This has been partly due to the importance
of this organism in relation to soil, and partly because it is more
convenient than, say, the tubercle bacillus for experimental work. In
1888 Grancher and Deschamps showed that the typhoid bacillus was
able to survive in soil for more than twenty weeks,f and Karlinski
arrived at a similar conclusion, J In 1894 Dempster published the
results of his work on the same subject, in which he obtained the
typhoid bacillus from sand after twenty-three days, from garden soil
after forty-two days, and from peat not later than twenty-four hours. §
Four years later came Dr Robertson's valuable researches into the

* Supplement to the Report of the Medical Officer of the Local Government Board,
1887, p. 7.

t Arch, de Med. Exp. et d"Anat. Path., 1889, 7th January.
% Arch. f. Hyg., Bd. xiii., Heft 3.
§ Brit. Med.' Jour., 1894, i., p. 1126.

TYPHOID AND SOIL 147

growth of the bacillus of typhoid in soil of an ordinary field. By
experimental inoculation of the soil with broth cultures, he was able
to isolate the bacillus twelve months after, alive and virulent. He
concluded that the typhoid organism is capable of growing very
rapidly in certain soils, and under certain circumstances can survive
from one summer to another. The rains of spring and autumn, or
the frosts and snows of winter, do not kill it off so long as there is
sufficient organic pabulum. Sunlight, the bactericidal power of
which is well known, had, as would be expected, no effect except
upon the bacteria directly exposed to its rays. The bacillus typhosus
quickly died out in the soil of grass-covered areas.*

Next came the experiments of Dr Sidney Martin, which were
undertaken to inquire into the extra-corporeal existence of the
bacillus of typhoid fever in soil. He found, after a prolonged
research, that certain cultivated soils, especially garden soils, when
sterilised are favourable to the vitality and growth of this bacillus,
whether the soil was kept at room temperature (19° C.) or blood-heat
(37° C). In such soils the B. typhosus was still alive after four
hundred and four days, and remained alive, though not for a long
period, if the soil were dried and reduced to dust. If, however, the
bacillus is added to a well-moistened but not sloppy cultivated soil,
it rapidly dies, and is usually not obtainable two days after being
sown in it, and its disappearance appears to be more rapid the higher
the temperature, which is probably due to the rapid growth of
ordinary soil bacteria. If the cultivated soil is not made very moisl
when the B. typhosus is added, the organism can be recovered from
the soil up to twelve days after it has been added Lastly, if this
bacillus is added to natural uncultivated soils which have not been
sterilised, it ceases to exist within twenty-four hours. Martin holds
that the reason of the rapid disappearance of the typhoid bacillus
from natural unsterilised soils is probably twofold. First, there is
the antagonism of the soil bacteria, many of which are putrefactive :
and secondly, the typhoid hacillus requires for its growth nitrogenous
substances, usually in the form of proteids. Cultivated soil is dis-
tinguished from uncultivated soil by containing more nitrogenous
organic matter in the form <»f nitrates and ammonia, and also more
partially changed pro teid suhstai ices. Hence it is a more favourable
environment for the typhoid bacillus. As a general result of these
investigations, it may be concluded that the typhoid bacillus has,
commonly, only a short existence in the soil, being destroyed by the
products of tho putrefactive bacteria which exist in most cultivated
soils.f

* Brit. Med. Jour.y 1898, i., pp. 69-71.

t Reports of Medical Officer to the Local. Government Hoard, 1898-99, pp. 382-41'i ;
1899-1900, pp. 525-548 ; 1900-01, pp. -4S7-510.

148 BACTERIA IN THE SOIL

Lastly, we have the results of the investigations of Firth and
Horrocks, who conclude that the typhoid bacillus is able to assume a
vegetative existence in ordinary soils and in sewage-polluted soils for
as long as seventy-four days. They further maintain that the
controlling factor is an excess or deficiency of moisture in the soil
rather than organic nutritive material. From dry fine sand the
bacillus was recovered after twenty-five days ; from moist fine sand,
after twelve days ; from damp (rain-water) ordinary soil, after sixty-
seven days ; from damp (sewage) ordinary soil, after fifty-three days ;
and from ordinary soil dried to the state of dust, after twenty-four
clays. In peat the bacillus lives apparently only a few days.* Firth
and Horrocks, therefore, arrive at a different conclusion from Martin,
namely, that the typhoid bacillus is able to assume a vegetative or
saprophytic existence for considerable periods outside the body; that
it can survive ordinary earth for over two months, whether the soil be
virgin or polluted with sewage, or frozen hard ; and that, therefore, it
follows that outbreaks of enteric fever may be due to the dissemina-
tion (for example by wind or flies) of infective soil dust. Pfuhl of
Berlin has arrived at results confirmatory of these experiments.
From moist garden earth he recovered the typhoid bacillus eighty-eight
days after inoculation, from dry sand after twenty-eight days, and
from moist peat twenty-one days.f

On the whole it would appear that whilst much valuable research
has been accomplished, it cannot be said that the relation of the
typhoid bacillus to soil is understood. -Some further light on the
subject is obtained from researches carried out in relation to the
bacterial condition of sewage-treated land and made-up refuse soil,
and brief reference may be made to such work.

Various workers have carried out experiments with the object of
ascertaining whether in the surface layers of soil, after it has had
sewage upon it, certain microbes characteristic of sewage retain their
vitality for any considerable length of time ; what, in short, was the
fate of such organisms as B. coli, B. enteriticlis sjwrogencs, etc., when
sown broadcast on soil. For if their fate be known, not only would
light be thrown upon questions of sewage treatment, and the pollu-
tion of soil which might in turn affect water supplies, but indication
would be obtained as to the possibility of disease germs maintaining
their existence in soil, and eventually infecting man. Chief among
such investigations in England have been those of Dr A. C. Houston^
whose conclusions are briefly as follows : —

(1) The addition of sewage to an ordinary garden soil does not

* Brit. Med. Journ., 1902, ii., pp. 936-943.
t Zeit. f. Hyg., 1902, Bd. xl., Heft 3, p 555.

t Report of Medical Officer to Local Government Board, 1900-01, App. No. 4,
p. 405 ; 1901-02, App. No. 6, p. 455.

POLLUTED SOIL 149

.seemingly lead to other than a temporary increase of the sewage
microbes at the expense of the soil microbes, the ordinary soil
1 tacteria ousting the sewage microbes in the struggle for existence.
But the addition of sewage to a sandy soil leads to an enormous
increase in the total number of microbes as compared with the
number originally present in the soil, which does not revert to its
original state for some months.

(2) The addition of sewage to garden soil tends primarily to
increase the ratio of total number of bacteria to spores of aerobic
bacteria, though the alteration is apt to be soon lost.

(3) The addition of sewage to a soil leads to an increase for a time
in the number of certain kinds of bacteria, namely: (a) inclol-
producing bacteria ; (b) gas-producing organisms ; (c) the spores of
B. enteritidis sjporogencs ; (d) B. coli communis and its allies ; and (e)
streptococci. The occurrence of true streptococci in soil indicates, in
Houston's opinion, extremely recent contamination. Whatever inter-
pretation be placed on these facts, it is evident that they indicate that
pathogenic organisms such as the typhoid bacillus do not maintain
their vitality in the surface layers of soil for more than a brief
period. Further, it is evident that some kinds of soils heavily
polluted with excremental matter tend to purify themselves, so far as
non-sporing bacilli of intestinal origin are concerned.

On the bacterial content of made-soil Dr Savage of Colchester has
carried out some work. The samples of such soil were collected with
a sterilised Frankel's borer, ancl the samples transmitted to the
laboratory in sterile Petri dishes. Each sample was then thoroughly
broken up and uniformly mixed in the Petri dishes by means of
sterile spatulas. Ten grammes were weighed on sterile glazed paper,
and added to 100 c.c. of sterile water in a large flask, and thoroughly
mixed. The contents of the flask were allowed to settle for five
minutes, and without disturbing the sediment 1 c.c. of the water was
taken up and added to further quantities of sterile water for dilution
purposes. The examination was then carried out in the ordinary way,
with a view of determining (a) total number of organisms present;
(b) number of B. coli ; (c) number of B. mycoides, and of Bismark-
brown Cladothrix ; and (d) the smallest quantity of soil producing
indol in one week at 37° C. grown in peptone water solution.

As a result of these experiments, Dr Savage reports that at a
depth of two feet in mounds of tip-refuse deposited on damp imper-
vious clay, putrefaction and concurrent purification takes place fairly
rapidly for the first two or three years. The organisms present in
the refuse as deposited rapidly decrease at the same time, but after
two to three years increase, apparently due to the invasion of ordinary
soil organisms. After two to three years, purification at tins depth
takes place extremely slowly, and samples nine to ten years old gi\e

150 BACTERIA IN THE SOIL

results very little different from four to five year old samples. The
B. coli readily dies out in such refuse heaps, from which Dr Savage
infers that the B. typhosus, being a less resistant organism, would still
more rapidly die out, and that therefore "the danger of specific
typhoid contamination from building on such made-soil can be
neglected." *

From what has been said, it will be seen that though a consider-
able amount of knowledge has been obtained respecting bacteria in
the soil, it may be conjectured that actually there is still a great deal
to ascertain before the micro-biology of soil is in any measure com-
plete or even intelligible. The mere mention of the bacilli of tetanus
and typhoid in the soil, and their habits, nutriment, and products
therein, not to mention the work of the economic bacteria, is to open
up to the scientific mind a vast realm of possibility. It is scarcely too
much to say that a fuller knowledge of the part which soil plays in
the culture and propagation of bacteria may suffice to modify
many views in preventive medicine. Tine, our knowledge at the
moment is rather a heterogeneous collection of isolated facts and
theories, some of which, at all events, require ample confirmation;
yet still there is a basis for the future which promises much con-
structive work.

Jour, of Sanitary Institute, 1903, vol. xxiv., pt. iii., pp. 442-458.

CHAPTEE VI

THE BACTERIOLOGY OF SEWAGE AND THE BACTERIAL
TREATMENT OF SEWAGE

Composition of Sewage— Quantity and Quality of Bacteria in Sewage— Treatment
of Sewage : (1) Disposal without Purification ; (2) Chemical Treatment ;
(3) Bacterial Treatment — Evolution of Bacterial Methods — Septic Tank
Method— Contact Bed Method— Manchester Experiments— Effect of Bacterial
Treatment on Pathogenic Organisms.

The relation of bacteria to sewage has during the last twenty-five
years assumed a position of the first importance. This is duo,
generally speaking, to three causes. In the first place, our knowledge
of the economic function of bacteria present in sewage has increased
in a very large measure in recent years. Secondly, as the population
has tended to gravitate to cities, the problem of a pure water supply,
free from sewage pollution, has become infinitely more complicated
than was the case in rural communities in the past. How often
sewage, from sewage or cesspools, gains access by means of direct
connection or percolation to drinking water, the history of typhoid
epidemics and similar outbreaks in this country only too fully records.
And thirdly, practical issues have now arisen in connection with the
bacterial treatment of sewage. In order to understand the bacteri-
ology of sewage and its practical lessons, we may first briefly consider
the quality and constitution of sewage as regards its bacterial content,
and then proceed to discuss its biological treatment.

The Constitution of Sewage

It is impossible to lay down any exact standard of the chemical
and bacterial quality of sewage. The quality will differ according to
the size of the community, the inclusion or otherwise of trade-effluents
and waste products, the addition of rain and storm water, and other

151

152 BACTERIAL TREATMENT OF SEWAGE

similar physical conditions.* Moreover, the sewage itself is con-
stantly undergoing rapid changes owing to fermentation, and the
competition of micro-organisms and the effect of their products. It
is clear that they are the chief agents in setting up fermentative and
putrefactive changes, for if sewage be placed in hermetically sealed
flasks and sterilised by heat it will be found that these changes do
not occur. Hence it will be at once apparent that no exact or hard-
and-fast formula can be laid down. ^Respecting the chemical con-
dition, with which we have but little to do here, we may shortly say
that the chief characteristic of sewage is its enormous amount of
contained organic matter (yielding saline and albuminoid ammonia,
etc.) in suspension or in solution. But there are in addition various
inorganic substances, and hence it is customary to subdivide the
chemical constituents into (a) organic matter in suspension ; excreta,
etc. ; (b) organic matter in solution ; (c) inorganic matter in suspen-
sion, such as sand, grit, street and road washings, gravel, etc. ; and
(d) inorganic matter in solution, which is not great in amount,
but includes phosphates, one of the favouring agencies of sewage
fungus. We may summarily classify the constituents of sewage as
follows : —

(a) Excretory substances, composed of solid excreta and urine.
The former consist of nitrogenous partly-digested matter, together
with vegetable non-nitrogenous residues of food. The former are
easily broken down ; but the latter are only gradually attacked
(chiefly by the anaerobic bacteria), and broken clown into soluble
compounds fcetidly odorous, and into small black masses, which
slowly deposit as black sludge.

(b) Household waste, solid substances, washings, etc.

(c) Rain and storm water of varying amount, according to season.

(d) Grit, gravel, sand, etc.

(e) Manufacturing waste products in certain localities.

Turning to the bacterial content, it will at once occur to us that
such a large quantity of organic matter as sewage contains, and in
which decomposition is constantly taking place, will afford an almost
ideal nidus for micro-organic life. There is, indeed, but one reason
why such a medium is not absolutely ideal from the microbe's point
of view, and that reason is that in sewage the vast numbers of
bacteria present make the struggle for existence exceptionally keen.
The source of the organisms is most largely the organic dejecta chiefly
constituting the sewage, but there are in addition the organisms of
the air and extraneous fluids and substances found in sewage. The
result of Jordan's + investigations into sewage gave an average of
708,000 living bacteria per c.c, his highest result being 3,963,000

* Analyst, 199, xxiii., 1898.

f Report of State Board of Health, Massachusetts, 1890.

BACTERIA IN SEWAGE 153

per c.c. He obtained higher figures during the summer months
than at other times ; but in any case his average was extremely low.
Laws and Andrewes * found that London crude sewage varied
from 2,781,650 to 11,216,666 micro-organisms per c.c. "It will
thus be seen," they conclude, " that very wide variations exist in the
total number of micro-organisms present in sewage at different times
and in different places. Temperature is one important factor in
determining the rapidity of their reproduction, and hence their
increase in numbers; dilution of the sewage by rainfall must also
exert a marked influence." Houston f has also examined the sewage
of London, and found, in 1898, that the Barking crude sewage con-
tained an average of nearly four millions of organisms per c.c. and
the Crossness crude sewage three and a half millions per c.c. In
1899 the same observer j reported 7,357,692 bacteria per c.c. as the
average in the sewage at the Crossness outfall. On one occasion he
records 19,500,000 micro-organisms as present in one cubic centi-
metre. In 1900, Houston reported similar figures, and on occasion
as many as 1,900,000 B. coli per c.c. in crude sewage. He further
added some records as to the number of bacteria from crude sewage
growing at blood - heat and room temperature. In the former
case he found as many as 6,830,000, and in the latter 11,170,000
per c.c.§

Not only are the numbers incredibly large, but we also find an
extensive representation of species, including both saprophytes and
parasites, non-pathogenic and pathogenic. Many of these are known
as "liquefying" bacteria (from the power which they possess of
liquefying or peptonising nutrient gelatine used as a culture medium),
and this is one of the features of putrefactive bacteria. Bacilli pre-
ponderate over micrococci in actual numbers, and in numbers of
species present. There are also many spores. Dr Houston has
tabulated these results in his Third Eeport (1900) from which it
appears that there are about 340 spores per c.c, and 1,076,923 lique-
fying bacteria per c.c. Moulds are but rarely found in sewage,
though common in sewer air.

It is probable that the investigations made into the contained
bacteria of sewage have up to the present, excellent though they
have been, only revealed those species of bacteria which occur in
considerable abundance. • So though it is impossible to make any
very complete record as regards the species of bacteria present in

Report on the Result of Investigations on the, Micro-organisms of Sewage, London
County Council, 1894.

+ " Filtration of Sewage," Report on the Bacteriological Examination of London
Crude Seioage (First Report), London County Council, 1898.

% " Bacterial Treatment of Crude Sewage " (Second Report), London County
Council, 1899. ' '

§ Ibid. (Third Report), 1900, p. 59.

154

BACTERIAL TREATMENT OF SEWAGE

sewage, we may attempt a provisional list of normal types of sewage
bacteria* as follows: —

1. Bacillus coli communis and all its varieties and allies. Houston
reports as many as 600,000 B. coli per c.c. in London sewage.

2. The Proteus family — Proteus vulgaris, P. Zenheri, P. mirabilis,
and P. cloacinus, first isolated from putrid meat by Hauser, isolated
from sewage by Jordan, etc. Houston also reports that frequently
there may be 100,000 "sewage proteus" present in one c.c. This is
an aerobic, non-chromogenic, actively motile, and rapidly liquefying
bacillus with round ends, one flagellum, and no spore formation. It
differs in essential particulars from the P. vulgaris. Some of the
cultures were pathogenic to guinea-pigs (Plate 14).

3. Bacillus entcritidis sporogenes of Klein. The number of spores
of this organism found in London sewage by Houston varied from
10 to 1000 per c.c, thus often exceeding in number the total number
of spores of aerobic bacilli. The relative numbers of B. coli and the
spores of B. entcritidis sporogenes in crude sewage have been demon-
strated by Klein and Houston in the following table : —

No. of Spores

Sample of Crude Sewage.

No. of Bacteria
per c.c.

No. of B. coli
per c.c.

of

B. ent. sporog.

per c.c.

1. Chiefly domestic sewage

14,240,000

260,000

2000

2. Mixed sewage ....

7,800,000

180,000

200

3. Chiefly domestic sewage .

4,800,000

500,000

2000

4. Mixed sewage and trade-effluent

86,000,000

1,100,000

400

5. Hospital sewage ....

2,800,000

200,000

30

6. Domestic sewage and trade-effluent

4,100,000

500,000

56

7. Domestic sewage ....

28,100,000

2,000,000

50

8. Mixed sewage

21,100,000

1,000,000

35

* The methods adopted for making a quantitative and qualitative examination
of sewage are precisely analogous to those used in milk research. Dilution with
sterilised water previous to plating out on gelatine in Petri dishes is essential (1 c.c.
to 10,000 c.c. of sterile water, or some equally considerable dilution), otherwise the
large number of germs would rapidly liquefy and destroy the film. The plate
should be incubated at a definite temperature, which is usually 20° C. Special
methods must be used for the isolation of special organisms, phenol-gelatine ( '1 c.c.
of a 5 per cent, solution of phenol to every 10 c.c. of gelatine). Eisner medium,
Parietti broth, indol-reaction, and "shake" cultures in gelatine (for testing gas-
production) must often be restored to for certain species. Spores in sewage may
be isolated by adding 1 c.c. of diluted sewage (1-10) to 10 c.c. of melted gelatine in
a test-tube, and heating the mixture to 80° C. for ten minutes before pouring out
into the Petri dish. This temperature kills all the bacilli, but leaves the spores
untouched. The same plan is adopted in principle for separating B. enteritidis
sporogenes : a small quantity of sewage is added to 15 c.c. of fresh sterile milk, which
is heated at 80° C. for ten minutes, and then incubated at 37° C. anaerobically in a
Buchner tube. B. coli communis is grown in phenol-broth for twenty-four hours,
and then plated out on phenol-gelatine.

PLATE 14.

Sewage Proteus (Houston).

Film preparation from agar culture, 24 hours at 20° C.
X 1000.

Sewage Proteus (Houston).
Gelatine plate culture, 48 hours' growth at 20° C. (Natural size).

[To face page 154.

SEWAGE BACTERIA 155

4. Sewage Streptococci. — Laws and Andrewes, Houston, Horrocks,
and others have isolated streptococci from crude sewage, which
appear to be normal sewage organisms, and as such may be taken,
when present in water, to indicate contamination, and, if accompanied
by B. coli, recent and dangerous contamination. Staphylococci have
also been frequently isolated. Houston has described some twenty
streptococci as present in London sewage. They are generally
present in crude sewage in numbers not less than 1000 per c.c.
These sewage streptococci are delicate, and readily lose their vitality
and die. They are probably little prone to enter on a saprophytic
phase or to multiply to any great extent, if at all, under such condi-
tions as prevail in sewage. They are present in the intestinal dis-
charge of animals, and comprise highly pathogenic organisms. They
.are usually absent in pure waters and virgin soils, and waters
recently polluted with excremental matters. They stain well by
Gram's method. The majority form short chains, which sometimes
cohere in masses. They grow well at blood-heat in the ordinary
media, producing acid in milk without clotting it.* Some streptococci
from sewage coagulate milk (Plate 15).

5. Liquefying bacteria, e.g. Bacillus superficialis (Jordan), B.
frondosus (Houston), B. hyalinus (Jordan), B. delicatulus (Jordan),
B. cloacm (Jordan), B. fluorescens stercoralis, B. membraneus patulus,
B. capillar eus (Houston), B. cloacm fluorescens (Laws and Andrewes),
various forms of Clostridium and the typical B. mesentericus (Plate 16).

6. Non-liquefying bacteria, e.g. Bacillus subtil issimus, B.fusiformis
(Houston), B. rubescens (Jordan), B. pyogenes doacinus (Klein f).

We have not included in the above classification any bacteria
virulently pathogenic to man. \ Doubtless, such Bpecies {e.g. Bacillus
typhosus) not infrequently find their way into sewage. But they are
not for various reasons normal habitants, and though they struggle
for survival, the keenness of the competition among the dense crowds
of saprophytes makes existence for a continuous period in sewage
almost impossible for them. In the investigation to which reference
has already been made, Laws and Andrewes devoted some attention
to the behaviour of B. typhosus in sewage. They found that this
bacillus was unable to grow, indeed quickly perished, in sewage
sterilised by filtration and heat, whereas the />'. coli is n 1 >le to increase
and multiply in such a medium. Sewage, therefore, even in the
absence of the normal micro-organisms which ii contains, is an

* Bacterial Treatment of Crude Sewage— Third Report to the London County
Council, by Dr Houston, 1900, pp. 60-68 ; Royal Commission on Sewage Disposal,
Second Report, 1902, p. 25.

t See British Medical Journal, 1899, vol. ii., p. 69.

X Bacillus enf nitidis sporogenes, II. pyogenes doacinus, and other organisms have
been held responsible for diarrhoea, abscess formation, etc., but they cannot yet
be compared \vith\B. typhosus as regards pathogenic effect.

156 BACTERIAL TREATMENT OF SEWAGE

unfavourable medium for the growth of the typhoid bacillus, which
in all probability would die out in a few days' time. In crude
unsterilised sewage it is clear that owing to competition and the
inimical effect some of the non-pathogenic species have upon B.
typhosus* that the death of that organism is, in sewage, " probably
only a matter of a few days or at most one or two weeks." MacConkey
found that in sterilised crude sewage inoculated with the B. typhosus,
this bacillus is recoverable in seventeen days, though it does not appear
to multiply. In ordinary crude sewage so inoculated, the bacillus
was recoverable after thirteen days.f

Of the organisms which we have named as normally present in
sewage, it is unnecessary to speak in detail, with the exception of the
Bacillus enteritidis sporogenes of Klein. J This bacillus is credited
with being a causal agent in diarrhoea, and has been isolated by
Dr Klein from the intestinal contents of children suffering from
autumnal diarrhoea, and from adults having " English cholera." It
has readily been detected in sewage from various localities, and also
in some sewage effluents. It has been separated from ordinary
milk, even from what was termed by the trade " sterilised " milk.
The biological characters of this bacillus are briefly as follows. It
is in thickness about equal to the bacillus of quarter-evil, thicker and
shorter than the bacillus of malignant oedema, and standing therefore
between the latter and the bacillus of anthrax. It is motile and
possesses flagella, but does not assume a thread form. It readily
forms spores, which develop, as a rule, near the ends of the rods,
and are thicker than the bacilli. They can withstand a tempera-
ture of 80° C. for fifteen minutes. The bacillus takes the Gram
stain. In various media it produces gas rapidly. Particularly
is this so in milk. It is an anaerobe, and may be isolated by the
following method. A small quantity of the suspected matter is
added to a tube of fresh sterilised milk, which is then heated in a
water-bath to 80° C. for fifteen minutes. It is then cooled and
incubated at blood-heat in a Buchner's tube (see p. 478). In twenty-
four hours the milk is coagulated into white stringy masses and small
casein coagula, whilst a large portion of the test-tube is filled with
gas or a thick, watery, slightly-turbid whey. It is necessary to
differentiate the B. enteritidis sporogenes from the bacilli of malignant
oedema and symptomatic anthrax and the Bacillus butyricus of Botkin.
For such differentiation it is important to remember that the
enteritidis organism (a) stains by Gram's method, (b) in gelatine culture

* Klein reports that although B. typhosus can live in crude sewage, it is only
for a short period. When sewage is diluted with large quantities of water the case
is different.

f Royal Commission on Sewage Disposal, Second Report, 1902, p. 62.

% Annual Report of the Medical Officer of the Local Government Board , 1897-98,
pp. 210-250.

PLATE 15.

•

v.*

(

J I

> - J>

Sewage Streptococcus, from Effluent. (Houston.)

From broth culture, 48 hours at 38° C. Stained by

Gram's method, x 1000.

Streptococcus pyogenes.

Film culture from broth culture. Stained by Gram's
method, x 1000.

\

Sewage Streptococcus, from Crude Sewage. (Houston.)
From broth culture, 48 hours at 3S C. Stained by Grain's method.

x 1000.

[To face page 156.

SEWAGE BACTERIA 157

.shows no lateral offshoots, and (c) possesses different pathologieal
characters on inoculation. If 1 c.c. of whey from a milk culture be
inoculated into a guinea-pig (200-300 grammes) a swelling appears
in the groin after six hours, which extends to the abdomen and thigh.
The animal is usually dead in eighteen to twenty-four hours with
gangrene of the subcutaneous tissue and offensive sanguineous exuda-
tion. These characteristics, coupled with the morphological and bio-
logical features, are sufficient for differentiation purposes (see also
p. 307).

Houston has shown that the cholera bacillus, B. pyocaneus, and
Staphylococcus pyogenes aureus are capable of retaining their vitality
in crude sewage in competition with the very numerous bacteria
normally present.* The bacillus of anthrax, and still more so its
spores, can also live in sewage and sewage effluents (Houston).f

Whilst we cannot here enter more fully into an account of the
bacteria found in sewage or their functions, it is necessary to remark
upon one important feature. A large number of these organisms
which we have named as normal inhabitants of sewage fulfil as their
main function the process of decomposition and denitrification, that
is to say, their role is to break down, by means of putrefaction, the
organic compounds constituting sewage. For example, urea which
is abundantly present in sewage is thus transformed with extra-
ordinary rapidity by several different forms of bacteria.

By way of summary we may quote Houston's account of the
" standard " of crude sewage. Crude sewage usually contains, at least,
in one cubic centimetre —

(a) 1-10 million bacteria.

(b) 100,000 B. coli or closely allied forms.

(c) 100 spores of B. entcritidis sporogenes.

(d) 1000 streptococci.

W two c-c- is usually sufficient to produce "gas" in gelatine
shake cultures in twenty-four hours at 20° C.

if) The inoculation of animals with crude Bewage always leads
to a local reaction, and not uncommonly results in deatkj

As we have already said, when dealing with the Bacteria of the
Soil, Nature is dependent upon the services of the "economic"
organisms. Dead organic matter is broken down as the result of
the vital activity of putrefactive bacteria (decomposing and denitri-
fying). The ammonia which is thus liberated becomes oxidised first
to nitrous and then to nitric acid by the agency of the nitrifying
bacteria, and the acids by their action upon bases, always present,
produce nitrites and then nitrates. It is upon those Bubstances that

luiri. Trail mi >ii ofCrudi Setoage, Third Report, 1900, p. 75.
i Royal Commission on Sewage Disposal* Second Report, 1902, p. 31.
X Ibid., 1902, p. 126.

158 BACTERIAL TREATMENT OF SEWAGE

plant life finds nutriment. That the carbon is converted into C0.2,
the hydrogen into water (H20), and the " lost " nitrogen refixed in
the soil, we have already seen.

Now just as soil contains these Economic Organisms, whose role
is to complete the cycle of nature, removing the dead remains of
plants and animals, and assimilating them in such a way as to add
to the fertility of the soil and recommence the cycle of life, so also
in sewage we have all the required organisms normally 'present, whose
business it is to render soluble the solid matters, and to split up the
organic compounds into their simple elements, and then as a final
stage in the process to oxidise these elements and so produce an
effluent free from putrescible matter, but containing nitrates and
other mineral substances.* For practical purposes these two main
groups of bacteria, the breakers-down and the builders-up, are looked
upon as anaerobic or aerobic. The former are active in the absence
of air, and their activity effects a decomposition of complex organic
matter and allied substances. The aerobes are most active in the
presence of oxygen, and part of their business is to convert urea into
ammonia and ammonia into nitrate.

From this brief recital of the functions of many of the sewage
bacteria we learn that they have important operations to perform,
and that their presence in sewage, even in very large numbers, is
not matter for regret, but far otherwise. We see also a remarkable
adaptation of those fermentations discovered by Schloesing and
Miintz, in 1878, to be of such inestimable economic value in soil.

We are now in a position to consider the treatment, especially
the biological treatment, of sewage.

The Biological Treatment of Sewage

Almost from time immemorial there has been adopted one of
three great methods of disposal of sewage : —

1. Disposal without purification.

2. Mechanical and chemical separations.

3. Biological methods.

It may be convenient to add here that the complete purification
of sewage involves three processes : — First, the process of clarifica-
tion, that is to say, the removal of suspended solid matters ; secondly,
an alteration of the chemical constitution of organic putrescible

* The following have been considered as the general conditions which an effluent
ought to fulfil : (a) It must contain practically no solids in suspension ; (b) it must
not contain in solution a quantity of organic matter sufficient to seriously absorb the
oxygen from the stream water into which it is discharged ; (c) it must not be liable
to putrefaction or secondary decomposition ; (d) it must contain nothing inimical to
microbial growth and activity, therefore it must not be treated with strong anti-
septics ; (e) it must not contain pathogenic organisms.

PLATE 16.

Bacillus mesentericus, Sewage variety (No. i.) (Houston.)
Film preparation from agar culture, 20 hours at 20° C. x 1000.

Bacillus mesentericus, Sewage variety (No. i.) (Houston.)
Gelatine plate culture, 20° C. (Natural size.)

[To face page 158.

SEWAGE DISPOSAL 159

matter in solution in the sewage, so that such putrescible matter
appears in the effluent in a form which will not undergo any further
putrefactive change ; and thirdly, the removal of disease-producing
bacteria, which will be present in practically all crude town sewage.

These results are not obtained equally by the various methods
employed, but it will be best to consider each of these separately.

1. Disposal without Purification.

Various antiquated forms of carrying out this mode have been
used. Seaside places have often been content to carry their un-
treated sewage out to sea. Towns situated on the banks of rivers
have frequently by means of a conduit conveyed sewage into the
running stream. There is nothing necessarily objectionable in this
mode of disposal, for both in the sea and in running river water the
sewage matter will become disintegrated and dissolved. Yet the
method is liable to give rise to very serious nuisance, unless the
conditions requisite for solution are carefully studied. Nuisances
may arise in respect to the pollution of bathing grounds, or actually
injurious effect upon the health of the population on the banks of
the river, or by injury to fish (by reducing the oxygen in the water,
destroying the food of fish, admitting poisonous matters into the
water, or by suspended matters clogging the gills of fish). In a
general way it may be said that before the admission of sewage into
any body of water is permissible, two points require consideration,
namely, the removal as far as practicable of the matters in suspension
in the sewage, and the sufficiency of dissolved oxygen in the water
completely to prevent any putrefaction. Broadly, also, it may
be said that for towns situated on non-tidal streams some form of
bacterial treatment is preferable. Towns on tidal rivers require as
a rule a chemical precipitation process.*

* Foulerton has recently drawn attention to a modified chemical preeipitation
proeess treatment of sewage which is to be discharged into a tidal water, which
may be carried out as follows : — The effluent from an ordinary chemical precipita-
tion process is distributed continuously over a coarse "filter-bed" by means of a
sprinkler. In this way a thorough aeration of the effluent before its discharge into
the stream is ensured, and provision is made tor the complete removal of all traces
of solid suspended matter. As the effluent from the sedimentation tanks trickles
slowly through the coarse interstices of the filter-bed, any solid suspended matter
which has escaped precipitation m the previous part of the process will be deposited,
and then dealt with by bacteria. And in the result an effluent, fully oxygenated,
free from the solid suspended matter of the crude sewage, and with the bacteria
originally present in the crude sewage considerably decreased in numbers, will be
discharged into the stream. The somewhat higher proportion of dissolved organic
putrescible matter in the effluent from such a chemical process, as compared with
the proportion which may obtain in a good bacterial process, is probably not a matter
of considerable importance in the case of tidal waters. — Report on Pollution of Tidal
Fishing Waters by Sewage, 1903, p. 8.

160 BACTERIAL TREATMENT OF SEWAGE

2. Mechanical and Chemical Separations.

Methods in which this principle is applied are numerous ; they
have generally been of the nature of a " precipitation " process. Six
to twelve grains of quicklime have been added to each gallon of
sewage, forming a precipitate of carbonate of lime, which carries
down with it the light, flocculent suspended matter of the sewage.
The process is simple and cheap ; it does not, however, remove the
organic matter in solution, but merely the solid matters in suspension.
On the one hand it does not produce a valuable manure; on the
other it fails to purify the effluent. A score of other methods have
been tried {e.g. the lime and ferrous sulphate treatment, Hanson's
process, " ferozone," amines, electrolysis, etc.), but with the exception
of electrolysis, all based on the addition of chemical substances able
to precipitate or otherwise change the organic matter of the sewage.
All these methods produce large quantities of sludge, the removal
of which, by pressing, digging into the land, or sending out to sea,
presents many difficulties. But the chemical processes have this
advantage, that they remove disease-producing organisms more per-
fectly than the bacterial process, though the latter carries further
the purification of dissolved organic putrescible matter.

3. Biological Methods.

The biological methods, though very various, all have two common
features. In the first place, the injurious and putrescible substances
in the sewage are not merely " disposed of " nor yet only " separated."
They are destroyed. There is a destruction of sewage as sewage,
and a building-up of new substance in its place. Secondly, this
desired effect is achieved, not by adding anything to the sewage, but
by the organisms normally present in the sewage or in the medium —
the land or the " filtering " agent — upon which the sewage is treated.
In short, all biological processes depend upon the employment of
bacteria in some shape or form. Hence each is a bacterial treat-
ment of sewage. It may appear at first sight that such a process,
involving, as it does, encouragement to the growth of putrefactive
bacteria, is not without danger. But we shall be satisfied that this
is not really so, when it is remembered that the bacterial treatment
of sewage is under control, and may be regulated at will. Moreover,
the processes of decomposition and nitrification ultimately destroy
the pabulum upon which the organisms in question depend for their
existence, and hence lead to their death when they have fulfilled
their function.

Two applications of this principle have long been in vogue,
namely, the intermittent downvjard, filtration and broad irrigation.

SEWAGE DISPOSAL 16l

The former may be defined as "the concentration of sewage at short
intej vals, on an area of specially-chosen porous ground as small as
will absorb and cleanse it, not excluding vegetation, but making the
product of secondary importance" (Metropolitan Sewage Commis-
sion). The interinittency is essential, and the process is partly
mechanical and partly bacterial, that is to say, clue in part to the
nitrification set up by the bacteria in the superficial layers of soil.
For successful filtration a porous soil is requisite, a proper inclination
of the land to allow of distribution, and a division into areas, in
order that each part may receive sewage for, say, six hours, and then
have eighteen hours' rest. Soil pipes carry off the effluent. Broad
irrigation (sewage-forms) is the " distribution of sewage over a large
surface of ordinary agricultural ground, having in view a maximum
growth of vegetation (consistently with clue purification) for the
amount of sewage supplied." To ensure success, the area must be
large (say, about one acre to every 100 of the population), the sewage
passed on intermittently to allow of aeration of the soil, and the soil
itself must be light and porous. Like the former, there is a bacterial
influence at work here. Both of these methods are much to be
preferred to chemical treatment (and were recommended by the
Sewage Commission of 1865); yet, on account of space and manage-
ment, as well as on account of the tendency of the land to clog or
become, as it is termed, "sewage sick," their success has not been
all that could be desired.

In 1868, a Commission was appointed to inquire into the best
means of preventing the pollution of rivers. They made several
reports, the fifth and last being made in 1874. The opinion of this
Commission on the comparative merits of the three classes of pro-
cesses for the treatment of sewage, viz : — chemical precipitation,
intermittent filtration, and broad irrigation, may be stated thus : — (1)
All these processes are to a great extent successful in removing pollut-
ing organic matter in suspension. But intermittent filtration is best,
broad irrigation ranks next, and the chemical precipitation processes
are less efficient. (2) But for removing organic matters in solution the
processes of downward intermittent filtration and broad irrigation are
greatly superior to upward filtration and chemical processes.

The last Commission was appointed in 1882. They were directed
to inquire into and report upon the system under which sewage was
discharged into the Thames by the Metropolitan Board of Works,
whether any evil effects resulted therefrom, and if so, what measures
could be applied for remedying or preventing the same. In
November 1884 they issued their final Report. They found that
evils did exist "imperatively demanding a prompt remedy," and that
by chemical precipitation a certain part of the organic matter of the
sewage would be removed. They reported, however, " that the liquid

L

162 BACTERIAL TREATMENT OF SEWAGE

so separated would not be sufficiently free from noxious matters to
allow of its being discharged at the present outfalls as a permanent
measure. It would require further purification, and this, according
to the present state of knowledge, can only be done effectually by its
application to the land."

The present Eoyal Commission has recorded its view in a pre-
liminary report on land treatment, " that peat and stiff clay lands are
generally unsuitable for the purification of sewage, that their use for
this purpose is always attended with difficulty, and that where the
depth of top soil is very small, say six inches or less, the area of such
lands which would be required for efficient purification would in cer-
tain cases be so great as to render land treatment impracticable." On
the subject of effluents they state in the same preliminary report : —

" We may, however, even at this stage, point out that as a result
of a large number of examinations of effluents from sewage farms and
from artificial processes, we find that while in the case of effluents
from land of a kind suitable for the purification of sewage, there are
fewer micro-organisms than in the effluents from most artificial pro-
cesses, yet both classes of effluents usually contain large numbers of
organisms, many of which appear to be of intestinal derivation, and
some of which are of a kind liable, under certain circumstances at
least, to give rise to disease.

" We are of opinion, therefore, that such effluents must be
regarded as potentially dangerous, and we are considering whether
means are available and practicable for eliminating or destroying
such organisms, or, at least, those giving rise to infectious diseases."

Until comparatively recent years, the above methods of treating
sewage were the only ones available, or, at all events, practised. But
now, as is well known, some new applications of the biological treat-
ment of sewage have been introduced, which call for consideration.
Their popularity has been due to it being possible to adopt them
where suitable soil did not exist for the other biological methods, and
to the fact they have been on the whole less expensive to work.
These new departures depend upon bacteria contained in the sewage.
The process may depend mainly upon anaerobes (Cameron's Septic
Tank or Scott-Moncrieff's process) or aerobes (Duckett's filter or
Dibdin's filter). These may be conveniently dealt with in more or
less chronological order.

The Bacterial Treatment of Sewage

In 1872 the Berlin Sewerage Commission reported that sewage
matter was converted into nitrates, not simply by molecular processes
but by organisms present in sewage and soil. Muntz, Mueller, and
others demonstrated this in various ways. Mueller, indeed, had shown

PIONEER WORK 163

this to bo so in 1865, naming spirilla, vibrios, and a " protococcus " as
the organisms in question. Then in 1881 M. Louis Mouras, of Vesoul
(Haute-Saone), published an account of an hermetically-sealed, in-
odorous, and automatically discharging cesspool, in which sewage was
anaerobically broken clown by " the mysterious agents of fermenta-
tion." The effluent, a homogeneous and scarcely turbid fluid, was
produced in a tolerably short time and without any addition of
chemical ingredients. It was surmised that the agents of fermenta-
tion might possibly be the "anaerobes of M. Pasteur." This, it would
seem, is the first record we have of the treatment of sewage by simply
allowing Nature to fulfil her function by means of bacteria.

The next step — and indeed as regards the problem in England the
first step — in the new bacterial treatment of sewage was inaugurated
by the workers (Jordan, and others) under the State Board of Health
of Massachusetts, who have carried out a laborious series of experi-
ments upon sewage purification during the last fifteen years. The work
undertaken at this station may be briefly divided into three main
classes : first, purification of unfiltered crude sewage by means of
intermittent filtration through sand filters ; secondly, rapid filtration
of sewage, from which a certain amount of sludge has been removed,
by different methods and through different materials; thirdly, purifi-
cation by dependence upon rapid oxidation or burning of sludge
either by forced aeration or other method of introducing air into the
filter. Various methods have also been devised with the object of
getting rid of the insoluble matter in sewage. The result of this
extremely valuable work by the Massachusetts Board clearly demon-
strated the accuracy of the fundamental principle that on prepared
beds or "intermittent downward filtration" the contained bacteria
were the potent agency.*

Whilst this work opened up a new. prospect for the bacterial

treatment of sewage, it still left the question in an experimental stage,-

and then it was that Scott-Moncrieff, Dibdiu, Cameron, Ducat, and

others carried forward the work.f It was in 1892 that Mr Scott-

* In his evidence before Lord Bram well's Commission, 1883, Dr Sorby had
pointed out that the destruction of the organic sewage matter in the Thames was due
to bacteria, and that it was only when these were unable to exert their functions to
the full extent by reason of deficient aeration of the water that putrefaction set in.
f The following general classification will serve to show the nature of the pro-
cesses adopted by various workers : —

Closed septic tank and contact beds.

Open septic tank and contact beds.

Chemical treatment, subsidence tanks, and contact beds.

Subsidence tanks and contact beds.

Contact beds alone.

Closed septic tank followed by continuous nitration.

Open septic tank followed by continuous nitration.

Chemical treatment, subsidence tanks, and continuous filtration.

Subsidence tanks followed by continuous filtration.

Continuous filtration alone.

164 BACTERIAL TREATMENT OF SEWAGE

Moncrieff introduced his cultivation beds filled with flint, coke, and
gravel. In this system the crude sewage passes into the bottom of
the bed, the liquid portion rises through the bed, and the suspended
matter is kept back at the bottom, where it undergoes solution by the
action of bacteria present upon the surfaces of the flints, etc. In order
to complete the process highly oxygenated water was added to the
effluent, which was then passed down a "nitrification channel," where
further oxidation was secured. The results of this process were
superior to anything previously obtained in this country.

Between the years 1891-95 Mr Dibdin experimented with sewage
from which solid organic matter, had been previously removed by
screens or chemical sedimentation. Such sewage he passed through
filter-beds of coke breeze, and was able by intermittent filtration
through such beds (i.e. allowing rest periods between charging the
filters) to obtain a purification of more than 70 per cent.

Dr Clowes has carried on this work on behalf of the London
County Council, and he reports that the sewage is allowed to flow
into large tanks which contain fragments of coke about the size of
walnuts. As soon as the level of the liquid has reached the upper
surface of the coke-bed, its further inflow is stopped, and it is
allowed to remain in contact with the bacteria coke surface for about
three hours. It is then allowed to flow slowly away from the
bottom of the coke-bed. After an interval of about seven hours, the
processes of emptying and filling the coke-bed are repeated with a
fresh portion of sewage. The coke-bed is usually filled in this way
twice in every twenty-four hours. The aeration of even the lowest
portions of a deep coke-bed seems to be satisfactory in the above
method of working, since the air present in the interstices of the
coke, between two fillings with sewage, usually contains 75 per cent,
of the amount of oxygen present in the open air.

In dealing with the sewage of the metropolis, Dr Clowes holds
that it is best to allow the roughly-screened raw sewage to undergo a
somewhat rapid process of sedimentation, in order to permit these
matters to subside ; and then to pass the sewage direct into the coke-
beds. The dissolved matters and the small amount of suspended
matters which are still present in the sewage are then readily dealt
with by the bacteria of the coke-bed, and practically no choking of
the bed occurs. The sewage effluent from the coke-bed is entirely
free from offensive odour, and remains inoffensive and odourless even
after it has been kept for a month.

The chemical character of this effluent may be briefly indicated
by stating that on an average 51 '3 per cent, of the dissolved matter
of the original sewage, which is oxidisable by permanganate, has
been removed by the bacteria, and that the portion which has
been removed is evidently the matter which would become rapidly

SEPTIC TANK METHOD 165

offensive, and would rapidly lead to cleoxygenation of the river water
if it were allowed to pass into the river. The above percentage
removal (51*3) was effected by coke-beds varying from 4 to 6 feet in
depth. A similar bed, 13 feet in depth, has proved more efficient,
and has for some time produced a percentage purification of 64 per
cent., while an old bed, 6 feet in depth, has given a percentage
purification of 86 per cent. A repetition of the treatment of the
effluent in a second similar coke-bed has produced an additional
purification of 19*3 per cent., giving a total purification of 70*6 per
cent. Effluents from chemical treatment would show a total
purification of under 20 per cent. It should be noted that the above
purification is reckoned on the dissolved impurity of the sewage ;
the suspended solid matter is not taken into account. The bacterio-
logical condition of the effluent corresponds in the main with that
of the raw sewage. The total number of bacteria undergoes some
reduction in the coke-beds, but the different kinds of bacteria which
were present in the sewage are still represented in the effluent.

From these and many other similar experiments, it has come to
be understood that the bacterial purification depends, as we have
seen, upon two main groups of organisms, namely, those that are
able to break down and liquefy solid organic matter, and those that
deal with it when in solution. Of the former group, some act besl
under anaerobic conditions. No strict line of demarcation can be
drawn as to where one group begins and the other absolutely ends.
It is a complex co-operation, shared in by a large variety of
organisms classified roughly into these two groups. Systems may
be introduced in which the anaerobes are encouraged (as at Exeter),
or systems may be established on an aerobic basis (as at Sutton and
Manchester). Hence it may be accepted as finally settled ili.it the
bacterial treatment may be mainly an anaerobic one (Cameron, Scott-
Moncrieff, and others), or mainly an aerobic one (Dibdin, Fowler, and
others), or a mixture of the two. Whatever system is used, the two
great agencies of breaking down and oxidation must be allowed
ample opportunity. Probably we shall most clearly recount the
application of these principles by considering in some detail two
examples of the two typical methods of bacterial treatment. These
two examples are furnished in Cameron's Septic Tank Installation
(anaerobic), and at the Davyhulme Works, Manchester, in the
Multiple Contact Bacteria-beds (aerobic).

1. Septic Tank and Cultivation-bed Method (Cameron).— This
method has been adopted at Exeter and other places. The plant is
twofold, namely, a septic tank and several cultivation or bacteria beds.

The septic tank is a large underground vault constructed of
Ancrete, cemented on exposed surfaces, and having a capacity of
thousands of gallons, according to the population. That at Exeter

166 BACTERIAL TREATMENT OF SEWAGE

has a capacity of 53,800 gallons, and takes the average sewage of
1500 inhabitants in twenty-four hours. Near the entrance is a
submerged wall, forming a grit chamber for the arrest of gravel
and coarser detritus. The remaining solid matter passes into the
tank itself. The inlet and outlet being below the level of the
sewage, light and air are excluded as far as possible. Both in the
sediment at the bottom of the tank and in the thick scum on the
surface the organic compounds are broken down and made soluble.
In the former position this is accomplished by anaerobic bacteria, in
the latter, on the surface, by aerobic bacteria. It need hardly be
added that these are denitrifying and putrefactive bacteria, and that
those at the bottom of the tank perform greater service than those at
the top. What are the changes taking place in the tank ? On
every side throughout the tank innumerable small masses of organic
matter may be observed rising and falling. At first the masses fall
to the bottom by gravity ; here they are attacked by countless bacteria
which generate numerous gases in the small masses which are thus
caused to rise again to the surface; the pressure being then reduced,
the gases expand and burst in bubbles, leaving the particles to sink
again and commence a similar cycle. Thus the sewage is rapidly
broken clown by a process of peptonisation and digestion (anaerobic
hydrolysis) until all the organic matter is in solution (soluble nitro-
genous compounds, phenol derivatives, gases, ammonia, nitrites, etc.).
No rest is necessary, for the supply of organisms is unlimited, being
perpetually replenished by incoming sewage. It is contended, and
probably with some truth, that most pathogenic organisms would not
be able to survive in the competition which must be present in the
septic tank. When the liquid sewage passes out of the tank, it
differs from the crude sewage entering the tank, in the following
particulars : — (a) The gravel and particular debris have been removed ;
(b) the organic solids in suspension are so greatly diminished that
they are almost absent ; (c) there is an increase of organic matter in
solution ; (d) the sewage is darker in colour and more opalescent ;
(e) compounds like albuminoids, urea, etc., have been more or less
completely broken down, reappearing in more elementary conditions,
like ammonia, methane, carbonic acid gas, and sulphuretted hydrogen.
These latter bodies may be in solution, or may have escaped as gas.

The cultivation beds are five or six " filters," to which the sewage
from the tank flows, and by an automatic arrangement is distributed
to each bed in turn. Each filter may thus be full, say, about six
hours, and has from ten to twelve hours' rest. The depth of the
filtering medium is 4 or 5 feet, and is composed as follows from the
bottom upwards : —

(a) About 1 foot in depth of broken furnace clinker, which will
pass a 3-inch mesh, but not a 1-inch.

SEPTIC TANK METHOD

167

168 BACTERIAL TREATMENT OF SEWAGE

(b) Two feet or so of screened clinkers to pass a f-inch mesh,
but not a J-inch mesh.

(c) Three inches of residue from above, which will pass a J-inch
mesh.

Collecting drains are laid on the bottom of the beds, joining main
collectors, which terminate in discharging wells.*

The change occurring in these bacteria beds is of the nature of
oxidation, with the result that the proportion of oxidised nitrogen
increases (as nitrites and nitrates), the ammonia becomes less and
the total solids and organic nitrogen almost disappear. It will thus
be seen that the work of these " filters " is not merely a straining
action. It is true that particulate matter in the effluent from the
tank is caught on the surface by the film (resulting from previous
effluents), but the real work of the bed is nitrification, an oxidation
of ammonia into nitrites and nitrates. This change obviously begins
when the tank effluent flows over on to the beds and the oxygen
then obtained by the effluent is carried down in solution into the
coke breeze. Upon the surfaces of the filtrant are oxidising bacteria.
When the effluent is on the bed they oxidise its contained products ;
when the bed is empty and "resting" they oxidise carbon. An
advantage arising from the periodical emptying and filling of the
" filter " is that the products of decomposition which would eventually
inhibit the action of the aerobic bacteria are washed away and pass
into the nearest stream, or on to the land direct, where they become,
of course, absolutely innocuous. The " filter " is perhaps more
correctly termed a cultivation bed, for its purpose is to furnish a
very large surface upon which nitrifying organisms, present as
we have seen in all soils, may flourish, and these feeding upon
the organic matter of the sewage, may perform their function of
oxidation.

The solid matter has plenty of time to settle in the tank and be
fully operated on by the bacteria, which are not only contained in
the sewage, but also grow and multiply in the tanks. This growth
is, of course, a question of time; and just as the growth of the
nitrifying layer is necessary to a water filter-bed, so the growth of
the necessary organisms is required in the septic tank and on the
filter-beds. In the tank, however, no rest is necessary, for the supply
of organisms is continuous and unlimited, both in supply and in
reproduction. Not so the beds. Here there is only a limited
amount of oxygen to start with, and consequently a definite limita-
tion to the amount of work the filters can perform. Hence the need
of rest, in order that the oxygen may be replenished.

The amount of sludge in the chemical processes has always been

* An account of the Septic Tank method will be found in the Brit. Med. Jour,,
1900, i., pp. 83-86.

SEPTIC TANK METHOD

169

a difficulty. In the bacterial processes it is reduced to one-third, and
often is so little as to be a negligible quantity.

It is not possible to lay down exact limits as to where denitrifica-
tion ends and oxidation begins. To a certain extent, and in varying
degree, they overlap each other. But, generally speaking, we may
say that in the tank there is a breaking-down (denitrification and
decomposition) and in the filter-beds a building-up (nitrification).

The case is precisely parallel to similar

changes occurring m

soil.

Metal;
Screen!

.Conduit to Beds

Filter Bed full of
Burnt Clay

Exit Pipe

Filter Bed empty

of Filtrant, showing1

Exit Pipe at the

bottom

Fio. 22.— Contact Beds (aa used at Sutton).

and with which we have already dealt. It is hardly necessary to add
that there is a marked reduction in the number of bacteria present
in the crude sewage, and the tank and cultivation-bed effluents. One
investigation has shown thai a sample of crude Bewage contained
4,084,827 bacteria per c.c. ; the sewage precipitate, 1,344,925; the
tank effluent, 398,695 ; and the cultivation-bed effluent, 45,755 bacteria
per e.c.

2. Multiple Contact Bacteria Beds. — This method in its
simplest form lias been applied, for instance, at Sutton, and in a

170 BACTERIAL TREATMENT OF SEWAGE

more advanced form at Davyhulme, Manchester. At Sutton there is
no tank. A metal screen holds back part of the solids from being
carried on to the beds. The filtrant is burnt clay, and it is forked
over occasionally to let in oxygen. The beds are 3J feet deep. The
bottom of the bed is provided with a 6-inch main drain with tributary
drains. The crude sewage, after passing through a roughing screen
to intercept floating paper, etc., is run directly upon the filter
without the addition of any chemicals. The filter is charged to
within 6 inches from the surface, and the sewage remains in contact
for a period of two hours, after which the outlet valve is opened and the
filtrate is drawn off to be further purified on fine-grain bacteria beds,
after which the effluent is in a fit condition to be discharged into the
stream, and is uniformly superior to the effluent obtainable by
chemical treatment. The sludge is absorbed by bacterial agency in
the beds, and does not accumulate or manifest itself. The beds are
free from any offensive odour. At first the beds were seeded with
Micrococcus candicans, but it is now known that the necessary bacteria
are in the sewage, and seeding is not required. For more effective
screening of the sewage an automatic rotary screen may be fixed.
This screen may be driven by a Poucelet water-wheel, actuated by
the sewage.

Experiments seem to prove that coarse-grain beds worked on the
contact principle may be constructed of a numerous class of materials,
and that different districts may adopt materials which are obtainable
locally, and often at a small cost, although it may be observed that
porous coarse-grained materials such as coke and burnt ballast effect
a greater degree of purification than do fine-grained impervious
material such as granite, slate, etc. The cost of such a system would
be in many cases one-quarter of a chemical precipitation and
irrigation system, and yet more effective. It will be understood that
the absence of a septic tank does not mean an entire absence of the
anaerobic action of the process. It simply takes for granted that
this portion of the process has been in part performed in the sewers.

The Manchester bacterial system is practically the same in
principle as at Sutton. But there are two important and interesting
differences. First, the quantity of sewage to be dealt with is very
much greater. Secondly, there is the added complication that the
Manchester sewage contains large quantities of trade-effiueiit from
breweries, dyeing and bleaching works, galvanising works, tanneries,
and derivatives from coal-tar, colour, and naphthalene works. It has
been frequently suggested that such chemicals in the sewage would
prevent the contained bacteria from fulfilling their role in the purifi-
cation. Consequently, the trial of the two chief methods of bacterial
treatment of Manchester sewage has been followed with much
interest. In 1898 three experts were appointed, and requested to

CONTACT BEDS 171

furnish a report as to the system best adapted for Manchester.
Various points demanded elucidation which had previously escaped.
These were chiefly (1) whether trade-refuse in the sewage impaired
the efficiency of bacterial purification ; (2) whether a portion, at any
rate, of the sludge can be destroyed by bacterial agency ; (3) whether
chemical precipitation, as in Dibdin's first method, before bacterial
treatment could be dispensed with ; (4) whether an aerobic process
alone or a combination of anaerobic and aerobic processes is the more
effective. With these objects in view a septic-tank (Cameron)
method and a filter-bed method were installed, and under the super-
intendence of Dr G-. J. Fowler, the superintendent and chemist of
the Corporation Sewage Works, the observations were carried out.*

That the experts' view of the bacterial treatment of sewage was
similar to that set forth above may be gathered from a preliminary
note : " The bacteria already existing in the sewage," they state,
" are brought by it on to the bacteria beds, or into the septic tank.
The former, by providing an enormously extended surface for the
development of bacterial growths, furnish an ideal habitat for the
aerobic micro-organisms which require air for the display of their
powers, whilst the septic tank, by confining a large volume of liquid
which is but superficially in contact with air, enables the anaerobic
micro-organisms to work to great advantage. It will be understood
that some time must elapse before the bacterial life attains its
maximum, either in the bacteria bed or in the septic tank, and
consequently the amount of sewage which can be purified therein
will gradually increase as time goes on.""f*

The contact beds at Manchester were five in number, and of different
area. In principle they were of similar construction to those described
already. The filtering medium was clinkers laid to a depth of 3 feet,
and of varying size in the five beds, but uniform in each bed.
Clinkers which passed j-inch mesh, rejected by ^-inch mesh, were
found to be the best grade.J

Arrangements are made by which it is possible to admit sewage
to the beds in three different conditions, namely, raw, screened
sewage; sewage which has undergone settlement in a small settling
tank ; sewage which has undergone both settlement and anaerobic
action. Eventually a plan was adopted by which the sewage was

* A full record of the work done at Manchester will be found in the Rivera
Department Reports for 1901, 1902, 1903, and 1904.

f City of Manchester, Rivers Department. Experts' Report on Treatment of
Manchester Seioaye, 1899, p. 12. (Mr Baldwin Latham, Professor Percy Frankland,
F.R.S., and Professor W. H. Perkin, Junr., F.R.S.).

X It appears that the initial capacity of a contact bed is uninfluenced by the
grade of clinker. At first there is a rapid decrease in capacity, due in part to
sinking of the surface and in part to bacterial growth on the clinkers, which must
necessarily occupy some space, even though relatively little. In a comparatively
short time the beds acquire a constant average capacity.

172 BACTERIAL TREATMENT OF SEWAGE

passed through a settling tank prior to its being brought on to the
contact beds. In short, the method resolved itself into one of an
open septic tank and multiple contact following the settlement or
screening out of grosser solids. The results surpassed even the most
sanguine expectations, even though the beds were filled four times
daily. If the two methods — namely, the closed septic tank and
single contact, and the open tank and multiple contact — are com-
pared, it is the opinion of the three experts named that " where both
systems are dealing with the same volume of sewage on the same
area, the advantage as regards efficiency belongs indisputably to the
double contact system" Boyce subsequently confirmed this conclusion
in favour of a combination of the anaerobic and aerobic processes,
provided that the septic process was perfected, and the suspended
sludge did not pass over on to the beds.*

Before summarising the main conclusions which may now be
legitimately drawn from the Manchester experiments, a word or two
may be said concerning the characters of an efficient bacteria bed
and the management of storm-waters in sewage treatment.

The material, or filtrant, of which the bed is composed may vary
within wide limits. Burnt clay, coke, clinkers, cinders, or various
forms of gravel may all be efficient, provided there is ample aeration
and porosity. The required organisms exist, of course, mainly in the
sewage, but they require abundant oxygen in order to perform their
function. To assist in maintaining this aeration the surface should
be raked over from time to time. It has been suggested that in
times of frost a layer of ice would prevent the action of the bed.
But in point of fact such obstruction would rarely occur, the
temperature of the sewage being sufficient both to prevent such
stoppage of the beds and also to maintain the necessary activity of
the bacteria, which, as we have already seen, require for their vitality
nutriment, oxygen, moisture, and a favourable temperature. From
December to April the average daily temperature of the Manchester
sewage was 55*5° F., whilst the average temperature of the sur-
rounding air was 45 "3° F. Hence the ice difficulty is naturally
overcome. Another point of importance in connection with aeration
is the allowance of sufficiently frequent and prolonged periods of
rest. Without such intervals the beds would of course become
clogged, and eventually inactive, because lacking in aerobic bacteria.
Though not absolutely a character of the beds, there is one further
point always to be borne in mind in securing their efficiency. It
is, that the sewage being applied to the bed should be as far as
possible uniform in consistence and freed from suspended matters
by sedimentation. Any suspended matter not so removed should be
retained as far as possible on the surface of the bed.

* Royal Commission on Sewage, Disposal, 1902, p. 11.

CONTACT BEDS 173

A moment's reflection will make it evident that the problem may
be seriously complicated at short notice by the great increase in
volume of the sewage following rain-storms. To this matter the
experimenters at Manchester have also directed their attention.
They draw the necessary distinction between the first flush of a
storm and the highly-diluted sewage which follows, designating the
latter only as " storm water." They decide that pro vision must be
made for the storage or separate treatment of " first flush " of sewage
at the beginning of a storm, and that about two hours after the
augmented flow is the time to commence accelerated treatment, the
exact procedure varying according to the character and duration of
the storm. Short double contacts, or even a single contact, is
sufficient to purify storm water, and there is no decrease in the
purifying capacity of the bed.

Summarily, the final conclusions arrived at by Latham, Frank-
land, and Perkin were as follows : —

" 1. That the bacterial system is the system best adapted for the
purification of the sewage of Manchester.

"2. That any doubts which may have arisen in the first instance
as to its suitability, owing to the presence in Manchester sewage of
much manufacturing refuse, have, through the convincing results of
our experimental inquiry, been entirely banished.

"3. That inasmuch as a bacterial contact bed can only effect a
definite amount of purification in a single contact, it becomes
necessary, in order to carry the purification beyond tliis limit, to
apply the effluent to a second bed, in which again a further definite
amount of purification can be effected. Hence, for obtaining a high
degree of efficiency in bacterial purification of sewage, a system of
multiple contact is generally necessary. Thus it may be taken
broadly that in the first contact 50 percent. «»!' the dissolved impurity
is removed, and that in the second contacl 50 per cent. of the
impurity still remaining in the effluent is disposed of, and so on."*

In subsequent experiments these conclusions were amply con-
firmed, and the Manchester Corporation eventually extended their
sewage works, laying down five additional tanks, and a Largenumber
of contact beds (primary and secondary). These included 92 half-acre
primary beds, with 26 acres storm-water filter-beds at Davyhulme,

* Hirers Department of Manchester, Experts' Report, 1899, p. 53. The
Borough Surveyor of Leicester (Special Report, 1900) examined various bacterial
systems for the disposal of the sewage of the Belgrave district, and finally
recommends as the best method the following: (1) Crude sewage passed into an
open or closed detritus tank to remove suspended mineral matters; (2) then on to
clarifying bacteria beds of 1 feet 6 inches depth, and containing crushed and
screened clinkers (coarse and fine) from the refuse destructors ; three fillings a day ;
(3) finally, land purification of the effluent on old pasture. (Total purification of
suspended matter of sewage 99*12 per cent. ; of albuminoid ammonia, 86 '70 per
cent. ; and of oxygen absorbed 91 'OS per cent.).

174 BACTERIAL TREATMENT OF SEWAGE

and 46 acres of secondary contact beds on land at Flixton.* Dr
Fowler, the superintendent and chemist, concludes that the bacterial
process is best conducted in three stages: — (a) Settlement and
screening out of the grosser solids ; (b) Anaerobic decomposition in
the septic tank ; and (c) Oxidation on bacteria beds. He concludes
in respect of the septic tanks : —

That the effluents from closed and open septic tanks are practically indentical in
composition, and that with a tank space equal to half the daily flow of Manchester
sewage, it is possible to digest about 25 per cent, of the total suspended matter in
the sewage. The suspended matter in the septic-tank effluent is of a granular char-
acter, and readily separates out on standing, and when arrested on the surface of a
bacteria bed does not seriously impede the free flow of the water into the bed. The
organic matter in solution is much more easily nitrified than that present in fresh
sewage, so that it is possible with one contact to constantly obtain non-putrefactive
filtrates. The blending which takes place in the septic tank is of value in minimis-
ing the effect of excessive amounts of manufacturing refuse, and in producing an
effluent of fairly constant composition.

In respect to contact beds he points out that the capacity of contact beds suffers
a rapid initial decrease, but afterwards, with careful working, the rate of decrease is
very much less.

The causes of loss of capacity appear to be five, namely — (a) Settling together of
the material ; (6) Growth of organisms ; (c) Impaired drainage ; (d) Insoluble
matter entering bed ; and (e) Breaking down of material. These matters will be
found fully discussed in the annual reports of the Rivers Department, and we have
not space to enter into them here. We may, however, briefly refer to the conditions
of the successful working of contact beds as arrived at as a result of the Manchester
experience : —

(1) The bed should be worked very slowly at first, in order to allow it to settle
down and the bacterial growths to form. In this way there will be less danger of
suspended matter finding its way into the body of the bed, while the material is
still loose and open. (2) The burden should not be increased till analysis reveals
the presence of surplus oxygen, either dissolved or in the form of nitrates in the
effluent. (3) Analyses of the air in the bed may usefully be made from time to time
during resting periods. (4) The variations in capacity should be carefully recorded.
If the capacity is found to be rapidly decreasing, a period of rest should be allowed.
(5) Long periods of rest should be avoided during winter, as when deprived of the
heat of the sewage the activity of the organisms decreases. If necessary, the
burden on the bed should then be decreased by reducing the number of fillings per
day, rather than by giving a long rest at one time. (6) The insoluble suspended
matter should be retained on the surface by covering the latter with a layer of finer
material not more than three inches in depth. The suspended matter thus arrested
should not be raked into the bed, but when its amount becomes excessive it should
be scraped oif. This should be done if possible in dry, warm weather, after the bed
has rested some days. By placing the inlet and outlet penstocks as close together
as possible, the suspended matter will tend to concentrate in their vicinity, and its
removal will be facilitated, f

How far the various applications of the bacterial agency in puri-
fication will pass the scrutiny of the Eoyal Commission on Sewage
Treatment, now sitting, it is impossible to say. But there can be no

* The particulars as to these new works, their construction, materials, capacities,
etc., will be found in the Manchester Elvers Department Report , 1902, with plates
and charts, pp. 18-24, and more recent extensions in subsequent Reports (1903-4).

f City of Manchester Rivers Department, Annual Report, 1902, p. 15.

RELATION TO DISEASE ORGANISMS 175

longer any doubt that some form of such agency is the only efficient,
because the only natural, means of disposing of sewage.

The Effect of the Bacterial Treatment upon Disease-
Producing- Organisms

It has been urged from time to time by the advocates of the vari-
ous methods of bacterial treatment, that pathogenic organisms are
destroyed during the purification in many of these processes. It is
clearly a matter of importance to know how far an effluent, in
addition to being non-putrescible and fully nitrified, also possesses
no disease-producing capacity. We have already seen, from the
researches of Klein, and Laws and Andrewes, that sewage is not a
favourable medium for B. typhosus. The bacillus of Asiatic cholera
is known to be but little less favoured by sewage. The spread of
diphtheria by sewage is at least a matter of doubt, and Shattock's
experiments tend to prove that in any case the virulence of the
Bacillus diphtherias is not increased by sewer air.

Anything like exhaustive researches into the effect of the septic
tank or cultivation beds upon pathogenic germs has not been under-
taken up to the present, and we can only conjecture as to their fate.
Dr Houston has made a cautious declaration upon this matter, and at
present we have not evidence to justify a more certain statement.
" The balance of evidence," he says, "points to the probability that
some, at all events, of the pathogenic organisms are crowded out in
the struggle for existence in a nutritive medium containing a mixed
bacterial flora, their vitality being weakened or destroyed by the
enzymes of the saprophytic species.''* * He further adds : " It must
be distinctly understood that I do not imply that such organisms as
the typhoid bacillus or the cholera vibrio would necessarily lose their
vitality, or even suffer a diminution in virulence under the conditions
prevailing in a biological filter. In the absence of actual experi-
ments, I am not prepared to say more than that I believe that if these
germs did gain access to the sewage they would suffer a diminution
in numbers primarily in the sewers [or septic tank], and secondarily
in the coke-beds [or cultivation beds]." Subsequently, as a result of
further experience of the effluent from the Crossness Sewage Works,
Houston wrote, " However satisfactory the process may be from the
chemical and practical point of view, the effluents from the bacterial
beds cannot reasonably be assumed to be more safe in their possible
relation to disease than raw sewage." f

Indirectly connected with this point, a word or two may be added
concerning some recent investigations made by Dr Houston upon the

* Bacterial Treatment of Crude Sewage, 1899 (Second Report), p. 19.
t Edin. Med. Jour., Feb. 1901.

176 BACTERIAL TREATMENT OF SEWAGE

deposit which accumulates on coke fragments used in the beds at
Barking and Crossness.* The coke was found to be coated with a
black-coloured slimy deposit, free from objectionable smell, and
almost odourless. On examination of the deposit, diluted with
sterile water, and making cultures, it was found that the number of
bacteria per gramme of the deposit was 1,800,000. This number,
large as it may seem, would weigh only a minute fraction of a
gramme,! so that it is evident that the number of living bacteria do
not in themselves account in any way for the deposit. As to the
nature of these organisms, Dr Houston adds : " The character of the
microbes appearing in the cultures differed somewhat from those
found in crude sewage. For example, there was an increase in the
number of spores of Bacillus entcritidis sporogenes (Klein), and a
decrease in the number of B. coli. Proteus-like germs were present
in abundance, many being of P. mirabilis type. Further, B.
arborcsccns and an allied form were present in considerable numbers.
An organism apparently identical with B. prodigiosm was also
isolated."^: The deposit also contained a number of bacilli with pre-
cisely similar staining properties as those of tubercle bacilli (acid-
fast). They were also morphologically indistinguishable from the
tubercle bacillus. In one instance Houston isolated a virulent
tubercle bacillus from a sewage effluent. Such facts are of evident
practical importance in relation to the final disposal of the effluent,
whether it is discharged into a stream used for drinking purposes or
otherwise.

Many of the researches having for their object the fate of
pathogenic organisms in sewage have been based upon the typhoid
bacillus as a type. Laws and Andrewes, in addition to demonstrat-
ing that this organism could only live in sewage a short time, showed
that one sewage bacillus (B. fluorescein stcrcorcdis) possessed the chief
powers of antagonism, and it is probable that the contained bacteria
rather than the chemical products of sewage act as unfavourable
conditions for the typhoid bacillus. Horrocks found that in sterilised
sewage the typhoid bacillus could live for sixty days.§ Houston has
thrown light upon the fate of B. typhosus by his work on the
occurrence of B. coli and B. entcritidis sjooivgcncs in effluents,! | and
Miss Chick has furnished evidence in respect of B. coli, tending in
the direction of showing that after sewage had passed over double
contact beds about 75 per cent, of the B. coli were removed, and

* Bacterial Treatment of Crude Sewage (Supplement to Second Report), 1890,
p. 4.

1 Dr Houston finds that 1,800,000 typhoid bacilli weigh only 0 '0000147
gramme.

% Ibid. , p. 4.

§ Jour, of Sanitary Institute, January 1900.

|| First, Second, and Third Reports to the London County Council {vide supra).

PLATE 17.

/

;

X^'Tmm ' '

Bacillus anthracis. Smear preparation from splenic juice of guinea-pig that died after inoculation
with 2 c.c. of Yeovil septic tank liquor, x 500.

Bacillus mitli racis.

mpression" preparation from a surface gelatine plate culture, 24 hours at 20 C.
Stained with methylene blue. x 1000.

[To face page 176.

RELATION TO DISEASE ORGANISMS 177

after land filtration practically an effluent might be free from B.
coll*' Pickard has recently shown that the typhoid bacillus vanishes
from crude sewage in about fourteen days. If the percentage of
original amount of typhoid bacillus introduced into the sewage for
experimental purposes be 100 ; in twenty-four hours it has fallen to
76 per cent., in thirty-two hours to 71 per cent., in forty-eight hours
to 60 per cent., in seven days to 8 per cent., and in fourteen days to
0*73 per cent. He also demonstrated that a large proportion (90
per cent.) of typhoid bacilli are actually destroyed in filter-beds such
as are used in the bacterial treatment of sewage, f

Houston likewise has found the B. anthracis in septic-tank
liquor and sludge, and in the secondary beds and general effluent. He
also found the anthrax bacillus in the mud of the banks of the river
Yeo at Yeovil within 150 feet of the main sewer. The spores of
anthrax are peculiarly resistant, and it is in this form that the bacillus
can pass through sewage unaffected J (Plate 17). The same author
has demonstrated that the B. pseudo-tuberculosis of Pfeiffer may be
present in the effluent of various sewage processes, and the same is
true of B. pyocyaneus which, however, occurs more rarely. Both
these organisms are highly pathogenic to lower animals, and are
also related to morbid processes occurring in the human subject. §
Houston, who has made extended inquiries on this question of the
effect of bacterial treatment of sewage on pathogenic organisms,
summarises his conclusions by stating that biological treatment on
land, or by artificial processes, does not necessarily remove patho-
genicity from the sewage effluent ; that the absence of pathogenic
result when sewage has been filtered shows that the products of
pathogenic bacteria in sewage are not of a markedly poisonous
nature ; and that the pathogenicity of sewage may depend on spores
rather than bacilli. || These conclusions must, however, be accepted
with reserve, and in a relative sense only, at present. Broadly, it
may be said that if sewage contains pathogenic bacteria, and is then
treated by bacterial methods, the effluent cannot be certainly
assumed to be safer in this respect than the raw sewage slightly
diluted ; or, expressed in other words, Houston's work indicates " the
inadvisability of relying on septic tanks, contact beds, or continuous
filters to remove altogether the element of potential danger to health
associated with the discharge of effluents from these processes of
sewage treatment into drinking-water streams." 1T

* Thompson- Yates Laboratory Report, vol. iii., part i., 1900.

t Jour, of State Medicine, 1903, pp. 203-210.

\ Royal Commission on Sewage IHsposal, Second Report, 1902, p. 39.

§ Ibid., p. 54. || Ibid., p. 58.

IT Ibid., Fourth Report, 1904, vol. iii., pp. 77-96.

M

CHAPTEE VII

BACTERIA IN MILK AND MILK PRODUCTS

General Principles — Sources of Pollution — Number of Bacteria in Milk — Influence
of Time and Temperature — Species of Bacteria found in Milk — Fermentations
of Milk — Pathogenic Organisms in Milk — Milk-borne Disease : Tuberculosis,
Typhoid Fever, Scarlet Fever, Sore-Throat Illnesses, Cholera, Epidemic
Diarrhoea — Preventive Measures — Protection of Milk Supply — Control of
Milk Supply : Refrigeration, Straining, Sterilisation, Pasteurisation — Special-
ised Milk — Bacteria in Milk Products— Cream-Ripening — Butter-Making —
Cheese-Making — Abnormal Cheese-Ripening — Poisonous Cheese.

Injurious micro-organisms in foods are, fortunately for the con-
sumers, usually killed by cooking. Vast numbers are, as far as we
know, of no harm whatever. Alarming reports of the large numbers
of bacteria which are contained in this or that food are generally as
irrelevant as they are incorrect. Bacteria, as we have seen, are
ubiquitous. In food we have abundance of the chief thing necessary
to their life and multiplication — favourable nutriment. Hence we
should expect to find in uncooked or stale food an ample supply of
saprophytic bacteria. There is much wholesome truth in the
assertion that good food as well as bad frequently contains large
numbers of bacteria, and often of the same species. It is well that
we should become familiarised with this idea, for its accuracy cannot
be doubted, and its acceptance at the present time may not be with-
out beneficial effect.

Nevertheless, it is well we should know the bacterial flora of
good and bad foods, for at least two reasons. First, there is no doubt
whatever that a considerable number of cases of poisoning can be
traced every year to food containing harmful bacteria or their products.
To several of the more illustrative cases we shall have occasion to
refer in passing. Secondly, we may approach the study of the
bacteriology of foods with some hope that therein light will be found

T78 '"

GENERAL PRINCIPLES 179

upon some important habits and effects of microbes. There can be
little doubt that food-bacteria afford an example of association and
antagonism of organisms to which reference has already been made.
Any information that can be gleaned to illumine these abstruse
questions would be very welcome at the present time. But there is
a still further, and possibly an equally important, point to bear in
mind, namely, the economic value of microbes in food. In a short
account like the present it will be impossible to enter into hypotheses
of pathology, but we shall at least be able to consider some of those
interesting experiments which have been conducted in the sphere
of beneficial bacteria.

The injurious effects of organisms contained in foods has been
elucidated by the excellent work of the late Dr Ballard. From the
careful study of a number of epidemics due to food poisoning, he
was able, without the aid of modern bacteriology, to arrive at a
simple principle which must not be forgotten. Food poisoning is
due either to bacteria themselves or to their products, which are
contained in the substance of the food. In cases of the first kind,
bacteria gaining entrance to the human alimentary canal set up their
specific changes and produce their toxins, and by so doing in course
of time bring about a diseased condition, with its consequent
symptoms. On the other hand, if the products, sometimes called
'ptomaines, are ingested as such, the symptoms set up by their action
in the body tissues appear earlier. From these facts Dr Ballard
deduced the simple principle that if there is no incubation period or,
at all events, a comparatively short space of time between eating
the poisoned food and the advent of disease, the agents of the disease
are products of bacteria. If, on the other hand, there is an incuba-
tion period, the agents are probably bacteria.

It is necessary to mention two other facts. Dr Cant lev has
isolated from poisoned foods some of the different species of bacteria
present.* It would appear that these are limited, as a rule, to two or
three kinds. As regards disease, the organisms of suppuration arc
the most common. Liquefying or fermentative bacteria are fre-
quently present, the Proteus family being well represented. In
addition there are, according to circumstances, a number of common
saprophytes. Now, as we have pointed out, these organisms may
act injuriously by some kind of co-operation, or they may by them-
selves be harmless, and pathological conditions be due to the
occasional introduction of pathogenic species.

The other fact requiring recognition from anyone who proposes
to study the bacteriology of milk, or indeed of other foods, is, that a
not inconsiderable amount of the evil results of food poisoning
depends upon the tissues of the individual ingesting the food.

* Report of Medical Officer to Local Government Board, 1895-96, Appendix.

180 BACTERIA IN MILK AND MILK PRODUCTS

There is ample evidence in support of the fact that not all the
persons partaking of infected milk suffer equally, and occasionally
some escape altogether. We know little or nothing of the causes of
such modification in the effect produced. It may be due to other
organisms, or chemical substances already in the alimentary canal of
the individual, or it may be due to some insusceptibility or resistance
of the tissues. Be that as it may, it is a matter which must not be
neglected in estimating the effects of food contaminated with bacteria
or their products.

There are few liquids in general use which contain such enor-
mous numbers of germs as milk. To begin with, milk is in every
way, physical and physiological, admirably adapted to be a favourable
medium for bacteria. It is constituted of all the chief elements of
the nutriment upon which bacteria live.

Briefly, we may summarise the full diet of bacteria as : — nitro-
genous matter (proteids) ; non-nitrogenous matter containing carbon
and hydrogen (carbohydrates); calcium, potassium, phosphates, etc.
(salts) ; and, for some species, oxygen. When we turn our atten-
tion to milk as a medium for bacteria, we find a complete bacterial diet
— proteids represented by casein and lactalbumin, 4 per cent, in
total — carbohydrates represented by lactose, the most readily affected
of all the sugars by bacteria; fat as palmitin and olein; salts,
potassium and calcium largely as phosphates, the calcium phosphate
being united with casein. Even the normal reaction of milk,
neutral or amphoteric, is favourable to the growth of bacteria, most
of which find a definitely acid or a definitely alkaline reaction
inimical to their growth. It is true that changes, mostly of a
fermentative nature, rapidly set in, which affect milk as a medium
for bacteria. But in its fresh, normal, untreated condition we have
theoretically an almost ideal medium for both saprophytic and
parasitic bacteria. Notwithstanding the truth of this general
statement, we must not pass over the experiments of Fokker,
Freudenreich, Cunningham, and others, which appear to demonstrate
that freshly-drawn milk possesses for certain species of bacteria a
germicidal power.

In the healthy condition of animals we have, generally speaking,
no micro-organisms whatever in their secretions, whatever may be the
condition of their excretions. Hence, though milk affords, from its
constitution, such an ideal nidus for the growth and multiplication of
bacteria, it is, as secreted, a perfectly sterile fluid. This was demon-
strated more than twenty years ago by Lister, who states that
"unboiled milk as coming from a healthy cow, really contains no
material capable of giving rise to any fermentative change, or to the
development of any kind of organism which we have the means of

SOURCES OF POLLUTION 181

discovering/' * Subsequent experiment has only confirmed the
general truth of this statement.*)" With efficient precautions it is
possible to draw from the udder of a healthy cow perfectly sterile
milk, which retains its sterility unchanged for long periods of time
in a sterilised and sealed flask. Yet we know by practical experi-
ence as well as by ultimate changes in the milk, that, generally
speaking, the presence therein of bacteria is very marked.

Sources of Pollution of Milk

These are various, and depend upon many minor circumstances
and conditions. For all practical purposes there are four chief
opportunities between the cow and the consumer when milk may
become contaminated with bacteria: —

1. At the time of milking and during manipulation at the farm.

2. During transit to the town, or dairy, or consumer.

3. At the milkshop.

4. In the home of the consumer.

Pollution at the time, of milking arises from the animal, the
milker, or unclean methods of milking. It is now well known that
in tuberculosis of the cow affecting the udder the milk itself shows
the presence of the bacillus of tubercle. In a precisely similar
manner all bacterial diseases of the cow which affect the milk-
secreting apparatus must inevitably add their quota of bacteria to
the milk. To this matter we shall have occasion to refer again.
There is a further contamination from the animal when it is kept
unclean, for it happens that the unclean coat of a cow will more
materially influence the number of micro-organisms in the milk than
the popularly supposed fermenting food which the animal may eat.
It is from this external source rather than from the diet that
organisms occur in the milk. The hairy coat offers many facilities
for harbouring dust and dirt. The mud and filth of every kind that
may be habitually seen on the hinder quarters of cattle all contribute
largely to polluted milk. Nor is this surprising. Such filth at or
near the temperature of the blood is an almost perfect environment
for many of the putrefactive bacteria.

The milker is also a source of risk. His hands, as well as the
clothes he is wearing, can and do readily convey both innocent and
pathogenic germs to the milk. Clothed in dust-laden garments, and
frequently characterised by dirty hands, the milker may easily act
as an excellent purveyor of germs. Not a few cases are also on
record where it appears that milkers have conveyed germs of

* Transactions of Pathological Society of London, 1878, p. 440.
t See also Rotch, Pediatrics, the Hygiene and Medical Treatment of Children.
London, 1896. J

182 BACTERIA IN MILK AND MILK PRODUCTS

disease from some case of infectious disease, such as scarlet fever,
in their homes. But under the more efficient registration of such
disease, which has recently characterised many dairy companies,
the danger of infection from this source has been reduced to a
minimum.

Professor Eussell recounts a simple experiment, which clearly
demonstrates these simple but effective sources of pollution : " A
cow that had been pastured in a meadow was taken for the
experiment, and the milking done out of doors, to eliminate as much
as possible the influence of the germs in the barn air. Without
any special precaution being taken, the cow was partially milked,
and during the operation a covered glass dish, containing a thin
layer of sterile gelatine, was exposed for sixty seconds underneath
the belly of the cow, in close proximity to the milk-pail. The udder,
flank, and legs of the cow were then thoroughly cleaned with water,
and all of the precautions referred to before were carried out, and
the milking then resumed. A second plate was then exposed in
the same place for an equal length of time, a control also being
exposed at the same time at a distance of ten feet from the animal
and six feet from the ground to ascertain the germ contents of the
surrounding air. From this experiment the following instructive
data were gathered. Where the animal was milked without any
special precautions being taken, there were 3250 bacterial germs per
minute deposited on an area equal to the exposed top of a ten-inch
milk-pail. Where the cow received the precautionary treatment as
suggested above, there were only 115 germs per minute deposited
on the same area. In the plate that was exposed to the surrounding
air at some distance from the cow there were 65 bacteria. This
indicates that a large number of organisms from the dry coat of
the animal can be kept out of milk if such simple precautions as
these are carried out." *

The influence of the byre air, and the cleanliness or otherwise of
the byre, is obviously great in this matter. As we have seen, moist
surfaces retain any bacteria lodged upon them ; but in a dry barn,
where molecular disturbance is the rule rather than the exception,
it is not surprising that the air is heavily laden with microbic life
derived from dust, dried manure, hay, straw, fodder, etc. Here again
many improvements have been made by sanitary cleanliness in
various well-known dairies. Still there is much more to be done in
this direction to ensure that the drawn milk is not polluted by a
microbe - imjDregnated atmosphere. Lastly, it should not be
forgotten that during the straining and cooling of milk there are
many opportunities of contamination.

The risks in transit differ according to many circumstances.

* H. L. Russell, Dairy Bacteriology \ p. 46.

SOURCES OF POLLUTION 183

Probably the commonest source of contamination is in the use of
unclean utensils and milk-cans. Any unnecessary delay in transit
affords increased opportunity for multiplication ; particularly is this
the case in the summer months, for at such times all the conditions
are favourable to an enormous increase of any extraneous germs
which may have gained admittance at the time of milking. Thus
we have (1) the milk itself affording an excellent medium and
supplying ideal pabulum for bacteria ; (2) a more or less lengthened
railway journey or period of transit giving ample time for multi-
plication ; (3) the favourable temperature of summer heat. We shall
refer again to the rate of multiplication of germs in milk. It has
been shown that milk brought into large cities, such as London,
Paris, or New York, has been travelling often for as long as two to
ten hours, often under conditions favourable to pollution or at least
under conditions of temperature favourable to the multiplication of
bacteria.

Pollution at the Milkshqp. — Many are the advantages given to
bacteria when milk has reached its commercial destination. In milk-
shops there are not a few risks to be added to the already imposing
category. Water is occasionally, if not frequently, added to milk to
increase its volume, either at the farm or the milkshop. Such water of
itself will make its own contribution to the flora of the milk, unless
indeed, which is unlikely, the water has been recently and thoroughly
boiled before addition to the milk. Again, it is impossible to suppose
that in small milkshops, perhaps of a general nature — where the
milk stands for several hours, pollution is avoidable. From a hundred
different sources such milk runs the risk of being polluted. The
dust of the shop and the street gain access to the pan of milk on
the counter, which, commonly, is uncovered. The " dipper " and the
vendors1 hands and clothes contribute bacteria. Flies also increase
the pollution.

Pollution in the Home. — Lastly, there is pollution from dust and
dirt, inorganic and organic, in the home. More than a million of
the population of London live in tenements of two rooms or less.
Cooking, eating, sleeping, cleaning, and sometimes even trade employ-
ments in the form of " home-work," are all conducted under conditions
of overcrowding and lack of space. Often there is no pantry or
larder, and consequently the days' supply of milk stands in a dirty
uncovered vessel in the midst of dirty surroundings. It is evident
that this is but one more opportunity for pollution.

Fore-Milk. — Before proceeding, a word must be said respecting
the first milk which flows from the udder in the process of milking,
and which is known as the fore-milk. This portion of the milk is
always rich in bacterial life, on account of the fact that it has
remained in the milk-ducts since the last milking. However thorough

184 BACTERIA IN MILK AND MILK PRODUCTS

the manipulation, there will always be a residue remaining in the
ducts, which will, and does, afford a suitable nidus and incubator for
organisms. The latter obtain their entrance through the imperfectly
closed teat of the udder, and pass readily into the milk-duct, some-
times even reaching the udder itself and setting up inflammation
{mastitis). Professor Eussell states that he has found 2800 germs
in the fore-milk, in a sample of which the average was only 330 per
c.c. Schultz found 83,000 micro-organisms per c.c. in the fore-milk,
and only 9000 in the mid-milk. As a matter of fact, most of this
large number belong to the lactic acid fermentation group, and the
fore-milk rarely contains more than two or three species, and still
more rarely any disease-producing bacteria. Still, bacteria occur in
such enormous numbers that their addition to the ordinary milk
very materially alters its quality. Bolley and Hall, of North Dakota,
report sixteen species of bacteria in the fore-milk, twelve of which
produced an acid reaction. Dr Veranus Moore, of the United States
Department of Agriculture,* concludes from a large mass of data
that freshly-drawn fore-milk contains a variable but generally
enormous number of bacteria, but only a few species, the last milk
containing, as compared with the fore-milk, very few micro-organisms.
The bacteria which become localised in the milk-ducts, and which are
necessarily carried into the milk, are for the greater part acid-
producing organisms, i.e., they ferment milk-sugar, forming acids.
They do not produce gas. Nevertheless their presence renders it
necessary to " pasteurise " as soon as possible. Dr Moore holds that
much of the intestinal trouble occurring in infants fed with ordinarily
"pasteurised" milk arises from acids produced by these bacteria
between the drawing of the milk and the pasteurisation. Prof.
MacFaclyean has given a full account of the ways in which milk
becomes pathogenic, and his views have received further support from
Prof. Delepine.f

The Number of Bacteria in Milk

From all that has been said respecting the sources of pollution
and the favourable nidus which milk affords for bacteria, it is not
surprising that a very large number of germs are almost always
present in milk. The quantitative estimation of milk appears more
alarming than the qualitative. It is true some diseases are conveyed
by bacteria in milk, but on the whole most of the species are non-
pathogenic. Nor need the numbers, though serious, too greatly
alarm us, for, as we shall see at a later stage, disease is due to other
agencies and conditions than merely the bacteria, which may be the

* Bureau of Animal Industry Reports, 1895-96.
t Jour. o/Comp. Path., 1897, vol. x., pp. 150-189.

TIME AND TEMPERATURE

185

vera causa. In addition to the fact that the high numbers have but
a limited significance, we must also remember that there is no
uniformity whatever in these numbers. The conditions which chiefly
control them are (1) time, and (2) temperature.

The Influence of Time and Temperature.— We have already
noticed, when considering the general conditions affecting bacteria,
how potent an agent in their growth is the surrounding temperature.
Generally speaking, temperature at or about blood-heat favours bac-
terial growth. Freuclenreich has drawn up the following table which
graphically sets forth the effect of temperature upon bacteria in milk: —

3 hours.

G hours.

9 hours.

24 hours.

59° F.
77° F.
95° F.

1 +

2

4

2-5
18-5
1290

5

107

3800

163

62,100

5,370

It will be noticed that at 59° F. there is very little multiplication.
That may be accepted as a rule. At 77° F. the multiplication, though
not particularly rapid at the outset, results finally, at the end of
the twenty-four hours, in the maximum quantity. These were
probably common species of saprophytic bacteria, which increase
readily at a comparatively low temperature. During the subsequent
hours, after the twenty-four, we should expect a temporary decline
rather than an increase in 62,000, owing to the keen competition con-
sequent upon the limitation of the pabulum. From a consideration
of these facts, we conclude that a warm temperature, somewhat below
blood-heat, is most favourable to multiplication of bacteria in milk ;
that the common saprophytic organisms multiply the most rapidly;
that, in the course of time, competition kills off a large number.

Another example may be taken from Professor Conn : —

Number of Bacteria per cubic centimetre in milk kept al different

temperatures.

No. at
outset.

In 12 hrs.

at 50°.

In 12 hrs.
at 70°.

In 50 hrs.

at 50°.

In 50 hrs., or at

time of curdling,

at 70°.

No. of hrs.

to curdling

at 50°.

No. of hrs.

to curdling

at 70°.

46,000
47,000
50,000

39,000
44,800
35,000

249,500
360,000
800,000

1,500,000
127,500
160,000

542,000,000

792,000,000
36 hours
2,560,000,000
42 hours

190

289
172

56
36
42

186 BACTERIA IN MILK AND MILK PRODUCTS

So strongly convinced is Conn of the exceptional influence of
temperature on the increase of bacteria in milk, and the subsequent
souring, that he holds that " the keeping of milk is more a matter of
temperature than of cleanliness." The cooling of milk immediately
after milking, and keeping it at a low temperature, will do more for
its preservation than any other practical device. Conn has also
pointed out that lactic organisms flourish in milk when it is kept at
tenrperatures above 50° C. He summarises the influence of tempera-
ture as follows : —

(1) Variations in temperature have a surprising influence upon
the rate of multiplication of bacteria. At 50° F. these organisms may
multiply only five-fold in twenty-four hours, while at 70° they may
multiply seven hundred and fifty-fold. (2) Temperature has a great
influence upon the keeping property of milk. Milk kept at 95° (heat
of the cow's body) will curdle in eighteen hours, while the same milk
kept at 70° will not curdle for forty-eight hours, and if kept at 50°,
the temperature of an ice-chest, may sometimes keep without curdling
for two weeks or more. (3) So far as the keeping property of milk
is concerned, the matter of temperature is of more significance than
the original contamination of the milk with bacteria. (4) Milk pre-
served at 50° or lower will keep sweet for a long time, but it becomes
filled with bacteria of a more unwholesome type than those that grow
at higher temperatures.*

The influence of time is not less marked than that of temperature,
as the following table will show : —

Milk drawn at 59° F. = 153,000 m.o. per cub. inch.

After 1 hour = 616,000

2 = 539,000

„ 4 „. = 680,000

„ 7 „ = 1,020,000

9 = 2,040,000

„ 24 „ = 85,000,000

Freudenreich gives another example, as follows : —

Milk drawn at 15 '5° C. = 27,000 m.o. per c.c.

After 4 hours = 34,000

9 „ = 100,000

„ 24 „ = 4,000,000

Concerning these figures little comment is necessary. But here
again, also, we may remember that this rapid multiplication only
continues up to a certain point, after which there is a marked reduc-
tion owing to products of activity.

Quite recently further investigations have been made in milk
maintained at a standard temperature by various workers. For the

* Storr's Agric. E.vpt. Sta. Conn. Ball. 26 (H. W. Conn).

TIME AND TEMPERATURE

is?

sake of comparison with other statistics, we may take two series
recorded by Park. In the first the temperature was 90° F., a tem-
perature common in New York in hot summer weather, and the
samples of milk were of three degrees of quality, namely, fresh and
good, fair, and bad. The result was as follows : —

Number of Bacteria per 1 c.c.

Good Fresh Milk.

Fair Milk from

Store.

Bad quality from
Store.

Original number of Bacteria .
After 2 hours

„ 4 „ .

„ 6 „ . . . .

„ 8 „ .

5,200

8,400

12,400

68,500

654,000

92,000

184,000

470,000

1,260,000

6,800,000

2,600,000

4,220,000

19,000,000

39,000,000

124,000,000

The second series of Park was milk taken from cows in common
dirty stalls, twenty-four, thirty-six, and forty-eight hours after
milking. The milk was cooled to 52° R, three hours after milking,
and maintained at that temperature, for the forty-eight hours of the
experiment. The result, therefore, shows the effect of time even
more exactly than the first series : —

Average Number of Bacteria per 1 c.c. of Milk at 52" F. (six samples).

After 3 hours.

30,366

After 24 hours.

69,433

After 3G hours.

348,883

After -IS hours.

1,668,:;:;:;

* The figures at 30 hours were estimated from the test of one sample only.

Even a cursory examination of these figures with those already
given will have shown how intimately the two influences of time and
temperature act and interact in relation to the multiplication of
micro-organisms in milk. They are scarcely separable, and no hard-
and-fast line can be drawn by way of comparison of these two
influences.

Eeference may also be made to two investigations made, one by
Park of New York, and the other carried out by Swithinbank and
the writer.

The following figures obtained by Park show the development
of bacteria in two samples of milk maintained at different tempera-
tures for twenty-four, forty-eight, and ninety-six hours respectively.
The first sample was obtained under the best conditions possible, the
second in the usual way (the iigures of this sample are underlined).

188

BACTERIA IN MILK AND MILK PRODUCTS

When received, Specimen No. 1 contained 3000 bacteria per c.c., and
Specimen No. 2, 30,000 per c.c.

Temperature.

Time which elapsed before making the test.

24 hours.

48 hours.

96 hours.

168 hours.

32° F. (0°C.) .
39° F. (4°C.) .
42° F. (5-5° C).
46° F. (6° C.) .
50° F. (10° C.) .
55° F. (13° C.) .
60° F. (16° C.) .
68° F. (20° C.) .
86° F. (30° C.) .
94° F. (35° C.) .

2,400
30,000

2,500
38,000

2,600
43,000

3,100
42,000

11,600
89,000

18,800
187,000

] 80,000
900,000

450,000
4,000,000

1,400,000,000
14,000,000,000

25,000,000,000
25,000,000,000

2,100

27,000

3,600
56,000

3,600
210,000

12,000
360,000

540,000
1,940,000

3,400,000
38,000,000

28,000,000
168,000,000

25,000,000,000
25,000,000,000

1,850
24,000

218,000
4,300,000

1,400
19,000

4.200,000
38,000,000

500,000
5,760,000

1,480,000
12,200,000

There are two points in this table which may be noted. First, it
may be seen that at 32° F. (0° C.) there is a decline in the number of
organisms both in good and bad milk during the first 168 hours. At
all the other temperatures, to which there is no exception, there is a
rise in the number of organisms. Secondly, the numbers of bacteria
at 20° C. in forty-eight hours are equal to the numbers at 35° C. in
twenty-four hours, and in both instances the number is phenomenally
high.

In 1900, Mr Swithinbank and the writer conducted a series of
experiments as part of an inquiry into the behaviour of bacteria in
milk, during which careful observation was made of a certain milk
from the time it was drawn from the udder up to thirty days, and then
subsequently after two years. Further, the observations were made
at three different temperatures. Broadly speaking, the conclusions
were as follow : —

First, there was an extremely rapid increase in the number of
organisms in the first four hours, particularly at 37° C. At the

TIME AND TEMPERATURE 189

commencement the milk contained 812,000 bacteria per c.c. After
four hours it contained 2,066,000 (at 5° C), 3,650,000 (at 15° C),
and 6,116,000 (at 37° 0.).

Secondly, speaking in a general way, the following great principle
became evident, namely, that there is at each temperature (a) a
sudden rise, (&) a sudden fall, (c) a steady rise to maximum, and (d)
a steady fall ultimately to sterility. In other words, there are tides
of organisms, and this was found to occur invariably in our study
of " natural " milks. It is a variable phenomenon in ordinary milks,
but is the rule in respect to " natural " milk examined immediately
after milking. It is obvious that if we had commenced our examina-
tion, as is frequently the case in the study of town milks, twelve
or twenty hours after milking, we should, even if we had obtained
the same figures, have drawn very different deductions, because the
initial rise and initial fall would have been lost sight of. The
sudden fall occurred in forty-eight hours at 5° C, in twelve hours
at 15° C. and 37° C.

Thirdly, the maximum number of bacteria occurred in ten days
at 5° C. (406,400,000 bacteria per c.c), in six days at 15° C.
(84,000,000), and in seventy-two hours at 37° C. (8,360,000). The
maximum was lowest at blood-heat and highest at 5° C. It is
evident, therefore, that what occurs in a short time at a high
temperature occurs in a longer period at a low temperature, but at
a low temperature the bacteria eventually become most numerous.
These facts are of great importance in relation to the time which
milk is kept before use, and to the injurious properties which it
may acquire during such a period in the direction of increased
bacterial toxin production.

Fourthly, marked acidity commenced between the twelfth and
sixteenth hours in the sample at 37° ; between the twentieth and
twenty-fourth hours at 15°; and between the seventy-second and
ninety-sixth hours at 5° C. At the end of these particular stages
it will be noticed that there is a rising tide following the " low-water
mark " of organisms at each temperature. The relation which the
degree of acidity bears to the bacterial content is an intimate one.
As far back as 1878 Lister pointed out the marked inhibitory effect
which the presence of a high degree of lactic acid had upon common
moulds and ordinary saprophytic bacteria.* When the lactic acid
declines, these other forms commence growth, and eventually
enormously preponderate.

Fifthly, as the flasks of milk were kept intact, we were able to
repeat the experiment in every particular after the lapse of exactly
two years from the commencement. The milk was the same milk,

* Path. Soe. Trans., 1878, p. 440.

190 BACTERIA IN MILK AND MILK PRODUCTS

and the experiment was repeated as at thirty days. During the
intervening period the flasks had been kept, hermetically sealed, at
the three temperatures. The result was that the flasks were found
to be germ-free, with the exception of an abundant growth of Oidium
lactis and other moulds.

Mr S withinl )ank and the writer came to the conclusion that the
explanation of the results of this investigation could only be found
in a glance at the life-history of such a milk as that under
consideration.

" At the time of milking there is, as we have seen, an introduction into the warm
milk of vast numbers of common saprophytic and parasitic bacteria. Finding
themselves in an ideal nidus, they multiply with almost incredible rapidity. Hence
the first rise in numbers of bacteria. Competition and exhaustion of pabulum
soon produce inevitable effects, and we obtain the first decline. At this stage it
may be said that the common extraneous bacteria, whether putrefactive or simple
saprophytes, practically die out, and that for a very simple reason, namely, that
they cannot live in the presence of the new tide of acid-forming bacteria. Although
the lactic acid group of organisms do not multiply as rapidly as ordinary sapro-
phytes, they reach a much higher maximum in the end. It is to this family of
bacteria that the second and maximum rise is due. In time, also, the same
inimical conditions begin to act, and the lactic acid bacteria decline owing to the
acidity and to the lack of pabulum. Eventually, the medium, which twenty days
before was an ideal one for any organism, and mostly so for those which came
first, and which ten days before was favourable to lactic acid organisms, is now
favourable to no bacteria at all. Accordingly, bacteria of all descriptions gradually
die out, and the medium is eventually left in possession of Oidium lactis and the
common moulds. That the destruction of large quantities of solid albuminous
substances may occur simply through bacterial agencies has been conclusively
shown in the so-called septic tank method of sewage disposal. The death of
bacteria under these circumstances always follows shortly after their enormous
multiplication, and how much is due to starvation or how much to poisoning by the
products of their own activity it is impossible to say. It is, however, clear that the
decomposition of large quantities of albuminous substance is first accompanied by
great bacterial reproduction, and this is invariably followed by a season of speedy
and extreme mortality of bacteria. In a general way that represents, we believe,
the changes taking place as represented in the record we have considered. That
there are two rises and two falls in the number of bacteria, the first rise being due
to extraneous organisms, and the second rise to lactic acid organisms, we believe to
be the almost universal rule in untreated ' natural ' milk. " ;:

The effect of temperature and time has been illustrated by Dr
Buchanan Young's researches into the numbers of bacteria in milk
according to season, and the results of which were laid before the Eoyal
Society of Edinburgh. He estimated in the Edinburgh milk supply that
three hours after milking there were 24,000 micro-organisms per c.c.
in winter ; 44,000 in spring ; 173,000 in late summer and autumn.
Again, he found that five hours after milking there were 41,000
micro-organisms per c.c. in country milk, and more than 350,000
per c.c. in town milk. Many London milks would

Bacteriology of Milk (Swithinbank and Newman), 1903, p. 135.

TIME AND TEMPERATURE 191

exceed 500,000 per c.c.* In summer the writer has found as many as
4,000,000 organisms per c.c. in fresh London milk obtained at first-
rate milkshops.f

There is no standard or uniformity in the numerical estimation
of bacteria in milk. A host of observers have recorded widely-
varying returns due to the widely-varying circumstances under
which the milk has been collected, removed, stored, and examined,
and due also to the two dominating influences of time and tempera-
ture. Nor is it possible to establish any standard which may be
accepted as a normal or healthy number of bacteria, as is done in
water examination. Bitter has suggested 50,000 micro-organisms
per c.c. as a maximum limit for milk intended for human consump-
tion, but actual experience shows that, at present at all events, such
standards are impracticable.

Owing to differences of nomenclature and classification, in addition
to differences in mode of examination, at present existing in various
countries, it is impossible to state even approximately how many
bacteria, and how many species of bacteria, have been isolated from
milk. Until some common international standard is established,
mathematical computations are practically worthless. They are need-
lessly alarming and sensational. And it should be remembered that
great reliance cannot be placed upon these numerical estimations, for
they vary from day to day, and even hour to hour. Furthermore, vast
numbers of bacteria are economic in the best sense of the term, and
the bacteria of milk are chiefly those of a fermentative kind, and
not disease-producers. \

The effects of time and temperature upon the bacteria of milk do
not only concern the numbers of organisms present. As long ago as
1897 Delepine showed that toxicity of milk was increased by a rise
in temperature, and this is as we should expect, for it stands to
reason that conditions favourable to the multiplication of bacteria in
milk must of necessity tend to increase the products of bacteria in
milk, and is likely to increase its virulence. §

The following tables summarise the more detailed figures given
in 1897, and in which the effects which length of keeping and
of temperature have upon the noxious effects of the milk arc
indicated.

* Brit. Med. Jour., 1895, ii., p. 322.

t Report on Milk Supply of Finsbury, 1903.

X At the same time it is important to remember that comparative series of
estimations as to the number of bacteria per c.c. in milk may be of value as
indication of unclean dairying. Leigh ton of Montclair, U.S.A., has shown that
numbers increase in direct proportion to unclean management. See The Milk
Supply of Two Hundred Cities and Toivns (Alvord & Pearson), U.S. Dep. of
Agriculture, Bull. 46, 1903, p. 117.

§ Jour. ofComp. Path, and Therapeutics, 1897.

192

BACTERIA IN MILK AND MILK PRODUCTS

Delepine showed that mixed milks coming from a distance of over
40 miles, and generally kept for from twenty-four to sixty hours, and
even more in a few cases (tuberculous samples excluded), gave the
following returns : —

Mean Temperature in the

Shade (Manchester)

during Time the Specimens

were kept.

Specimens

producing

no Noxious

Effects.

Noxious
Specimens.

Totals.

Percentage

of Good
Specimens.

Deg. Fahr.
30 to 35
35 to 40
40 to 45
45 to 50
50 to 55
55 to 60

7

7
2
1

6

5

11

3

4

2

12

18

5

5

2

58-0
38*5
40'0
20-0

o;o

17

25

42

39-0

Mixed milks coining from a short distance (generally under 20
miles), most of them kept for less than ten hours (with the excep-
tion of five out of the seven bad specimens, and four out of the
twenty-two good specimens, which had been kept somewhat longer),
(tuberculous samples excluded), gave the following results : —

Mean Temperature in the

Shade (Manchester)

during Time the Specimens

were kept.

Specimens
producing

no Noxious
Effects.

Noxious
Specimens.

Totals.

Percentage

of Good

Specimens.

Deg. Fahr.
50 to 55
55 to 60
60 to 65
65 to 70
70 to 75

1

8
11

2

0
1
4

"i

1

9

15

4

lOO'O
88-8
73-2

50-0

22

7

29

75-68

Whilst unmixed, milks kept for various lengths of time, but
collected from the udder in sterilised vessels (tuberculous samples
excluded), resulted as follows : —

TIME AND TEMPERATURE

193

Mean Temperature in the

Shade (Manchester)

during Time the Specimens

were kept.

Specimens

producing

no Noxious

Effects.

Noxious

Specimens.

Totals.

Percentage

of Good

Specimens.

Deg. Fahr.
35 to 40
40 to 45
45 to 50
50 to 55
55 to 60
60 to 65

6
3

5

0

0
2
2

3

6
5

7

*8

lOO'O
60-0
71-5

o:o

14

7

21

67-2

"The influence of time," Prof. Delepine adds, "is well shown by
the number of specimens remaining good, even at a high temperature,
when the milk had been kept only half a day. On the other hand,
the influence of temperature is still more evident, for in every
category the number of good specimens is almost inversely propor-
tional to the height of the temperature. Still, it is important to keep
the two factors of time and temperature in mind. What is produced
in a few hours in summer may also occur in winter, when the milk has
been kept a long time"

The converse is also true, namely, that if the temperature of milk
be reduced by refrigeration, the toxicity of the milk is lessened.
Professor Delepine has shown that the mortality from all causes in
guinea-pigs inoculated with refrigerated milk is considerably less
than it is if unrefrigerated milk be inoculated : —

Unrefrigerated Milk Examined during the years 1896 and 1897.

Number
of

Samples.

Number of Samples
causing the Death
of Two Animals in-
oculated in less than
Ten days.

Number of Samples
causing the Death
of One of the In-
oculated Animals in
less Uian Three Days.

Total.

1896-97

148

Per cent.
5 . .3-3

Per cent.

11 . . 7-1

Per cent.

10-7

Refrigerated Milk Examined from 1898 to 1901.

1898
1899
1900
1901

111
175

802
694

0 . .0

1 . . 0-57
4 . . 0-50
1 . . 0-14

3 . . 2-7

1 . . 0-57
25 . .3-1
8 . .1-1

2-7
1-14
3*60
1-24

N

194 BACTERIA IN MILK AND MILK PRODUCTS

" The inference to be drawn from these gross results is clear : a
certain proportion of the samples of milk contained bacteria which,
under favourable circumstances, gave to the milk noxious properties,
the development of which could be checked in many cases by pre-
venting the growth of these bacteria. The difference between
refrigerated and non-refrigerated milk would have been very much
greater, if the milk had invariably been cooled immediately after the
milking of the cows " (Delcpine).

Therefore, it may be said that to refrigerate milk immediately
after drawing it from the cow is to reduce the number of bacteria and
to diminish the potential toxicity of the milk. Finally, Professor
Dele'pine writes : —

" When the clear relation existing between time of keeping, plus
temperature and the noxious properties of a certain number of
samples of milk, is contrasted with the ambiguous results obtained
when an attempt is made to connect these noxious properties with
disease of the udder (tuberculosis being excluded), it is difficult not
to feel convinced that infection of the milk outside the udder, and
the conditions under which milk is kept, are the most important
factors causing it to acquire infective properties."*

Species of Bacteria found in Milk

The kinds of bacteria occurring in milk may for purposes of con-
venience be classified in the following four divisions ; though of the
first two groups it is not necessary to say much here : —

1. Ordinary bacteria of air, soil, or water.

2. Bacteria of sewage or intestinal origin.

3. Bacteria concerned in fermentation.

4. Pathogenic bacteria, in particular those associated with tuber-
culosis, enteric fever, cholera, scarlet fever, diphtheria, sore-throat
illnesses, and epidemic diarrhoea.

1. Ordinary bacteria of soil, air, or water readily gain access to
milk from their natural media. It is unnecessary to consider them
here.

2. Bacteria of sewage and intestinal origin occur from time to time
in milk. The two chief representatives are B. coli and B. enteritidis
sporogencs. In Liverpool, from 1900-1902, 788 " country " milks were
examined, and 55 per cent, contained B. coli and 9 per cent, contained
B. enteritidis sporogencs ; of 722 " town " milks, 23 per cent, contained
the former bacillus, and 4 per cent, the latter. -f" Chick found B. coli
present in 17 out of 239 new milks, and Balfour Stewart found B.
enteritidis sporogenes in 49 samples out of 213. When it is considered

* Jour, of Hygiene, 1903, pp. 80-84.
f Bacteriology of Milk, p. 216.

(1) Water

87*5 per cent

(2) Milk-sugar

4'9

(3) Fat

3'6

(4) Proteids (casein, etc.)

3-3

(5) Mineral matter

07

COMPOSITION OF MILK 195

how filthily many cows are kept, it is not to be wondered that many
intestinal organisms find their way to milk.

3. Bacteria concerned in fermentations in milk cannot well be
understood without some appreciation of the different elements of
milk which are most affected by the changes of fermentation. It is
therefore necessary, before proceeding, to consider shortly what are
the constituents of milk upon which living ferments of various kinds
exert their action, for without these facts the action of fermentation
bacteria is not evident. A tabulation of the chief constituents of
milk may be stated as follows : —

Ordinary-
fresh milk =
100 per cent.

^ {p) Mineral matter

100-0

Or the average milk constitution may be expressed thus : —

Fat . . . . . .4*1 per cent.

Solids not fat . . . . . 8'8 ,,

Total solids .... 12*9
Water 87*1

It is not necessary to remark that milks vary in standard, and the above figures
can only be taken as fair averages.

Milk-sugar, or Lactose, (C^H^O,.,), is an important and constant constituent of
milk. It forms the chief substance in solution in whey or serum, and is a member
of the cane-sugar group. Milk-sugar is found in varying quantities in the milk of
mammals. About 5 per cent, is present in human milk, and somewhat less in that
of the cow. It is very resistant to fermentation by yeast, and therefore undergoes
alcoholic fermentation very slowly. It is not acted upon by rennet, pepsin, or
trypsin. But of all the sugars it is most readily acted upon by micro-organisms.

Fat occurs in milk as suspended globules of varying size. It forms the cream,
and by churning is, of course, made into butter, though both cream and butter con-
tain other constituents besides fat. Lloyd has shown that it is the large globules
that form the cream, and he has also made observations- upon the size of fat globules
in relation to breed of cattle. The decomposition and breaking down of milk-fat by
fermentation is the chief cause of gross abnormalities of cream and the rancidity of
butter.

The Proteids of Milk include casein, lactalbumin, and lactoglobulin. Casein is
by far the most abundant and the most important. When milk separates naturally
into its constituent parts the fat rises and the casein falls, leaving a clear fluid, the
milk plasma or serum, between the two substances. The changes set up in casein
by bacteria are various, and furnish a means of diagnosis.

Mineral Matter.— -The ash of milk, obtained by careful ignition of the solids,
contains calcium, magnesium, potassium, sodium, phosphoric acid, sulphuric
acid, chlorine, and iron— phosphoric acid and lime being present in the largest
amounts.

We may now consider the fermentations of milk and the pathogenic
organisms associated with milk.

196 BACTERIA IN MILK AND MILK PRODUCTS

1. The Fermentations of Milk

(1) Lactic Acid Fermentation: the Souring of Milk. — If milk is
left undisturbed, it is well known that eventually it becomes sour.
The casein is coagulated, and falls to the bottom of the vessel ; the
whey or serum rises, carrying to the surface flakes or lumps of fat.
In fact, a coagulation analogous to the clotting of blood has taken
place. In addition to this, the whole has acquired an acid taste.
Now this double change is not due to any one of the constituents we
have named above. It is, in short, a fermentation set up by a living
ferment introduced from without. The constituents most affected
by the fermentation are (a) the milk-sugar, which is broken down
into lactic acid, carbonic acid gas, and other products, and (b) the
casein, which is curdled and becomes suspended in a semi-colloidal
form.

For many years it has been known that sour milk contained
bacteria. Pasteur iirst described the Bacillus acidi lactici, which
Lister isolated in 1877, and obtained in pure culture by the dilution
method. In 1884 Hueppe contributed still further to what was
known of this bacillus, and pointed out that there were a large
number of varieties, rather than one species, to be included under
the term B. acidi lactici. We have already dealt with the chief
characters of this family of organisms. When a certain quantity of
lactic acid has been formed, the fermentation ceases. It will
recommence if the liquid be neutralised with carbonate of lime, or
if pepsine be added. Since Pasteur's discovery of a causal bacillus
for this fermentation, other investigators have added a number of
bacteria to the lactic acid family. Some of these in pure culture
have been used in dairy industry, as we shall subsequently have
occasion to notice.

We have already seen that milk as it leaves the healthy udder
is generally sterile, and immediately gains bacteria from air, dust,
etc. Whilst the exact origin of lactic acid bacilli is not known,
many bacteriologists hold that they gain entrance to the milk from
the surrounding air of byre or dairy. Others maintain that some
species, at any rate, are soil bacteria, and associated with certain
geographical localities. Eussell states that, under ordinary con-
ditions, the organisms found in the teat of the udder are those which
produce lactic fermentation. He quotes Bolley and Hall as finding
twelve out of sixteen species in the teat of the udder to be lactic acid
producers.* Veranus Moore has arrived at very similar results.^
Eollin Burr has recently investigated this subject with a different

* Outlines of Dairy Bacteriology, H. L. Russell, 1898, p. 43.
t Twelfth and Thirteenth Reports of the Bureau of Animal Industry, U.S.A., 1895
and 1896, p. 265.

LACTIC ACID FERMENTATION 197

result.*" He finds that when milk is drawn from the cow in such a
manner as to exclude from it dirt and dust from the air, the stalls, and
the cow, such milk may contain none of the organisms capable of
producing a normal souring of milk. This also has been the
experience of the writer. The lactic acid organisms are a secondary
contamination of the milk from some external source. None of the
species of lactic organisms characteristic of the locality in which
Burr worked, could be found in the udder. This is in accordance
with the results of others who have had the opportunity of examining
the udder or milk ducts for lactic bacilli. Out of 300 examinations
made of fore-milks drawn directly from the udder into sterile flasks,
Burr found only 2 per cent, contained ordinary lactic acid bacteria,
and in these cases the origin was probably outside contamination.
Conn found the acid organisms present in 5 cases out of 200
examinations, involving 75 cows. He also maintains that the origin
of lactic acid bacteria is in external conditions.*)- Further, there is
the recognised fact which has been pointed out by Conn and Esten,
and frequently met with by Swithinbank and the writer, namely,
that lactic acid organisms are not the predominant species in freshly-
drawn milk, as they undoubtedly would be were they organisms
of the udder. Hence there can, we think, be little doubt that the
origin of lactic acid organisms is to be found in some external
condition or conditions.

It follows from what has been said that cleanliness of byre, dairy,
and general manipulation is an important factor in the presence,
both actual and in degree, of lactic acid organisms.

(2) Butyric Acid Fermentation. — This form of fermentation is
also one which we have previously considered. Botli in lactic and
butyric fermentation we must recognise that in the decomposition of
milk-sugar there are almost always a number of minor products
occurring. Some of the chief of these are gases. Hydrogen, carbonic
acid, nitrogen, and methane occur, and cause a characteristic effect
which is frequently deleterious to the flavour of the milk and its
products. Most of the gas-producing ferments are members of the
lactic acid group, and are sometimes classified in a group by them-
selves. In butyric fermentation of milk the three chief products
are butyric acid (which causes the bitterness), hydrogen, and carbonic
acid gas.

(3) Coagulation Fermentations without Acid Production. — Of these
there are several, caused by different bacteria. What happens is
that the milk coagulates, but no acid is produced, the whey being-
sweet to the taste rather than otherwise. The condition is in the

* Store's Agricultural Expt. Sta. Rep. for 1900, pp. 66-81. Centralb. f. Bakt.,
Abth. ii., 1902," p. 236.

y Store's Agricultural Expt. Sta. Rep., 1899, p. 23.

198 BACTERIA IN MILK AND MILK PRODUCTS

main one of milk-clotting rather than milk-curdling. The two chief
examples are the rennet fermentation of milk and the production of
casease.

(4) The Alcoholic Fermentations of Milk — Lactose is not readily
acted upon by yeasts though they have the power of breaking it up
and producing alcohol and carbonic acid gas. When it does occur the
percentage of alcohol is very small. The first change is the inversion
of the milk-sugar into dextrose and galactose, and the second is
fermentation of these sugars.

Occasionally, alcohol is present in the milk of a dairy, as a sort of
by-product accompanying lactic fermentation, and alcoholic fermenta-
tion may, under exceptional circumstances, cause serious trouble to
the dairyman. But the chief illustrations of this fermentation in
milk are the well-known examples of the artificial beverages known
as koumiss (or kumiss, kumys) and kephir (or kefyr, kefr), the former
a fermentation of mare's milk, the latter of cow's milk. Matzoon
and Leben are two other examples of similar changes.

Koumiss is made on the Steppes of South-Western Siberia and
European Eussia, by nomadic Tartars. It is not a simple process
nor a single fermentation. There is first a lactic fermentation
producing lactic acid, and secondly, a vinous fermentation result-
ing in alcohol. The former is produced by bacteria, the latter by
yeasts. In neither case is the process set up by a pure culture.
"The net change which has taken place in the original milk may
be summed up by saying that the sugar has been to a large extent
replaced by lactic acid, alcohol, and carbonic acid gas ; the casein has
been partly precipitated in a state of very fine division, and partly
predigested and dissolved, while the fat and salts have been left
much as they were." * The total proteid in koumiss is hardly less
than in mare's and cow's milk ; the fat is practically the same as in
mare's milk, and the sugar is reduced from about 5 per cent, to 1/5
per cent. The amount of alcohol in koumiss is as little as 17 per
cent., and there is not as much as 1 per cent, of lactic acid.

Kephir, the second example named, is an effervescent alcoholic
sour milk prepared by inhabitants of the Caucasus from the milk of
goats, sheep, and cows. The process of fermentation is a double
one, and precisely parallel to that occurring in the production of
koumiss. Its method of manufacture is simply to add to milk a few
" kephir grains," allow the milk to stand for. twenty-four hours at a
temperature of 17° to 19° C, pour off the milk and mix with fresh
volumes, and so on. Fermentation is complete in two or three days'
time, and the resultant fluid contains about 2 per cent, of alcohol,
being slightly more than in koumiss.

* Food and the Principles of Dietetics, by R. Hutchison, M.D., F.R.C.P., 1902,
p. 136.

ANOMALOUS FERMENTATIONS 199

(5) Anomalous Fermentations of Milk. — There are a number of
changes, mostly clue to fermentation, which occur in milk, and to
which reference must be made. These conditions have been termed
" diseases " of milk, but it is not altogether a satisfactory term.

(a) Bitter Fermentation. — Some bitter conditions of milk are due
to irregularity of diet in the cow. Similar changes occur in con-
junction with some of the acid fermentations and proteid decomposi-
tions. Weigmann and Conn have, however, shown that there is a
specific bitterness in milk due to bacteria, which appear to produce
no other change. Hueppe suggests that it may be due in part
to a proteid decomposition resulting in bitter peptones. Such
bodies are produced by bacteria from the albuminoids of milk,
and hence the bitterness does not appear immediately after milk-
ing, but only after an incubation period. Some nine or ten different
micro-organisms have been credited with this power, and such
organisms may infect a farm, a byre, or a dairy, for months or even
years, contaminating the milk. In all probability, most outbreaks
of this bitter fermentation are due to Weigmann's bacillus of bitter
milk or Conn's micrococcus. There seems to be evidence for sup-
posing that some of the " bitter " bacilli produce very resistant
spores, which make them resistant against conditions in the milk
itself or externally.

(b) Slimy Fermentation. — This graphic but inelegant term is
used to denote an increased viscosity in milk, and its tendency when
being poured to become ropy and fall in strings. Such a condition
deprives the milk of its use in the making of certain cheeses, whilst
in other cases it favours the process. In Holland, for example, in
the manufacture of Edam cheese, this " slimy " fermentation is
desired. Tcettemoelk, a popular beverage in Norway, is made from
milk that has been infected with the loaves of the common butter-
wort, Pinguicula vulgaris, from which -Weigmann separated a bacillus
possessing the power of setting up slimy fermentation. There are,
perhaps, as many as a dozen species of bacteria which have in a
greater or less degree the power of setting up this kind of fermenta-
tion. In 1882, Schmidt-Miihlheim isolated the Mierococcus viscosus,
which occurs in chains and rosaries, affecting the milk-sugar. It
grows at blood-heat, and is not easily destroyed by cold. Its effect
on various sugars is the same. M. Frcttdenreiehii, one of the specific
micro-organisms of "ropiness" in milk, is a large, non-motile, lique-
fying coccus, which can produce its result in milk within five hours.
On account of its resistance to drying, it is difficult to eradicate
when once it makes its appearance in a dairy. The organism used
in making Edam cheese is the Streptococeus Hollandictcs, and in hot
milk it can produce ropiness in one day. A number of bacilli have
been detected by several observers, and classified as slime fermenta-

200 BACTERIA IN MILK AND MILK PRODUCTS

tion bacteria. The Bacillus ladis pituitosi, a slightly curved, non-
liquefying rod, which is said to produce a characteristic odour, in
addition to causing ropiness, brings about some acidity. B. ladis
viscosus of Adametz, B. actinobacter of Duclaux, B. Hessii of Guille-
beau, and other bacilli are similar agents. Many of the above
organisms, with others, produce " slimy" fermentation in alcoholic
beverages as well as in milk.

(c) Soapy Milk is another form of fermentation, the etiology of
which has been elucidated by Weigmann. The Bacillus lactis
saponacei imparts to milk a peculiar soapy flavour. It was detected
in the straw of the bedding and hay of the fodder, and from such
sources may infect the milk. There is little or no coagulation,
but a certain amount of sliminess and ropiness, with a peculiar
soapy taste to the milk.

(d) Ghromogenic Changes. — We have already remarked that
colour is the natural and apparently chief product of many of the
innocent bacteria. They put out their strength, so to speak, in the
production of bright colours. The chief colours produced by germs
in milk are as follows : —

Bed Milk — Bacillus prodigiosus, in the presence of oxygen, causes
a redness, particularly on the surface of milk. It was possibly the
work of this bacillus that caused "the bleeding host" which was one
of the superstitions of the Middle Ages. B. lactis erythrogenes pro-
duces a red colour only in the dark, and in milk that is not strongly
acid in reaction. When grown in the light this organism produces a
yellow colour. There is a red sarcina (Sarcina rosea) which also
has the faculty of producing red pigment. One of the yeasts is
another example. It must not be forgotten that redness in milk
may actually be due to the presence of blood from the udder of the
cow.

It is of importance clearly to differentiate between milk reddened
by the admixture of blood from the mammary gland, and that pro-
duced by the organism isolated and studied by Hueppe and Groten-
feldt — Bacillus lactis erythrogenes — the presence of which in the
milk is now looked upon as the active causation of the disease. In
the former case the coloration is apparent immediately after milking,
is uniform, and if the milk is allowed to rest the flocculent blood
coagulum causing the coloration will gradually sink and deposit
itself in the form of a precipitate at the bottom of the milk
receptacle. In the latter the red spots do not appear until later, the
infection of the milk is comparatively slow, and the milk serum is
alone affected, the cream layer not taking the red coloration.
This is probably due to the simple fact that the cream layer being
at the surface is exposed to the light, which inhibits the coloration.
A general coagulation of the milk takes place accompanied by a

ANOMALOUS FERMENTATIONS 201

distinctive sickly sweetish odour. The red coloration will not
occur if the milk is exposed to light or has an acid reaction. The
Bacillus lactis erythrogenes of Hueppe (Bacterium erythrogenes of
Grotenfeldt) is an aerobic, liquefying, non-motile, non-sporing,
chromogenic bacillus of 1 to 1/5 mm. in length by '3 to *4 mm. in
breadth, at times attaining, especially in broth cultures, a length of
4 to 5 mm. in the form of filaments. It takes readily all ordinary
stains and holds the Gram. In sterile milk a gradual precipitation
of the casein takes place with a neutral or slightly alkaline reaction
of the medium. The resultant serum, in the absence of light, absorbs
the red colouring matter produced by the organisms, taking a deep
red tint provided the medium has no acid reaction. The coagulation
by rennet of milk infected with the organism has the effect of
producing a marked dirty red coloration, changing to a reddish-
brown and finally to blood red.

Blue Milk is due to the growth of the Bacillus cyanogenes
(Bacterium syncyaneum of Ehrenberg), or as Hueppe originally
termed it, the Bacillus lactis cyanogenus, an anaerobic, non-liquefying
bacillus, motile, bi-polar, flagellated, chromogenic, and round-ended,
with a varying average length of from 1 to 4 mm. by '3 to *5 mm. in
breadth. Spore formation has been claimed by Hueppe, but
denied by Heim, who describes the so-called spores of Hueppe as
involution forms only. In liquid cultures curious involution forms
are often observed, which are especially noticeable if the organism is
grown in mineral media, as those of Conn and Nsegeli. The organism
does not liquefy gelatine and grows freely on all the usual laboratory
media at room temperature, the dark purplish blue or in some cases
brownish coloration of the medium being very characteristic, but
this freedom of growth becomes less as the temperature advances to
37° C, and the cultures themselves die at 40°. The reaction is
invariably alkaline, although the medium itself may have been in the
first place acid. It stains well with all the ordinary stains but does
not hold the Gram. In milk the bluish tint would appear to be
dependent upon certain unknown conditions, and in the sterile milk
used for the laboratory purposes it is not easy to obtain it.

Yellow Milk. — Bacillus synxanthus is held responsible for curdling
the milk, and then at a later stage, in redissolving the curd, produces
a yellow pigment.

In addition to the bacteria of fermentation occasionally present
in milk, there is a group of Various Unclassified Bacteria. In
milk this is a comparatively small group, for it happens that
those bacteria in milk which cannot be classified as fermentative or
pathogenic are few. B. coli communis occurs here as elsewhere,
and might be grouped with the gaseous fermentative organisms on
account of its extraordinary power of producing gas and breaking up

202 BACTERIA IN MILK AND MILK PRODUCTS

the medium (whether agar or cheese) in which it is growing. What
its exact role is in milk it would be difficult to say. It may act, as
it frequently does elsewhere, by association in various fermentations.
Some authorities hold that its presence in excessive numbers may
cause epidemic diarrhoea in infants (Delepine).

Several years ago a commission was appointed by the British
Medical Journal to inquire into the quality of the milk sold in some
of the poorer districts of London. Every sample was found to
contain B. coli, and it was declared that this particular microbe
constituted 90 per cent, of all the organisms found in the milk.*
We record this statement, but accept it with some reserve. The
diagnosis of B. coli eight or nine years ago was not such a strict
matter as to-day. Still, undoubtedly, this particular organism is not
uncommonly found in milk, and its source is uncleanly dairying. In
the same investigation, Proteus vulgaris, B. fluorescens, and many
liquefying bacteria were frequently found. Their presence in milk
means contamination with putrefying matter, surface water, or a
foul atmosphere.

A number of water bacteria find their way into milk in the
practice of adulteration, and foul byres, and dirty dairies and milk
shops, afford ample opportunity for aerial pollution.

Another unclassified group occasionally present in milk is repre-
sented by moulds, particularly Oidium lactis, the mould which causes
a white fur, possessing a sour odour. It is allied to the Mycoderm<<
albicans (0. albicans), which also occurs in milk, and causes the
whitish-grey patches on the mucous membrane of the mouths of
infants (thrush). These and many more are occasionally present in
milk.

2. The Disease-producing Power of Milk

The general use of milk as an article of diet, especially by the
younger and least resistant portion of mankind, very much increases
the importance of the question as to how far it acts as a vehicle of
disease. Recently, considerable attention has been drawn to the
matter, though it is now a number of years since milk was proved to be
a channel for the conveyance of infectious diseases. During the last
twenty years, particular and conclusive evidence has been deduced
to show that milch cows may themselves afford some measure of
infection. The extensive work on tuberculosis by three Eoyal Com-
missions has done much to obtain new light on the conveyance of
that disease by milk and meat. The enormous strides in the know-
ledge of the bacteriology of diphtheria and other germ diseases have
also placed us in a better position respecting the conveyance of such
diseases by milk. Generally speaking, for reasons already given,
* Brit. Med. Jour., 1895, vol. ii., p. 322.

MILK-BORNE TUBERCULOSIS 203

milk affords an ideal medium for bacteria, and its adaptability there-
fore for conveying pathogenic organisms is undoubted. We shall
speak shortly of the outstanding facts of the chief diseases carried by
milk.

Milk-borne Tuberculosis

It is a well-known fact that tuberculosis is a common disease of
cattle. Probably not less than 20 to 30 per cent, of milch cows in
this country are affected with it. Therefore, at first sight it might
appear that the consumption of milk from such animals would lead
to considerable spread of the disease. But in point of fact there are
two limiting conditions. The first has relation to the question of
the communicability of the disease from the cow to man. The
second concerns the degree of disease in the cow which can affect the
milk. It is necessary that both points should be discussed some-
what fully. But the former question will be discussed in the chapter
dealing with Tuberculosis (see pp. 338-346). The latter condition
only will be discussed here. It has reference to that which limits the
transmissibility of tuberculosis from the cow to man by means of
milk has relation to the well-established fact that the milk of tuber-
culous cows is only certainly infective when the tuberculous disease
affects the udder* This is not necessarily a condition of advance. I
tuberculosis. The udder may become infected at a comparatively early
stage. The presence or absence of tubercle bacilli in the milk o\' cows

• This is the generally accepted view, but it should be added that various
workers have shown that cows having generalised tuberculosis, but apparently un-
affected udders, may yield tuberculous milk. Quite recently (1903) Mohler, as the
result of a long series of experiments, arrives at the following important con-
clusions : — (1) That the tubercle bacillus may be demonstrated in milk from tuber-
culous cows when the udders show no perceptible evidence of disease either
macroscopically or microscopically ; (2) that the bacillus of tuberculosis may be
excreted from such an udder in sufficient numbers to produce infection in experi-
mental animals, both by ingestion and inoculation ; (3) that in cows suffering from
tuberculosis the udder may, therefore, become affected at any moment ; (4) that the
presence of the tubercle bacillus in the milk of tuberculous cows is not constant, but
varies from day to day ; (5) that cows secreting virulent milk may be affected with
tuberculosis to a degree that can be detected only by the tuberculin test; (6) that
the physical examination or general appearance of the animal cannot foretell the
infectiveness of the milk ; (7) that the milk of all cows which have reacted to the
tuberculin test should be considered as suspicious, and should be subjected to
sterilisation before using; and (8) that it would be better still that tuberculous cows
should not be used for general dairy purposes.* Experiments by Lydia Rabino-
witsch have given somewhat similar results. She relics on tuberculin as a test of
infectivity, and the animal experiment as proof of tuberculosis in milk.t Ravenel
also maintains that cows which show no evidence of tuberculous udders, but which
react to tuberculin, may yield tubercle bacilli in their milk. J. H. Young of Aberdeen
maintains, on the other hand, that cows free from udder disease though reacting to
tuberculin yield milk free from tuberculosis.;:

* Bureau of Animal Industry, Washington, U.S.A., Bulletin 41, 1903, p. 93.
I See also Zeitschr ift f u'r Thiermedicin, 1904, p. 202,
+ Brit. Med, Jour,, 1903, i., p. 81(5.

204 BACTERIA IN MILK AND MILK PRODUCTS

with udder tuberculosis is greatly dependent on the extent of the dis-
ease in the udder. But to make the milk infective the udder must be
affected, and milk from such an udder possesses a considerable degree
of virulence. When the udder is thus itself the seat of disease, not
only the derived milk, but the skimmed milk, butter-milk, and even
butter, may all contain tuberculous material. Furthermore, tuber-
cular disease of the udder spreads in extent and degree with extreme
rapidity. From these facts it will be obvious that it is of first-rate
importance to be able to diagnose udder disease. This is not always
possible in the early stage. The signs upon which most reliance may
be placed are the enlargement of the lymph-glands lying above the
posterior region of the udder ; the serous, yellowish milk which later
on discharges small coagula ; the partial or total lack of milk from
one quarter of the udder (following upon excessive secretion) ; the
hard, diffuse nodular swelling and induration of a part or the whole
wall of the udder ; and the detection in the milk of tubercle bacilli.
The whole organ may increase in weight as well as size, and on post-
mortem examination show an increase of connective tissue, a number
of large nodules of tubercle, and a scattering of small granular bodies,
known as " miliary " tubercles.

The udder is affected in about 2 per cent, of the cows in the
milking herds in this country (MacFadyean).* In London about
0'2 per cent, of the cows are so affected, as judged by clinical
observation. It will be remembered that in the country generally
between 20-30 per cent, of the cows suffer from tuberculosis. In
London 15-20 per cent, are tuberculous. The higher standard in
London is due to better-class animals being housed in London, to
more thorough inspection, and to the fact that there is no inbreeding'.
But let us take the generally-accepted figure of 2 per cent, of the
cows in the United Kingdom as having tuberculous udders, and
therefore yielding, in greater or less degree, tuberculous milk, leaving
altogether out of account cows which may be tuberculous but are
not affected in the udder. In the United Kingdom in 1901 there
were 4,102,000 milch cows.f If we take 2 per cent, of these as having
tuberculous udders, it gives us 80,000. The average annual yield of
milk per cow may be taken as, at least, 400 gallons, J which means that
from these 80,000 tuberculous udders 32,000,000 gallons of milk are
obtained. It is not asserted that this large amount of milk is actually
virulent with tuberculous matter, but it will be admitted that the
entire amount of it is open to grave suspicion, if not absolute con-
demnation.

* Trans, of Brit. Congress on Tuberculosis ■, 1902, vol. i. , p. 84.
f In the United States of America there were in 1901, 18,112,000 milch cows in
actual dairy use.

X 420 gallons is the generally accepted figure.

1 ' 'S v

•' r.fe

.>/?• 0^:'

-, ' i (,'

V

PLATE 18.

Bacillus tuberculosis. Film preparation from glycerine glucose, agar culture. Fuur months old-
2 months at 37° C, and 2 months at 20° C. Stained with carbol fuchsin. x 1000.

Bacillus tuberculosis, in Udder of Cow. Stained with carbol fuchsin and methylene blue.

x 1000.

['Jo face page 204.

MILK-BORNE TUBERCULOSIS

205

There are a variety of conditions in addition to the vera causa, the
presence of the bacillus of tubercle, which make the disease common
amongst cattle. Constitution, temperament, age, work, food, in-
breeding, and prolonged lactation, are the individual features which
act as predisposing conditions ; they may act by favouring the pro-
pagation of the bacillus or by weakening the resistance of the tissues.
To this category must further be added conditions of environment.
Bad stabling, dark, ill- ventilated stalls, high temperature, prolonged
and close contact with other cows, all tend in the same direction.

The danger from drinking raw tuberculous milk only exists for
persons who use it as their sole or principal food, that is to say, young
children and certain invalids. With adults in normal health, the
danger is greatly minimised, as the healthy digestive tract is relatively
insusceptible. Moreover, dairy milk is almost invariably mixed milk ;
that is to say, that if there is a tubercular cow in a herd yielding tubercle
bacilli in her milk, the addition of the milk of the rest of the herd so
effectually dilutes the whole as to render it in some degree innocuous.

It should not be forgotten that milk may become tubercular
through the carelessness or dirty habits of the milker. Such a
common practice as moistening the hands with saliva previously to
milking may, in cases of tubercular milkers, effectually contaminate
the milk. Again, it may become polluted by dried tubercular matter
getting into it from dust or infected dried excreta. Such convey-
ances must be of rare occurrence, yet their possibility should not be
forgotten.

The Tubercle Bacillus in Market Milk. — Investigations have been
made in many cities as to the actual occurrence of the tubercle

Table shotting the Total Number of Milks Examined Bacteriological^ for

Tubercle Bacilli from August 1896 to 2>lst December 1903.

(Liverpool.)

Year.

Total

number

of Samples

taken.

Town Samples.

Country Samples.

Number
taken.

Tubercular.

Percentage
Tubercular.

Number
taken.

Tubercular.

Percentage

Tubercular.

1896
1897
1898
1899
1900
1901
1902
1903

119
150
112
352
560
566
595
582

83
63
84
167
255
254
213
231

4
4
7
1
4
2
1
2

4-8
6-3
8-3
0-6
1-5
0-7
0-4
0*8

36
87
28
185
305
312
382
351

5

5

5

15

5

20

32

19

14'0
5*7

17-9
8-1
1-6
6*4
8-3
5-6

bacillus in milk as placed on the market. It has been found in

206 BACTERIA IN MILK AND MILK PRODUCTS

varying percentage,* but, as a rule, the milk coming in to the
cities from the country has contained more tubercle bacilli than
the milk obtained from the town cows. This characteristic has
been found to occur in London, Manchester, Liverpool, and other
cities.

It should also be added that there are a number of cases on record
where the tubercle bacillus has been found in butter.

The Virulence of the Tubercle Bacillus in Milk. — Martin and
Woodhead concluded as the result of their investigations for the
Koyal Commission on Tuberculosis that tuberculous milk possessed
a high degree of virulence for man. Sir Eichard Thorne held that
tabes mesenterica (alimentary tuberculosis) of children had not
declined as phthisis (pulmonary tuberculosis) had done in recent
years on account of the conveyance of the virus of tubercle in milk.f
If the occurrence of primary lesion in the intestine is indication of
infection through the alimentary tract, then it is instructive to notice
that of all the tuberculosis in children in this country about 25 per
cent, is alimentary in origin, and in 60 to 70 per cent, of the cases
the mesenteric glands are affected. Both of these figures deal with
deaths only, but as Eaw has pointed out, no doubt a number of cases
of alimentary tuberculosis recover, the infection having been mild.
Amongst 269 tuberculous children under twelve years of age whom
Dr Still examined post mortem, he found it possible to determine
the channel of infection with some degree of certainty in 216
cases. In 138 (63'8 per cent.) infection entered through the lung ;
in 63 (291 per cent.) primary infection occurred, in all prob-
ability through the intestine. Of children up to two years of age
he found 65 per cent, contracted infection through the lung, and
22 per cent, through the intestine. In infants under one year of
age apparently only 13 per cent, contracted tuberculosis through the
intestine. %

It is recognised that, owing to the great tendency to generalisa-
tion of tuberculosis in children, it is a matter of extreme difficulty
to determine which was, in fact, the primary channel of infection,
and this must be taken into consideration in estimating the
significance of the frequency in the above figures. It should also
be remembered that the tubercle bacillus may, and probably

* In 1898, 14 per cent, of the milks examined in Berlin by Petri contained the
tubercle bacillus. In 1899, in Islington, the percentage was 14*4; in 1893, St
Petersburg, 5 per cent. ; in 1901, in London, 7 per cent. (Klein) ; in 1901, at Croydon,
67 per cent., and Manchester, 9*5 per cent. ; in 1902, at Woolwich, 10 per cent. ;
in Camberwell, 11 per cent. ; in the City of London and in Finsbury, nil. These
percentages must not be accepted as anything but passing figures and illustrations
of what various investigators have found under varying conditions.

t The Administrative Control of Tuberculosis (Harben Lectures), 1899, pp. 5-7
and 28-32.

% Practitioner, 1901 (July), p. 94.

MILK-BORNE TYPHOID 207

does, pass through the intestinal wall into the nearest lymphatic
glands, leaving no visible trace on the intestine. Further, owing
to the fact that children swallow their pulmonary expectoration,
secondary infection of the intestine may rapidly follow primary
infection of the lungs. Hence it comes about that, in many cases,
the intestine and mesenteric glands are affected, and yet such a
condition cannot be taken as evidence of the infection by food. Dr
Still concludes that (a) the commonest channel of infection with
tuberculosis in childhood is through the lung ; (b) infection through
the intestine is less common in infancy than in later childhood;
(c) milk, therefore, is not the usual source of tuberculosis in infancy ;
and (d) inhalation is much the commonest mode of infection in the
tuberculosis of childhood, and especially in infancy. Dr Still has
placed on record 5 cases of tuberculous ulcer of the stomach in
children.

Taking a broad view of the facts, it would appear that whilst
tuberculosis is not chiefly spread by means of milk, there is
unmistakable evidence, derived from pathological and clinical
experience, proving that tuberculous milk can, and does on occasion,
set up some form of tuberculosis (bovine or human) in the bodies
of man and other animals consuming the milk.

Milk-borne Typhoid Fever

Dr Michael Taylor of Penrith was the first to establish the now
well-known fact that milk may act as a vehicle of the virus of
enteric fever. That was in 1857.* Since that date more than 160
epidemics of this disease have been traced to a polluted milk supply.
Schuder states that 17 per cent, of all typhoid epidemics arc clue to
the consumption of infected milk.

The steps in the process of infection are briefly as follow. Enteric
fever affects the intestine, and hence the excreta, especially in the
early stages of the disease, are charged with large numbers of the
causal bacilli. It is now known that the sweat, expectoration from
the lungs, and the urine of a typhoid patient may also contain the
typhoid bacillus. Indeed, the urine in 25 per cent, of the cases
generally contains large numbers of the bacillus (Horton Smith).f
There is also evidence to show that the bacilli may remain in the
urine for long periods after convalescence, even for months and
possibly for years. The bowel discharges and the urine are, therefore,
the two chief channels by which the typhoid bacillus is excreted.
It, therefore, readily gains access to the soil, to drains, and eventually

* Edin. Med. Jour., 1858, pp. 993-1004.
t Lectures on Typhoid Fever, 1900.

208 BACTERIA IN MILK AND MILK PRODUCTS

on occasion to the water supply, and thus into milk and back again
to man. The virus does not always pass in the discharges to water
and milk, but may reach them by becoming dried dust. A small
pollution may in this way set up widespread disease. (For the
behaviour of the typhoid bacillus in soil, see pp. 145-150.)

The most common way for milk to become infected by the
typhoid bacillus is through infected water. Such water may be
added to the milk by way of adulteration or by accident; or the
milk vessels may have been " cleansed " with polluted water (in 29
per cent, of milk-borne outbreaks according to Schuder). Another
source of infection of the milk is when persons suffering from a
mild attack of typhoid fever continue to work a dairy or otherwise
deal with milk, and this has proved a frequent means of infection.
Flics doubtless convey the germ of the disease not infrequently, as
was shown in the Spanish-American War of 1898 * and the Boer
War of 1900-1901.f

Though the typhoid bacillus appears not to have the power of
rapid multiplication in milk, it has the faculty of existing in milk
for a considerable time (twenty days or longer) even when milk has
curdled or soured, and may thus infect milk products, such as butter
and cheese. But infection by milk products may be eliminated as
of too rare occurrence to deserve attention. The bacillus does not
coagulate milk like its ally the B. coli communis, which is a much
more frequent inhabitant of milk. It flourishes in milk at room
temperature and blood-heat, and does not produce acid or alter the
appearance of the milk.

Several typical milk-borne outbreaks of typhoid fever may be
cited : —

1. Infection from Personal Contact with Typhoid Patients. — At
Penrith, it appears that about the beginning of September, 1857, a
young servant girl, E. 0., returned home to Penrith from Liverpool
suffering fom typhoid fever. The family of which she was a
member consisted of father, mother, and five children, of whom she
was the eldest. The cottage in which they lived consisted of two
ill- ventilated and ill-lighted rooms, a kitchen or sitting-room, and a
bedroom opening out of it. The father possessed three cows, and
carried on a small milk business dealing with some fourteen
families. The mother milked the cows, and the milk was brought
into the kitchen, direct from the byre, and in due course dis-
tributed in tin measures amongst the customers. After her return
home the girl continued ill for about a fortnight, during which
period she was nursed by her mother in the kitchen or common

* American War Department, Official Report, 1900.

f Brit. Med. Jour., 1901, i., p. 642 et seq. ; ibid., 1902, ii., pp. 936-941 (Firth
and Horrocks).

MILK-BORNE TYPHOID 209

sitting-room. At the end of the fourth week in September she was
convalescent, and began to help at once in the distribution of the
milk. Two other children of the family sickened and passed through
the fever. The mother nursed all three patients, and continued to
milk the cows and attend to the distribution of the milk. In
October and November some 13 cases of typhoid fever occurred
in seven families dealing with the infected cottage, and from these
primary cases a number of persons, over a somewhat wide area, were
infected by contact. By most careful observation and reasoning, Dr
Taylor arrived at the conclusion that the milk became contaminated
in the kitchen of this cottage, from the typhoid patients there
being nursed.*

2. Infection from Washing Milk Vessels with Polluted Water. — At
Clifton, Bristol, in October 1897, an outbreak affected 244 persons,
31 of whom died. Ninety-six per cent, of the patients consumed sus-
pected milk. It happened that a brook received the sewage of thirty-
seven houses, the overflow of a cesspool serving twenty-two more, the
washings from fields over which the drainage of several others was
distributed, and the direct sewage from at least one other, and then
flowed directly through a certain farm. In September it seems that
some excreta from a man suffering from typhoid fever gained access
to the brook. The water of this stream supplied the farm pump, and
the water itself, it is scarcely necessary to add, was highly charged
with putrescent organic matter and micro-organisms. This water
was used for washing the milk-cans from this particular farm,
otherwise the dairy arrangements were efficient. Part of the milk
was distributed to fifty-seven houses in Clifton ; in forty-one of
them cases of typhoid occurred. Another part of the milk was
sold over the counter; twenty households so obtaining it were
attacked with typhoid fever, and a number of further infections
arose in the course of a third delivery. f

3. Infection from Water added to Milk. — At Moseley, in 1873,
96 persons in fifty families contracted typhoid fever from milk.
Boy at milkman's house fell ill of typhoid fever, suffered there for a
fortnight and died. Two wells were polluted from a privy into which
typhoid excreta had been thrown. The water of the well was added
accidentally or intentionally to the milk. Dr Ballard summed up
his view of the causation in this outbreak as follows : — (1) Two
wells upon adjoining premises occupied by milk sellers became
infected early in November with the infectious matter or virus of
enteric fever, through the soakage from a privy into them of
excremental matters containing that matter of infection. (2)
Through the medium of water drawn from these wells the milk

* Edin. Med. Jour., 1858, pp. 993-1004.

t Trans. Epidem. Soc. of London, vol. xvii., pp. 78-103 (Dr D. S. Davies).

O

210 BACTERIA IN MILK AND MILK PRODUCTS

supplied by these milk sellers became infected, and many of their
regular customers who drank the milk suffered from the disease.
(3) The same infected milk having been sold to two other milk
purveyors, some of the persons using the milk supplied by these
milkmen also suffered in a similar manner. (4) There is no
evidence that the disease spread in these districts in any other
way than through this milk.*

4. Infection from the Air by Dried Typhoid Excreta. — At
Millbrook, in Cornwall, in 1880 (July — September), an outbreak
occurred having a total number of cases of 91. In this instance part
of a slaughter-house not used as such but as a wash-house was
boarded off to constitute a dairy. On a shelf of this dairy the milk
was habitually set in pans, exposed to the air. In one corner of the
slaughter-house, nearest the dairy, was a badly trapped and very
offensive drain inlet. Close to this inlet ran the wooden partition
between the slaughter-house and the dairy, which near the inlet had
been long broken away. The drain was in communication with an
old square drain which had received typhoid excreta, so that the
infected sewer air from the inlet had free access to the dairy and the
exposed milk which stood in the dairy. There was evidence to show
that the drain was in a dry and " waterless " condition. Six cases of
typhoid occurred in the butcher's family, f A similar outbreak
occurred in county Durham in 1893.

5. Infection from Contaminated Cloths and Clothes. — At Barrow-
ford in Lancashire there occurred a typhoid epidemic in 1876. The
total number of cases was 57, all of whom drank the suspected milk.
The farmer had had typhoid fever in his house for two or three
weeks before the outbreak, and no precautions had been taken to
prevent the spread of the disease. The milk was left for some time
in the farm-house before being sold. The milk-tins were wiped
with the same " dish cloth " as that used among the fever patients.
The farmer himself nursed his children, and then went immediately
without disinfection amongst his cattle and milked them in the same
clothes he had worn whilst nursing his children. The cases occurred
within a very short space of time, and every one of them without
exception drank the milk from this farm. Twenty-five of the
patients were under ten years of age. There was no other typhoid
in the district.

6. Infection owing to Cooling Milk in Water. — At Springfield,
Mass., U.S.A., in 1892, an outbreak affecting 150 persons (25 of
whom died). Upon the farm supplying the implicated milk there
was one, and probably more than one, case of typhoid fever. The
farmer submerged his sealed milk-cans when full of milk, in the well

* Report of Local Government Board, 1874, p. 92.
t Brit. Med. Jour., 1881, i., p. 20.

MILK-BORNE DIPHTHERIA 211

adjoining the cow yards, with the object of keeping the milk cool.
The water in this was polluted, and it was found that four of nine
milk-cans leaked when inverted. Hence it became evident that
water could gain access if the cans were submerged as they had
been. The investigators suggest that as the typhoid excreta of the
patient were placed, undisinfected, in the privy, and the contents of
the latter spread over the tobacco field, the germs of typhoid may
have gained access to the well by dirt from the labourers' boots, who
both worked in the field and at the milk. Coliform organisms were
found in the well water.*

Milk-borne Diphtheria

Milk is a favourable medium for the B. diphtherial. The organism
both lives and multiplies in ordinary sterilised milk, but it thrives
better in milk at comparatively low temperatures than at 37° C. In
ordinary milk, unsterilised and unprepared, the commoner organisms
multiply much more rapidly, and so the diphtheria bacillus is in
all probability soon crowded out.

The cases, however, in which the B. dvphtherice has been actually
isolated from market milk are extremely few. In the outbreak of
diphtheria at Senghenydd in South Wales, in 1899, Bowhill f
isolated a diphtheria bacillus from the suspected milk. The culture
of the bacillus in broth proved fatal to a guinea-pig in two days.
In the same year, Eyre \ isolated a virulent diphtheria bacillus
from some milk implicated in an outbreak of diphtheria in a
school. In 1900, Klein § also reported the isolation of a genuine
diphtheria bacillus in an examination of 100 samples of milk in
London. Lastly, Dean and Todd, isolated the B. diplitherice from cow's
milk in 1901. || These are the only four authentic cases of actual
detection of the B. diphtherias in ordinary milk with which we have
met.

There is a question which must now be considered, viz.: the
relationship existing between diphtheria in man and animals and the
milk supply. How does the milk become infected ?

(1) In the first place, it is now generally held that the B.
diphtherias has a comparatively wide distribution in nature ; whilst
it appears not to be conveyed by water, it is believed that certain
conditions of soil favour its growth as a saprophyte. But this is
not proved. (2) In the second place, it has been proved that persons

* Boston Med. and Surg. Jour.* 1893, ii., p. 485 (Sedgwick and Chapin).
I Veterinary Record, 8th April 1899, No. 561. Jour, of State Medicine, 1899,
pp. 705-710.

X Brit. Med. Jour., 1899, vol. ii., p. 586.
§ Jour, of Hygiene, 1901, vol. i., p. 85.
|| Ibid., 1902, Vol. ii., pp. 194-205.

212 BACTERIA IN MILK AND MILK PRODUCTS

suffering from diphtheria are foci of infection. The exact channels
of infection differ under varying circumstances ; but, in general, the
source of infection is the throat and mouth of the patient. Anything
which comes into contact with the mucous membrane becomes
infected. Thus handkerchiefs, sweets, children's toys, etc., may act
as the vehicles of contagion. The mucus and saliva may also be
infective, and in speaking, kissing, coughing, or expectorating such
mucus (probably rich in bacilli) may be disseminated in very fine
particles, and so carry the disease. It is by such means that the
disease is spread. Moreover, there is the fact of the long period
during which the human throat may remain infective. Professor
Sims Woodhead has recently stated that the persistence of the
diphtheria bacillus for periods up to eight weeks is of very common
occurrence. (3) Eichardiere and Tollemer * and others have proved
that the dust floating in the air of a diphtheria ward may contain
large numbers of diphtheria bacilli, and in this way milk and other
foods may become contaminated.

Between 1878 and 1883, certain outbreaks of diphtheria due to
milk appeared to suggest that the cow itself might suffer from
diphtheria. The discovery of the Klebs-Loffler bacillus in 1883
furnished the basis for experimental work, and in 1886 Dr Klein
undertook some experiments to ascertain whether or not diphtheria
was inoculable into cows. He took for the experiment two healthy
milch cows which had calved some three or four weeks previously.
One c.c. of broth culture of B. diphtheria (derived from human
diphtheritic membrane) was injected under the skin into the sub-
cutaneous tissue of the left shoulder in each of the two cows. Two
or three days after the inoculation (a) the temperature rose (to
40*6°), and the animals suffered from slight general malaise. On
the third day (b) a tumour appeared at the site of inoculation, which
steadily increased in size to the seventh day. On the fifth day (c)
a papular eruption appeared on the udder and hind teat. In addition
to the papules there were half a dozen vesicles, and some round
patches covered with brown crusts. The process of eruption was
mature by the eighth day. In the lymph of the vesicles and pustules
the B. diphtheria? could be demonstrated, according to Klein, both
microscopically and by culture. He therefore concluded that the
B. diphtheria:, as such, inoculated into the shoulder of the cow, was
received into the general system of the cow, and produced its effects
not in the viscera, but on the udder as a specific eruption, and that
before the end of five days after inoculation, was finally excreted in
the cow's milk. " The presence of the diphtheria bacillus," he wrote,
" could with certainty, by microscopic and culture observations, be
demonstrated in the milk of the cow collected under all precautions ;
* Gazette des Maladies Infantiles, 1899, No. 10.

MILK-BORNE DIPHTHERIA 213

the number of bacilli present on that day in the milk amounted to
32 per c.c. Scrapings from vesicles on the sixth day were inoculated
into two calves, which then suffered from a like disease." *

During 1890 and 1891, Dr Klein repeated these experiments
on milch cows, and in two further instances, out of six cows, an
eruption was produced on the udder and teats, and in one of these
positive cases the B. diphtherias was found in the milk about a week
after inoculation.

It must be admitted that positive results did not always follow
these experimental researches. Loftier, Abbott, Bitter, and others,
including many veterinarians, criticised the experiments, and held
that there was no conclusive evidence that diphtheria was a bovine-
disease. Since that time some twenty milk-diphtheria outbreaks
have been investigated, with the result that, with one or two
exceptions, the infectivity of the milk was certainly derived from
human sources and not from bovine. In the Croydon outbreak in
1890, at Worcester in 1891, and at Glasgow in 1892, evidence was
obtained which appeared to support Klein's views.

Up to the present it may, however, be said that the evidence
forthcoming points in the direction of human rather than bovine
infection as the origin of the diphtheria bacillus in milk.

An interesting investigation has recently been made by Dean and
Todd, respecting a small outbreak of diphtheria occurring in 1901.+
In this outbreak several individuals suffered from diphtheria, and
several others in the same households suffered from sore throat,
probably diphtheritic. These individuals obtained their milk from
two cows suffering from a contagious eruptive disease of the udder,
from which Dean and Todd isolated a bacillus indistinguishable from
Klebs Loftier bacillus of diphtheria. The case was a very interesting
one. But the whole matter of bovine diphtheria is sub judice.

It was then in 1878, that evidence was forthcoming in support
of the view that diphtheria, like typhoid fever, might on occasion be
spread by means of milk. In that year, Mr W. H. Power made an
inquiry into an outbreak of diphtheria in North London, chiefly in
Kilburn and St John's Wood. There were as many as 264 persons
attacked, and 38 died. The infection invaded some 118 different
households. The epidemic was most severe in May (first four weeks),
when about 190 cases occurred. The outbreak terminated abruptly.
The area infected, and time of infection, clearly showed that there
was some factor at work over a circumscribed area, and operative

* See A Treatise on Hygiene and Public Health (Stevenson and Murphy), vol. ii.,
pp. 161-164 (Klein). Also Local Government Board Report, 1889, p. 167 et seq.

f Jour, of Hygiene, April 1902 (vol. ii., No. 2, p. 194). Experiments on the
relation of the Cow to Milk Diphtheria, by George Dean, M.B., and Charles Todd,
M.D.

214 BACTERIA IN MILK AND MILK PRODUCTS

during a limited time. There was no antecedent throat illness, and
no school infection or contact contagion traceable. The houses were
sanitarily good, and had a good water supply. There was but one
thing common to most of the cases, and this was the milk supply.
It was found that within the area of the greatest prevalence of
throat illness, about one-fifth of the households were supplied by a
common milk supply. The incidence of the disease fell, actually and
relatively, upon consumers of the suspected milk.

Inquiry into the milk supply elicited no evidence of human
disease pollution, nor contamination by water or air. Nor was there
any definite disease of cows present at the time as far as could be
judged. But by a process of exclusion, Mr Power suggested that
" there may have been risk of specific fouling of milk by particular
cows suffering, whether recognised or not, from sj)ecific disease."

Since that time there have been some 30 outbreaks of milk-borne
diphtheria. In most cases the milk appears to have been infected
directly by persons suffering from the disease, recognised or un-
recognised.

Milk-borne Scarlet Fever

There arc some seventy milk-borne epidemics of scarlet fever on
record, and yet comparatively little is known as to the bacteriology
of the disease (see p. 296). In almost all the outbreaks which have
occurred there is evidence, more or less conclusive, that persons
suffering from scarlet fever have been concerned in milking or in the
distribution of milk. But in 1882 Mr W. H. Power, in investigating
an outbreak of milk-borne scarlet fever in North London, came to the
conclusion that the cow had been the exciting cause of the epidemic,
and was suffering from some diseased condition which could convey
to the milk the virus of scarlet fever.* In 1885 occurred the
" Hendon outbreak " of scarlet fever, which affected North London
districts and Hendon. After inquiry, it was believed that the scarlet
fever in these districts followed the consumption of milk from a
particular farm at Hendon. Further, in these four districts wherein
scarlatina had shown an extravagant incidence upon the milkman's
customers, the disease had begun its peculiar incidence about the end
of November or beginning of December. In one of those districts
(South Marylebone), scarlatina continued day by day, and with
increasing force up to the date of the inquiry, to attack the customers
of the retail business. In two other districts (Hampstead and St
Pancras), after attacking in some numbers, for a few days at the end
of November and beginning of December, the customers of the busi-
ness, the disease showed no fresh attacks for about ten days (a short
but clearly defined intermission), and then about the middle of

* Supplement to the Report of Local Government Board, 1882, p. 65.

MILK-BORNE SCARLET FEVER 215

December attacked them again in larger numbers, and continued to
do so up to date of inquiry.

The chief facts concerning the distribution of the milk may be
set out as follows : (a) The Marylebone customers suffered at the end
of November and up till the end of the third week in December, (b)
The Hampstead cases occurred in two groups, one small group at the
end of November and a larger group in the latter part of December,
(c) The St Pancras customers suffered like the Hampstead ones, but
in a less degree. They obtained milk from the same vendors, (d)
The St John's Wood customers did not suffer until after the end of
the year, (e) The few persons affected at Hendon suffered early in
December, having consumed milk which had been returned from
Marylebone, and at the same time new cases of scarlet fever ceased
to occur in Marylebone. Examination was then made to ascertain
if there had been any possible infection of the milk to explain this
incidence and intermission.

When Mr Power came to inquire as to the movements of the
cows, he learned that on 15th November three newly-calved cows
arrived at the Hendon farm from Derbyshire, this event shortly pre-
ceding the first attack of scarlatina. It happened that these three
animals were placed in a shed by themselves, and their milk was dis-
tributed in part to South Marylebone, Hampstead, and St Pancras,
immediately preceding the outbreak of scarlatina in those districts.
On examination it was found that the implicated cows were suffering
from some kind of disease of the udders, which had spread to other
cows in the herd. It would appear that the diseased condition, what-
ever it was, had been introduced by one of the Derbyshire cows, and
had then spread through various sheds at the Hendon farm. Mr Power
was able by the most minute inquiry to trace the movements of those
cows and the various sheds in which they were placed from time to
time, and he held that the various recrudescences of the outbreak in
North London corresponded with the movements of the affected cows.

The exciting cause, then, of this outbreak was believed by Mr
Power to be a condition of certain milch cows which had for its outward
manifestation an eruption on teats and udders, and which was com-
municable from cow to cow. Subcultures of the ulcerous discharges
of the affected animals inoculated into calves produced a disease
having unmistakable affinities, under some conditions, with the
disease in the milch cows, and under other conditions with scarlet
fever in the human subject (Klein). Now, it must be added, that
scarlet fever appeared simultaneously in all but one of the five
localities to which the milk was distributed. The exception received
none of the milk from the affected cows until later, when the disease
also appeared in this district, owing to some of the milk from the
affected cows being sent there. When the sale of the milk was pro-

216 BACTERIA IN MILK AND MILK PRODUCTS

hibited in London, some of it was clandestinely distributed amongst
the poor of Hendon. Amongst those served, half a dozen families
were invaded by scarlatina at a time when the disease had ceased to
exert its influence in the London districts. The intermission which
had occurred in the scarlatina in Hampstead and St Pancras during
the ten days referred to above, was at the time when the infective
cows had been moved into a shed sending milk only to Marylebone.
A few days later they were again moved into a shed from which
milk was distributed to the two former districts.

Thus the investigation showed the Hendon farm to be the main
source, and, as far as could be judged, the cows referred to, the
particular source, of the implicated milk, for the disease followed the
distribution of their milk. The further inquiry was with regard to
the nature of the disease or influence appertaining to these cows.*

Sir George Buchanan summarised for the Local Government
Board his interpretation of the facts, and concluded that the Hendon
disease was "a form occurring of the very disease that we call
scarlatina when it occurs in the human subject."f His views found
acceptance with a large number of persons, but most veterinarians
and certain pathologists were not prepared to accept the matter as
proved. In consequence, further investigations and inquiries were
instituted, and a controversy arose. Sir George Brown, then head of
the Privy Council's Agricultural Department, held (1) that the
Hendon disease was not confined to the Hendon farm from which the
implicated milk was derived, but occurred elsewhere, and was followed
by no scarlet fever; (2) that the probable source of infection at
Hendon was human ; and (3) that the alleged bovine scarlet fever
was cowjoox.\

As a matter of fact, the exact origin of the London epidemic at
that time has not yet been, and now probably never will be, demon-
strated. Even at the present time the specific micro-organism which
is the causal agent of human scarlet fever is not established or
proved. That is to say, no micro-organism has yet been isolated in
human scarlet fever which fulfils the postulates of Koch. Much less
was this the case sixteen years ago, when bacteriological methods
were less perfect than they are even to-day. From this it follows
that the vera causa was obscure, and yet without this link it was
impossible to complete the chain of evidence by which it could be
definitely known that any disease of the cow was responsible for the
epidemic. The probabilities might be strong or weak, but proof was

* Local Government Board Report, 1885, pp. 73-111 (W. H. Power).

f Seventeenth Annual Report of the Local Government Board, 1887-88 (Medical
Officer's Supplement), pp. 13, 14.

% For a discussion of the whole subject, see Bacteriology of Milk (Swithinbank
and Newman), 1903, pp. 279-304. .

MILK-BORNE SCARLET FEVER 217

wanting. The inoculation experiments* in so far as they yielded
positive results, were also open to the same unreliability. Unfortu-
nately, too, there was, on the other hand, circumstantial evidence of
various kinds, which, while it proved little, opened up a variety of
possibilities by which the milk consumed in London might have
become infected. The case was therefore unproved. Nevertheless, it
raised many important questions and stimulated much valuable
inquiry. When milk becomes infected with scarlet fever the infec-
tion is almost invariably derived directly from some person suffering
from the disease, recognised or unrecognised.

Scarlet Fever, in not a few milk-epidemics, has shown certain
modifications of a more or less marked character. The disease is
generally mild, and simultaneously with an outbreak of the specific
disease due to milk, there will not infrequently be found a large
number of " ordinary sore throats." Even in the scarlatinal cases,
the disease has a tendency to remain localised to the throat (Power).
The rash may be evanescent, and the desquamation is scanty
(Parsons). There is also a marked absence of post-scarlatinal
nephritis or any other kidney complication (Parsons, Buchanan, and
others). A characteristic which has been frequently noted, and is
readily to be understood, is the frequency of vomiting and diarrhcea,
rather particularly at the commencement of the disease (the Fallow -
field epidemic, 1879, is an illustration). It is probable that these signs
of alimentary irritation or poisoning are due to poisonous organismal
products contained in the milk. On more than one occasion they
have led to an appearance of intoxication rather than infection.
Finally, there is a clinical feature, to which reference has already
been made, and which may bear a significance not at first appreciated,
namely, the comparative indisposition of the disease to spread by
contagion. This may be attributable to the mildness of the disease,
to the small amount of skin eruption and desquamation commonly
present, and possibly to the fact that the poisonous properties of the
milk are to a certain extent eliminated from the system by the
vomiting and purging. Every clinical sign which has been noted
leads to the conclusion that the disease as conveyed by milk is
frequently mild, and therefore has both a small mortality, and no
tendency to spread by contact. There is one other point deserving
of mention. Sir George Buchanan noticed, in a scarlet fever
epidemic with which he had to deal, that in persons who had had
scarlet fever at some previous time, and who drank the implicated
milk, almost the only symptom of ill-health which they presented
was a sore throat. There was no rash, no vomiting, no pyrexia,
although other members of the family under precisely similar circum-
stances suffered from typical scarlet fever. Many other workers have
confirmed the occurrence of aberrant forms of milk-borne scarlatina.

218 BACTERIA IN MILK AND MILK PRODUCTS

Characteristics of Milk-borne Epidemics

The following are the chief characteristics of infectious disease
carried by milk : —

(a) There is a special incidence of disease upon the track of the
implicated milk supply. It is localised to such areas.

(b) Better-class houses and persons generally suffer most.

(c) Milk drinkers are chiefly affected, and those suffer most who
are large consumers of raw milk.

(d) Women and children suffer most, and frequently adults suffer
proportionately more than children.

(e) Incubation periods are shortened.

(/) There is a sudden onset and a rapid decline.
(g) Multiple cases in one house occur simultaneously.
(h) Clinically, the attacks of disease are often mild, contact infec-
tivity is reduced, and the mortality rate is lower than usual.

Other Diseases Conveyed by Milk

In addition to the above, there are other diseases spread by
means of polluted milk. From time to time exceptional cases have
occurred in which diseases like anthrax, or some forms of foot-and-
mouth disease have been spread by this means. But it is not to
such rare cases that we refer. There are three very common diseases
in which milk has been proved to play a not inconsiderable part,
viz., thrush, sore throat, and diarrhoea.

Thrush. — The mould which gives rise to the curd-like patches in
the throats of children, and which is known as Oidium albicans,
frequently occurs in milk. Soft, white specks are seen on the
tongue and mucous membrane of the cheeks and lips, looking not
unlike particles of milk curd. If a scraping be placed upon a glass
slide with a drop of glycerine, and examined by means of the micro-
scope, the spores and mycelial threads of this mould will be seen.
The spores are oval, and possess a definite capsule. The threads are
branched and jointed at somewhat long intervals. Milk affords an
excellent medium for the growth of this parasite. Thus undoubtedly
we must hold milk partly responsible for spreading this complaint.
Penicillium, Aspergillus, and Mucor are also frequent moulds in
milk.

Sore -Throat Illnesses. — The obscure milk-borne epidemics of
which sore throat has been the chief symptom, are among the
most interesting of all the diseases conveyed by milk. The usual
symptoms are congestion of the tonsils and mucous membrane of
the throat, with sometimes ulceration, enlargement of the cervical
glands, and some pyrexia, and general malaise. In not a few

MILK-BORNE SORE THROAT 219

instances there has been observed various kinds of rash which
have generally been of an evanescent character. Where the throat
illnesses have occurred contemporaneously with outbreaks of scarlet
fever or diphtheria, it is not unlikely that the condition was in reality
scarlet fever or diphtheria. In South Kensington, in 1875, there
was an outbreak of disease which attracted much attention at the
time, and was officially investigated.* The illness was traced to
some cream consumed at a dinner party, and in all twenty persons
suffered, some of whom had scarlet fever, and others only sore
throat. But the investigation showed that in the district from
which the cream was obtained 119 cases of sore throat had occurred.
Dr Darbishire described an outbreak of 18 cases of sore throat at
Oxford in 1882, the condition being characterised by marked severity
of throat symptoms and a disproportionate amount of constitutional
symptoms.f

Similar outbreaks occurred in 1881 at Aberdeen (300 persons
affected), and Eugby school (90 boys) in three school-houses supplied
by one milkman, who did not supply any other houses in the school.
But he supplied houses in the town, of which nearly 50 per cent,
were attacked with sore throat. Inquiry showed that some of the
milk used had been obtained from a cow suffering from mastitis. J
A similar outbreak took place in Edinburgh in 1888, and was
investigated by Cotterill and Woodhead; and another at Dover in
1884, where there was a sudden and severe outbreak of sore throat
in a localised area of good-class houses, affecting 205 persons, who
all obtained milk from one particular farm. The chief symptoms
were local inflammation of the throat, enlargement of lymphatic
glands in neck, and vesicular eruptions preceding and accompanying
the inflammation. The dairyman obtained his supply from 12 cows
of his own, and from three farms in the country. On one of these
latter apthous fever had broken out, and it was from this farm that
the dairyman obtained his implicated milk and cream. Moreover,
when the supply from this farm was diverted temporarily, it set up
a simultaneous second outbreak of sore throat. § In 1890 there
occurred an epidemic of sore throat at Craigmore, which was referred
to milk infection. The number of cases was 80. The disease
manifested chiefly in the form of severe sore throat, but in a
number of the cases erysipelas developed in addition. The milk
appears to have been infected by two milkmaids who were suffer-
ing from sore throat. Those attacked most severely had drunk
most of the implicated milk. A dog and cat which had a good

* Report of Medical Officer of Local Government Board, 1875, vol. vii., p. 80.

f St Bartholomew's Hospital Reports, vol. xx.

t Brit. Med. Jour., 1881, vol. i., p. 657; vol. ii., p. 415.

§ Practitioner, 1884, vol. i., p. 467 (Dr M. K. Robertson).

220 BACTERIA IN MILK AND MILK PRODUCTS

deal of the milk became very ill with " severe inflammation of the
throat." *

In 1892 there was an extensive outbreak of sore throat in Upper
Clapton. Dr King Warry, describing the symptoms in the
Practitioner, at the time held that the pathological condition was
scarlet fever in a mild form. His reasons for this view were three :

(a) scarlet fever attacked one member of a family, and the sore
throat disease other members who had previously had scarlet fever ;

(b) both scarlet fever and sore throat patients were isolated together,
but those suffering from sore throats did not contract the scarlet
fever ; and (c) some of the patients suffering from sore throat had
at the same time certain symptoms simulating scarlet fever, such
as desquamation of the skin, kidney disease, and rheumatic symptoms.
"With this view of the specificity of throat illness under similar
circumstances Dr Parsons agrees.f In the Upton and Macclesfield
scarlet-fever outbreak of 1889, there were 40 cases of sore throat
which were apparently related to scarlet fever for the following
reasons: — (a) The sore throat occurred in older persons in whom
rashes are less prone to occur, and who had had scarlet fever ; (b)
in some cases there was skin desquamation ; (c) when the children
suffered from scarlet fever the adults in the same house suffered
from sore throat; (d) two cases of diphtheria at the same period
showed symptoms simulating scarlet fever; and (c) pyrexia and
delirium were present in the worst cases.

Two comparatively small milk-borne outbreaks of "follicular
tonsillitis" were reported in 1897, one in Anglesey J (15 cases), and
the other at Surbiton § (30 cases). The milk was bacteriologically
examined, and Staphylococcus pyogenes and Streptococcus pyogenes were
found, but no B. diphtheria:. Bacteriological examinations of the
patients' throats yielded precisely similar results. A man whose
business it was to milk the cows was found to be out of health, with
well-marked tonsillitis and suppurating whitlows on both hands.

In April and May 1900 an outbreak of septic sore throat occurred
in North Hackney affecting 151 persons in eighty-eight households,
85 per cent, of which were supplied by one milkman.

A sore-throat outbreak at Brighton in November 1901 was
investigated by Dr Newsholme. Out of a total of 29 persons in a
private girls' school, 18 were affected. The chief symptoms were
high temperature, rapid pulse, tonsillitis with fibrinous exudation
locally except on the soft palate. In two cases there was an
evanescent rash. Dr Newsholme was able, after minute inquiry, to

* Glasgow Med. Jour., 1890, vol. xxxiv., pp. 241-258.

f Report to Local Government Board, 1889.

I Brit. Med. Jour., 1897, vol. ii., p. 339 (Dr C. Grey-Edwards of Beaumauris).

§ Annual Report of Medical Officer of Health, 1897 (Dr Coleman).

MILK-BORNE SORE THROAT 221

trace the cases at the school to one of their number, who had come
into the way of infection derived from a milk supply contaminated
by infectious disease in three families connected with the dairy.*

In 1902 an outbreak of milk-borne sore throat occurred at
Lincoln (199 cases). Of the total 168 or 85 per cent, had consumed
the suspected milk. The outbreak commenced suddenly, lasted a
few days, and then suddenly terminated. The majority of the
victims were adults or persons over twelve years of age. Females
were much more affected than males. The symptoms of the disease
simulated scarlet fever. There was marked sore throat and swelling
of the tonsils, which were in many cases furred. On the third or
fourth day of the disease there was enlargement of the cervical
glands, rash (like rotheln), and fever. The commonest complications
were gastritis and rheumatism, but there were a number of irregular
conditions and varieties of rash. The poison in the milk seems to
have existed in the highest degree in the cream, and Klein isolated
a yeast which he considers related to a yeast known heretofore to
have been associated with throat illness and thrush. It has been
suggested that some relationship, may exist between this yeast and
the spores of rusts, smuts, and mushroom fungi consumed by the
cows. The whole circumstances of the case of this outbreak furnish
one of the most interesting modern chapters in milk epidemiology, j"
In 1903 another outbreak occurred (56 cases) at Lincoln of a somewhat
similar kind. In 1902 an outbreak occurred at Bedford (42 cases)
consisting of sore throat, malaise, headache, giddiness, etc. Here also
the cream seemed more infective than the milk. Indeed, in several
families only those who had taken the cream suffered. The incidence
was chiefly upon young adults.J

A somewhat similar outbreak occurred in October and November
1903, at Woking, in which persons were infected in ninety-eight
different houses. The illness was sore throat, with glandular enlarge-
ment and general symptoms. Of the ninety-eight households affected,
seventy-six obtained their milk directly from a source open to
criticism. Dr Pierce, the medical officer of health, examined four cows
yielding the milk, and a bacteriological examination was made of the
milk. In the result it was found that two of the cows suffered from
suppurative mammitis, and the liquid yielded by these two cows
" consisted of the most part of pus such as would be contained in an
abscess." This was the character of the milk which had been con-
sumed by the persons suffering from the illness. §

* Jour, of Hygiene, 1902, vol. ii., pp. 150-169. Annual Report of Medical Officer
of Health of Brighton, 1901.

t Report to 'the Local Government Board, No. 190, Oct. 1903 (Dr L. W. Darra
Mair).

% Report of Medical Officer of Bedfordshire County Council, 1902, pp. 60-62.

§ Brit. Med. Jour., 1903, ii. p. 1492.

222 BACTERIA IN MILK AND MILK PRODUCTS

The most recent sore-throat outbreak due to consumption of
infected milk occurred in Finchley in 1904. Some 500 cases came
to the knowledge of the medical officer of health (Prof. Kenwood)
of the district, and another fifty occurred in the outlying neighbour-
hood. The incubation period was twenty-four to forty-eight hours,
followed by enlargement of submaxillary glands, sore throat, fever,
and general malaise. In a few cases there was a measles-like eruption
on the lower limbs. Professor Kenwood formed the opinion that the
epidemic was due to disease in the cows furnishing the milk, but no
specific organism was discovered.* A somewhat similar outbreak
occurred in the same district in 1894.

Pus in Milk. — It may here be stated that not infrequently milk
contains pus cells, and there can be little doubt that such milk
may set up disease in persons consuming it.

Stokes and Wegefarth made an inquiry into the subject some
years ago, counting the number of pus cells in the field of the micro-
scope in milk from cows kept under various conditions of insanitation.
Taking one pus cell in the field as a standard, Stokes and Wegefarth
found 25 per cent, of the milks of country cows, kept under good
conditions, and 88 per cent, of town cows, kept under bad con-
ditions, contained pus cells. Eastes, who made an examination of
186 London milks, found pus cells present in 30 per cent., muco-pus
in 48*7 per cent., and streptococci in 75 '2 per cent.f An in-
vestigation of milk in St Pancras in 1899 yielded 24 per cent, of
samples containing pus cells. \ Foulerton, examining a series of
milks from Finsbury in 1903, found pus and allied cells in 32 per cent,
of them, staphylococci in 28 per cent, and streptococci in 32 per
cent. Forty per cent, of the samples examined contained "foreign
dirt."§ Mucous threads are commonly found in milk containing pus.
Such threads probably consist of nucleo-albumin, and when occurring
with pus cells, the condition of " muco-pus " is present. This is held
to indicate inflammatory lesion of the ducts of the udder, and not
abscess formation in the substance of the gland. Blood corpuscles
are not rare in milk, particularly soon after lactation. The last and
least important kind of cell is that of the epithelium. Such scales
may be derived, either from the hand of the milker or from the
teats of the udder. Epithelial cells are large and nucleated. Milk
containing many blood cells, mucous threads, and leucocytes, and
milk containing any pus cells, should be looked upon as unfit for
human consumption. Eastes, Hoist, Mven, Stokes, Bergey, Hirsch,
and others, have drawn attention to the ill effects which streptococcal

* Special Report of Medical Officer of Stoke Newington, 1904.
f Brit. Med. Jour., 1899, vol. ii., p. 1342.
X Report on Health of St Pancras, 1899, pp. 61-66 (Dr Sykes).
§ Report on Milk Supply of Finsbury, 1903, p. 44.

MILK-BORNE DIARRHCEA 223

milk has upon persons consuming it. In the main these are twofold,
namely, gastro-intestinal diseases and sore throats. The evidence
implicating streptococcal milk is empirical and circumstantial, and
yet it appears to be growing in force and volume. On the other
hand, streptococcus has been found in the fresh milk derived from
healthy udders (Eeecl and Ward).

Milk-borne Cholera

The cholera bacillus is unable to live in an acid medium. Hence
its life in milk is a limited one, and generally depends upon some
alkaline change in the milk. Heim found that the organism of
cholera would live in raw milk from one to four days, depending
upon the temperature. D. D. Cunningham, from the results of a
large number of investigations in India, concludes that the rapidly
developing acid fermentations normally or usually setting in, con-
nected with the rapid multiplication of other common bacteria and
moulds, tend to arrest the multiplication of cholera bacilli, and
eventually to destroy their vitality. Boiling milk appears, on the
contrary, to increase the suitability of the milk as a nidus for
cholera bacilli, partly by its germicidal effect upon the acid -producing
microbes, and partly that it removes from the milk the enormous
numbers of common bacteria, which in raw milk cause such keen
competition that the cholera bacillus finds existence impossible.

Professor W. J. Simpson, sometime the Medical Officer of Health
for Calcutta, has placed on record an interesting series of cholera
cases on board the Ardenclutha, in the port of Calcutta, which arose
from drinking milk which had been polluted with one quarter of its
volume of cholera-infected water. This water came from a tank
into which some cholera dejecta had passed. Of the ten men who
drank the milk, four died, five were severely ill, and one, who drank
but very little of the milk, was only slightly ill. There was no
illness whatever among those who did not drink the milk. In 1894,
a milk-borne outbreak of cholera occurred in the Gaya Gaol, affecting
some twenty-six persons.

Milk-borne Epidemic Diarrhoea

In 1892, Gaffky recorded an instance in which three men con-
nected with the Hygienic Institute at Giessen were suddenly taken
ill. They had chills, fever, diarrhoea, and general symptoms. The
only article of diet of which they had all partaken was milk, which
was traced to a cow suffering from enteritis. The milk of this cow as
it left the udder contained no bacteria. But bacteria gained access
during the milking from the dried particles of faecal matter on the

224 BACTERIA IN MILK AND MILK PRODUCTS

posterior portion of the udder. In these particles was found a
bacillus which proved very pathogenic for mice and guinea-pigs, and
which corresponded to an organism isolated from the stools of the
patients.*

In 1894 an outbreak occurred at Manchester,*)- characterised by
diarrhsea, sickness, and abdominal pains. The cases numbered 160
in forty-seven houses, or just 50 per cent, of the houses served by one
and the same milk-seller. Eaw-milk drinkers were the chief sufferers,
and those not drinking the implicated milk did not suffer. Dr
Niven visited the farm whence the milk came, and found that it
was the milk from the farm itself, and not the added milk from a
more distant farm, which supplemented the farmer's stock that had
caused the epidemic, the home-farm milk only being sent into the
affected district. Near the farm were 40,000 tons of privy-midden
refuse. Two streams ran near the farm, meeting below, one fouled
by the drainage of the tip, and the other being contaminated with
sewage and with matter from a tripe-boiling place. The water used
for washing the milk-pails was tepid, and kept in a foul cistern.
The cows drank from a pool which received the drainage from the
cowshed midden. The stored milk could be readily contaminated
from emanations from the cowshed. Professor Delepine examined
the milk, and found B. coli communis abundantly present, and Dr
Niven elicited the fact that a cow affected with inflamed udder
(" garget ") had been removed from the farm and slaughtered. The
outbreak was attributed to milk in any case, and to the probable
infection of it by the diseased cow. But Delepine has pointed out
that it is more probable that the milk was contaminated with fsecal
pollution rather than infectious disease of the cow.J

In 1895 § and 1898 1| three outbreaks of epidemic diarrhoea
occurred amongst the patients at St Bartholomew's Hospital, London,
traceable in the first two instances to milk, and in the third, to rice
pudding made with milk.1T On Sunday night, 27th October 1895,
an outbreak of diarrhoea affected 59 in-patients, all of whom had
recently taken milk, and from the evacuations the spores of B. enteri-
tidis sporogenes was isolated by JQein. The patients suffered quite
irrespective of whether or not the milk had been boiled. Some
milk also, derived from the same source as the milk which had
caused the poisoning, was examined by Klein, and found to contain
the spores of the same organism. On Sunday, 6th March 1898, a
second outbreak of severe diarrhoea occurred in this hospital, affect-

* Deut. Med. Woch., vol. xviii., p. 14.

f Annual Report of Medical Officer of Health of Manchester, 1894 (Dr Niven).
% Jour, of Hygiene, 1903, vol. hi., No. 1, pp. 76, 77.

§ Report of the Medical Officer of Local Government Board, 1895-96, pp. 197-204.
|| Ibid., 1897-98, p. 235.
H Ibid., 1898-99, p. 336. Lancet, 7th January 1899.

MILK-BORNE DIARRHOEA 225

ing 146 patients, and there was evidence on this occasion also that
the medium of infection had been milk. On 5th August 1898, a
third outbreak affecting 84 patients and 2 nurses took place at the
same hospital, the vehicle of infection in this instance being some
rice pudding made with milk, also said to contain an organism
similar or identical with the B. enteritidis sporogenes. There can be
no doubt that milk was the agent of infection in each of these three
outbreaks. It was in these outbreaks that the B. enteritidis sporogenes
of Klein was isolated and held to be the specific organism. Dr Klein
has produced evidence in behalf of this bacillus being the true cause
of epidemic diarrhoea.*

Other similar outbreaks are on record traceable to contaminated

milk. Nor is the evidence on this subject derived only from epidemics.

Newsholme has shown that of the fatal cases of diarrhoea in children

only 9-4 per cent, occur in children which have been breast-fed.f

In Fmsbury 20 per cent., in Kensington 35 per cent., and in Croydon

12 per cent., were breast-fed. From such figures it is evident that

most of the deaths of infants from diarrhoea occur in children who

have been hand-fed. This fact is now widely accepted. In one of his

official reports+ Dr Hope, of Liverpool, states that "the method of

feeding plays a most important part in the causation of diarrhoea :

when artificial feeding becomes necessary, the most scrupulous

attention should be paid to feeding-bottles." Careless feeding, in

conjunction with a warm, dry summer, invariably results in a high

death-rate from this cause. These two causes interact upon each

other A warm temperature is a favourable temperature for the

growth of the poisonous micro-organism; a dry season affords

ample opportunity for its conveyance through the air. Unclean

feeding-bottles are obviously an admirable nidus for these injurious

bacteria, for m such a resting-place the three main conditions

necessary for bacterial life are well fulfilled, viz., heat, moisture, and

pabulum. The heat is supplied by the warm temperature, the

moisture and food by the dregs of milk left in the bottle; and the

dry air ot summer assists in transit. It becomes clear that diarrhoea

is, m part at all events, due to polluted milk, polluted by dust or

flies directly or indirectly, at the farm or in the home

Delepme has urged that milk is infected at the farm or in transit,
as many of the milks which he examined and proved to be virulent
had not been exposed to any influence attributable to a consumer's
home, but were in fact infective before they reached the consumer.§

1898-99?°^ °fMedical0^Ger°f Local Government Board, 1895-96, 1896-97, 1897-98,
f Report on Health of Brighton, 1902, p. 50
X Report of Health of Liverpool, 1899, p 11
§ Jour, of Hygiene, 1903, p. 86.

226 BACTERIA IN MILK AND MILK PRODUCTS

He considers the injurious properties of such milk is due to faecal
pollution and the action of B. coli (in particular those coliform
bacilli which produce a large amount of acid and do not coagulate
milk). Newsholme considers such contamination may be responsible
for setting up epidemics of diarrhoea occurring in connection with
a particular milk supply, as in the analogous case of epidemics of
infectious diseases, such as typhoid. But he holds that the ordinary
sporadic cases of diarrhoea, which carry off single children in large
numbers in urban districts, are clue " chiefly to domestic infection
of milk or other foods, or to direct swallowing of infective dust."*
Probably, we have a double pollution of milk in actual practice,
one originating at the farm, one brought about subsequently. The
latter may be produced by flies, or from manure heaps (Waldo), or
from dust in roads and yards of towns (Richards), or from the
generally filthy manipulation of the milk from the time when it
becomes the property of the milk -seller to the moment of con-
sumption. It should not be forgotten in this relation that stale
milk contains toxic properties altogether apart from, and in addition
to, actual bacteria. It is possible that the products of organismal
action have a much greater effect in the causation of diarrhoea than
is generally supposed.f

Preventive Measures

It is not possible in the present state of our knowledge in respect
of milk bacteriology to lay down very exact limits as to what is, and
what is not, unsatisfactory milk. A numerical standard of contained
organisms is not practicable at present. But we think, it may be said,
that, in any case, milk should not be considered as marketable if it
contains (a) numerous pus cells; (b) pathogenic organisms; or (c)
"organisms of indication" of contamination. The presence of vast
numbers of bacteria, such as millions per cubic centimetre, also
indicates unclean manipulation.

1. Among the preventive measures which these conditions indi-
cate, cleanliness of cows, dairy-hands, and milk-cans or other milk
vessels, stands first in importance. Refrigeration of the milk, being
more easily obtained than cleanliness, should be insisted upon without
delay. Similar measures are also needed with regard to all things
or persons coming in contact with the milk. Absolute bacterial
cleanliness is most difficult to obtain, if not practically impossible.
Occasional infection must, therefore, occur.

2. To guard against the effects of accidental fsecal infection, milk
should be consumed fresh, when possible. When it cannot be con-

* Report on Health of Brighton, 1902, p. 50.

f For a fuller discussion of the whole question of the disease-producing power
of milk, see Bacteriology of Milk, 1903, pp. 210-391.

CONTROL OF THE MILK SUPPLY 227

sumed fresh it should be refrigerated, i.e. kept at a temperature
below 4° C, for this inhibits the rapid multiplication of bacteria.
When milk cannot be used fresh or refrigerated, it should be sterilised
by heat.

3. Greater domestic and municipal cleanliness is an essential
requirement. This must include the cleanly preparation of food,
and particularly the handling and storage of milk; the cleansing
of milk-bottles; reduction of dust in houses, courts, and streets,
and protection of milk from dust in shops and houses ; _ impervious
roads and paving; and the substitution of wet-cleansing for dry
cleansing, and frequent cleansing for infrequent.

4. Lastly, there is needed "a crusade against the domestic fly,
which is most numerous at the seasons and in the years when
epidemic diarrhoea is most prevalent, and probably plays a large
part in spreading infection " (Newsholme).

Methods of Protecting and Purifying Milk

After the consideration of the somewhat extensive category of
diseases which may be milk-borne, it will be desirable to make brief
reference to some of the means at our disposal for obtaining and
preserving good, pure milk.

We considered at the commencement of this chapter the most
frequent channels of contamination. If these be avoided or pre-
vented, and if the milk be derived from cows in good health and
well kept, the risk of infection is reduced to a minimum. The two
things necessary are clean, healthy cows and clean methods of milking
and manipulation. What the Danes can do, other dairy workers
can do. The cow byre, the udder, the milk vessel,* and the milkers
should each be thoroughly clean.f But we have seen that much,

* Probably the best method of cleansing dairy utensils is by using steam or
boiling water and soda. The advantage of boiling water is obvious. The addition
of soda enhanees its value, as the soda unites with the lactic acid present, forming a
soluble lactate of soda, and also with grease, a fat forming an easily soluble soap.
Nor does it injure or rust the metal with which it comes into contact.

t It may be well to add in a footnote an account of the Danish method as
carried out in England, for it illustrates in concrete form the practical way of
reducing pollution of milk to a minimum : —

The principles and practice of the Copenhagen Milk Supply Company have
been introduced into England, and arc being carried out by Mr C. W. Sorensen
at the White Rose Dairy, West Huntington, York. Mr Sorensen is a nephew of
Mr Busck, of the Copenhagen Company, and has been trained in the Danish
methods. His dairy farm at York is carried on in a similar manner to the
Copenhagen Company's work, with this difference, that whilst the latter obtain
their milk from contributory farms, Mr Sorensen works his own farm, and the
control and management of the cows is under his direct and immediate supervision.
The writer had an opportunity recently of visiting this dairy farm near York, and
a brief description of the most important points may be added here.

1. The health of the cows is secured by a special monthly inspection by the

228 BACTERIA IN MILK AND MILK PRODUCTS

if not most, of the pollution of milk arises after the milking process
and during transit and storage preparatory to use. Bacteria are
so ubiquitous that to prevent the entrance of any at all is futile.
It is, therefore, well to bear in mind the extreme importance of
careful straining and immediate cooling. Straining or screening milk
removes the grosser particles of dust, dirt, hairs, etc., and these, it
will be remembered, are the "rafts" and "vehicles" of bacteria.
If they are at once removed therefore, many bacteria will be removed
with them.*

York Corporation Veterinary Officer, Mr William Fawdington, M. R. C. V. S. , who
has authority to order the disposal of any unhealthy or even suspected animal,
and whose reputation and experience affords a guarantee of efficiency in this
important point. There are about 50 cows in all, 10 of which are Jerseys. The
feeding of the cows is scientifically carried out. No brewers' grains, turnip-tops,
or other unsuitable foods are used, and especial care is exercised in the selection
and feeding of the cows supplying " Table or Nursery Milk," so as to maintain a
high standard of richness and flavour. To ensure an abundant supply of pure
water for the cows to drink, as well as for cleansing purposes, the farm has been
connected with the York City Water Supply, which is provided in a continuous
trough at the head of the stalls. The cleanliness and ventilation of the cow-houses
receives special attention, and is in every way excellent.

2. While no money has been wasted on fancy fittings (which make the milk no
better, but simply increase the cost), the proprietor's aim has been to keep every-
thing sweet and clean from the cows themselves down to the smallest utensil. A
high-pressure boiler has been put in for sterilising all utensils, cans, etc., with
steam.

The udders of the cows are cleansed before milking. The milkers are clothed
in over-alls, and wash before, and if necessary, during milking. The operation of
milking is carried out under cleanly conditions and with clean utensils. After
milking the milk is strained by a " Ulax" strainer.

3. Immediately after the milk is strained, prompt and efficient cooling is obtained
by allowing it to flow in a thin layer over a corrugated copper cylinder, inside which
cold water and ice are passed, thus reducing the temperature in a few seconds to a
point at which germ life cannot develop. Clean milk, so treated, needs no " pre-
servation." If kept in a cool place it will remain perfectly sweet for several days,
even in the hottest weather. Therefore, no preservation or sterilisation is necessary.

4. The usual practice of slopping milk about from one can to another in the
street — exposed to contamination from clouds of dust, the not always clean hands,
or, in wet weather, the dripping garments of the driver — is one so objectionable
that only long usage and the absence of anything better has made it tolerated.
The ideal system, without doubt, is delivery in glass bottles, filled and sealed at
the dairy, and placed straight on the table without the intervention of jugs, basins,
or what not. Next comes delivery in cans, likewise filled and sealed at the dairy.
After that comes the system of drawing the milk by tap from a sealed can, which,
however, is much preferable to the plan of dipping into an open can. The entire
system at this dairy farm is so arranged as to supply a clean whole milk from
healthy cows kept under hygienic conditions, and protected from dust and infection.
This desirable object is attained by (a) clean milking, (b) straining, (c) cooling, and
(d) bottling promptly, efficiently, and at the dairy farm. On the whole, Mr
Sorensen's methods appear to represent the high tide of dairy farm work in England.
But nothing is done by him which could not be done by every dairy farmer in the
country.

* One of the most satisfactory strainers in the market is that known as the
"Ulax." This apparatus consists of a metal sieve through which the milk is first
passed. Then a finer double sieve with a thin layer of sterilised cotton-wool placed
between the two metal sieves acts as a secondary filter (see Fig. 23).

CONTROL OF THE MILK SUPPLY

229

Low temperatures, it is true, do not easily destroy life, but they
have a most beneficial effect upon the keeping quality of milk. It
has been suggested that at the outset of the process of cooling,
currents of air, inimical to bacteria, are started in the milk. If,
however, the temperature be lowered sufficiently, the contained
bacteria become inactive and torpid, and eventually are unable to
multiply or produce their characteristic fermentations. At about
50° F. (10° C.) the activity ceases, and at temperatures of 45° F.
(7° C.) and 39° F. (4° C.) organisms are practically deprived of their
injurious powers. If it happens that the milk is to be conveyed
long distances, then even a lower temperature is desirable. The
most important point with regard to the cooling of milk is that it
should take place immediately. Various kinds of apparatus are

Pig. 23.~"Ulax" Filter.

effective in accomplishing this. Perhaps those best known are
Lawrence's cooler and Pfeiffer's cooler, the advantage of the latter
being that during the process the milk is not exposed to the air.
It must not be forgotten that cooling processes are not sterilising
processes. _ They do not necessarily kill bacteria; they only
inhibit activity, and under favourable circumstances torpid pathogenic
bacteria may again acquire their injurious faculties. Hence, during
the cooling of milk greater care must be taken to prevent aerial
contamination than is necessary during the process of sterilising
milk. No cooling whatever should be attempted in the stable;
but, on the other hand, there should be no delay. Climate makes
little or no difference to the practical desirability of cooling milk,
yet it is obvious that less cooling will be required in the cold season.
The final treatment of milk lias in practice comprised the addition
of preservatives, filtration, and sterilisation.

230 BACTERIA IN MILK AND MILK PRODUCTS

Preservatives are widely used, especially in town milks. They
do not, as a rule, kill bacteria in milk, but merely stifle them, and
prevent rapid multiplication and increasing acidity. They disguise
the true condition of the milk in which they exist. It is to be feared
that their systematic addition to milk tends to place a premium on
uncleanly and improper dairying. There is evidence, also, to show
that by a cumulative process preservatives may be injurious to
persons consuming the milk. The most commonly used antiseptics
in milk are borax, formalin,* carbonate of soda, and salicylic acid.f

Secondly, it is possible to remove in part the pollution of milk
by filtration. Filtration has been practised for some time by the
Copenhagen Dairy Company, by Bolle, of Berlin, and various milk
companies. The filters used consist of large cylindrical vessels
divided by horizontal perforated diaphragms into five superposed
compartments, of which the middle three are filled with fine sand
of three sizes. At the bottom is the coarsest sand, and at the top
the finest. The milk enters the lowest compartment by a pipe
under gravitation pressure, and is forced upwards, and finally is run
off into an iced cooler, and from that into the distribution cans. By
this means the number of bacteria are reduced to one- third. The
difficulty of drying and sterilising enough sand to admit a large
turnover of milk is a serious one. This, in conjunction with the
belief that filtration removes some of the essential nutritive
elements of milk, has caused the process to be but little adopted.
Dr Seibert states that if milk be filtered through half an inch of
compressed absorbent cotton, seven-eighths of the contained bacteria
will be removed, and a second filtration will further reduce the
number to one-twentieth. One quart of milk may thus be filtered
in fifteen minutes.

* S. Rideal and A. G. R. Foulerton conclude from a series of experiments that
boric acid (1-2000) and formaldehyde (1-50,000) are effective preservatives for milk
for a period of twenty-four hours, and that these quantities have no appreciable
effect upon digestion or the digestibility of foods preserved by them {Public Health,
1899, pp. 554-568).

f The Departmental Committee on Preservatives and Colouring Mattel's in Food,
1901 (Report, pp. xxiv.-xxv.) recommend: —

1. That the use of Formaldehyde in food and drink is absolutely prohibited, and
the Salicylic Acid be not used in greater proportion than one grain per pint or
pound respectively for liquid or solid food, its presence in all cases to be declared.

2. That the use of any preservatives or colouring matter in milk be made an
offence under the Sale of Food and Drugs Acts.

3. That Boric Acid preservatives only be allowed in cream, the amount not to
exceed 0 *25 per cent. , and be notified on a label.

4. That Boric Acid preservatives only be allowed in butter, the amount not to
exceed 0'5 per cent.

5. That chemical preservatives be prohibited in all dietetic preparations for the
use of children and invalids.

6. That the use of Copper Salts for " greening " be prohibited.

7. That a Court of Reference be established to supervise the use of preservatives
and colouring matters in foods.

CONTROL OF THE MILK SUPPLY 231

Sterilisation and Pasteurisation

Sterilisation means the use of heat at or above boiling-point, or
boiling under pressure. This may be applied in one application
of one to two hours at 212°-250° F., or it may be applied at stated
intervals at a lower temperature. The milk is sterilised — that is to
say, contains no living germs— is altered in chemical composition,
and is also boiled or " cooked," and hence possesses a flavour which
to many people is unpalatable.

Now such a radical alteration is not necessary in order to secure
non-infectious milk. The bacteria causing the diseases conveyable
by milk succumb at much lower temperatures than the boiling-
point. Advantage is taken of this in the process known as
'pasteurisation. By this method the milk is heated to 167-185° F.
(75-85° C). Such a temperature kills harmful microbes, because
75° C. is decidedly above their average thermal death-point, and
yet the physical changes in the milk are practically nil, because
85° C. does not relatively approach the boiling-point. There is no
fixed standard for pasteurisation, except that it must be above the
thermal death-point of pathogenic bacteria, and yet below the
boiling-point. As a matter of fact, 158° F. (70° C.) will kill lactic
acid bacteria as well as most disease-producing organisms found in
milk. If the milk is kept at that temperature for ten or fifteen
minutes, we say it has been " pasteurised." If it has been boiled,
with or without pressure, for half an hour, we say it has been
"sterilised." The only practical difference in the result is that
sterilised milks have a better keeping quality than pasteurised, for
the simple reason that in the latter some living germs have been
unaffected.

Sterilisation may, of course, be carried out in a variety of
modifications of the two chief ways above named. When the
process is to be completed in one event an autoclave is used, in
order to obtain increased pressure and a higher temperature. Milk
so treated is physically changed in greater degree than in the slower
process. The slow or intermittent method is, of course, based on
Tyndall's discovery that actively growing bacteria are more easily
killed than their spores. The first sterilisation kills the bacteria,
but leaves their spores. By the time of the second application the
spores have developed into bacteria, which in turn are killed before
they can sporulate.

The application of sterilisation to milk is now very widely

adopted by corporations, dairy companies, etc. Eecently the writer

has had the opportunity of studying an excellent system in vogue in

Essex,* and which may be mentioned because it seems to emphasise

* J. Carson, Crystalbrook Farm, Theydon Bois, Essex.

232 BACTERIA IN MILK AND MILK PRODUCTS

principles which might be practised all over England. Briefly, it
may be said that Mr Carson's system lays emphasis on five chief
points : —

1. The cows used are carefully selected for milking purposes, are
regularly inspected, and have all been tested with the tuberculin
test. Their feeding is also kept under strict control, no brewers'
grains being used. In summer the cows feed on grass, linseed oil
cake, and a small quantity of cotton cake and bran ; in winter they
have hay, mangolds, maize, germ meal, and linseed and cotton cake.
The farm is well kept, and maintained under strict sanitary control,
the health and cleanliness of the cows being the one thing aimed at.

2. The cows are milked in the byre adjoining the sterilising
plant. Both cows and cowsheds are continually maintained in
cleanliness. The udders are cleansed before milking, and it is required
that milkers also shall be clean in person and management of
milking.

3. Immediately after milking, the milk is removed into an
adjoining room, strained through a metal screen, and at once
separated by an ordinary Laval separator. This separation is
adopted for purposes of purification only. The milk and cream
are again mixed, and poured by means of a mechanical automatic
bottle-filler into bottles.

4. The milk in bottles is then, within a few minutes of leaving
the udder, placed in the steriliser and maintained at 212° F. for
twenty minutes. The bottles have been previously sterilised at
220° F. for sixty minutes.

5. After sterilisation the milk is cooled to 53° F., and kept at
that temperature till required for delivery.

We have examined this milk chemically and bacteriologically,
and have found it to be of excellent quality. It is unquestionably
an advantage to have milk which is to be sterilised brought under
treatment at once, after milking. This cannot always be done, and
hence it is the custom of some dairy companies and institutions to
sterilise milk on its delivery. But it is of extreme importance to
avoid this if practicable. Whatever treatment milk receives, be it
refrigeration or sterilisation, such treatment should be applied
immediately after the milk is drawn from the udder. There are a
large number of appliances and different forms of apparatus now on
the market, having for their object the sterilisation of milk. Our
object has not been the recommendation of any apparatus or process,
but the principles underlying the efficient pasteurisation and ster-
ilisation of milk.

The methods of pasteurisation are continually being modified
and improved, especially in Germany, Denmark, and America.
Most of the variations in apparatus may be classed under two

CONTROL OF THE MILK SUPPLY 233

headings. There are, first, those in which a sheet of milk is allowed
to flow over a surface heated by steam or hot water. This may be a
flat, corrugated surface or a revolving cylinder. The milk is then
passed into coolers. Secondly, milk is pasteurised by being placed
in reservoirs surrounded by an external shell containing hot water or
steam. Dr H. L. Eussell * has described one apparatus consisting of
a pasteuriser, a water cooler, and an ice cooler. The pasteuriser is
heated by hot water in the outside casement. To equalise rapidly
the temperature of the water and milk, a series of agitators must be
used. These are suspended on movable rods, and hang vertically in
the milk and water chambers. By this ingenious arrangement, the
heat is diffused rapidly throughout the whole mass, and as the
temperature of the milk reaches the proper point, the steam is shut
off and the heat of the whole body of water and milk will remain
constant for the proper length of time. The somewhat difficult
problem of drawing off the pasteurised milk from the vat without
reinfecting it by contact with the air is solved by placing a valve
inside the chamber, and by means of a pipe leading the pasteurised
milk directly and rapidly into the coolers. These are of two kinds,
which may be used separately or conjointly. In one set of cylinders
there is cold circulating water, in the other finely-crushed ice.

In England, many methods (including a number of patents) are
in common use where milk is pasteurised. For instance, at the
Hospital for Sick Children in Great Ormond Street, which is in
advance of other London hospitals in this respect, milk is received
from a well-known metropolitan dairy company in quantities of 200
quarts daily, some being delivered in the morning, and a smaller
quantity in the evening. The milk is derived from healthy cows,
and sanitary cowsheds, the farms being placed under strict super-
vision. On receipt, the milk is filtered through muslin, by downward
and upward filtration, and passed directly into a bottle-filling
machine. Clean, stoppered bottles are kept in readiness. When
filled, the bottles are placed in circles in the cage at the bottom of
the pasteuriser. Into the centre of the apparatus is placed the
thermometer. The lid is closed down and clamped, and the steam is
admitted from below. The temperatures used are 160° F. (or 71° C.)
for twenty minutes in winter, and 180° F. (or 82° C.) for twenty
minutes in summer. After the elapse of this period, the lid is
removed, the stoppers of the bottles are fixed down, and hot water
is admitted into the floor of the apparatus. To this hot water is
slowly added cold water, and in about forty minutes the pasteurised
milk has been cooled down, and is ready for use in the wards. The
apparatus is readily cleansed after use, and the various parts, includ-
ing the bottles, stoppers, etc., are cleaned daily. A somewhat

* Report from Wisconsin Agricultural Experiment Station, 1896.

234 BACTERIA IN MILK AND MILK PRODUCTS

similar apparatus is in use by a Health Association at York,* which
has recently started (1903) the York Infants' Milk Depot, after the
manner of the Liverpool and Battersea system. The apparatus pro-
vided for this work is one of the latest construction. It consists of
an ordinary oval cylinder disinfecting chamber, having doors at both
ends. The apparatus is lagged, and with outside steel casing, pro-
vided with a steam distributor inside, steam gauge, safety valve,
thermometers, ready for steam supply from boiler. In connection
with this apparatus there is also provided a convenient size trolley,
upon three wheels, together with a steel frame holding three separate
platforms, which can be taken apart to suit bottles lor vessels of
larger sizes. This frame is mounted also upon wheels running in
grooves, and channels are fitted inside steriliser to correspond. The
steam rises around the bottles from the bottom of the cylinder. The
trolley is fitted for both ends, and when duplicated, a " charge " can
be taken from one end of the apparatus, and a fresh one inserted at
the other. This apparatus can be used as a steriliser or a pasteuriser.

Domestic pasteurisation can be accomplished readily by heating
the milk in vessels in a water-bath raised to the required tempera-
ture for half an hour, or Aymard's milk sterilisers may be used.

Without entering into a long discussion upon the various
pasteurising methods adopted, we may summarise the chief essential
conditions. It need scarcely be said that the operation must be
efficiently conducted, and in such a way as to maintain absolute con-
trol over the time and temperature. The apparatus should be simple
enough to be easily cleaned and sterilised, and economical in use.
Arrangements must always be made to protect the milk from rein-
fection during and after the process. The entire preparation of
pasteurised milk for market may be summed up in four items : —

1. Pasteurisation in heat reservoir.

2. Rapid cooling in water or ice coolers.

3. All cans, pails, bottles, and other utensils to be thoroughly
sterilised in steam before use.

4. The prepared milk to be placed in sterilised bottles, and
sealed up.

The quality of the milk to be pasteurised is an important point.
All milks are not equally suited for this purpose, and those contain-
ing a large quantity of contamination, especially of spores, are
distinctly unsuitable. Such milks, to be purified, must be sterilised.
Dr Eussell has laid down a standard test for the degree of contamina-
tion which may be corrected by pasteurisation by estimating the
degree of acidity, a low acidity {e.g. 0'2 per cent.) usually indicating

* The York Health and Housing Reform Association, established 1901.
Secretary, Miss Hutchinson, 63 Gillygate, York. Apparatus by Wyttenbach : a
central depot in Gillygate, and branch depots elsewhere in the city.

CONTROL OF THE MILK SUPPLY

235

a smaller number of spore-bearing germs than that which contains a
high percentage of acid.

Lastly, while the heating process is, of course, the essential feature
of efficient pasteurisation, it must not be forgotten that rapid and
thorough cooling is almost equally important. As we have seen,
pasteurisation differs from complete sterilisation in that it leaves
behind a certain number of microbes or their spores. Cooling
inhibits the germination and growth of this organismal residue. If
after the heating process the milk is cooled and kept in a refrigerator,
it will probably keep sweet from three to six days, and may do so
for three weeks.

Results of Pasteurisation. — Before leaving this subject, we may
glance for a moment at the bacterial results of pasteurisation and
sterilisation. The two chief of these are the enhanced keeping
quality and the removal of disease-producing germs. The former is
due in part to the latter, and also to the removal of the lactic acid
and other fermentative bacteria. As a general rule, these bacteria
do not produce spores, and hence they are easily annihilated by
pasteurisation. True, a number of indifferent bacteria are untouched,
and also some of the peptonising species. The cooling itself con-
tributes to the increased keeping power of the milk, especially in
transit to the consumer.

Pasteurised milks have the following three economical and com-
mercial advantages over sterilised milks, namely (a) they are more
digestible, (b) the flavour is not altered, and (c) the fat and lact-
albumen are unchanged. Professor Hunter Stewart, of Edinburgh
compiled from a number of experiments the following instructs
and comprehensive table: —

No. of

Experi-
ments.

Average No.
of Microbes
per c.c. in
Milk before
Treatment.

Temperature
and

Duration of
Pasteurisation

m minutes.

5

4

136,262
53,656

10' 60° C.
30' 60° C.

12

78,562

10' 65° C.

12

132,833

30' 65° C.

13
9
2

.'!

1
1

49,867
38,320
77,062
48,250
1,107,000
1,107,000

10' 70° C.
30' 70° C.
10' 75° C.
30' 75° C.
10' 80° C.
30' 80° C.

No. of Microbes

Soluble

Albumen

Soluble
Albumen

per c.c. m

In

Pasteurised Milk
after '-'4 hours.

Fresh Milk

Pasteurised

Milk

per cent.

per cent.

1722 average

0-423

0*418

1 sterile

0-435

0-427

3 averaged ! '.">.">

6 sterile

0-395

0-362

6 averaged 686

9 sterile

0-395

0-333

3 averaged J.;:;

sterile

0-422

0-269

, ,

0-421

0-253

,,

0-38

0-07

,,

0-38

0-05

,,

0-375

o-oo

"

0-375

o-oo

Taste of
Pasteurise! Milk.

Unaffected

Not appreci-
ably affected

Slightly boiled
Boiled

236

BACTERIA IN MILK AND MfLK PRODUCTS

It will be admitted that this table exhibits much in favour of
pasteurisation ; yet the crucial test must ever be the effect upon
pathogenic bacteria. Fliigge has conducted a series of experiments
upon the destruction of bacteria in milk, and he states that a
temperature of 158° F. (70° C.) maintained for thirty minutes will
kill the specific organisms of tubercle, diphtheria, typhoid, and
cholera. MacFadyen and Hewlett have demonstrated,* by sudden
alternate heating and cooling, that 70° C. maintained for half a
minute is generally sufficient to kill suppurative organisms, and such
virulent types of pathogenic bacteria as B. diphtherial, B. typhosus,
and B. tuberculosis.

Eespecting the numerical diminution of microbes brought about
by pasteurisation and sterilisation respectively, we may take the
following series of experiments. Dr H. L. Eussellf tabulates the
immediate results of pasteurisation as follows : —

Number of Micro-organisms per c.c.

Unpasteurised.

Pasteurised.

Minimum.

Maximum.

Average.

Min.

Max.

Av.

Full cream

milk . .

Cream . .

25,300 18,827,000
425,000 32,800,000

1

3,674,000
8,700,000

0
0

37,500
57,000

6,140
24,250

As regards the later effect of the process, he states that in fifteen
samples of pasteurised milk examined from November to December,
nine of them revealed no organisms, or so few that they might
almost be regarded as sterile ; in those samples examined after
January, the lowest number was 100 germs per c.c, while the average
was nearly 5000. With the pasteurised cream a similar condition
was to be observed. Other workers hold that from 95 to 99 per
cent, of all bacteria are removed by pasteurisation.

Summary of Practical Control of Milk Supply

Briefly, it may be said that the requirement is a 'pure milk supply,
that is : —

(1) A clean, whole milk, unsophisticated and without preserva-

tion

* Jenner Institute of Preventive Medicine (First Series Transactions).
f Centralhlatt fur Bakteriologie, etc., Abth. ii.

SPECIALISED MILK 237

(2) To be derived from healthy cows, guaranteed free from
tuberculosis by the tuberculin test, and living under clean and
sanitary conditions ;

(3) To be obtained by clean methods of milking, to be strained,
and to be. protected from contamination by dust or dirt, or from
infection by disease of milker;

(4) To be kept cool by means of refrigeration from the time it
leaves the cow to the time it reaches the consumer, and not to be
exposed to dust or uncleanliness in any way from the vessels in
which it is placed or from the persons by whom it is handled.*

Specialised Milk Supplies for Infants. — The movement for 'the supply of
modified milk for the use of infants, particularly of the artisan class, has now become
a considerable one, both in Europe and America. Broadly speaking, the systems repre-
sented in England are (i.) the Municipal Milk Depot (Liverpool, Battersea, Bradford,
etc.), and (ii.) the Rotch system (Walker-Gordon Laboratories), (i.) There can be
little doubt that this kind of milk supply may be of great service for the children of
the poor, in the reduction of infantile mortality due to the use of contaminated or
infected milk, and in special cases calling for special treatment. It is not, however,
of the nature of control of the milk supply, but rather, of a specialised supply, to meet
special needs. There is evidence to show that at Liverpool, Battersea, and other
places, it has had beneficial results in this special direction. It has, however, several
limitations, unless properly managed. Its object being the saving of life and pre-
vention of infant diseases, it is necessary that the system should be individualised.
Each mother must be separately advised, each infant inspected and weighed periodi-
cally, each home supervised, the condition of the milk regularly tested, and the
source of the milk kept under control, the cows and cowsheds from which the milk
is derived being supervised by a veterinary surgeon and the Medical Officer of
Health. And here, in any event, the quality of the milk used must reach a high
standard, chemically and bacteriologically. If these conditions are not fulfilled, it
would appear that a municipal sterilised milk supply can only be a palliative measure
of transient usefulness. The chief desideratum is a naturally pure milk supply, rather
than an artificially purified and humanised supply. Tlie latter question is one
certainly requiring careful consideration, but of a different nature to the former. If
undertaken by a Local Authority, it would appear desirable to do it very thoroughly,
after the manner of Budin and Variot in Paris, each case being under strict medical
supervision.

A typical municipal milk depot in this country is described in the following
words : —

The milk is supplied by a local dairyman, and arrives in the early morning. It is
guaranteed free from chemical preservatives, and to contain not less than 3*25 per
cent, butter fat, and 8*75 percent, of solids not fat, and cream which must contain not
less than 50 per cent, of butter fat. The milk must be drawn from healthy cows,
stabled and milked under clean and sanitary conditions. Utensils, etc., used must
be thoroughly clean. These and other requirements are set out in the contract. The
first proeess is the modification or Immunisation. Three modifications are employed.
The first contains one part milk to two of water, seven ounces of cream and seven of
lactose being added to each gallon of the mixture. This modification is given to
infants under three months old. The second modification, which is given to infants
between three and six months old, consists of equal parts of milk and water, with
five ounces of cream and lactose added per gallon. The third consists of two parts

:: For details respecting control of milk supply, see Bacteriology of Milk, pp.
452-599 ; also, Report on Health of City of Manchester, 1902 ; Report 'on Milk Supply
of Finsbury, 1903; The Milk Supply of 200 Cities and Towns {U.S.A. Depart, of
Agric, 1903, Bulletin 46); Brit. Med. Jour., 1904, vol. ii., pp. 421-429 (Newman).

238

BACTERIA IN MILK AND MILK PRODUCTS

milk to one of water, with three ounces of cream and lactose added per gallon, and
is given to infants over six months old. ;;

The milk having been modified, it is bottled, and the number of bottles and the
quantities contained are set out below : —

No. of

Amount

1

Amount

No.

Age of Infant.

Bottles

per

per

per day.

Bottle.

day.

1

Below 2 weeks old ....

9

1 \ oz.

13.V oz.

2

Between 2 weeks and 2 months old .

9

2.1 ,,

221 „

3

,, 2 and 3 months old

8

3 ,,

24

4

,, 3 and 4 months old

7

4 „

28

5

,, 4 and 5 months old

7

41 „

31V

6*

,, 5 and 6 months old

7

5 ,,

35

7

,, b' and 8 months old

6

6 „

36

8

„ 8 and 12 months old .

6

7 „

42

'

After bottling, the milk is heated by steam under pressure for some 20-30 minutes,
and kept at a temperature of 212° F. for from five to ten minutes. It is allowed to
cool, and then is supplied to the consumers. Each mother, on first coming to the
depot, is given a leaflet of instructions as to the proper method of using the milk.
The method is very simple. When feeding time arrives, all she has to do is to place
the bottle, unopened, in some warm water till the milk has reached body tempera-
ture. The bottle is then opened, a small teat put on the mouth of the bottle, and
the baby takes its milk from the sterilised bottle direct. The use of the long-tube
feeding-bottle is obviated.

This process of humanisation adapts the milk to the infant's digestive organs, the
sterilisation kills the germs, and as the bottle is not opened — or should not be opened
from the time it enters the steriliser until the infant is ready to take milk from it
direct (no feeding-bottle should be used) — home contamination, unless it is wilful or
due to extreme carelessness, is prevented.

It may be convenient briefly to summarise the chief advantages and disadvantages
of this system : — Advantages. — (1) Infants fed on depot milk receive a modified milk
suited to their digestive capacities. (2) The milk is free from injurious and other
bacteria. (3) The mother is saved labour, and mistakes are prevented, as the
preparation of food for infants, etc., is not necessary if modified and sterilised milk
is obtained ready prepared. (4) Home contamination of milk is avoided. These
four advantages are, in my judgment, of great value. Disadvantages. — (1) The object
being the saving of life and the prevention of infant disease, it is necessary that such
a system should be individualised and placed under direct medical supervision.
Indiscriminate use of such milk is undesirable, and the hand-feeding of infants
requires so much intelligent care that it should not be generally recommended.
(2) There must inevitably result from any success of this system a tendency or risk
for mothers to feed their infants on such milk, instead of nursing them in the natural
way. Therefore, such milk should only be provided for those infants which for some
adequate reason cannot be nurtured on human milk. Such infants are the exception
and not the rule, and it is undesirable to adopt any method which tends to lessen
maternal feeding or maternal responsibility. There are cases, and these not a few,
where depot milk solves the problem of infant feeding, especially in large towns. But
anything which tends in the direction of placing the responsibility of rearing infants
on the municipality is to be deprecated. It is not a question of sentiment, but of
fact, to say that the great need of the present time in respect of this infant problem
is a better home life, more maternal care and feeding rather than less, and a more

* These quantities are of course dependent on the proportions of fat, sugar, and
albuminoid substances originally present in the milk dealt with.

SPECIALISED MILK 239

intelligent, cleanly, and simple mode of rearing infants. Infants should be breast-fed,
and anything which relieves the healthy mother from this duty should not be looked
upon favourably until it has absolutely justified its worth. (3) The cost is at present
in many places prohibitive. (4) The evidence of benefit is not yet of a conclusive
or sufficient nature to form an opinion as to how far the use of depot milk reduces the
infant death-rate. But there can be no doubt that, indirectly, benefit is derived.

Infantile mortality has a definite relationship to (a) the feeding of infants ; (b)
personal care of infants by parents ; (c) housing accommodation ; and (d) certain
meteorological conditions affecting temperature and the dissemination of dust. Other
elements enter into the problem, but, so far as municipal action is concerned, those
are the four main elements. If we can succeed in raising the quality, as regards
purity, of the milk on which infants are fed, we shall at the same time educate and
improve the sense of duty towards their infants on the part of parents. The mischief
lies in polluted milk. The sources of the pollution are not only in unsatisfactory
methods of milking, and in storing and conveying the milk supplied, but also in
dirty domestic conditions, and particularly in carelessness in the use of feeding-
bottles. Successfully to attack, by municipal administration, all the sources of
pollution, is at present impossible, but the ideal of public health administration in
respect of infant feeding is a pure milk supply which needs no sterilisation ; and
towards that end all our efforts should be directed. Modification of such cow's milk
for infant use will still be necessary. Meanwhile we must do the best possible
under existing conditions, and that involves sterilisation, modification, and protection
from home contamination. It is also essential that such a milk supply should be
under medical supervision, and adopted only in suitable cases. Without entering
into unnecessary details, it would appear that there are five possible means of
supplying such a suitable and pure milk for infants : — (1) By means of municipal milk
depots (vide supra) ; (2) by one or more milk-vendors or dairymen undertaking, by
private enterprise, to furnish such modified milk (certified) under medical super-
vision ; (3) by obtaining such a supply from some central institution, company, or
society, such, for example, as the Walker-Gordon Laboratory ; (4) by means of
medical milk commissions, as is done, in part, by the milk commissions established
in the United States of America ; and (5) by means of a voluntary health society
supplying such milk under necessary supervision and control of sources and usage,
as is done, in part, by the York Health and Housing Association.

The essential points requiring attention are such modification of the cow's milk
as will make it, like human milk, suitable for infant consumption, absolute control
of its source and handling prior to its modification, and prevention of home contami-
nation by delivery in sealed bottles. At present it would also be necessary to
pasteurise or sterilise such milk.

(ii.) The Rotch system was introduced by Dr Thomas Morgan Rotch of Harvard
University, and is now in operation in some eighteen or twenty cities in the United
States and Canada, and has also a centre in London. The system, which has for
its object the betterment of infant feeding, consists in controlling the milk supply by
controlling the farms, and establishing a chain of protection from the time the milk
leaves the cow until it arrives at the mouth of the infant. But in addition to this
scheme of protection, there is also combined with it a scheme of modification of the
milk to make it meet more exactly the requirements of infant feeding. The two-
fold function of the Rotch system may be briefly referred to : — (a) Prof eel ion of llir.
Milk. — With this object in view Rotch made a number of recommendations similar
to those laid down by various Milk Commissions, which latter indeed took many of
Rotch's proposals for their model. At the farms supplying milk under this system,
the breed of the cow and its food are matters which receive primary attention. In
America the Holstein has been found to be the best for its adaptability for infant
feeding. The cow itself must be regularly and wisely fed on the basis to which
reference has been made. There is regular grooming and good housing. The cow-
house has cemented walls, ceilings, and floors, and is properly drained and
frequently cleansed. A most careful supervision of the cow's health is maintained,
and if in any way abnormal, the cow is isolated until in normal health. Careful
tuberculin testing is made of each cow used, and the milk of each cow also under-
goes microscopical examination for the purposes of detecting pus cells, colostrum

240 BACTERIA IN MILK AND MILK PRODUCTS

cells, bacteria, etc. The milkers are under strict medical supervision, and regula-
tions are enforced in respect of " cleanliness." Cows are milked in their own stalls,
but immediately after milking the milk is taken in closely-covered vessels to the
milk-room, where it is cooled and screened. The milk-room is a specially prepared
chamber, having smooth surfaces of polished cement, and specially constructed
ventilators with cold-water sprays to moisten the air and prevent dust gaining
access, asepsis being the requirement. The milk is now ready for the laboratory.
(/>) Modification. — The object of the Walker-Gordon Laboratories is first to insure
and distribute the naturally pure milk; and secondly, to provide a place where
different combinations of milk may be put up, according to the prescriptions of
medical men, with accuracy, and under such conditions of cleanliness and asepsis as
to insure the best possible food for infant feeding. The necessity for modification
arises from two facts, namely, that milk varies in constituent percentages, and to
obtain a regular and uniform constitution, modification is necessary ; and secondly,
some children require, for one reason or another, a milk containing certain per-
centages of the various constituents. Thus the patient can receive on the physician's
order a mixture of the percentages called for, made up of either separated cream or
gravity cream, separated milk or whole milk. Twelve years' experience in Boston,
U.S.A., seem to indicate the practicability of this system in preventing the summer
diarrhoea of infants due to contaminated milk.

In addition to these two types, there are various similar methods in vogue, each
of which has its points of advantage. *

Bacteria in Milk Products

Cream is generally richer in bacteria than milk. Set cream
contains more bacteria than separated cream, but germs are abundant
in both. The number of organisms found in cream is enormous.
Probably no other natural medium contains more. We have
frequently examined fresh cream in the country, and found it to
contain more than 100,000,000 bacteria per c.c. It is not only a
favourable medium. It is the filter, so to speak, of milk. For, as
the cream rises, the milk parts with more than 90 per cent, of its
contained bacteria. Conn and Estenf found 110,000,000 of bacteria
per c.c. in unripened cream (average of four examinations) and
284,000,000 in the same cream ripened (average of four examinations).
Cream obtained from a creamery gave an average on eight examina-
tions of 56,000,000 organisms per c.c. unripened, and 350,000,000
organisms per c.c. ripened. Other examples of unripened cream
averaged more than 90,000,000, but when ripened averaged over
300,000,000. Normally ripened cream probably averages four or
five hundred millions of bacteria per c.c, which is greatly in excess
of any other natural media. The number of organisms in unripened
cream varies widely.

The most characteristic feature of cream-ripening is the growth of
the acid-producing organisms, chiefly B. acidi lactici, and the decline
of the liquefying and extraneous organisms. B. acidi lactici is found
in very small numbers in fresh milk, as we have already pointed out,

* For full account, see Jour, of Hygiene, 1904, p. 329 (McCleary).
t Thirteenth Annual Report of the Stow's Agricultural Expt. Sta., Connecticut,
1900, pp. 13-33.

BUTTER AND CHEESE 241

and also in cream. But as the ripening process proceeds with
uniform regularity, the numbers of this organism rapidly increase.
Kollin Burr considers these lactic bacteria gain access to the milk
from outside sources* Buttermilk and whey vary much in their
bacterial content.

Butter necessarily follows the standard of the cream. But, as
the butter fat is not well adapted for bacterial food, the number of
bacteria in butter is usually less than in cream. Butter, when first
made, may contain many million bacteria per gramme. After a
few days only two or three million may be found, and if butter
is examined after it is several months old, it is often found to be
almost free from germs; yet in the intervening period a variety
of conditions are set up directly or indirectly through bacterial
action.f Rancid butter is largely due to organisms. Putrid butter
is caused, according to Jensen, by various putrefactive bacteria, one
form of which is named Bacillus fcetidus lactis. This organism is
killed at a comparatively low temperature, and is, therefore, com-
pletely removed by pasteurisation. Ill-flavoured butter may be due
to germs or an unsuitable diet of the cow and a retention of the bad
quality of the resulting milk. Lardy and oily butters have been
investigated by Storch and Jensen, and traced to bacteria. Lastly,
bitter butter occasionally occurs, and is due to fermentative changes
in the milk, or the presence of acid-producing organisms in the
butter such as B. fluorescens liquefaciens, Oidium lactis, and Clado-
sporium butyri (Jensen). Butter may also contain pathogenic
bacteria, like tubercle. The B. coli can live for a month in butter.
Cheese suffers from, very much the same kind of " diseases " as
butter, except that chromogenic conditions occur more frequently.
Most of the troubles in cheese originate in the milk.} The number
of bacteria in cheese is naturally less than that present in milk or
cream. The closer texture and consistence of cheese, coupled with
the lessened degree of moisture, are sufficient factors to account for
this. Nevertheless, cheese contains a considerable number of organ-
isms. Adametz found that freshly precipitated curd, moulded in
the press and freed from excess of whey, contained between 90,000
and 140,000 micro-organisms per gramme, a comparatively large
number of them having the power of liquefying gelatine, or, in
other words, they possessed a peptonising ferment. During the
period of ripening, the bacterial content of the cheese gradually
rose to 850,000 in Emmenthaler cheese, and 5,600,000 per gramme,

* Thirteenth Annual Report of Storr's Agricultural Expt. Sta., Connecticut,
1900, pp. 66-81; also Centralh. f Bakt.> Abth. ii., 1902, p. 236.

^Keeping Quality of Butter (L. A. Rogers), U.S. Dep. of Agriculture, 1904,
Bull. 57.

X Board of Agriculture Report on Cheddar Cheese Making (F. J. Lloyd), 1899,
pp. 78, 103. '

242 BACTERIA IN MILK AND MILK PRODUCTS

in a soft household cheese. Only a small percentage of these are
of the peptonising species. Tides of organisms occur in cheese, as
in butter and milk.

Method of Examination of Butter and Cheese. — Several grams of
the butter or cheese should be placed in a large test-tube, which is
then two-thirds filled with sterilised water and placed in a water-
bath at about 45° C. until the butter or cheese is melted or " washed."
A small quantity may then be added to gelatine or agar and plated
out on Petri dishes, or in flat-bottomed flasks, in the usual way.
After which the tube may be well shaken and returned to the bath
inverted. In the space of twenty or thirty minutes the butter or
cheese has separated from the water with which it has been
emulsified. It is then placed in the cold to set. The water may be
now either centrifugalised or placed in sedimentation flasks, and the
deposit examined for bacteria.

The Uses of Bacteria in Dairy Produce

In considering the relation of bacteria to milk, we found that
many of the species present were injurious rather than otherwise,
and when we come to consider bacteria in dairy products, like butter
and cheese, we find that the dairyman possesses in them very powerful
allies. Within recent years almost a new industry has arisen owing
to the scientific application of bacteriology to butter and cheese
making.

1. Bacteria in Butter-Making-

As a preliminary to butter-making, the general custom in most
countries is to subject the cream to a process of " ripening." As we
have seen, cream in ordinary dairies and creameries invariably
contains some bacteria, a large number of which are in no sense
injurious. Indeed, it is to these bacteria that the ripening and
flavouring processes are due. They are perfectly consistent with
the production of the best quality of butter. The aroma of butter,
as we know, controls in a large measure its price in the market.
This aroma is due to the decomposing effect upon the constituents
of the butter of the bacteria contained in the cream. In the months
of May and June the variety and number of these types of bacteria
are decidedly greater than in the winter months, and this explains
in part the better quality of the butter at these seasons. As a
result of these ripening bacteria the milk becomes changed and
soured, and slightly curdled. Thus it is rendered more fit for butter-
making, and acquires its pleasant taste and aroma. It is then
churned, after which bacterial action is reduced to a minimum or
absent altogether. Sweet-cream butter lacks the flavour of ripened

BUTTER-MAKING 243

or sour-cream butter. The process is really a fermentation, the
ripening bacteria acting on each and all of the constituents of the
milk, resulting in the production of various bye-products. This
fermentation is a decomposition, and just as we found when dis-
cussing fermentation, so here also the action is only beneficial if
it is stopped at the right moment. If, for example, instead of being
stopped on the second day, it is allowed to continue for a week, the
cream will degenerate and become offensive, and the pleasant ripen-
ing aroma will be changed to the contrary. Speaking generally,
about 25 per cent, of cream bacteria exert a favourable effect on
butter, and 10-15 per cent, a deleterious effect. Many of the former
are acid-producers, and are widely distributed.

Bacteriologists have demonstrated that butters possessing
different flavours have been ripened by different species of bacteria.
Occasionally, one comes across a dairy which seems to be impregnated
with bacteria that improve cream and flavour well. In other cases
the contrary happens, and a dairy becomes impregnated with a
species having deleterious effects upon its butter. Such a species
may be favoured by unclean utensils and dairying, by disease of the
cow, or by a change in the cow's diet. Thus it comes about that the
butter-maker is not always able to depend upon good ripening for
his cream. At other times he gets ripening to occur, but the flavour
is an unpleasant one, and the results correspond. It may be bitter or
tainted, and just as certainly as these flavours develop in the cream,
so is it certain that the butter will suffer. Fortunately, the bacterial
content of the cream is generally either favourable or indifferent in
its action. Thus it comes about that the custom is to allow the
cream simply to ripen, so to speak, of its own accord, in a vat
exposed to the influence of any bacteria which may happen to bo
around. This generally proves satisfactory, but it has the great
disadvantage of being indefinite and uncertain. Occasionally it
turns out wholly unsatisfactory, and results in financial loss. Shortly,
it may be said that cream-ripening assists the making of butter in
four ways : —

1. Churning is easier and more effectual.

2. The yield of butter is increased.

3. The butter has better keeping qualities.

4. The flavour and aroma are more satisfactory.

Control of Ripening Process. — There are various means at our
command for improving the ripening process. Perfect cleanliness
in the entire manipulation necessary in milking and dairying, com-
bined with freedom from disease in the milch cows, will carry us a
long way on the road towards a good cream-ripening. Recently,
however, a new method has been introduced, largely through the

244 BACTERIA IN MILK AND MILK PRODUCTS

work and influence of Professor S torch in Denmark, which is based
upon our new knowledge respecting bacterial action in cream-
ripening. We refer to the artificial processes of ripening set up by
the addition of pure cultures of favourable germs* If a culture of
organisms possessing the faculty of producing in cream a good flavour
be added to the sweet cream, it is clear that advantage will accrue.
This simple plan of starting any special or desired flavour by intro-
ducing the specific micro-organism of that flavour, may be adopted
in two or three different ways. If cream be inoculated with a large,
pure culture of some particular kind of bacteria, this species will
frequently grow so well and so rapidly that it will check the growth
of the other bacteria which were present in the cream at the com-
mencement and before the "starter" was added. That is, perhaps,
the simplest method of adding an artificial culture. But secondly,
it will be apparent to those who have followed us thus far, that if
the cream is previously pasteurised at 70° C, these competing bacteria
will have been mostly or entirely destroyed, and the pure culture,
or " starter," will have the field to itself. There is a third modifica-
tion, which is sometimes termed ripening by natural starters. A
" natural starter " is a certain small quantity of cream taken from a
favourable ripening — from a clean dairy or a good herd — and placed
aside to sour for two days until it is heavily impregnated with the
specific organism which was present in the whole favourable stock
of which the " natural starter " is but a part. It is then added to
the new cream, the favourable ripening of which is desired. Of the
species which produce good flavours in butter, the majority are found
to be members of the acid-producing class ; but probably the flavour
is not dependent upon the acid. The aroma of good ripening is
also probably independent of the acid production.

Artificial Ripening. — Of all the methods of ripening — natural
ripening, the addition of " natural starters," the addition of pure
cultures with or without pasteurisation — there can be no doubt that
pure culture after pasteurisation is the most accurate and reliable.
The use of " natural starters " is a method in the right direction ;
yet it is, after all, a mixed culture, and therefore not uniform in
action. In order to obtain the best results with the addition of
pure cultures, Professor Eussell has made the following recom-
mendations : —

1. The dry powder of the pure culture must be added to a small
amount of milk that has been first pasteurised, in order to develop
an active growth from the dried material.

2. The cream to be ripened must first be pasteurised, in order to

* Such pure cultures for such purposes are in the United States termed
" starters," because they start the process of special ripening. For the sake of
convenience the term will be used here.

BUTTER-MAKING 245

destroy the developing organisms already in it, and thus be prepared
for the addition of the pure culture.

3. The addition of the developing "starter" to the pasteurised
cream, and the holding of the cream at such a temperature as will
readily induce the best development of flavour.

4. The propagation of the " starter " from day to day. A fresh
lot of pasteurised milk should be inoculated daily with some of the pure
culture of the previous day, not with the ripening cream containing
the culture. In this way the purity of the " starter " is maintained for
a considerable length of time. Those " starters " are best which grow
rapidly at a comparatively low temperature (60-75° P.), which produce
a good flavour, and which increase the keeping qualities of the
butter. Now, whilst it is true that the practice of using pure
cultures in this way is becoming more general, very few species have
been isolated which fulfil all the desirable qualities above mentioned.
In America, " starters " are preferred which yield a " high " flavour,
whereas in Danish butter a mild aroma is more common. In this country,
as yet, very little has been done, and that on an experimental scale
rather than a commercial one. In 1891 it appears that only 4 per
cent, of the butter exhibited at the Danish butter exhibitions was
made from pasteurised cream plus a culture " starter " ; but in 1895,
86 per cent, of the butter was so made. Moreover, such butter
obtained the prizes awarded for first-class butter with preferable
flavour. Different cultures will, of course, yield differently flavoured
butter. If we desire, say, a Danish butter, then some species like
" Hansen's Danish starter " would be added ; if we desire an American
butter, we should use a species like that known as " Conn's Bacillus,
No. 41." But whilst these are two common types, they are not the
only suitable and effective " starters." On many farms in England
there are equally good cultures, which, placed under favourable
temperatures in new cream, would immediately commence active
ripening. A good lactic acid culture for dairy use (a) should
sour cream strongly in a short time, (b) should be able to thrive at
low temperatures, and (c) should produce a favourable taste and flavour
in the butter (Jensen).

Professor H. W. Conn, who, with Professor Eussell, has done so
much in America for the advancement of dairy bacteriology, reports
a year's experience with the bacillus to which reference has been
made, and which is termed No. 41.* It was originally obtained
from a specimen of milk from Uruguay, South America, which was
exhibited at the World's Pair in Chicago, and proved the most
successful flavouring and ripening agent among a number of cultures
that were tried. The conclusions arrived at after a considerable
period of testing and experimentation appear to be on the whole

* Report of Starrs Agricultural Expt. Sla., State of Connecticut, 1895.

246 BACTERIA IN MILK AND MILK PRODUCTS

satisfactory. A frequent method of testing has been to divide a
certain quantity of cream into two parts : one part inoculated with
the culture, and the other part left uninoculated. Both have then
been ripened under similar conditions, and churned in the same way ;
the differences have then been noted. It is interesting to know
that, as a result of the year's experience, creameries have been able
to command a price varying from half a cent to two cents a pound
more for the " culture " butters than for the uninoculated butters. The
method advised in using this pure culture is to pasteurise (by heating
at 155° F.) six quarts of cream, and after cooling, to dissolve in this
cream the pellet containing bacillus No. 41. The cream is then set
in a warm place (70° F.), and the bacillus is allowed to grow for two
days, and is then inoculated into twenty-five gallons of ordinary
cream. This is allowed to ripen as usual, and is then used as an
infecting culture, or " starter," in the large cream vats in the pro-
portion of one gallon of infecting culture to twenty-five gallons of
cream, and the whole is ripened at a temperature of about 68° F. for
one day. The cream ripened by this organism needs to be churned
at a little lower temperature (say 52-54° F.), but to be ripened at
a little higher temperature than ordinary cream to produce the best
results. Cream ripened with No. 41 has its keeping power much
increased, and the body or grain of the butter is not affected. More
than 200 creameries in America used this culture during 1895, which
proves that its use for the production of flavour in butter is feasible
in ordinary creameries, and in the hands of ordinary butter -makers,
provided they will use proper methods and discretion. More recently,
pasteurisation has fallen into abeyance, and the use of artificial
cultures is said to have declined in America. In England, with few
exceptions, practically nothing has been done in a commercial way
in the direction of artificial " starters."

2. Bacteria in Cheese-Making

The cases where it has been possible to trace bacterial disease to
the consumption of butter and cheese have been rare. Notwith-
standing this fact, it must not be supposed that therefore cheese
contains few or no bacteria. On the contrary, for the making
of cheese bacteria are not only favourable, but actually essential,
for in its manufacture the casein of the milk has to be separated
from the other products by the use of rennet, and is then col-
lected in large masses and pressed, forming the fresh cheese. In
the course of time this undergoes ripening, which develops the
peculiar flavours characteristic of cheese, and upon which its value
depends.

We have said that the casein is separated by the addition of

CHEESE-MAKING 247

rennet, which has the power of coagulating the casein. But this
precipitation may also be accomplished by allowing acid to develop
in the milk until the casein is precipitated, as in some sour-milk or
cottage cheeses. The former method is, of course, the usual one in
practice. It has been suggested that the bacteria contained in the
rennet exert a considerable influence on the cheese, but this, although
rennet contains bacteria, is hardly established. It is not here, how-
ever, that bacteria really play their role. After this physical separa-
tion, when the cheese is pressed and set aside, is the period for the
commencement of the ripening process.

That bacteria perform the major part of this ripening process,
and are essential to it, is proved by the fact that when they are
either removed or opposed the curing changes immediately cease. If
the milk be first sterilised (Freudenreich), or if antiseptics, like
thymol, be added (Adametz), the results are negative. It is not yet
known whether this ripening process is clue to the influence of a
single organism or not. The probability, however, is that it is to be
ascribed to the action of that group of bacteria known as the lactic
acid organisms. Nor is it yet known whether the peptonisation of
the casein and the production of the flavour are the results of one or
more species. Freudenreich believes them to be due to two different
forms.

However that may be, we meet with at least four common groups
of bacteria more or less constantly present in cheese-ripening, either
in the early or late stages. First, there are the lactic acid bacteria, by
far the largest group, and the one common feature of which is the
production by fermentation of lactic acid ; secondly, there are the
casein-digesting bacteria, present in relatively small numbers ; thirdly,
the gas-producing bacteria, which give to cheese its honeycombed
appearance ; lastly, an indifferent or miscellaneous group of extrane-
ous bacteria, which were in the milk at the outset of cheese-making,
or are intruders from the air or rennet. All these four groups may
bring about a variety of changes, beneficial and otherwise, in the
cheese-making.

Eussell divides the ripening process into three divisions : —

1. Period of Initial Bacterial Decline in Cheese. — Where the green
cheeses were examined immediately after removing from the press,
it was usually found that a diminution in numbers of bacteria had
taken place. This period of decline lasts but a short time, not
beyond the second day. Lower temperature and expulsion of the
whey would account for this general decline in all species of
bacteria.

2. Period of Bacterial Increase.— Soon after the cheese is removed
from the press a most noteworthy change takes place in green cheese.
A very rapid increase of bacteria occurs, confined almost exclusively

248 BACTERIA IN MILK AND MILK PRODUCTS

to the lactic acid group. This commences in green cheese about the
eighth day, and continues more or less for twenty days. In Cheddar
cheese it commences about the fifth clay, reaches its maximum about
the twentieth day, declines rapidly to the thirtieth day, and gradu-
ally for a hundred following days. During the first forty days of
this period the casein-digesting and gas-producing organisms are
present, and at first increasing, but relatively to only a very slight
degree. With this rapid increase in organisms the curd begins to
lose its elastic texture, and before the maximum number of bacteria
is reached the curing is far advanced. Freudenreich has shown that
acid inhibits the growth of the casein-digesting microbes, and vice
versd.

3. Period of Final Bacterial Decline. — The cause of this decline
can only be conjectured, but it is highly probable that it is due to a
general principle to which reference has frequently been made, viz.,
that after a certain time the further growth of any species of
bacteria is prevented by its own products. We may observe that
the gas-producing bacteria in Cheddar cheese last much longer than
the peptonising organisms, for they are still present up to eighty
days. Professor Eussell aptly compares the bacterial vegetation of
cheese with its analogue in a freshly-seeded field. " At first multi-
tudes of weeds appear with the grass. These are the casein-digesting
organisms, while the grass is comparable to the more native lactic
acid flora. In course of time, however, grass, which is the natural
covering of soil, ' drives out ' the weeds, and in cheese a similar con-
dition occurs." In milk the lactic acid bacteria and peptonising
organisms grow together; in ripening cheese the former eliminate
the latter.

Artificial Ripening. — We have seen that the conclusion generally
held respecting these lactic acid bacteria is that they are the main
agents in curing the cheese. Upon this basis a system of pure
" starters " has been adopted, the characteristics of which must be
as follows: — (a) The organism should be a pure lactic-acid-producing
germ, incapable of producing gaseous products ; (b) it should be free
from any undesirable aroma ; (c) it should be especially adapted for
vigorous development in milk. The " starter " may be propagated in
pasteurised or sterilised milk from a pure culture from the labora-
tory. The advantages accruing from the uses of this lactic acid
culture, as compared with cheese made without a culture, are that
with sweet milk it saves time in the process of manufacture ; that
with tainted milk, in which acid develops imperfectly, it is an aid to
the development of a proper amount of acid for a typical Cheddar
cheese ; and that the flavour and quality of such cheese is preferable
to cheese which has not been thus produced. Professor Eussell is of
opinion that the lactic acid organisms are to be credited with greater

CHEESE-MAKING 249

ripening powers than the casein-digesting organisms, but it must not
be forgotten that these two great families of bacteria are still more
or less on trial, and it is not yet possible finally to decide on either
of them. Lloyd holds that though " the greater the number of lactic
acid bacilli in the milk the greater the chance of a good curd," still
" this organism alone will not produce that nutty flavour which is so
much sought after as being the essential characteristic of an excellent
Cheddar cheese."

There are three difficulties to be encountered by dairymen
"starting" a ripening by the addition of a pure culture. First,
there is the initial difficulty of not being able to use pasteurised
milk for cheese, as such milk is uncoagulable by rennet (Lloyd).
Hence it is impossible to avoid some contamination of the milk
previous to the addition of the culture. Secondly, the continual
uncontaminated supply of pure culture is by no means an easy
matter. Thirdly, the maintenance of a low-temperature cellar to
prevent the rapid multiplication of extraneous bacteria will, in
some localities, be a serious difficulty. These difficulties have,
however, not proved insurmountable, and by various workers in
various localities and countries culture-ripening of cheese is being
carried on.*

* As regards the Cheddar cheese industry in this country, Lloyd arrives at the
following five conclusions as a result of investigation : —

1. To make Cheddar cheese of excellent quality, the Bacillus acidi lactici alone
is necessary ; other germs will tend to make the work more rather than less
difficult. Hence scrupulous cleanliness should be a primary consideration of the
cheese-maker.

2. No matter what system of manufacture be adopted, two things are necessary.
One is that the whey be separated from the curd, so that when the curd is ground
it shall contain not less than 40 per cent, of water, and not more than 43 per cent. ;
the other point is that the whey left in the curd shall contain developed in it before
the curd is put in the press at least 1 per cent, of lactic acid if the cheese is required
for sale within four months, and not less than *8 per cent, of lactic acid if the cheese
is to be kept ripening for a longer period.

3. The quality of the cheeses will vary with the quality of the milk from which
they have been made, and proportionately to the amount of fat present in that
milk.

4. " Spongy curd " is produced by at least five organisms, and one of these is
responsible for a disagreeable taint found in curd. They occur in water. Hence
the desirability of securing clean water for all manipulative purposes, and also for
the drinking purposes of the milch cow.

5. The fact that certain bacteria are found in certain localities and dairies is due
more to local conditions than to climatic causes.

It is needless to remark that these conclusions once more emphasise the fact
that strict and continual cleanliness is the one desideratum for bacteriologically
good dairying. That being secured in the cow at the milking, in the transit, and
at the dairy, it is a comparatively simple step, by means of pasteurisation and the
use of good pure cultures of flavouring bacteria, to the successful application of
bacteriology to dairy produce.

250 BACTERIA IN MILK AND MILK PRODUCTS

Abnormal Cheese-Ripening-

Unfortunately, from one cause or another, faulty fermentations
and changes are not infrequently set up. Many of these may be
prevented, being due to lack of cleanliness in the process or in the
milking ; others are due to the gas-producing bacteria being present
in abnormally large numbers. When this occurs we obtain what is
known as " gassy " or inflated cheese, on account of its substance
being split up by innumerable cavities and holes containing carbonic
acid gas or sometimes ammonia or free nitrogen. Some twenty-five
species of micro-organisms have been shown by Adametz to cause
this abnormal swelling. In severe cases of this gaseous fermentation
the product is rendered worthless, and even when less marked the
flavour and value are much impaired. Winter cheese contains more
of this species of bacteria than summer. Acid and salt are both
used to inhibit the action of these gas-producing bacteria and yeasts,
and with excellent results. The character of the gas holes in cheese is
not of import in the differentiation of species. If a few gas bacteria
are present, the holes will be largo and less frequent; if many, the
holes will be small, but numerous. (Swiss cheese having this
characteristic is known as Nissler cheese.) Many of these gas-
producing germs belong to the lactic acid groiq), and are susceptible
to heat. A temperature of 140° F. maintained for fifteen minutes
is fatal to most of them, largely because they do not form
spores.

The sources of the extensive list of bacteria found in cheese are
of course varied, more varied indeed than is the case with milk. For
there are, in addition to the organisms contained in the milk brought
to the cheese factory, the following prolific sources, viz., the vats
and additional apparatus ; the rennet (which itself contains a great
number) ; the water that is used in the manufacture.

In addition to the abnormalities due to gas, there are also other
faulty types. The following chromogenic conditions occur: red
cheese, due to a micrococcus; blue cheese, produced, according to
Vries, by a bacillus ; and black cheese, caused by a copious growth
of low fungi. Bitter cheese is the result of Tyrothrix geniculatus
(Duclaux), or the Micrococcus casei amari of Freudenreich, a closely
allied form of Conn's micrococcus of bitter milk. Sometimes cheese
undergoes a putrefactive decomposition, and becomes more or less
putrid. At other times it becomes " tainted." These latter condi-
tions, like the gassy cheeses, are due to the intrusion of bacteria
from without, or from udder disease of the cow.* Healthy cows,
clean milking, and the introduction of pure cultures, are the methods

* For a discussion on the duration of life of the tubercle bacillus in cheese, see
Nineteenth Ann. Rep. Bureau of Animal Industry, 1902, p. 217.

POISONOUS CHEESE 251

bo be adopted for avoiding "diseases" of eheese and obtaining a
well-flavoured article which will keep.

Finally, there is poisonous cheese which is of more importance to
the public health than all the other abnormal conditions of cheese
put together. In 1883 and 1884 there occurred in Michigan, U.S.A.,
an outbreak of cheese-poisoning. Three hundred persons in all were
affected, and the illness was traced by Professor V. C. Vaughan to a
poisonous ptomaine present in the cheese, and to which he gave the
name tyro-toxicon. It is not improbable that this ptomaine is a
product of bacterial fermentation. It is one of a large class of
substances said to be formed by the action of bacteria upon nitro-
genous compounds. It is unstable, and easily destroyed by the
action of heat and moisture, and even by exposure to the air. Being
present in small quantities only, it has never been isolated in suffici-
ently large quantities to allow of its composition being definitely
determined. Tyro-toxicon has been proved to be a violent poison
both to man and the lower animals. A minute portion consumed
by a child produced sickness and diarrhoea in a manner almost
identical with cholera infantum. (Vaughan). Similar symptoms were
obtained with cats and dogs. Vaughan found that three months are
required for the formation of tyro-toxicon in milk kept in tightly-
stoppered bottles; but under certain circumstances, and in the
presence of butyric fermentation in milk, the poison is produced in
about eight or ten days. Similar, and possibly identical, poisons
occasionally occur in cream, rancid butter and milk {lacto-toxiron and
cliazo-ocnzol). They have the same poisonous effects. Vaughan has
isolated a microbe growing readily on ordinary culture media and
upon fruit and vegetables. This micro-organism, it is considered,
may be the agent producing tyro-toxico?i, but the bacteriology of the
subject has not been worked out.

The writer investigated a similar outbreak due to tyro-toxicon in
Dutch cheese in London in 1901.* Seventeen persons were affected.
The symptoms of illness in all these 17 cases occurred in from two
to eight hours after eating the cheese in question, which came from
the same consignment. Moreover, the symptoms were similar,
namely, epigastric pain, rigors, vomiting, diarrhoea, prostration, and
some fever. The degree of sickness does not appear to have
depended upon the amount of cheese eaten. There was no death
attributed to the poisoning, and in general the symptoms appear to
have passed off in the course of forty-eight hours. A short incuba-
tion period suggests that the poison was "available" in the cheese,
as a product of previous changes, possibly bacterial, set up therein.
A long incubation period between eating the food and symptoms of
poisoning would suggest that the persons affected had consumed, not

; Report on the Public Health of Finsbury, 1901, pp. 110-116.

252 BACTERIA IN MILK AND MILK PRODUCTS

the products of bacteria, but the bacteria themselves, which had then
taken some little time to produce their injurious effects in the
persons eating the food. In the present case the incubation period
was comparatively short, and the acuteness of the symptoms did not
appear to have a direct relationship to the amount of cheese eaten.*

* For a general discussion and bibliography of this subject, see Die Milch
and Hire Bedeutmig fur Volkswirlschaft und Volksgesundheit, Hamburg, 1903,
pp. 345-357.

CHAPTEK VIII

BACTEEIA IN OTHER FOODS

1. Shell-fish : Oysters, Cockles, Clams, and their Relation to Disease ; Symptoms of
Oyster-borne Disease ; Channels of Infection ; Preventive Methods— 2. Meat
Poisoning; Tuberculous Meat— 3. Ice-cream and Ice— 4. Bacterial Infection
of Bread— 5. Miscellaneous Foods, Watercress, etc.

In this chapter the occurrence and significance of bacteria in shell-
fish, meat, ice-cream, and bread will be considered.

1. Shell-fish

Shell-fish have recently claimed the attention of bacteriologists,
owing to the outbreak of typhoid and other epidemics apparently
traceable to oysters.

Oysters. — It was not till 1880 (Cameron) that any substantial
evidence was forthcoming to establish the view, winch had previously
been promulgated (by Pasquier in 1816), that oysters and other
shell-fish might convey the infection of typhoid fever. In 1893
oysters came under the suspicion of Sir Richard Thome Thome
as concerned in the diffusion of scattered cases of cholera in Eng-
land, and he reported on the risks of consuming shell-fish culti-
vated at sewage outfalls.* In the spring of 1894 Dr Newsholme
reported to the Corporation of Brighton the particulars of a number
of cases of typhoid fever which were apparently attributable to the
consumption of oysters obtained from layings grossly contaminated
by sewage. At the end of the same year an outbreak of typhoid
fever occurred at the Wesleyan University, in the State of Con-
necticut, U.S.A., and an investigation was made by Professor H.

* " On Cholera in England in 1893," Local Government Board Report, 1894.

253

254 BACTERIA IN OTHER FOODS

W. Conn, who found that the only channel of infection was the
consumption of oysters served at certain college suppers. These
oysters had been obtained from dealers at Middletown, and had been
cultivated on oyster-beds in the region of a sewage outfall. The
facts briefly were these : Two cases of typhoid fever occurred in a
house discharging into a certain sewer ; the outfall of the sewer was
in immediate proximity to an oyster-bed, from which oysters were
taken for consumption at the college suppers ; 23 cases of typhoid
fever followed among the students who attended the suppers at
which the oysters were eaten, but these cases were limited to three
out of seven fraternities; the only article of food used by the
three implicated fraternities, and not by the other four, was raw
oysters from the polluted consignment; and lastly, some of
the same consignment were consumed at Amherst College, and an
outbreak of typhoid fever occurred among those who consumed
them.*

This outbreak furnished evidence almost equal to a series of
experiments designed with the object of proving the possibility of
the transmission of typhoid fever by oysters. It served also to
stimulate inquiry, and since its occurrence a number of outbreaks
have been traced to a similar source. Sir William Broadbent
described several such cases in 1895, and Dr Newsholme continued
to follow the matter up, and reported that in 1894 38*2 per cent.,
in 1895, 33*9 per cent., in 1896, 31*8 per cent., and in 1897, 307 per
cent., of the total cases of typhoid fever originating in Brighton
were caused by sewage-contaminated shell-fish. Between midsummer
1893 and the end of 1902, 630 cases of enteric fever occurred at
Brighton, of which Dr Newsholme states 226 or 36 per cent, were
caused by sewage-polluted shell-fish, 152 cases being traced to
oysters and the remainder to other kinds of shell-fish, j" In 1896 Dr
Bruce Law reported an outbreak of typhoid fever at Southend, in
which certain cases had apparently been due to the same vehicle
of infection. In the same year Chantemesse described to the
Academy of Medicine an outbreak at Saint-Andre, in the Medi-
terranean Department of Herault, which was caused by a barrel
of oysters derived from contaminated oyster - beds at Cette.
Fourteen persons eating these oysters in an uncooked condition
contracted typhoid fever, or a disease simulating it. Evidence of
the same character as that recorded in the above cases was forth-
coming from Brightlingsea (Buchanan), Chichester (Theodore Thom-
son), Belfast (Jarre), Southend-on-Sea (Foulerton, Nash), Yarmouth,

* Seventeenth Annual Report of the State Board of Health of Connecticut, U.S.A.,
1894 ; New York Medical Record, 1894 ; Report of Medical Officer to Local Govern-
ment Board, 1894-95.

t Report on Health of Brighton, 1902, p. 45.

OYSTERS AND TYPHOID 255

Exeter, Blackpool, and other places. In 1902 occurred the outbreaks
of typhoid fever and similar illnesses at Winchester and Southampton,
following upon the consumption (at mayoral banquets) of oysters
derived from some oyster-beds at Ernsworth. From the same beds at
the same time oysters were obtained which apparently caused cases of
the disease at Portsmouth, Brighton, Ventnor, Hove, and Eastbourne.
The matter was inquired into by Dr Timbrell Bulstrode, who found
that at Winchester, out of a total of 134 guests at the banquet, 62
or 46 '3 per cent, were attacked with illness ; and at Southampton
mayoral banquet, out of 132 guests, 55 or 41 "6 per cent, were attacked
with illness. Eleven of these cases were enteric fever, and 44 were
cases of gastro-enteritis. In the two outbreaks 266 persons were
guests, 21 (or 7*8 per cent.) were attacked with enteric fever, and 118
(or 44*3 per cent.) suffered from other illness. All those who had no
oysters escaped enteric fever. After a minute inquiry Dr Bulstrode
came to the following conclusion : — (a) Two mayoral banquets occur on
the same day in separate towns several miles apart ; (b) in connection
with each banquet there occurs illness of analogous nature attacking,
approximately speaking, the same percentage of guests and at cor-
responding intervals ; (c) at both banquets not every guest partook
of oysters, but all those who suffered enteric fever, and approxi-
mately all those who suffered other illness, did partake of oysters,
the exceptions to this rule appearing insignificant when all the facts
are marshalled; (d) oysters derived directly from the same source
constituted the only article of food which was common to the guests
attacked ; and (e) oysters from this source were at the same time
and in other places proving themselves competent causes of enteric
fever. It may be added that the oyster-beds at Ernsworth from
which the implicated oysters were obtained are in immediate
proximity to the outfall of the Ernsworth sewers, and had for
several years been known to be contaminated beds.*

In 1902 there also occurred an outbreak of typhoid fever at
Mistley and Bardfield in Essex, which was shown to be due to oysters,
in which only a small portion of the oysters appears to have been
capable of causing illness, and the nature of the illness varied from
a mere feeling of nausea and weakness to a fatal attack of typhoid.f

In 1902 and 1903 further evidence was forthcoming from various
sources which went to show the intimate and apparently causal
relationship between the consumption of polluted shell-fish and
typhoid fever. Dr Nash, of Southend-on-Sea, states that in 1902
only 0"4 per cent, of the cases (501) of notifiable infectious diseases

* Special Report to the Local Government Board, 14th May 1903, by Dr H.
Timbrell Bulstrode. See also Foulerton's Report on the Pollution of Tidal Fishing

Waters by Sewage, 1903, pp. 31-37.

I Report of Medical officer of Essex County Council, 1902, pp. 53-60.

256 BACTERIA IN OTHER' FOODS

other than typhoid fever occurred in persons who had eaten shell-fish,
whereas 54 per cent, of the cases of typhoid fever occurred in persons
who had recently eaten shell-fish. In 1903 the comparative figures
were 2 per cent, and 90 per cent, respectively.*

Evidence necessary to prove Contamination of Oysters. — Evidence
of contamination of oyster-layings by sewage must be sought in
three directions : —

(1) There must be personal inspection of the neighbourhood and
surroundings of the layings and storage ponds. The immediate
sanitary circumstances may be such that a definite conclusion can
be come to that the locality is dangerously unfit for the purposes of
the oyster industry, and no further examination by chemical or
bacteriological methods will be necessary. On the other hand,
local inspection may not reveal any probable source of dangerous
sewage contamination ; and to test the matter, it may be necessary
to make further examination in order to detect traces of pollution
which may have arisen from sources of contamination which were
not obvious to the eye.

(2) In the event of the results of inspection being satisfactory,
the next step is to examine the water in which the oysters are laid,
in order to ascertain whether the chemical or bacteriological evidence
of sewage contamination is sufficiently strong to enable one to say,
in spite of the local inspection, that the sewage is not sufficiently
diluted and purified to obviate all possible danger.

(3) A bacteriological examination of the molluscs themselves
must be made, in order to ascertain whether they contain those
bacteria which are ordinarily associated with contamination by
sewage.

It is hardly needful to add that, in order to establish the fact that
infection has occurred or may occur from the consumption of polluted
oysters, it is necessary to prove disease in persons or animals who
have eaten some of the oysters. In any inquiry of this kind it is
essential to take into consideration, (a) clinical evidence, (b) the
history and circumstances of each case, and (c) the exclusion of
all other possible causes.

Symptoms of Oyster-Poisoning. — Obviously, the diseased conditions set up by
the consumption of polluted oysters will vary according to circumstances. If the
pollution be the specific infection of typhoid fever, the clinical disease of typhoid
fever will supervene. The same applies to cholera. But in many cases on record
the illness resulting has been of a less specific character, and has simulated gastro-
enteritis, colic, certain nervous conditions, and so on. Hence it may be desirable
to make a provisional classification as follows : —

(a) Nervous Conditions, which Mosny likens to curare poisoning. This type is
rare, always severe, and generally fatal.

(6) Gastro-Enteritis. — This group, which is one of the most common and least

* Public Health, 1903 (November), pp. 81 and 82.

OYSTERS AND TYPHOID 257

fatal, includes colic, nausea, vomiting, with more or less prostration. The onset is
generally sudden, and the attack lasts a comparatively short time.

(c) Dysenteric Symptoms may occur which simulate the symptoms of group (6),
but are more severe. In this group, between it and (b), may be classified the cholera-
like conditions which sometimes occur.

(d) Specific Disease, such as typhoid fever or cholera.*

In all cases there is, of course, an incubation period, which is usually longer in
duration than that occurring in ptomaine poisoning.

Infection of Oysters. — The mode of infection of oysters by
pathogenic bacteria is briefly as follows: — The sewage of certain
coast towns is passed untreated into the sea. At or near the outfall,
oyster-beds are laid down for the purpose of "fattening" oysters.
Thus they become contaminated with saprophytic and pathogenic
germs contained in the sewage. It will be at once apparent that
several preliminary questions require attention before any deductions
can be drawn as to whether or not oysters convey virulent disease
to consumers.

The precise conditions which render one locality more favourable
than another in respect to oyster culture are not fully known. But
it has been observed that they do not nourish in water containing
less than 3 per cent, of salt. Hence they are absent from the
Baltic Sea, which, owing to the fresh river-water flowing into it,
contains a small percentage of salt. Oysters appear, in addition, to
favour a locality where they find their chosen food of small ani-
malcula and particles of organic matter. Such a favourable locality
is the mouth of a river, where tides and currents also assist in
bringing food to the oyster. Unfortunately, however, in a crowded
country like England, such localities round our coast are frequently
contaminated by sewage from outfalls. Thus the oysters and the
sewage come into intimate relation with each other.

Professor Giaxa carried out some experiments in 1889 at Naples
which appeared to show that the bacilli of cholera and typhoid
rapidly disappeared in ordinary sea- water. Other observers at about
the same time, notably Foster and Freitag, arrived at an opposite
conclusion. Klein also found the cholera bacillus four days after
the removal of the oysters from water purposely contaminated with
them. In 1894 Professor Percy Frankland, in a report to the
Royal Society, declared "that common salt, whilst enormously
stimulating the multiplication of many forms of water bacteria,
exerts a directly and highly prejudicial effect on the typhoid bacilli,
causing their rapid disappearance from the water, whether water
bacteria are present or not." Boyce and Herdman found that up
to a certain point oysters could render clear sewage-contaminated

* The general question of mollusc poisoning is treated of by Mosny in the Revue
d'Hyyiene, December 1899 to March 1900. Mosny also furnished the French
Government with reports on the subject.

R

258 BACTERIA IN OTHER FOODS

water, and could live for a prolonged period in water rendered opaque
by the addition of fseeal matter. They also proved that the number
of organisms in the pallial cavity and rectum of oysters which had
been in clear water was much less than occurred in oysters laid
down in proximity to a sewer outfall (10 bacteria in the former case
as against 17,000 in the latter). It was found that more organisms
were present in the pallial cavity than in the rectum. B. typhosus
could be identified in cultures taken from the water of the pallial
cavity and rectum fourteen days after inspection, but in diminishing
numbers.* Several important links in the chain of evidence
remained in obscurity. It was at this time, when the matter was
admittedly in an unsatisfactory stage, that Dr Cartwright Wood
made his experiments.-)- We have not space here to enter into this
work. But his conclusions seem to have been amply established,
and were to the effect that typhoid and cholera bacilli could, as a
matter of fact, exist over very lengthened periods in. ordinary sea-
water. The next step was to demonstrate the length of time the
bacilli of cholera remained alive in the pallial cavity and body of
the oyster. Dr Wood found they did so for eighteen days after
infection, though in greatly diminished numbers. This diminution
was due to one or all of three reasons : (a) the effect of the sea-
water already referred to as finally prejudicial to bacilli of typhoid ;
(b) the vital action of the body-cells of the oyster; (c) the wash-
ing away of bacilli by the water circulating through the pallial
cavity.

Broadly it may be said that the same principles apply to the
typhoid bacillus. It can live in sea-water, probably, for three to
five weeks (Klein, Boyce) although it does not appear to multiply
in this medium. Oysters infected with the typhoid bacillus can
retain their infective properties for two to three weeks, and even
if placed in running sea-water, may not lose their infective properties
for some days. Mosny suggested any period from one to eight
days. In cockles there is evidence to show that the typhoid bacillus
thrives and even multiplies, and these shell-fish are not rendered
free from infection by being laid in pure water.

It will have been noticed that up to the present we have learned
that typhoid bacilli can and do live in sea-water, and also inside
oysters up to eighteen days, but in ever-diminishing quantities.
The question now arises : What is the influence of the oyster upon
the contained bacilli? Under certain conditions of temperature
organisms may multiply with great rapidity inside the shell of the
oyster. Yet, on the other hand, the amoeboid cells of the oyster,

* Report of British Association for Advancement of Science, 1895; and Thompson-
Yates Laboratory Report, vol. ii.

t Brit. Med. Jour., 1896, ii., p. 760 et seq.

\

OYSTERS AND TYPHOID 259

the acid secretion of its digestive glands, or the water circulating
through its pallia! cavity, may act inimically on the germs. Proof
can be produced in favour of the third and last-named mode by which
an oyster can cleanse itself of germs. So far, then, we have met with
no facts which make it impossible for oysters to contain for a lengthened
period the specific bacteria of disease. Let us now turn to their oppor-
tunity for acquiring such disease germs. It is afforded them during
the process of what is termed " fattening." By this process the body
of the oyster acquires a plumpness and weight which enhances its
commercial value. This desired condition is obtained by growing the
oyster in " brackish " water, for thus it becomes filled out and mechani-
cally distended with water. But if this water contain germs of disease,
what better opportunity could such germs have for multiplication than
within the body-cavity of an oyster ? " The contamination of sea-
water, therefore, in the neighbourhood of oyster-beds may undoubtedly
lead to the molluscs becoming infected with pathogenic organisms "
(Wood). Yet we have seen that, apart altogether from the individual
susceptibilities or otherwise of the consumer, there are in the series
of events necessary to infection many occasions when circumstances
would practically free the oysters from infection. This explains the
absence of uniformity in degree of contamination, and, coupled with
individual susceptibility and degree of cooking, the absence of
uniformity in causing outbreaks of disease.

The sources of pollution of oysters are not the fattening beds
alone. The native beds also may afford opportunity for contamina-
tion. Then, in packing and transit, and in storage in shops and
warehouses, there is frequently abundant facility for putrefactive
bacteria to gain entrance to the shells of oysters.

Dr Klein's researches into this question have been largely con-
firmatory of the facts elicited by Dr Cartwright Wood.* Despite
the tendency of the bacilli of cholera and typhoid to die out quickly
in crude sewage, the sewage is sufficiently altered or diluted at the
outfall for these organisms to exist there in a virulent state. We
may give Dr Klein's conclusions : —

1. That the cholera and typhoid bacilli are difficult of demonstra-
tion in sewage known to have received them.

2. That both organisms may persist in sea-water tanks for two
or more weeks, the typhoid bacillus retaining its characteristics
unimpaired, the cholera bacillus tending to lose them.

3. That oysters from sources free from sewage pollution con-
tained no bacteria of sewage (e.g. B. coli communis). Subsequently
to these experiments, Klein examined 172 oysters from various
layings to which no sewage gained access, and B. coli was absent

* Special Report of the Medical Officer to the Local Government Board on Ouster
Culture, etc., 1896.

260 BACTERIA IN OTHER FOODS

in all cases.* The Massachusetts State Board of Health have
recently arrived at similar results, and conclude that the presence
of B. coli in shell-fish is abnormal and due to contamination either
by sewage or by uncleanly handling. The presence of B. coli is
therefore looked upon as " invariable aid " in determining the occur-
rence of pollution.*)-

4. That oysters from sources exposed to risk of sewage contami-
nation did contain colon bacilli and other sewage bacteria.

5. That in one case Eberth's typhoid bacillus was found in the
body and liquor of the oyster. Nor do typhoid bacilli lose activity
or virulence by passing through an oyster.

In 1902 Dr Klein had occasion to examine a number of oysters
in connection with the Winchester outbreak of typhoid fever, to
which reference has already been made. In all 18 oysters were
examined with the following results : — (a) Every one of the 18 con-
tained B. coli, (b) 3 out of the 18 contained a bacillus belonging to
the Gaertner-typhoid group, and (c) 3 out of 15 contained the spores
of B. entcritidis sporogenes. All these oysters came from the
Emsworth layings, and all showed contamination with excremen-
titious matter. At the end of 1902 and beginning of 1903, Dr Klein
examined 25 different sets of oysters, only 7 sets of which showed
no signs of pollution.^

Boyce examined 140 samples of shell-fish at Liverpool in 1902,
and found B. coli present in 104, and B. entcritidis sporogenes in 10
cases. The former was more frequently present in oysters and
mussels, and the latter in cockles. §

In 1903 Mr Fouler ton examined a number of oysters derived
from suspicious oyster-beds, with the object of detecting the presence
or absence of bacteria characteristic of sewage. The two sewage
organisms which he selected as " indication " bacteria were B. coli
and B. entcritidis sporogenes, and his results were as follows : Out of
65 oysters examined, in 48 neither bacillus was found; in 5, B.
entcritidis sporogenes was present alone; in 8, B. coli was present
alone ; and in 4, both organisms were present. Foulerton attaches
most importance, as indication of recent sewage contamination, to the
presence of B. coli, and he therefore concludes that out of 65 oysters
12 or 19 '4 per cent, showed evidence of recent sewage pollution. ||
In a second series of 27 oysters B. coli was found in 4 instances, or
a percentage of 14*7.

* Brit. Med. Jour., 1903, i., p. 419.

t Thirty-fourth Annual Report of the State Board of Health of Massachusetts,
1903, pp. 260-264, and 280.

% Report of Medical Officer of Health, City of London, 1902, pp. 150-157.

§ Report on Health of Liverpool, 1902, p. 173.

|| The Pollution of Tidal Fishing Waters by Sewage, 1903. A special report by
A. G. R. Foulerton; F.R.C.S., D.P.H., pp. 42-49.

OYSTERS AND TYPHOID 261

Lastly, the results of the investigations carried out by Dr
Houston for the Koyal Commission on Sewage, tend to prove that
the contamination of oysters by B. coli is widespread and not alto-
gether dependent on sewage contamination of the oyster. He
examined over 1000 oysters, and nearly all, from whatever laying
they were taken, contained B. coli or coliform Organisms. This did
not hold good as regards deep-sea oysters which were free from this
organism (and spores of B. enteritidis sporogenes) as was also deep-sea
water. Houston found the number of B. coli in an oyster varied
from 10 to 10,000 (in 10-15 c.c), and the contents of the stomach
of the oyster contained more B. coli than the liquor in the shell.
Fewer B. coli were found, as a rule, in oysters stored in pure waters,
but instances occurred where the number of such bacilli was as
great as in oysters from contaminated sources.*

Such is the bacteriological evidence down to recent date, and
whilst some of it may appear to be of a conflicting nature, there are
certain conclusions which may be drawn. First, the presence of B.
coli in oysters must be judged relatively. Secondly, topographical
evidence as to pollution must be taken in conjunction with bacterial
evidence. Thirdly, there is the broad general fact that oysters
ordinarily grown on oyster-beds contaminated with bacteria may,
and do on occasion, contain the virulent specific bacillus of typhoid,
which can live both in sea-water and within the shell of the oyster.
This being so, the risk of infection of typhoid by oysters is a real
one. Yet in actual occurrence many conditions have to be fulfilled.
For, in addition to the fact that the oysters must be consumed, as is
usual, uncooked, the following conditions must also be present : —

(a) Each infective oyster must contain infected sewage, which
presupposes that typhoid excreta from patients suffering from the
disease have passed into that particular crude sewage, and have not
been disinfected.

(b) The infective oyster must have fed upon infected sewage, and
still contain the virus in its substance.

(c) There must have been no period of natural cleansing after
" fattening."

(d) The oyster must then be eaten, uncooked or undercooked,
by a susceptible person.

Even to this formidable list of conditions we must add the
further remark that, owing to the vitality of the body-cells of the
oyster or to the lessened vitality of the bacilli of cholera and typhoid,
it is generally the case that the tendency of these organisms is
rather to decrease and die out than live and multiply.

* Royal Commission on Sewage Disposal: Fourth Report on Pollution of Tidal
Waters and Contamination of Shell-fish, 1904, vols. i. and iii. The latter volume
contains a large amount of information as to bacteriological technique, etc.

262 BACTERIA IN OTHER FOODS

We shall probably maintain a satisfactory balance of truth if we
place alongside these facts the summary of the Local Government
Board Eeport. " There can be no doubt," wrote Sir Kichard Thorne,
" that oysters which have been brought into sustained relation with
the typhoid bacillus are liable to exhibit that microbe within the
shell contents, and to retain it for a while under circumstances not
only permitting its rapid multiplication when transferred again to
appropriate media, but conserving at the same time its ability to
manifest its hurtful properties." * The Eoyal Commission on Sewage
Disposal concludes, that at the present time it is undesirable to con-
demn oysters only on bacteriological evidence of the presence of B.
coli. At present topography must stand before bacteriology, and the
condemnation or otherwise of oysters must be judged on broad,
common-sense lines. It should be borne in mind that the oyster
trade is a considerable industry, which should not be injured except
on proved and substantial evidence.

Means of Prevention, — In the special report issued by the
Local Government Board in 1896, on oyster culture, which had been
drawn up by Dr Timbrell Bulstrode, accounts are given with
diagrams of the layings, fattening beds, and storage ponds used in
oyster cultivation in various counties round the coast of England
and Wales. Various proposals were made by Dr Bulstrode for the
control of this industry. In some cases, particularly the larger
layings, altering the position of the fattening beds was considered
sufficient, but in other cases nothing short of a complete diversion of
the sewers and drains, or withdrawal of existing layings, could be
regarded as sufficient.! We may repeat that the Eoyal Commission
on Sewage Disposal urge that topographical conditions shall be taken
along with bacteriological evidence in arriving at a decision as to any
oyster layings, and under the present circumstances of our limited know-
ledge of the bacteriology of the subject, this is the right course and
should assist in indicating preventive methods.

From what has been said, such preventive treatment is obvious : —
(1) All oyster layings and shell-fish beds round the coast should be
registered, superintended, and inspected by the sanitary authority or
the Government. (2) Local Sanitary Authorities should have power
of control over oyster layings situated in their district, and should
be enabled to prevent the sale in their district of oysters and other
molluscs derived from sewage - contaminated sources. (3) The
importation of foreign oysters, grown on uncontrolled beds, should,

* Special Report to Local Government Board on Oyster Culture, etc., 1896. This
report by Dr Timbrell Bulstrode is probably the fullest statement yet written on the
question as it affects England.

+ For a brief record of the attempts at legislation on this subject, see Brit. Med.
Jour., 1903, ii., p. 296 (Newsholme).

COCKLES 263

if possible, be restricted or supervised. (4) Further, as a protective
measure of the first importance, oysters should be cleansed, after
fattening on a contaminated bed, by being deposited for several
weeks at some point along the coast which is washed by pure sea-
water. (5) Eetention in dirty- water tanks, in uncleanly shops and
warehouses, should also be prohibited.

Other shell-fish than oysters do, from time to time, cause
epidemics or individual cases of gastro-intestinal irritation, and prob-
ably contain various germs. These they acquire in all probability
from their food, which by their own choice is frequently of a doubt-
ful character.

In a preliminary inquiry into "Cockles as agents of Infectious
Diseases," Dr Klein detected the B. coli in 3 out of 8 cockles
which had been taken from a foreshore polluted with the discharge
from a sewer outfall, and also B. enteritidis sporogenes in 4 of them.
No typhoid bacilli were detected. In 8 raw cockles in their
shells bought from a street hawker, Dr Klein found no typhoid
organisms, but B. coli was found in 5 out of the 8 cockles and
B. enteritidis sporogenes in 4 out of the 8.* In subsequent experi-
ments Dr Klein came to the conclusion that a mussel immersed for
twenty-two hours in cholera-infected water retained the bacilli of
cholera for forty-eight hours after immersion in clean sea- water, and
the same may be said in respect of typhoid infection. Indeed, evi-
dence was obtained showing that the typhoid bacillus could multiply
in cockles. He also showed that merely pouring boiling water over
a heap of shell-fish did not necessarily destroy either cholera or
typhoid infection contained in them.f Since the time of these
investigations, a number of outbreaks of disease, including enteric
fever, have been traced to the consumption of mussels and cockles,
and it has been shown that the cooking which these shell-fish
undergo is not sufficient to rid them of poisonous pollution.

In 1902 several cases of typhoid occurred in Wandsworth due
to infected cockles, and a number of similar cockles being examined
by Dr Klein showed the presence of B. coli and other allied forms,
and by other workers (Lister Institute) during the same year the
typhoid bacillus itself is stated to have been isolated from cockles
derived from a sewage-polluted laying.^

In 1903 an outbreak of typhoid fever occurred in Glasgow,
which was traced to the consumption of sewage-polluted shell-fish
at a neighbouring seaside town. An examination of a number of
shell-fish from this particular locality was made by Dr R M.
Buchanan, the Corporation bacteriologist, who reported that — (1) All

* Report of Medical Officer to Local Government Board, 1899-1900, p. 574.

t find., 1900-01, pp. 564-71.

J Report of Medical Officer of Health of City of London, 1902, pp. 134-49.

264 BACTERIA IN OTHER FOODS

the edible shell-fish within the area of sewage contamination showed
by the presence of virulent B. coli excremental pollution, and their
consumption must therefore be regarded as highly prejudicial to
health. Cultures of the species of B. coli killed guinea-pigs within
eighteen hours. (2) The shell-fish beyond the range of sewage
contamination were found to be normal and perfectly safe for edible
purposes. (3) Certain shell-fish, cockles and " muscins " {My a
arenaria) within the area of sewage contamination showed, accord-
ing to Buchanan, the presence of the bacillus of typhoid fever in
great number, and in some cases almost in pure culture, and the
consumption of similarly infected shell -fish by holiday visitors in
the end of July would sufficiently explain the outbreak of typhoid
fever which occurred amongst them after return to their own homes
in Glasgow and elsewhere.

As regards the specificity of this bacillus, Buchanan reports that
it had all the microscopical and cultural characteristics of the
typhoid bacillus. Further, it gave the characteristic reaction with
human typhoid serum and with serum obtained from a typhoid
immunised guinea-pig. The finding of this bacillus in such numbers,
and in so many individual shell-fish, is so exceptional that it was
repeatedly subjected to reliable culture tests and to repeated serum
tests, and always with the result of proving its general identity with
the typhoid bacillus obtained from a case of typhoid fever.

A third apparent instance of finding the typhoid bacillus may
be quoted. In 1902-1903 Klein examined ten samples of Leigh
cockles, and found every one of them showing evidence of sewage
pollution, though six had been " cooked." One of the uncooked ones
contained B. typhosus, a typical typhoid bacillus agglutinating with
typhoid blood (Klein). The cooking to which some of these cockles
were submitted must have been perfunctory, as it is fairly well estab-
lished that 60-61° C. kills B. coli. Yet this organism was found in
two instances where the cockles in question had been "boiled con-
tinuously for one minute," or " put in boiling water and taken out
when the water boiled over, time in water three and a quarter
minutes." *

These facts reflect a new and not reassuring light upon the
possibility of cockle infection. But they must be accepted with
great reserve until very fully confirmed. Dr Bulstrode in his report
on Oyster Culture in 1896 suggested that the infection by cockles
was a remote contingency, because " in the first place these molluscs
are, as a rule, only eaten after being cooked ; and in the second place
it is seldom that extensive cockle industries are carried on in other
localities than those where large stretches of sand are exposed at
low tide, and such stretches are found chiefly on the actual seashore
* Report of Medical Officer of Health of City of London, 1902, pp. 134-149.

COCKLES 265

or quite at the mouth of estuaries far away from sources of sewage
contamination." The experience at Leigh-on-Sea, Southend, and
other places seems to tell a different tale, and it is evident that
the shell-fish may be grown on polluted beds. After growth, it is
true, they are raked into hand-nets, and taken to the cockle-sheds,
and here are plunged into coppers of boiling water in the nets, after
which they are riddled through wide-meshed sieves, which allow
the soft parts to pass through, retaining the shells, which are
deposited in heaps for sale to oyster cultivators. The cockles them-
selves are then washed in about five changes of water, to the last
of which a certain quantity of salt is added. Not infrequently, the
same water is used in all the washings. The so-called boiling is
evidently misleading. Though the water is actually at the boiling
point, the cockles are plunged in in a mass, and for a short time,
and it by no means follows that every part is exposed to a tempera-
ture of 212° F.

Dr Klein has shown that the usual method of cooking only
amounts to scalding, and cannot be relied on to sterilise micro-
organisms. The live fish, with shells tightly closed, are held in a
net and plunged en masse into a vessel containing boiling water.
The immersion of the cold mass immediately lowers the tempera-
ture, and when in the course of two or three minutes it begins to
boil again, the net is lifted out. The scalding kills the fish and
causes the shells to open, but it does not sterilise the contents.
Dr Klein found that the temperature of the water fell, on the im-
mersion of the fish, from 100° C. to 65° ; and that cooking for the
usual time was totally inadequate to kill the micro-organisms. Fish
that had been kept in typhoid polluted water were tested, and were
found to be swarming with live bacilli after cooking. Prolonged
boiling would, no doubt, be effective, but it causes the fish to
shrivel up and spoils them for sale.

Dr Klein then suggested that cooking by steam might be found
an efficient steriliser without spoiling the fish as food. It is well
established that current steam is much more penetrating than
boiling water for purposes of disinfection, and it is always used in
preference. The question was whether an exposure sufficient to
sterilise would amount to over-cooking, and recent experiments
carried out in the kitchen at Fishmongers' Hall were intended to
settle that point. Cockles and mussels were cooked in a steamer
under the direction of Dr Klein in the presence of several repre-
sentatives of the trade, who examined them afterwards. Two
batches were cooked, one for ten minutes and the other for five.
The steamer used was a fixed vessel some 2 feet deep, into which
steam is introduced by a pipe about an inch from the bottom. A
layer of cockles was placed at the bottom, and two other layers on

266 BACTERIA IN OTHER FOODS

trays above it. In the top tray mussels were also placed. Some of
the fish were spread out and others heaped up. The results were : —
Ten minutes — mussels pronounced spoilt, and useless to the trade ;
cockles "all right" in upper layers, but the bottom layer overcooked.
Five minutes — mussels " all right," and cockles better than the ten
minutes' batch ; the upper layers "could not be better" in appear-
ance and flavour, but the bottom layer was again pronounced
somewhat overcooked, or at any rate less satisfactory than the
others. No doubt the steam was hotter at the bottom of the vessel
and the exposure greater. The bacterial results were as follows : —
The cockles were found to be sterile in all cases : the mussels were
also found to be sterile, except in the case of those placed in heap on
the top layer and steamed for five minutes. Some of these still
retained living spores. It is probable that if exposed to the more
direct action of the steam even the heaped mussels would be
completely sterilised by five minutes' cooking, without impairing
their trade value. As a result of these experiments the Fish-
mongers' Company was reported as recommending to the trade the
substitution of steaming for boiling.

Many other similar foods have been implicated in the spread of
disease. Dr Hamer investigated outbreaks of typhoid fever in
London in 1900 and 1903. in which lie showed the extreme prob-
ability of fried fish acting as the vehicle of infection.* In 1900 Dr
Plowright traced similar infection in thirty persons to polluted
clams, shell-fish comparatively little known in this country as an
article of diet.f Derived from sewage-polluted layings, clams may
readily become contaminated, and if uncooked may convey disease to
the consumer.

* Ninth Annual Report of Medical Officer of Health of Administrative County of
London, 1900, p. 37, and Appendix; and Special Report, No. 719, issued 1904.

f The clam is a shell-fish comparatively little known in this country as an
article of diet except to the dwellers near those of our coasts on which it occurs.
Belonging to the Siphonidse division of the Conchiferae, the clam (Mya armaria"),
like its ally the cockle, is found abundantly round our shores. It has, however, a
wider geographical distribution, being found in the Arctic Regions, where it con-
stitutes an important article of food. In America it is largely consumed in Boston
and along the Massachusetts seaboard. The clam of New York is a different species
(Venus marcenaria). It has a remarkably developed syphon, tfae inhalent and
exhalent tubes being joined into a trunk-like body 3 or 4 inches in length, which the
animal protrudes in an upward direction towards the surface of the mud. The clam
itself lies buried in the mud, into which it has worked itself by the aid of its muscular
foot to a depth varying from 8 to 18 inches. The currents of water passing in and
out the syphon keep open the vertical burrow the creature has made, while the
surface of the mud is covered by the tide, but where this recedes and the mud
becomes dry, the position of the clam is shown by a small round depression on the
surface. In Great Britain it is regarded as a kind of inferior oyster, and like the
last-named is preferred uncooked by those persons who are really fond of it and to
whom it is a luxury. More generally, clams are cooked by having boiling water
poured over them, and being allowed to remain in it until the shells open.

ETIOLOGY OF MEAT-POISONING

26'

2. Meat

Since 1880, more than fifty outbreaks of disease have been traced
to the consumption of unwholesome or diseased meat. In 1880
occurred the well-known "Welbeck disease" epidemic. A public
luncheon was followed by severe and, in some cases, fatal illness.
Seventy-two persons were affected and four died. A specific bacillus
was isolated by Klein from the cold hams, the consumption of
which caused the outbreak. The incubation period varied from
twelve to forty-eight hours. This epidemic drew marked attention
to the whole question of food-poisoning, and subsequent epidemics
were very thoroughly investigated by the aid of bacteriology. Some
of the better-known outbreaks may be tabulated as follows : —

Date of
Occurrence.

Place of Occurrence.

No. of
Cases.

Period of

Incubation in

Hours.

rrobable Source of
Infection.

1880

Welbeck

72

1 2- 1 8

Cold boiled hams

1881

Nottingham

15

12-34

Pork

1882

Oldham

9

4

American Tinned
Pig's Tongue

-1882

Bishop Stortford

G

24

Beef

1882

Whitchurch

20

1-f)

Brawn

1886

Carlisle

20

6-40

Ham and game pie

1886

Ironbridge

12

6-12

Veal pies

1887

Retford

80

8-36

Pork brawn

1888

Middlesborongh

11 1

American bacon

1889

Carlisle

25

2 1

Pork pies

1891

Portsmouth

13

1 1-17

Cold meat pie

189G

Mansfield

265

1 8-2 1

Potted meat

1898

Oldham

54

48

Veal pies

1899

Nuneaton

42

12-48

Pork chitterlings

1902

Derby

22]

4-24

Pork pie

In nearly all these cases the genera] symptoms have been usually
one of two kinds, namely, conditions simulating gastro-enteritis, or
conditions simulating nervous disease. Each of the outbreaks have
shown more or less clearly the characters common to these epi-
demics:—

1. Simultaneous attacks.

2. Similarity of symptoms and post-mortem signs.

3. A history of infection and collateral circumstances.

The common symptoms have included rigors, faintness, vomiting,
diarrhoea, abdominal pain, and occasionally skin eruptions. As 'a
rule, certain nervous conditions have supervened, such as giddiness,
headache, paralyses, mental depression, etc., and occasionally these
symptoms have been predominant.

268 BACTERIA IN OTHER' FOODS

Meat-poisoning appears to depend not upon the number of
bacteria present in the meat, but upon the particular species and
their products. As we have already stated, a long incubation period
generally indicates poisoning by bacteria, and a short incubation
period poisoning by products (ptomaines, toxins, etc.). In 1888,
Gaertner of Jena investigated an outbreak of disease affecting 58
persons who had eaten uncooked meat. One of the unfortunate
victims died, and from his body, as well as from the meat, Gaertner
isolated a bacillus which he called the B. enteritidis, an organism
allied to the coli group. This was practically the starting-point of
accurate bacteriological investigation into this group of epidemics
(fleiscJivergiftung, Ger. ; and intoxications alimentaires, Er.). Since
that period, the B. botulinus of Ermengem, and certain putrefactive
bacteria, have been held responsible for causing such illnesses.
More than twenty different species of bacteria have been isolated
from tinned meats and hams. As is pointed out elsewhere in the
present volume, there is evidence that the infectious properties which
food acquires frequently in summer, and which give rise to the
ordinary type of epidemic diarrhoea, are due to bacilli belonging to
the colon group, of which the B. coli communis of Escherisch and the
B. enteritidis of Gaertner are the two extreme types. According to
Delepine, the varieties of those bacilli which are the most important
sources of infection are those which resemble the bacillus of Gaertner,
and which, therefore, produce no permanent acidity, coagulation, or
distinct smell when grown in milk. Very few infectious samples of
milk give a distinct acid reaction, so that absence of acidity in milk
is not, as generally believed, an index of safety. It is probable that
the most dangerous kind of foecal infection is that produced by matter
containing bacilli resembling Gaertner's bacillus. Such an infection
is probably connected with the existence of an infectious diarrhoeal
disease liable to occur in the lower animals as well as in man.

It is certain that bacilli presenting the characters of the ordinary
B. coli communis are seldom capable of producing such a rapid
infection as that produced by the B. enteritidis, or by closely-allied
bacilli, such as the B. enteritidis Beroiensis.

The last-named organism is a member of the Gaertner group
isolated by Delepine from pork pies, the consumption of which
caused the Derby illness in 1902. He considered the presence of
this bacillus in the pork pies was due to fsecal pollution of the
meat before it was cooked, and that the central parts of the pies
were not thoroughly cooked.* It frequently happens in these cases

* Report on Outbreak of Food-Poisoning in Derby, 1902 (Howarth and Delepine).
In this reference, and in Brit. Med. Jour., 1898, ii.' pp. 1456-58 and 1797-1801, and
ibid., 1899, vol. ii., pp. 791, 1367, will be found many particulars with regard to
meat-poisoning, its symptoms, prevention, investigation, etc.

MEAT 269

that some constituent part (such as jelly) of the manufactured article
or prepared dish is really the polluted portion.

Bacteria associated with Meat-Poisoning

The chief organisms, therefore, which have been considered as causally related to
meat (and " ptomaine ") poisoning are B. coli communis, B. enteritidis sporogenes,
B. enteritidis of Gaertner, and B. botulinus. The main facts respecting these organisms
must be mentioned here.

(a) B. coli communis (see p. 46).

(6) B. enteritidis sporogenes (see pp. 156 and 307).

(c) B. enteritidis of Gaertner. Isolated by Gaertner in 1888 from flesh of
diseased cow which had caused illness in persons eating it. Characters similar to
B. typhosus (morphology, motility, and staining properties), but grows more rapidly
in gelatine; fewer flagella; ferments lactose and sometimes dextrose; does not
produce indol or coagulate milk ; positive neutral-red reaction ; in litmus whey or
litmus broth, acid is first produced, and then the medium becomes distinctly alkaline.
Virulent to rodents and small animals (gastro-intestinal symptoms, haemorrhagic
enteritis, and swelling of lymph follicles). B. enteritidis Derbiensis of Delepine is one
of the members of the Gaertner group of enteritidis bacilli. B. enteritidis possesses
no spores, and therefore cannot stand very high temperatures. It produces
agglutinating properties in the blood of the patient.

(d) B. botulinus (Ermengem). This bacillus is held to be responsible for setting
up botulism. Van Ermengem describes, under the name of botulism, a state brought
about by the ingestion of various articles of food, such as ham, tinned or preserved
foods, oysters, mussels, etc., and which is characterised by comparatively slow
onset (twelve to twenty-four hours after infection), secretory troubles, paralysis of
certain muscles, particularly tongue and pharynx, dilatation of pupil, aphonia,
dysphagia, constipation, retention of urine, absence of unconsciousness and of fever,
etc. Van Ermengem has found that these symptoms were produced by a bacillus,
to which he has given the name of bacillus botulinus. Botulism differs considerably
from the more common form of food-poisoning with which we are acquainted in
England, and which is characterised by practically the same symptoms as those of
epidemic diarrhoea. B. botulinus is 4-9 /j. long and '9-12 \x broad; round, slowly
motile, 4-9 flagella. Polar spores ; killed in thirty minutes at 80° C. Liquefies
gelatine ; does not coagulate milk ; anaerobic ; in cultures often produces gas and
a sour, rancid odour. Pathogenic for guinea-pigs, rabbits, and the other small animals
(botulism).

Preventive methods. — Experience of meat-poisoning outbreaks
leads to the conclusion that the meat has contracted its poisonous
properties in either or both of two ways — (a) putrefaction or unsound-
ness in the meat itself; (b) unclean manipulation or storage in
insanitary conditions. Generally there has also been insufficient
cooking. The methods of prevention arc therefore obvious. Occasion-
ally tinned foods cause poisoning owing to metallic absorption, and
this must be differentiated from bacterial poisoning.

There is another class of meat conditions related to disease, to
which reference must now be made, viz., certain conditions occurring
in fresh meat, joints, or carcases. It is well known that the meat
substance itself does not frequently contain injurious bacteria.
They may nevertheless occur in the organs, glands, and tissues

270 BACTERIA IN OTHER FOODS

other than muscular, and when present set up during life the
bacterial diseases of animals, or after death putrefactive changes.
It is for these conditions that meat is "seized" under the Public
Health Acts as unfit for food of man. Such conditions may be
broadly divided into two kinds : —

1. Specific Diseases in Meat, such as tuberculosis, anthrax, swine
fever, actinomycosis, milk fever, etc.

2. Decompositions of Meat due to invasion by putrefactive
organisms after death. These conditions may be disposed of at
once by saying that they arise commonly as a result of keeping meat
too long under conditions likely to lead to putrefaction. Unclean
storage, insufficient preservation, summer weather, and similar
circumstances afford the opportunity for putrefactive organisms to
perform their function. The signs of decomposing meat do not require
explanation or elaboration. They are mainly three : — {a) Smell of
putrefaction ; (b) discoloration ; and (c) loss of elasticity of tissue
which becomes doughy, pits on pressure, or may even become slimy
or soapy.

The chief specific diseases which occur in meat are dealt with
briefly in the section treating of the relation between bacteria and
disease. It will, therefore, be unnecessary to make more than a
passing reference in this place.

Tuberculosis. — This disease is set up in animals by the tubercle
bacillus, which is either identical with or closely allied to the 1>.
tuberculosis of Koch. It may set up a generalised disease affecting
the body of the animal more or less completely, or it may set up
only a local disease.

The Koyal Commission on Tuberculosis, in the report which they
made in 1898, referred to the degree of tubercular disease which
should cause a carcase, or part thereof, to be seized, and which may
be accepted broadly as indicative of general and local tuberculosis.
They stated as follows : —

"We are of opinion that the following principles should be
observed in the inspection of tuberculous carcases of cattle : —

(a) When there is miliary tuberculosis

of both lungs

(b) When tuberculous lesions are present

on the pleura and peritoneum

(c) When tuberculous lesions are present

in the muscular system, or in the
lymphatic glands embedded in or
between the muscles

(d) When tuberculous lesions exist in

any part of an emaciated carcase . j

Generalised tuber-
culosis is present,
and the entire
carcase and all
the organs may
be seized.

TUBERCULOUS MEAT 271

(a) When the lesions are confined to the

lungs and the thoracic lymphatic
glands . . .

(b) When the lesions are confined to the

liver

(c) When the lesions are confined to the

pharyngeal lymphatic glands .

(d) When the lesions are confined to any

combination of the foregoing, but
are collectively small in extent

Localised tubercu-
losis is present,
and the carcase,
if otherwise
healthy,shall not
be condemned,
but every part
of it contain-
ing tuberculous
lesions shall be
seized.

" In view of the greater tendency to generalisation of tuberculosis
in the pig, we consider that the presence of tubercular deposit in
any degree should involve seizure of the whole carcase and of the
organs.

"In respect of foreign dead meat, seizure shall ensue in every
case where the pleura have been ' stripped.' "

The tubercle bacilli are most easily found in the glands. They
are scarce in the caseating nodules. In the pig it is difficult
to detect the bacilli as a rule. The Eoyal Commission on
Tuberculosis emphasised the absence of bacilli in the meat
substance: — "In tissues which go to form the butcher's joint,
the material of tubercle is not often found even where the
organs (lungs, liver, spleen, membranes, etc.) exhibit very advanced
or generalised tuberculosis; indeed, in muscle and muscle juice
it is very seldom that tubercle bacilli are to be met with;
perhaps they are somewhat more often to 1).' discovered in bone,
or in some small lymphatic gland embedded in intermuscular fat."*
The chief way in which such meat substance becomes infected with
tubercle appears to be through carelessness of the butcher, who
perchance smears the meat substance with a knife that has boon
used in cutting the organs, and so has become contaminated with
infected material. Very instructive also are the results at which
Dr Sims Woodhead arrived in furnishing evidence for the same
Commission on the effect of cooking upon tuberculous meat: —
"Ordinary cooking, such as boiling and more especially roasting,
though quite sufficient to sterilise the surface, and even the substance
for a short distance from the surface of a joint, cannot be relied
upon to sterilise tubercular material included in the centre of rolls
of meat, especially when these are more than three pounds or four
pounds weight. The least reliable method of cooking for tin's
purpose is roasting before a, fire; next comes roasting in an oven,
and then boiling." f From this statement it will be understood that

* Royal Commission on Tuberculosis, Report, 1S95, part i., p. 13.
t Ibid., p. 18.

272 BACTERIA IN OTHER FOODS

rolled meat may be a source of infection to a greater degree than
the ordinary fresh joint, and this is borne out by the experience
derived from epidemics due to meat-poisoning.

Tuberculous meat also finds its way occasionally into sausages,
and other similarly prepared meat foods.

Swine Fever is not an uncommon disease of pigs, and makes the
meat unfit for food. Schiitz and others have isolated a bacillus from
this disease. The chief post-mortem signs are the red punctiform
rash, generally becoming confluent on the back, extremities, and
ears ; the ulceration of the intestine, and the characteristic mottling
of the lymph glands.

Anthrax (see p. 315), Actinomycosis (see p. 321), and other condi-
tions are described elsewhere. Many parasitic diseases also make meat
unfit for food.

3. Ice-cream

In 1894 Dr Klein had occasion bacteriological ly to examine ice-
cream sold in the streets of London. In all six samples were
analysed, and in each sample the conclusions resulting were of a
nature sufficiently serious to support the view that the bacterial
flora was not inferior to ordinary sewage. The water in which the
ice-cream glasses were washed was also examined, and found to
contain large numbers of bacteria.

Since that date many investigations have been made into ice-
cream. It appears that this luxury is frequently manufactured
under extremely objectionable circumstances, and with anything but
sterilised appliances. Little wonder, then, that the numbers of
bacteria present run into millions per c.c. (varying from two to
twenty millions or more). In nearly all recorded cases, the quality
of the germs as well as the quantity has been of a nature to
cause some concern. B. coli communis has been very commonly
found, and in considerable abundance. The Proteus family, which
also possesses a putrefactive function, is common in ice-creams.
The common water bacteria are nearly always present. B. typhosus
itself, it is said, has been isolated from some ice-cream which
was held responsible for an outbreak of enteric fever. The
material had become infected during process of manufacture in the
house of a person suffering from unnotified typhoid fever.

The Manufacture of Ice-cream. — There are, practically speaking, three methods
of manufacture : —

(1) The real ice-cream, which cannot be sold at a low price, and which is made
simply by mixing cream (with a small proportion of milk), fruit or fruit pulp and
sugar. This mixture is then at once frozen.

(2) Milk is flavoured with fruit, or fruit essence and sugar, and has then added
to it a small quantity of dissolved gelatine, and at once frozen. In these two
processes there is no boiling, and both are frozen immediately after mixture.

MANUFACTURE OF ICE-CREAM 273

(3) Skimmed milk is boiled, either with or without the addition of some form
of starch (generally corn flour) with a certain number of eggs, sugar and flavouring.
Practically, the number of eggs used varies inversely with the amount of starch, the
effect of both being to thicken the mixture, and in some cases the practice has been
to at once freeze the mixture ; in others, in order to economise the consumption of
ice, the heated liquid has been allowed to cool naturally. Usually the number of
eggs is three or four to the quart. The whole is again boiled (for perhaps
twenty minutes), after which it is set aside to cool until the following morning,
when it is placed in the "freezers" and frozen. In short, the mixture contains
no cream, and is, in fact, a frozen custard. The milk and eggs are generally
obtained locally, and are usually of good quality. It is stated that if such
were not the case the ice-cream would be unpalatable, owing to ill-flavour,
and therefore unsaleable. The eggs used to be blown, but of late years
that practice has fallen into disuse, and they are now broken and mixed in the
ordinary way. The boiling is carried out in various utensils over the open fire. The
mixture stands for cooling purposes in the living room, or in any out-of-the-way
corner, sometimes in the open yard or area. The freezing takes place early on the
following morning in the " freezers." A " freezer " consists of a tin or galvanised iron
cylinder or container, in which the mixture is placed. The cylinder fits into an
outer vessel of wood or metal so loosely as to leave an inch or two of space all
round. In this space is placed broken ice and salt, and the inner cylinder is rotated
from time to time in this ice medium. No ice is added directly or indirectly to the
mixture itself, nor are colouring agents used as a rule. The utensils and materials
appear, as a general rule, to be clean.

From this description, which applies, generally speaking, to the street ice-cream
industry in London, it will be seen that the " ice-cream " is boiled for some time, and
in all probability sterilised, and in due course it undergoes, at least approximate,
freezing. These facts, added to the generally wholesome condition of the elementary
materials used, would appear, at first sight, to place the substance beyond risk of
contamination. But the critical period is the time of exposure between boiling and
freezing. Boiling sterilises, but freezing does not sterilise. Hence, if in the long
cooling process the substance is exposed to contaminated surroundings, the result
may be, in effect, a contaminated ice-cream. Such, in fact, frequently occurs."
Hence it would appear that it is not the process of manufacture that needs super-
vision so much as the general condition of the houses in which the substance is
made, and of the persons who make it, and the manufacture so far as length of time
between boiling and freezing is concerned.

The important stage of the operation is, therefore, that between the boiling and
freezing. Attention has been drawn to the fact that the majority of the specific
pathogenic bacilli discovered in ice-cream are of a non-sporing variety, and that,
therefore, if originally present in the material, would have been destroyed by boil-
ing, and, if found subsequently, must have gained access to the material while
cooling. The subsequent freezing, while it might inhibit such bacilli, would
certainly not destroy them, and on ingestion and melting their growth and develop-
ment would again commence.

Some dozen outbreaks of disease have been attributed to the
consumption of ice-creams. A typhoid epidemic occurred in
Liverpool (27 cases) in 1897 due to ice-cream, and an earlier epidemic
of the same disease traceable to the same cause occurred at Deptford
in 1891 (Turner). Kecently, a small outbreak occurred in the city
of London affecting 16 telegraph boys. The symptoms were colic
and diffuse abdominal pains, headache, vomiting, diarrhoea, and
nervous depression. Dr Collingridge's inquiry resulted in the
following conclusions : —

* See investigations by Klein, Cook, Wilkinson, Foulerton, and others.

S

274 BACTERIA IN OTHER FOODS

(1) That in a number of cases of illness occurring among young
persons of a susceptible age, the symptoms were strictly identical,
and were characteristic of poisoning by ingestion of toxic material.

(2) That the cases reported followed the ingestion of ice-creams.

(3) That ice-creams subsequently obtained at shops frequented
by the patients contained bacilli of a virulent character.

(4) That the symptoms observed were those generally following
the ingestion of material containing such bacilli.

(5) That where pathogenic bacilli were found, the ices had been
manufactured under insanitary conditions. The majority of the
manufacturers are aliens, and although the premises may be kept in
a fairly sanitary condition, their personal habits unfortunately leave
much to be desired where the preparation of food is concerned.

Dr Klein examined 24 samples of ice-cream from the same
locality, and found 13 (or 54 per cent.) to be poisonous to guinea-
pigs.* The writer traced 18 cases of typhoid fever in 1902 to the
consumption of contaminated ice-cream.*)- Owing to outbreaks of
this nature the London County Council (General Powers) Act, 1902
(sects. 42-45), has given powers for controlling this trade : —

(a) Ice-cream must be made and stored in sanitary premises.

(b) It must not be made or stored in living rooms.

(c) Strict precautions must be taken as to protection from con-
tamination.

(d) Cases of infectious disease must be reported.

(e) The name and address of the maker must appear on street
barrows. '

These regulations are new for London, though they have practi-
cally been in existence in Glasgow since 1895, and in Liverpool since
1898.

It should not be forgotten that ice-cream may have deleterious
effects on the consumer owing to its low temperature or to the
presence of alkaloidal poisons of the nature of tyro-toxicon which
have been detected in such substance as well as in milk (Mount
Morgan outbreak) and cheese (Michigan and Finsbury outbreaks).

Ice contains bacteria in varying quantities, from 20 per c.c. to
10,000 or more. Nor is variation in number affected alone by the
conditions of the water, for samples collected from one and the same
place differ widely. The quality follows in large measure the
standard of the water.

Water bacteria, B. coli, putrefactive and even pathogenic bacteria,
have been found in ice. Many organisms can live without much
difficulty, and are most numerous in ice containing air-bubbles.

* Report of Medical Officer of City of London, 1902, pp. 116-26.
t Report on Health of Finsbury, 1902, p. 67.

ICE 275

Dr Prudden, of New York, performed a series of experiments in
1887 to show the relative behaviour of bacteria in ice. Taking half
a dozen species, he inoculated sterilised water and reduced it to a
very low temperature for a hundred and three days, with the
following results : — B. prodigiosus diminished from 6300 per c.c. to
3000 within the first four days, to 22 in thirty-seven days, and
vanished altogether in fifty-one days; a liquefying water bacillus
numbering 800,000 per c.c. at the commencement, had disappeared
in four days ; Staphylococcus pyogenes aureus and B. fluorescens showed
large numbers present at the end of sixty-six and seventy-seven days
respectively ; B. typhosus, which was present 1,000,000 per c.c. after
eleven days, fell to 72,000 after seventy-seven days, and 7000 at the
end of one hundred and three days. Anthrax bacilli are susceptible to
freezing, but their spores are practically unaffected (Frankland). From
these facts it will be seen that bacteria live, but do not multiply, in ice.

Hutchings and Wheeler recently examined some ice suspected of
conveying typhoid fever at the St Lawrence State Hospital on the
river St Lawrence. The fragments were melted in a clean vessel at
room temperature, after which a considerable black sediment deposited
itself in the vessel. Cultures and plates were made in the usual
way, and B. coli and B. typhosus were both isolated.* The last-
named had the following characters : — On nutrient agar it grew
readily, in broth growth without pellicle, in lactose media no fermenta-
tion occurred, on potato the "invisible" growth, litmus milk became
alkaline without coagulation, and the bacillus was morphologically
identical with B. typhosus. With the serum of typhoid patients
characteristic agglutination occurred. Blumer found the number
of bacteria in some of the same ice was 30,400 per c.c. (agar) and
50,400 per c.c. (gelatine). Many colon bacilli were present.

Sedgwick and Winslow have also carefully studied the influence
of natural and normal conditions of cold upon the typhoid bacillus
in particular. The experiments were carried out with special refer-
ence to the danger of conveyance of the disease in question by
polluted ice, and with reference to the seasonal distribution of the
disease. The matter was undoubtedly one that called for investiga-
tion, and notably so in America where i<-«' and iced drinks arc in
such universal demand.

The apparent purity of ice is deceptive. It is true that water in
freezing undergoes a certain amount of purification. It loses, on
conversion into ice, saline constituents, contained air, and a certain
proportion of organic suspended matter. At the same time, it is not
entirely freed from microbes. The figures quoted by Sedgwick and
Winslow show that snow-ice may contain an average of more than
600 bacteria per cubic centimetre.

* American Jour, of Medical Sciences, Oct 1903, p. G83.

276 BACTERIA IN OTHER FOODS

Laboratory experiments have confirmed the conclusion that
a freezing process is not necessarily fatal to bacterial life
(see p. 18). We have instances of bacteria multiplying at zero,
and of their survival after six months' exposure to the temperature
of liquid air. It would appear that about 90 per cent, of the
ordinary water bacteria are eliminated by the process of freezing.
In the case of a specific pathogenic organism such as the
B. typhosus, less than 1 per cent, survive simple freezing for a
period of fourteen days. Complete sterility does not occur even
at the end of three months, whilst a process of alternate thawing
and freezing, if on the whole more fatal to the typhoid germs
than a simple freezing, is equally unsuccessful in effecting an
absolute sterilisation of the infected water. The reduction in
the number of typhoid bacilli in chilled water is approximately
as great as occurs in ice. Cold exercises an inhibitory action
as regards the typhoid bacillus, and in natural ice there is a
supplementary purifying influence to be taken into account, as, at
the time of freezing, 90 per cent, of the germs are thrown out by a
process of physical exclusion. Therefore, the danger of infection in
the case of ice, if it is minimised, is not abolished. A certain number
of typhoid bacilli do remain alive, and these may, on rethawing,
undergo a rapid multiplication outside as well as inside the human
body. And it has likewise to be remembered that it is notoriously
difficult to trace the exact channels of infection in sporadic cases of
typhoid fever. Sedgwick and Winslow have rightly drawn attention
to the unfavourable conditions furnished by natural ice for the
propagation of the typhoid organism.

In making a bacterial investigation into the flora of ice and ice-
cream, it is necessary to remember that considerable dilution with
sterilised water is required. The usual methods of examining water
and milk are adopted.

4. Bread

Bread forms an excellent medium for moulds, but unless
specially exposed the bacteria in it are few. Waldo and Walsh
have, however, demonstrated that baking does not sterilise the
interior of bread. These observers cultivated numerous bacteria
from the centre of newly-baked London loaves.* The writer has
recently made a series of examinations of the air of some nine or ten
underground bakehouses in central London. The general result of
these investigations was that the air of the typical underground
bakehouses examined — (1) contained 14'8 volumes per 10,000 of
carbonic acid gas, C02 (as compared with 4-9 in above-ground bake-
houses, and 43 in the street) ; (2) that it contained between 10 and
* Brit. Med. Jour., 1895, vol. ii., p. 519.

BREAD 277

24 per cent, less moisture than outside air surrounding the bake-
houses ; and (3) that it contained at least four times more bacteria
than surrounding street air, and three times more bacteria than the
air of a typical above-ground bakehouse.* (See also p. 86.)

The normal fermentation of bread with which all bakers are
familiar is due to the energy of the yeast plant growing in the
dough. Any other fermentation going on at the same time as the
normal one, or arising after the bread has left the oven, must be
looked upon as abnormal.

Flour or dough is open to infection by bacteria, and scrupulous
cleanliness is absolutely necessary to avoid unfavourable fermenta-
tions. Bacteria are especially numerous in low-grade flours ; in fact,
the poorer the flour the larger the number of injurious organisms.
Prescott has lately shown that flour may contain bacilli indistinguish-
able from the B. coli. This organism is more liable to be found
in poor than in high-class flours.

1. Sour Bread. — The commonest abnormal fermentation of bread
produces what is known as "sour bread," which means that the
odour and flavour of the bread are " sour " to the senses of smell and
taste. Lactic and butyric germs are commonly found in poor flours,
where they remain in a dormant condition until provided with the
essentials necessary for their growth — moisture, a sufficient tempera-
ture, and proper and adequate food supply. The food supply
naturally surrounds them, and when water is added to the flour,
and the temperature is raised to between 70° and 90° E, they are
able to reproduce and rapidly man il'est their presence by the products
they form. Dough, with considerable moisture present, or, as it is
termed, " slack," gives bacteria a better environment, and consequently
sourness is more apt to increase rapidly in such doughs.

Acid-producing germs are also present in many Bamples of
yeast. Analyses of a large number of yeast samples used for
bread-making purposes, have shown that many of them contain
injurious bacteria which may lessen the alcoholic fermentation.
If, on the other hand, the normal alcoholic fermentation is at
first vigorous, and then diminishes, it gives bacteria opportunity to
grow. Hence, overproved dough is especially liable to bee. .me sour.
Dirty utensils, tubs or troughs, harbour injurious bacteria which are
able to reproduce when given favourable conditions. All cracks and
crevices which harbour food are teeming with life, usually undesirable
from the bakers' standpoint, and, therefore, absolute cleanliness
should be the rule in every detail.

Acetic bacteria, which are often present in flours, sometimes
cause trouble, and as these bacteria, require a, plentiful supply of
oxygen, it has been suggested that all dough should be kept as much

* Special Report on Bakeliouses in Fiii.sburi/, 1902.

278 BACTERIA IN OTHER FOODS

as possible out of contact with the air. It is doubtful if such a
remedy is practical, as the lowering of the temperature follows the
removal of covers on the dough troughs and retards the whole course
of fermentation.

2. Sticky, Slimy, or Viscous Bread. — This affection is not nearly
as common as the preceding, yet the number of cases recorded is
quite large, and this abnormal fermentation is frequently met with
in country districts. As the name implies, the bread, usually the
crumb near the centre of the loaf, is slimy or sticky. The stringiness
increases with age, a proof of the living nature of the trouble.
Cases of sticky bread usually occur in the warm summer months,
the high temperature favouring the growth of the bacteria which pro-
duce the trouble. From this sticky bread it is comparatively easy to
isolate an organism which, when placed in sterilised bread, is able
to produce the stickiness met with under natural conditions, thus
proving the relation of bacteria to the trouble. The specific germ
Causing stickiness, known as the "potato bacillus" on account of
the frequency with which it is met with on potatoes, is also formed
in yeast cake. Harrison has repeatedly Pound this germ present in
both dried and compressed yeast cake. Given favourable conditions
for rapid growth, this organism might produce epidemics of slimy
bread at any time. The bacillus forms spores able to resist un-
favourable conditions. This germ is occasionally met with in milk.
Slimy bread may be controlled, by exercising absolute cleanliness
in the yeast tubs and kneading troughs, and by the proper sterilisa-
tion of the brew or ferment by the use of a certain quantity of hops.
In a number of experiments made with hop extracts it has been
found that even a small quantity of good hops (one half-ounce to
the gallon) has some antiseptic power and hinders the development
of the potato bacillus, without injuring the activity of the yeast.
The bread should be kept in a cool place after baking, for this
stickiness is most prevalent during hot weather, and a cool tempera-
ture prevents the rapid growth of the organism.

3. Musty or Mouldy Bread. — Musty or mouldy bread is, as a
rule, only met with after the bread has been cut and allowed to
stand several days. Occasionally, however, we find bread only one
day old affected with mustiness. The specific organism is the mould
Mucor mucedo, which has action on bread, producing a musty odour
without decomposing the bread. But the chemical composition of
the bread is changed by the growth of mould, and this change
favours the subsequent growth of any bacteria that may be present.
Flours which have become damp, or even very low-grade flours, may
have this mould present in large amount, and although the organism
is killed by the baking process, yet the musty flavour persists and is
present in the baked loaf.

WATERCRESS, ETC. 279

4. Red or " Bloody " Bread. — Bloody, or red bread, is not an
affection which often troubles bakers, but it sometimes makes its
appearance in the household. The microbe which produces this
affection is of great historical interest. Livy refers to its occurrence
in the Eoman army, and it is said to have appeared during the siege
of Troy. There are various records of its occurrence in England
during the Middle Ages, and early in the nineteenth century a large
quantity of red spotted bread occurred in the province of Padua, in
North Italy. It is possibly due to B. prodigiosus or other similar
chromogenic organism, and is traceable to contamination of the
bread.

5. Miscellaneous Foods

Watercress has frequently been found to be the vehicle of bacteria
if grown in polluted water. There are several instances on record
where the consumption of such contaminated watercress has caused
disease. In June and July 1903 an outbreak of enteric fever
occurred in Hackney, in N.E. London, in which there were 110
cases of the disease, of whom 55*5 per cent, had consumed watercress
which was shown to have been grown in polluted water. The latter
contained 50 B. coli per c.c, and the cresses themselves were markedly
contaminated with sewage organisms of intestinal type. Altogether,
17 samples of watercress were examined, and every one of them
revealed the bacteria of sewage. This was a fairly clear case of
conveyance of enteric infection, as (1) the excess of enteric fever
corresponded with the season for watercress, viz., June to September;
(2) the excess of cases of enteric fever was amongst watercress
eaters, viz., 55 per cent, for the whole period ; (3) watercress caters
suffered more than three times as much as non-watercress eaters,
who constituted only 27'5 per cent, of the entire population; (4)
samples of watercress, taken from the places where infected persons
market, were found on bacteriological examination, to bo sewage-
polluted; and, (5) a large proportion of the polluted samples were
found to be cultivated in beds fed by almost undiluted Bewage.*

Other foods and beverages (including aerated waters) have from
time to time been contaminated with bacteria to the injury of the
consumer, but the above represent the chief foods infected. Sausages
have frequently been found to be contaminated. In Liverpool,
Boyce found B. coli present in all 17 samples examined, and B.
enteritidis sporogenes in 2 out of 17. Porkpies, tinned meats and pastes,
chicken, jellies, etc., have been shown to harbour injurious organism s.f

* Report on Outbreak of Enteric Fever at Hackney, 1903 (Dr King Warry).
\ Report on Health of Liverpool, 1902, p. 172.

CHAPTEE IX

BACTERIA AND DISEASE

Growth of Knowledge of Bacteria as Disease Producers — Channels of Infection —
How Bacteria cause Disease — Diphtheria : Conditions of Infection — Scarlet
Fever, Typhoid Fever, Epidemic Diarrhoea : Conditions of Infection —
Suppuration and Abscess Formation — Anthrax — Pneumonia — Influenza —
Actinomycosis — Glanders.

Pkobably the most universally known fact respecting bacteria is
that they are related in some way to the production of disease.
Yet we have seen that it was not as disease-producing agents that
they were first studied. Indeed, it is only within comparatively the
latest period of the two centuries during which they have been
more or less under observation that our knowledge of them as
causes of disease has assumed any exactitude or general recognition.
Nor is this surprising, for although an intimate relationship between
fermentation and disease had been hinted at in the middle of the
seventeenth century, it was not till the time of Pasteur that the
bacterial cause of fermentation was experimentally, and finally,
established.

In the middle of the seventeenth century men learned, through
the eyes of Leeuwenhoek, that drops of water contained "moving
animalcules." A hundred years later Spallanzani demonstrated the
fact that decomposition and fermentation were set up in boiled
vegetable infusions when outside air was admitted, but when it
was withheld from these boiled infusions no such change occurred.
Almost a hundred years more passed before the epoch-making work
of Tyndall and Pasteur, who separated these putrefactive germs
from the air. Quickly following in their footsteps came Davaine

280

KOCH'S POSTULATES 281

and Pollender, who found in the blood of animals suffering from
anthrax the now well-known specific bacillus of that disease.
Improvements in the microscope and in methods of cultivation
(Koch's plate method in particular) soon brought an army of zealous
investigators into the field, and during the last thirty years one
disease after another has been traced to a bacterial origin. We may
summarise the vast collection of historical, physiological, and patho-
logical research extending from 1650 to 1904 in three great periods :
The period of detection of living, moving cells (Leeuwenhoek and
others in the seventeenth century) ; the period of the discovery of
their close relationship to fermentation and putrefaction (Spallanzani,
Schulze, Schwann, in the eighteenth century); and, thirdly, the
period of appreciation of the role of bacteria in the economy of
nature and in the production of disease (Tyndall, Pasteur, Lister,
Koch, in the nineteenth).

But we must look less cursorily at the growth of the idea of
bacteria as the cause of disease. More than two hundred years ago
Eobert Boyle (1627-91), the philosopher who did so much towards
the foundation of the present Eoyal Society, wrote an elaborate treatise
on The Pathological Part of Physic. He was one of the earliest
scientists to declare that a relationship existed between fermentation
and disease. When more accurate knowledge was attained respecting
fermentation, great advance was consequently made in the etiology
of disease. The preliminary discoveries of Fuchs and others between
1840 and 1850 had relation to the existence in diseased tissues of
a large number of bacteria. But this was no proof that these germs
caused disease. It was not till Davaine had inoculated healthy
animals with bacilli from the blood of an anthrax carcase, and had
thus reproduced the disease, that reliance could be placed upon that
bacillus as the vera causa of anthrax. Too much emphasis cannot
be laid upon the idea, that unless a certain organism produces in
healthy tissues the disease in question, it cannot be considered as
proven that the particular organism is related to the disease as
cause to effect. In order to secure a standard by which all investi-
gators should test their results, Koch introduced four postulates.
Until each of the four has been fulfilled, the final conclusion respect-
ing the causal agent in any bacterial disease must be considered sub
judice. The postulates are as follows : —

(a) The organism must be demonstrated in the circulation or
tissues of the diseased animals.

(b) The organism thus demonstrated must be cultivated in
artificial media outside the body, and successive generations of a
pure culture of that organism must be obtained.

(6') Such pure cultures must, when introduced into a healthy and
susceptible animal, produce the specific disease.

282 BACTERIA AND DISEASE

(d) The organism must be found and isolated from the circulation
or tissues of the inoculated animal.

It is evident that there are some diseases — for example, cholera,
leprosy, and typhoid fever — which are not communicable to lower
animals, and therefore their virus cannot be made to fulfil postulate
(c). In such cases there is no choice. They cannot be classified
along with tubercle and anthrax. Bacteriologists have little doubt
that Hansen's bacillus of leprosy is the cause of that disease, yet
it has not fulfilled postulates (b) and (c). Nor has the generally
accepted bacillus of typhoid fever fulfilled postulate (c), yet by the
majority it is provisionally accepted as the agent in producing the
disease. Hence it will be seen that, though there is an academical
classification of causal pathogenic bacteria according as they respond
to Koch's postulates, yet nevertheless there are a number of patho-
genic bacteria which are looked upon as causes of disease provisionally.
The bacilli of anthrax and tubercle, with perhaps the organisms of
suppuration, tetanus, plague, and actinomycosis, stand in the first
order of pathogenic germs. Then comes a group awaiting further
confirmation, which includes the organisms related to typhoid fever,
cholera, malaria, leprosy, epidemic diarrhoea, and pneumonia. Then
comes in a third category, a long list of diseases, such as scarlet
fever, small-pox, measles, rabies, and others too numerous to mention,
in which the nature of the causal agent is still unknown. Hence it
must not be supposed that every disease has its germ, and without
a germ there is no disease. Such universal assertions, though not
uncommonly heard, are devoid of accuracy.

In the production of bacterial disease there are two factors.
First, there is the body tissue of the individual ; secondly, there is
the specific organism.

Whatever may be said hereinafter with regard to the power of
micro-organisms to cause disease, we must understand one cardinal
point, namely, that bacteria are never more than causes, for the nature
of disease depends upon the behaviour of the organs or tissues with
which the bacteria or their products meet (Yirchow). Fortunately for
a clear conception of what " organs and tissues " mean, these have
been reduced to a common denominator, the cell. Every living
organism, of whatever size or kind, and every organ and tissue in
that living organism, contains and consists of cells. Further, these
cells are composed of organic chemical substances which are not
themselves alive, but the mechanical arrangement of which determines
the direction and power of their organic activity and of their resist-
ance to the specific agents of disease. With these facts clearly
before us, we may hope to gain some insight into the reasons for
departure from health.

The normal living tissues have an inimical effect upon bacteria.

RESISTANCE OF THE TISSUES 283

Saprophytic bacteria of various kinds are normally present on
exposed surfaces of skin or mucous membrane. Tissues, also, which
are dead or depressed in vitality from injury or previous disease,
but which are still in contact with the living body, afford an excellent
nidus for the growth of bacteria. Still these have not the power,
unless specific, to thrive in the normal living tissue. It has been
definitely shown that the natural fluids of the body have in their
fresh state protective substances (alexincs) which prevent bacteria
from flourishing in these tissues. Such protection depends in measure
upon the number of invading germs as well as their quality, for the
killing power of blood and lymph must be limited. Buchner has
pointed out that the antagonistic action of these fluids depends in part
possibly upon phagocytosis, but largely upon a chemical condition
of the serum. The blood, then, is no friend to intruding bacteria.
Its efforts are to a certain extent seconded by the lymphoid tissue
throughout the body. Rings of lymphoid tissue surround the oral
openings of the trachea and oesophagus, and the tonsils are masses
of lymphoid tissue. Composed as it is of cells having a germicidal
influence when in health, the lymphoid tissue may afford formidable
obstruction to invading germs.

All the foregoing points in one direction, namely, that if the
tissues are maintained in sound health, they form a very resistant
barrier against disease-producing germs. But we know from experi-
ence that a full measure of health is not often the happy condition
of human tissues. There are a variety of circumstances which
predispose the individual to disease. One of the commonest forms
of predisposition is that duo to heredity. Probably it is true that
what are known as "hereditary diseases" are due far more to a
hereditary predisposition than to any transmission of the virus itself
in any form. Again, an f credent disease predisposes the tissues to
form a nidus for bacteria, and conditions of environment or 'personal
habits act powerfully in the same way. Damp soils must be hold
responsible for many disasters to health, not directly, but indirectly,
by predisposition ; dirty houses and insanitary houses, dusty trades
and injurious occupations, have a similar effect. Anv one <>i" these
different influences may in a, variety of ways affecl the tissues and
increase their susceptibility to disease. Nol infrequently we may
get^ them combined. For example, the lull, .win- is not aii unlikely
series of events terminating in consumption (tuberculosis of the
lungs): — (a) The individual is predisposed by inheritance to
tuberculosis; (//) an ordinary chronic catarrh, which lowers the
resisting power of the lungs, may he contracted; (c) the epithelial
collections in the air vesicles of the lung-r-i.e. dead matter attached
to the body — afford an excellent nidus I'm' bacteria; (d) owing to
occupation, or personal habits, or surroundings, the patient comes

284 BACTERIA AND DISEASE

within a range of tubercular infection, and the specific bacilli of
tubercle gain access to the lungs. The result will be a case of
consumption more or less acute according to environment and
treatment.

Channels of Infection

The channels of infection by which organisms gaiii the vantage-
ground afforded by the depressed tissues are various, and next to
the maintenance of resistant tissues they call for most attention
from the physician and surgeon. It is in this field of preventive
medicine — that is to say, preventing infective matter from entering
the tissues at all — that science has triumphed in recent years. It
is, in short, applied bacteriology.

1. Pure Heredity. — This term is to be understood in this connec-
tion as concerned with actual transmission of germs of disease from
the mother to the child in utero. That such conveyance may occur
is admitted, but it is certainly not frequent, nor is bacterial disease
widely spread by this means. The transmission of tendency (diathesis)
is, of course, another matter, and there can be little doubt that ante-
natal conditions exert an influence on bacterial diseases of infancy.

2. Inoculation, or inserting virus directly through a broken sur-
face of skin, is a method of producing diseases in animals commonly
used in experimental work. Such inoculations may be subcutaneous,
intravenous, intracerebral, intraperitoneal, etc. In the natural pro-
duction of disease, inoculation is also a not uncommon channel of
infection. Injuries of the skin caused by instruments, gunshot
wounds, broken glass or china, etc., may serve as the point of intro-
duction of specific virus. Tetanus is commonly an inoculated disease.
Malaria must now also be so considered. Local tuberculosis is not
infrequently produced by inoculation through a broken skin surface.

3. Contagion indicates that a disease is transmitted by personal
contact, through unbroken skin surfaces. Small-pox, measles, ring-
worm, and other diseases may be thus contracted. It is not unlikely
that as our knowledge grows, the diseases to be defined as spread by
contagion will become less.

4 The Alimentary System. — Many diseases are spread by the
consumption of infected food or water, and in children the sucking
of dirty objects may introduce germs of disease into the alimentary
canal. Milk, cream, butter, cheese, ice-cream, oysters, shell-fish,
meats of various kinds, vegetables, water-cress, ice, and a large variety
of foods, have been the means of introducing pathogenic organisms
into the body, and in this way enteric fever, cholera, dysentery, and a
large number of acute and chronic diseases are originated. Water-
borne disease furnishes a large percentage of such cases.

5. The Respiratory Tract. — The air may become infected with

ACTION OF BACTERIA 285

pathogenic organisms, which may be inhaled, and thus gain entrance
to the body and set up disease. Diphtheria and pulmonary tubercu-
losis are two examples. In this channel of infection pathogenic
bacteria must, as a rule, be present in large numbers, or must meet
with devitalised and non-resisting tissues, to set up disease.

These, then, are the five possible ways in which germs gain
access to the body tissues. The question now arises, How do
bacteria, having obtained entrance, set up the process of disease ? For
a long time pathologists looked upon the action of these microscopic
parasites in the body as similar to, if not identical with, the larger
parasites sometimes infesting the human body. Their work was
viewed as a devouring of the tissues of the body. Now it is well
known that, however much or little of this may be clone, the
specific action of pathogenic bacteria is of a different nature. It is
twofold. We have the action of the bacteria themselves, and also
of their products or toxins. In particular diseases, now one and
now the other property conies to the front. In bacterial diseases
affecting or being transmitted mostly by the blood, it is the toxins
which act chiefly. The convenient term infection is applied to those
conditions in which there has been a multiplication of living-
organisms after they have entered the body, the word intoxication
indicating a condition of poisoning brought about by their products.
It will be apparent at once that we may have both these conditions
present, the former before the latter, and the latter following as a
direct effect of the former. Until intoxication occurs, there may be
few or no symptoms; but directly enough bacteria are present to
produce in the body certain poisons in sufficient amount to result in
more or less marked tissue change, then the symptoms of that tissue
change appear. This period of latency between infection and the
appearance of the disease is known as the incubation period. Take
typhoid, for example. A man drinks a typhoid-polluted water.
For about fourteen days the bacilli are making headway in his body
without his being aware of it. But at the end of that incubation
period the signs of the disease assert themselves. Prof essor Watson
Cheyne and others have maintained that there is some exact
proportion between the number of bacteria gaining entrance and the
length of the incubation period.

Speaking generally, we may note that pathogenic bacteria divide
themselves into two groups: those which, on entering the body,
pass at once, by the lymph or blood-stream, t<» all parts of the body,
and become more and more diffused throughout the blood and
tissues, although in some cases they settle down in some spot
remote from the point of entrance, and produce their chief lesions
there. Tubercle and anthrax would be types of this group. On
the other hand, there is a second group, which remain almost

286 BACTERIA AND DISEASE

absolutely local, producing only little reaction around them, rarely
passing through the body generally, and yet influencing the whole
body eventually by means of their ferments or toxins. Of such, the
best representatives are tetanus and diphtheria. The local site of
the bacteria is, in this case, the local factory of the disease.

Whilst the mere bodily presence of bacteria may have mechanical
influence injurious to the tissues (as in the small peripheral
capillaries in anthrax), or may in some way act as a foreign body
and be a focus of inflammation (as in tubercle), the real disease-
producing action of pathogenic bacteria depends upon the chemical
poisons (toxins) formed directly or indirectly by them. Though
within recent years a great deal of knowledge has been acquired
about the formation of these bodies, their exact nature is not at
present known. They are allied to the proteoses, and are frequently
described as tox-albumens. It may be found, after all, that they
are not of a proteid nature. Sidney Martin has pointed out that
there is much that is analogous between the production of toxins
and the production of the final bodies of digestion. Just as ferments
are necessary in the intestine to bring about a change in the food
by which the non-soluble albumens shall be made into soluble
peptones, and thus become absorbed through the intestinal wall, so
also a ferment may be necessary to the production of toxins. Such
ferments have not as yet been isolated, but their existence in
diphtheria and tetanus is, as we have seen, extremely likely.
However that may be, it is now more or less established that
there are two kinds of toxic bodies, differing from each other in
their resistance to heat. It may be that the one most easily
destroyed by heat is a ferment and possibly an originator of the
other. A second division which has been suggested for toxic bodies,
and to which reference will be made, is intracellular and extra-
cellular, according to whether or not the poison exists within or
without the body of the bacillus.

Lastly, we may turn to consider the action of the toxins on the
individual in whose body-fluids they are formed. It is hardly
necessary to say that any action which bacteria or toxins may have
will depend upon their virulence, in some measure upon their
number, and not a little upon the channel of infection by which
they have gained entrance. It could not be otherwise. If the
virulence is attenuated, or if the invasion very limited in numbers,
it stands to reason that the pathogenic effects will be correspondingly
small or absent. The influence of the toxins is twofold. In the
first place, (i.) they act locally upon the tissues at the site of their
formation, or at distant points by absorption. There is inflamma-
tion with marked cell-proliferation, and this is, more or less rapidly,
followed by a specific cell-poisoning. The former change may be

ACTION OF BACTERIA 287

accompanied by exudation, and simulate the early stages of abscess
formation ; the latter is the specific effect, and results, as in leprosy
and tubercle, in infective nodules. The site in some diseases, like
typhoid (intestinal ulceration, splenic and mesenteric change) or
diphtheria (membrane in the throat), may be definite and always the
same. But, on the other hand, the site may depend upon the point
of entrance, as in tetanus. The distant effects of the toxin are due
to absorption, but what controls its action it is impossible to say.
We only know that we do find pathological conditions in certain
organs at a distance from the site of disease, and without the
presence of bacteria. We have a parallel in the action of drugs ;
for example, a drug may be given by the mouth, and yet produce a
rash in some distant part of the body. In the second place, (ii.)
toxins produce toxic symptoms. Fever and many of the nervous
conditions resulting from bacterial action must thus be classified.
We have, it is true, the physical signs of the pathological tissue
change, for example, the large spleen of anthrax or the obstruction
from diphtheritic membrane. But, in addition to these, we have
general symptoms, as fever, in which after death no tissue change
can be found.

We may now consider briefly some of the more important forms
of disease produced by bacteria.*

Diphtheria

Diphtheria is an infective disease characterised by a variety
of clinical symptoms, including a severe inflammation usually
followed by a fibrinous infiltration (constituting a membrane) of
certain parts. The membrane ultimately breaks down. The parts
affected are the mucous membrane of the fauces, larynx, pharynx,
trachea, and sometimes wounds, or the inner wall of the stomach.
Diphtheritic conjunctivitis may also occur. The common sign
of the disease is the membrane in the throat; but muscle

- Bacterial diseases may be classified as follows : —

(1) Diseases common to man and certain animals, and presumably trans-

missible from animals to man, and vicever.su, e.g. bubonic plague and
tuberculosis.

(2) Diseases common to man and animals, but not known to be directly

transmissible, e.g. actinomycosis, tetanus. Diphtheria, belongs to this
class, or Group (1) or (5).

(3) Diseases transmitted from animals to man, but not as a rule communicated

from man to man owing to interfering conditions, e.g. anthrax, glanders,
rabies, vaccinia, foot-and-mouth disease, meat-poisoning, psittacosis,
and possibly infections due to pus bacteria.

(4) Certain specific symbiotic relations requiring two hosts for the complete

cycle of life of the micro-organisms, e.g. malaria, trichinosis, tape-worm
infection.

(5) Diseases occurring in man, but not, as far as known, in animals, e.g.

typhoid fever, gonorrhoea, leprosy.

288 BACTERIA AND DISEASE

weakness, syncope, albuminuria, post-diphtheritic paralyses, con-
vulsions, and many other symptoms guide the physician in
diagnosis and the course of the disease. It begins as a local
disease, and the greyish-white membranous deposit, already referred
to, is produced. The toxins or poisons resulting from the growth
and multiplication of the bacillus are absorbed into the blood stream,
and general symptoms follow. The incubation period is from two
to seven days.

Although diphtheria owes its name to the false membrane seen
in the throats of typical cases, it is now almost universally recognised
that in many cases of undoubted diphtheria no membrane is formed.
The occurrence of a nasal form of diphtheria has, too, in recent years
been recognised, and as such cases are not easily recognisable without
a bacteriological examination, they are very liable to remain undi-
agnosed and be left free to spread the infection.

The fons et origo of the disease is the specific bacillus. Without
the presence of that organism it is not possible to have diphtheria.
Yet that organism may exist in the healthy throat without producing
the recognised clinical symptoms of diphtheria. It may be conveyed
to the human throat in a variety of ways, for example, by kissing
and other forms of contact, or by drinking milk and other con-
taminated foods. In a perfectly healthy throat it may do no mischief.
But in a sore throat or in the throat of a weakly person, it might
readily set up severe and even fatal disease. Anything, therefore,
which tends to lower the vitality of the individual may play an
important part in propagating diphtheria, and must be as carefully
considered as any agency which might directly or indirectly introduce
the bacillus to the human throat. Some epidemics have been due
to school influence; other epidemics have been brought about
through an infected milk supply ; and yet other outbreaks are due to
the introduction of a case of diphtheria into a susceptible community,
weakened by insanitary surroundings or the prevalence of previous
sore throat.

Further, there is reason to suppose that Bacillus diphtherial may
retain its virulence — and possibly spend a stage of its cycle of
existence as a saprophyte — in the soil, in dust, and even in the
throat for months. Three or four weeks is the average length of
time for its presence in the throat, but, as a matter of fact, all the
conditions in the' throat — mucous membrane, blood-heat, moisture, and
air — are extremely favourable to the bacillus, and it may linger
there far beyond the time of disappearance of clinical symptoms of
the disease.

The Bacillus diphtherial was isolated from the many bacteria
found in the membrane by Loffler (1884). Klebs had previously
identified the bacillus as the cause of the disease (1883). It is a

PLATE 19.

* y-

^ _ ur^

1 — ^5

Bacillus diphtherias.

Film preparation from serum culture, 24 hours at '67 C.
x 1000.

Bacillus von Hofmann.

Film preparation from serum culture, 4S hours at 37 C. Boux'a stain.

X 1000.

To face page 288.

DIPHTHERIA 289

slender rod, straight or slightly curved, and remarkable for its
beaded appearance ; there are also irregular and club-shaped forms.
It differs in size according to its culture
medium, but is generally 3 or 4 ^ in
length. Gobbett and Graham-Smith re-
cognise five morphological types of diph-
theria bacilli on young serum cultures:
— (1) Oval bacilli, with one unstained
septum ; (2) long, faintly-stained, irregu-
larly-beaded bacilli; (3) long, regularly-
beaded bacilli — " streptococcal " forms ;
(4) segmented bacilli ; and (5) uniformly-
stained bacilli. All these forms are
longer than the pseudo-diphtheria bacilli,
more curved, and generally with clubbed fig. 24.— Diagram .or Bacillus
ends. Their arrangement to each other (%pit

is generally likened to Chinese characters. In the membrane
which is its strictly local habitat in the body — indeed, with the
exception of the secretions of the pharynx and larynx, and occasion-
ally in lymph-glands and the spleen, the bacillus is found practically
nowhere else in the body — it sometimes shows parallel grouping,
lies on the surface of the exudation, and is separated from
the mucous membrane by the fibrin. It is mixed with other
organisms, which are performing their part in complicating the
disease, or are normal inhabitants of the mouth. The bacillus
possesses five negative characters: namely, it has no spores, no
threads, and no power of motility ; it does not liquefy gelatine, nor
does it produce gas. It is stained with Loffler's methylene blue, and
shows metachromatic granules and polar staining. It is generally
best to use the stain in dilute form. The favourable temperature is
blood-heat, though it will grow at room temperature. It is aerobic,
and, indeed, prefers a current of air. Loftier contrived a medium
for cultivation which has proved most successful. It is made by
mixing three parts of ox-blood serum with one pail of broth contain-
ing 1 per cent, of glucose, 1 per cent, of peptone, and \ per cent, of
common salt; the whole is coagulated. Upon this medium the
Klebs-Loffler bacillus grows rapidly in eighteen or twenty hours,
producing scattered, "nucleated," round, white colonies, becoming
yellowish. Horse serum is used by some bacteriologists instead of
ox serum. Lorrain Smith and Marmorek also devised excellent
serum media. The bacillus grows well in broth, but without
producing either a pellicle or turbidity; it can grow on the
ordinary media, though its growth on potato is not readily
visible; on the white of egg it flourishes extremely well. In a
moist condition, as a rule, the bacillus has a low degree of

T

290 BACTERIA AND DISEASE

resistance, but when in a dry condition it can survive for long
periods.

The bacillus is secured for diagnostic purposes by one of two
methods : (a) Either a piece of the membrane is detached, and after
washing carefully examined by culture as well as the microscope ;
or (b) a " swab " is made from the infected throat, cultured on serum,
and incubated at 37° C. for eighteen hours, and then microscopically
examined. Both methods — and there is no further choice — present
some difficulties owing to the large number of bacteria found in the
throat. Hence a negative result must be accepted with reserve.
Indeed the rule to follow is three examinations before deciding that
a throat is free from infectivity, and it should be remembered that
about 20 per cent, of all cases of diphtheria offer no bacteriological
evidence of infection.* It therefore comes back to the point of
broad judgment and common-sense. The clinical condition is the
main fact for guidance, and the bacteriological must not usurp

it.t

Locally, the bacillus produces inflammatory change with fibrinous
exudation and some cellular necrosis. In the membrane a ferment
is probably produced which, unlike the localised bacilli, passes
throughout the body, and by digestion of the proteids, produces
albumoses and an organic acid which have the toxic influence. The
toxins act on the blood-vessels, the nerves, the muscle fibres of the
heart (hyaline degeneration and even fatty changes), and many of the
more highly specialised cells of the body. Thus we get degenerative
changes in the kidney (cloudy swelling, and, clinically, albuminuria)
in cells of the central nervous system, in the peripheral nerves
(post -diphtheritic paralysis), and elsewhere, these pathological condi-
tions setting up, in addition to the membrane, the symptoms of the
disease. The bacillus is pathogenic for the horse, ox, rabbit, guinea-
pig, cat, and some birds. Cases are on record of supposed infection
of children by cats suffering from the disease. Koux and Yersin
in 1888-89 showed that the diphtheria bacillus is capable of pro-
ducing the various phenomena associated with the disease, including
the formation of false membrane and diphtheritic paralysis. They
also succeeded in separating and studying the toxin, which they
found to be capable of producing all the effects produced by the
bacillus. In 1890 appeared the great work by Behring, to which
reference will be made subsequently ; and the observations in regard
to diphtheria made in that work were extended and strengthened in
a paper by Behring and Wernicke in 1892. At the Medical Congress
at Buda-Pesth in 1894, Eoux read a paper on the treatment of diph-
theria by diphtheria antitoxin, which first proved to the medical

* Jour, of Hygiene, 1903, p. 217.

j See also Brit. Med. Jour., 1900, ii., p. 907 (Andrewes).

DIPHTHERIA 291

world that this was the one method of successfully combating the
disease. The experimental and clinical data, and the favourable
statistics brought forward by Koux, at once put this method in a
secure position from the practical standpoint.*

Conditions Favourable to the Spread of Diphtheria.—

Sir Kichard Thorne Thorne held that soil, and especially surface soil,
when considered in connection with relative altitude, slope, aspect,
and prevailing winds, plays a part in the maintenance and diffusion
of diphtheria, and possibly has relation with its beginnings. He
believed that where a surface soil retained moisture and organic
refuse, and where buildings founded on such soil were exposed to
cold wet winds, there you had conditions likely to foster diphtherial
Dr Newsholme of Brighton considers that such conditions act as
"vital depressants," favouring sore throat and catarrh, and thus
preparing the way for the inroads of the diphtheria bacillus. He
concludes as the result of extended investigations that " diphtheria
only becomes epidemic in years in which the rainfall is deficient, and
the epidemics are on the largest scale when three or more years of
deficient rainfall immediately follow each other. Occasionally, dry
years are unassociated with epidemic diphtheria, though usually in
these instances there is evidence of some rise in the curve of
diphtheritic death-rate. Conversely, diphtheria is nearly always at
a very low ebb during years of excessive rainfall, and is only
epidemic during such years when the disease in the immediately
preceding dry years has obtained a firm hold of the community and
continues to spread presumably by personal infection." J Newsholme
thinks the specific micro-organism has a double cycle of existence :
one phase passed in the soil, saprophytic ; another in the human
organism, parasitic.

Insanitary surroundings necessarily act prejudicially. Damp and
ill-constructed houses and bad drainage, have undoubtedly played a
part, and that not a small part, in diphtheria outbreaks. The
position of many houses must inevitably lead to dampness; there is
also the dampness arising from imdrained and unpaved curtilages ;
and lastly, there is damp and steamy atmosphere. Now these con-
ditions cannot but affect the health of children and give rise to sore
throats and similar complaints. Wheu to this dampness of houses
is added the pollution of the soil, the undrained condition of towns,
and the nuisances readily arising from ash-pits, cess-pits, and similar
methods of refuse removal in close proximity to the houses, there is

:: See also Brit. Med. J oar., 1900, ii., pp. 658-62 (Marsden) ; Metropolitan
Asylums Board Reports, 1869-1902. See also pp. 425-431 of present volume.
t The Natural History of Diphtheria, p. 17.
X Epidemic Diphtheria (Newsholme), p. 157.

292 BACTERIA AND DISEASE

ample ground for concluding that the sanitary circumstances of a
town may be such as to depress the physical vitality of children, and
lessen their powers of resistance to infectious disease once introduced
among them. Thus the insanitary conditions named weaken the
physique of the children, as well as preparing favourable external
circumstances for the growth and multiplication of the germs of
disease. Hence it must come about that from time to time a disease
like diphtheria will take on an increased virulence as well as a higher
measure of epidemicity.

But general conditions do not wholly account for the occurrence
of diphtheria. Apart from these general conditions personal infection
is the chief means by which diphtheria is spread.

Infection has been proved to be conveyed by nasal discharge of
infected persons, or by kissing infected persons, or by sucking sweets,
pencils, pens, slates, and other articles in schools. School influence
as an agency in the dissemination of diphtheria was shown as far
back as 1876 by Mr W. H. Power, and since that date abundant con-
firmatory evidence has been forthcoming. In 1894 Mr Shirley
Murphy, medical officer to the London County Council, reported
that there had been an increase in diphtheria mortality in London
at school ages (three to ten) as compared with other ages since the
Elementary Education Act became operative in 1871 ; that the
increased mortality from diphtheria in populous districts, as com-
pared with rural districts, since 1871, might be clue to the greater effect
of the Education Act in the former ; and that there was a diminution
of diphtheria in London during the summer holidays at the schools
in 1893, but that 1892 did not show any marked changes for August.
In 1896 Professor W. E. Smith, the medical officer to the London
School Board, furnished a report* on this same subject of school
influence, in which he produced evidence to show that the recru-
descence of the disease in 1881-90 was greatest in England and
Wales at the age of two to three years, and in London at the age of
one to two years, in both cases before school age; that age as an
absolute factor in the incidence of the disease is enormously more
active than any school influence, and that personal contact is another
important source of infection.

Although it is said that "statistics can be made to prove
anything," there can be little doubt that both of these reports
contain a great deal of truth ; nor are these truths wholly incompatible
with each other. They both emphasise age as a great factor in the
incidence of the disease, and whatever affects the health of the child,
population, like schools, must play, directly or indirectly, a not
unimportant part in the transmission of the disease.

* Jour, of State Med., 1896, vol. iv., p. 169; see also L.C.C. Education Meport,
1904, No. 718.

DIPHTHERIA 293

Infection can also be conveyed by means of milk (see p. 211).
Clothing and other articles which have been in contact with infected
persons may carry the bacillus. Birds and cats also have been, as
far as can be judged, channels of infection (Klein, Bruce Low, and
others). But there is from a bacteriological point of view another
body of facts altogether which affect the spread of the disease, namely,
the behaviour of the bacillus in the throat.

The Diphtheria Bacillus in the Human Throat.— Since 1896

it has been known that the diphtheria bacillus may remain for long
periods in the throat.

"The persistence of the diphtheria bacillus for periods up to eight
weeks is of very common occurrence whether antitoxin be given or
not; indeed, the majority of cases appear to retain bacilli in the
throat for from two to nine weeks." * After the ninth week, the
number falls off very rapidly, but not infrequently the bacillus
remains in the throat for 100 days, and it has been known to
remain more than 200 days. This persistence of diphtheria bacilli
in the throat may play an important part in determining the spread
of the disease by means of such cases which are supposed to be no
longer infective. "For it is now a matter of common experience
that so long as these diphtheria bacilli, even the loss virulent forms,
remain in the crypts of the tonsils, etc., so long is the patient a
centre of infection, the diphtheria bacilli present resuming, under
favourable conditions, their move virulent form" (Woodhead). In
this way diphtheria bacilli can be readily transmitted by patients
who are apparently no longer suffering from the effects of the disease,
to those who have weak or ulcerated throats. In precisely a similar
manner, may the bacillus be conveyed to ail ides of at lire and articles
of food, such as milk (as at Leeds in 1903).

Further, whilst in 72 per cent, of notified definite eases of
diphtheria the bacillus may be found, it has been shown (hat in
ii] )parently healthy persoi is who have nol suffered Prom diphtheria,
the B. diphtheria may be present. Loftier found diphtheria bacilli in
the throats of 4 out of 160 healthy children, and Park and Beebe
found similar virulent bacilli in 8 out of 3o0 "healthy" throats.
Hewlett and Murray found 15 per cent, of the children in a general
hospital had diphtheria bacilli in their throats.f Kober} examined
diphtheria cultures from two series of healthy persons. The first
series comprised 128 individuals known to have been in recent contact
with patients suffering from diphtheria, The diphtheria bacillus was

* Report on the Bacteriological Diagnosis and Antitoxic Serum Treatment of Cases
admitted to the Hospitals of the Metropolitan Asylums Board, !S9f>-96, by Professor
Sims Woodhead, sect. 2, 1902, pp. 14, 28, 31.

t Brit. Med. Jour., 1901, vol. i., p. 1474.

% Revue des Maladies de VEnfance, Juillet, 1900.

294 BACTERIA AND DISEASE

found in the throat of 8 per cent, of these. The second series com-
prised 600 individuals who had not recently come into contact with
any diphtheria cases — from 5 of these a bacillus similar to diphtheria
was isolated. It was rather short, with swollen ends, and was not
pathogenic to guinea-pigs. Denny of Brooklyn examined 235
healthy persons, and found the diphtheria bacillus in one case. Biggs
met with 32 cases out of 330 healthy persons, and the Committee of
the Massachusetts Board of Health reported that 1-2 per cent, of
healthy persons amongst the general public are infected with diph-
theria bacilli. Only 17 per cent, of such bacilli appear, however,
to be virulent. Goadby examined 100 healthy school children, and
found 18 with diphtheria bacilli, but the disease was prevalent in
the neighbourhood.*

It is certain that, as a rule, " healthy " throats do not yield the
true B. diphtheria? unless those examined have been in contact with
infected persons. But that raises the real difficulty in practical
public health work. The definitely diseased and the definitely
healthy persons can be arranged for. It is the apparently healthy
person, who coming into contact with the infected person acts as a
" carrier " of infection, that creates the problem. The actual per-
centage of such persons varies widely. In infected families it may
be 50 per cent. (Park and Beebe), or 100 per cent. (Goadby). In
schools and institutions the percentage is, of course, lower. Goadby
found it to l)e 34 per cent, in the Poplar Union Schools, but it may be
as low as 7 (Thomas). Aaser found that 19 per cent, of the " contacts "
in a soldiers' barracks contained the bacillus in their throats, and
Denny found in a truant school that the percentage was 11. In
hospital wards it is commonly above 20, and among the general
public above 10 per cent.f These figures at once explain the spread
of diphtheria. They also suggest the methods of prevention.

In the ordinary diphtheria epidemic, whether large or small, these
methods are mainly five. First, the actual cases of diphtheria must
be effectually and promptly isolated. Secondly, the throats of contact
persons should be bacteriologically examined, and those persons in
whose throats the bacillus is found — "carriers" — should be isolated,
and their throats treated. Thirdly, sore throats in the immediate
vicinity of the diphtheria infection should be similarly examined.
Fourthly, thorough disinfection should take place in respect of in-
fected houses, and inquiry made as to school influences, social habits,
etc., of infected households. Fifthly, antitoxin should be used as
prophylactic in infected families.

The new facts respecting the persistence of the bacillus in the

* See Jour, of Hygiene, 1903 (Graham-Smith), p. 216 ; and 1904, p. 255.
t See Jour, of Hygiene, 1904, pp. 258-328 (Graham-Smith); and Practitioner,
1903, vol. ii., pp. 715-21 (Newman).

PSEUDO-DIPHTHERIA BACILLUS 295

throat indicate the importance of throat treatment which there has
been a tendency to ignore on account of the increased use of antitoxin.
But antitoxin has little direct effect upon the bacilli in the throat,
which should, therefore, be treated by painting with perchloride of
mercury (1-500), or washed with chlorine water or permanganate of
potash (1-300). The methods to adopt in order to clean the throat
of the diphtheria bacillus are three, namely, (a) complete isolation of
the patient, coupled with open-air treatment; (b) application of
antiseptics to throat ; and (c) antitoxin.

The Pseudo-diphtheria Bacillus.* — In this group should not
be included non-virulent forms of the diphtheria bacillus, but allied
forms.f Loffler and Hofmann described a bacillus having similar
morphological characters as the true B. diphtheria?, except that it had
no virulence. It is frequently found in healthy throats, and is
believed by some to be a common inhabitant of the mouths of the
poorer classes, especially children. The chief differences between the
real and the pseudo-bacillus are —

1. The pseudo-bacillus is thicker in the middle than at the poles,
and not so variable as the B. diphtheria?. Polar staining is generally
less marked. It appears as an oval bacillus of variable length, gener-
ally having one narrow unstained septum. In broth cultures it often
more closely resembles B. diphtheria?, but under all other conditions
is shorter. It forms no toxins (Cobbett).

2. Hofmann's bacillus forms no acid in glucose culture media.

3. It does not give Neisser's reaction (see pp. 476 and 481) when
grown for twenty-four hours on Loffler'sox serum.

4. The colonies produced by Hofmann's bacillus on blood serum
usually become after a few days larger, more opaque, and whiter
than those of the diphtheria bacillus. They also grow a little more
rapidly than the true bacillus.

5. No pathogenic change is produced in guinea-pigs inoculated
with this bacillus (1 c.c. of a twenty-four hours' broth culture), and
it appears to be innocuous to man. In forming an opinion, all the
facts, including the clinical, if possible, must be taken into considera-
tion. But on the whole, recent evidence appears to support the view
that Hofmann's bacillus possesses little, if any, pathological signifi-
cance in man. It does not agglutinate like B. diphtheria?.

Attempts, followed by some degree of success, have been made to
convert a virulent diphtheria bacillus into Hofmann's bacillus, and
Hofmann's bacillus into a virulent form (Eoux and Yersin,}

* For a fuller statement, see Trans. Jenner Inst. (First Series), pp. 7-32.

t For a discussion of the three forms (a) the true virulent bacillus, (b) the true
non-virulent bacillus, and (c) the pseudo-bacillus, see Report of Metropolitan Asylums
Board 1901 (Gordon Pugh).

\ Ann, de Vlnst. Past., 1890, vol. iv.

296 BACTERIA AND DISEASE

Hewlett and Knight,* Kichmond and Salter,-]- Ohlmacher,J and others).
Evidence in support of the view that Hofmann's bacillus is an
attenuated variety of the true diphtheria bacillus has been brought
forward. But it can only be accepted provisionally. Graham-Smith,
Thomas and others consider the pseudo-bacillus to be absolutely
innocuous. In practice, it is the right course at present to look upon
the presence of the Hof mann bacillus as indicating a suspicious throat.
It should not be forgotten that there are a number of other bacilli
from which the true diphtheria bacillus must be differentiated. § These
include the B. coryzce segmentosus, the bacillus of Hofmann, B. xerosis,
and a number of diphtheroid bacilli, and organisms from nasal and
aural discharge. Similar organisms occur in birds and other animals.
There are, as summarised by Gordon, live chief characters by which
the true diphtheria bacillus may be known: — (a) The macroscopic
and microscopic appearance of the growth on blood serum; (b) the
behaviour of the bacillus to Lofiier's blue, Gram's stain, and Neisser's
stain for granules ; (c) the reaction to litmus of a culture in alkaline
broth, containing 2 per cent, of dextrose after 48 hours at 37° C. ;
(d) the pathogenic test — 1 c.c. of broth culture, 48 hours' growth at
37° C, injected subcutaneously into 200-300 gramme guinea-pig, pro-
duces death generally in 48 hours, whilst post-mortem hemorrhagic
necrosis and oedema are found locally, the internal organs are con-
gested, the pleural, pericardial, and peritoneal fluids are increased,
and the supra-renal capsules are enlarged and engorged with blood ;
(a) the virulence of the organism or its toxin is completely neutralised
by a simultaneous dose of diphtheria antitoxin. For purposes of rapid
diagnosis, (a) and (b) are relied upon.

Scarlet Fever

That the essential cause of scarlet fever is a micro-organism there
can be little doubt. But up to the present time no organism has
been definitely isolated which fulfils the postulates of Koch in respect
to specificity of bacteria. Various organisms have, however, been
described as associated with the disease. Edington, Frankel, Freud-
enberg, Klein, Kurth, Gordon, Baginsky, Class, and others have
described organisms which they believed to be etiologically related to
the disease. At present, however, it can only be said that these
bacteria have been found associated with scarlet fever, but are not yet
proved to be its cause. The organism which appears at present to be

* Trans. Jenner Inst. Prevent. Med. (First Series), 1897, p. 7 et seq.
t Guy's Hospital Report, 1898, vol. liii., p. 55.
X Jour, of Med. Research, 1902, vol. ii., p. 128.

§ Rep. of Local Govt. Board, 1901-02, pp. 418-39 (Gordon); Jour, of Hi/g., 1904,
pp. 299-316,

SCARLET FEVER 297

the most likely cause of the disease is the Streptococcus scarlatina? of
Gordon. Probably the organisms isolated by Baginsky and Class are
different forms of the same streptococcus.

As regards dissemination, it has long been known that scarlet
fever, like small-pox, is most commonly spread by direct infection
through the medium of infected clothing and other articles, or
materials handled by the patient. The means by which infection
has thus been carried are manifold, and need not claim our attention
here. As we have seen, in 1870 a wider field of conveyance of scarlet
fever was revealed by the investigations of Dr M. W. Taylor of
Penrith. While studying an outbreak of scarlet fever, he observed
that the main incidence of the disease fell upon customers of a certain
milk-shop where scarlet fever was existent. Since that date abundant
evidence has been forthcoming to show that to the channels of
infection previously recognised, that of conveyance by milk must be
added. Scarlet fever is disseminated in many ways from person to
person, and also by the vehicle of "fomites." The virus is not
diffusible, but is evidently tenacious of life. Infected garments that
have been put aside for months have been known to originate an
outbreak of the disease. Linen has been known on many occasions
to infect laundresses. There is no evidence that the virus can be
conveyed by water. As a rule, probably the infection of scarlet fever
is not greatly spread by aerial connection, but by articles (toys, books,
bed-clothes, letters, etc.), and such infected articles if set aside in
stagnant air, at a moderate temperature, and in the absence of day-
light, may retain the infection, like garments, for months.

Infectivity begins at the earliest stage of the attack, but is prob-
ably greatest when the fever is at its highest. In most cases the
patient is free from infection at the end of six weeks. There is now
strong evidence that at least the later desquamation is not infective.
Probably the infection lingers longest in the nasal, tonsillar, buccal,
and pharyngeal mucus, and especially in any chronic discharge from
those mucous membranes. Discharges from the car may retain
infection for months.*

It is most probable that milk obtains its infection of scarlet fever
from being brought into contact with persons suffering, as a rule,
from the early and acute stages of the disease.

Streptococcus Scarlatinae (Klein and Gordon). Streptococcus Conglomer-
atus (Kurth). — The organism was isolated from the blood, nasal and tonsillar
discharge of persons suffering from scarlet fever in its earlier and later stages. Not
from urine or skin. It has been isolated from blood of persons dying from scarlet
fever. Assumed to be identical with streptococcus isolated from diseased udders of
cows and from their milk. Found by Klein in ulcerations of teats and udders of

: See also Report to Metropolitan A si/lums Board on lid urn Cases of ScarUt Fever,
by W. J. Simpson, M.D., 1901, p. 24; also 7,7/7. Med, Jour . 1902," vol. ii., p. 44f>
(M. H. Gordon).

298 BACTERIA AND DISEASE

certain cows. Morphology. — A streptococcus; polymorphic; showing- tendency to
oval and rod-shaped elements, especially in impression preparations. Presence of
wedge-shaped, spindle-like, rod-shaped elements in agar and gelatine, and the
characteristic of coherent conglomeration differentiate this streptococcus from others
of the same genus. Irregularity in size and shape of elements ; every transition
between coccus and bacillus. Coccus shape prevails in bouillon, the bacillary being
more common on agar. The streptococcus is stained by simple stains and Gram's
method. Cultural characters: Bouillon. — At 37° C. after 24 hours, the medium
remaining clear, a single, thick, white-grey mass, or several smaller masses, appear
at the foot of the tube ; coherent on shaking the tube, floats through the medium as
a flattened bun-like body. Kurth pointed out that when this mass was examined
under the microscope, a conglomerate appearance was present. The mass is co-
hesive. Gelatine plates and tubes. — Slow growth, forming small grey colonies,
circular or oblong, with firm edge, and consisting of closely set coherent mass of
cocci. Older colonies become nodular. Non-liquefying. In gelatine at 37° C. the
same appearances occur as in bouillon, but often more marked. Chain formation from
these cultures is more marked than in ordinary streptococcus. In gelatine at 37° C. this
organism grows similar to S. long us. Agar plates and tubes. — Three types of colonies
occur after 24 hours : (a) grey, granular, irregularly-outlined tuberculated colonies ;
{b) colonies of similar kind, but having confluent appearance without tubercles ;
(c) younger and smaller colonies which have a fine frilling of chains around a more
compact coherent centre. The most useful feature for differential purposes is the
granular, glossy, coherent centre, combined with tuberculatum. Grows more slowly
than S. pyogenes, and on the whole its colonies on agar are smaller, more opaque, and
more irregular than those of the other streptococci present. Milk. — Rapid coagula-
tion ; produces acid. Sometimes fails to clot milk. A firm, solid clot forms Lit inns
milk, as a rule, within the first 2 days at 37° C. After 24 hours the acid-production
is very strong, and commonly, when there is a clot as well, the lower half of the tube
is yellow-white — the top layer being pink. This decolorisation of lower half of litmus
milk is due to a reducing action of the streptococcus. Chain formation occurs more
than in bouillon. The four chief diagnostic features are : (1) the sediment growth in
broth cultures ; (2) the rapid coagulation of milk ; (3) the acid reaction in litmus
milk; (4) the character of the agar colonies. Pathogenesis. — Pathogenic for mice
and rabbits. After passing through the mouse, the streptococcus takes on a bacillary
form (Gordon), and other modifications, including the diminution of conglomeration,
occur. Its virulence differentiates this streptococcus from streptococci present in
non-scarlatinal throats, except S. pyogenes, which is more virulent to white mice than
S. conglomeratus. Klein holds that this £. conglomeratus is causally related to
scarlet fever in man, and is wholly distinct from S. pyogenes. Gordon has isolated
the latter from the secretion on the surface of the tonsil in a case of clinically mild,
uncomplicated scarlatina. It has also been found like the S. conglomeratus in the
nasal and aural discharge of scarlet fever patients. Gordon believes that both strep-
tococci may play a part in the causation of scarlet fever, but that S. conglomeratus is
the more important of the two, and that it occupies a position in the bacteriological
kingdom between 8. pyogenes and B. diphtherial. *

Typhoid Fever

Typhoid fever is an acute infectious disease characterised clini-
cally by continuous fever, with diarrhoea and other symptoms, and
anatomically by a more or less extensive ulceration of the Peyer's
patches in the intestine (ileum), with swelling of the mesenteric glands
and enlargement of the spleen. The lesion of importance is the ulcera-
tion of the bowel, partly on account of its origin, partly on account of

* For full record of Gordon's researches, see Reports of Medical Officer to Loc.
Gov. Bd., 1898-99, pp. 480-93 ; 1899-1900, pp. 385-457 ; and 1900-01, pp. 353-404.

TYPHOID FEVER 299

its result. The pathogenetic action of the bacilli is obscure, but there
can be no doubt that the ulcers in the intestine are directly or in-
directly the result of the specific bacillus. When the bacilli reach the
intestine they multiply, and, penetrating the mucous and submucous
coats, set up the changes, which lead, first to hyperemia, then to infil-
tration, and finally to ulceration of Peyer's patches. Some of the bacilli
pass into the blood, collecting in the spleen and other glands. Whether
in the bowels or in the organs of the body, the bacilli produce their
toxins, and as a result of their action, inflammation and fever follow.
The inflammation in the intestine leads, in conjunction with the
irritation produced by the ulcers, to increased peristalsis, and there-
fore diarrhoea. Hence the excreta of a typhoid patient have two
characteristics. They are usually abundant and frequent : and they
are charged with large numbers of the bacilli of typhoid fever. It
is, however, necessary to guard against the idea that typhoid fever is
a local disease of the intestine, or even chiefly so. In ordinary cases,
it is true, the intestinal lesions form the starting-point of the disease,
but the bacilli rapidly become generalised, and are found in the most
varied parts of the body, and not uncommonly in the blood itself.
Such a state of things leads to a condition not remote from septi-
caemia, and this may occur with little or no local lesion in the
intestinal tract. The reason why the bacilli of typhoid are not found
in greater number in the blood, is probably in part due to the fact
that in ordinary cases the blood is not a favourable medium for their
growth, and in part to the fact that they are rapidly eliminated or
excreted. "Any conception of the disease," writes Dr Horton-Smith,
" which regards it merely as affecting the alimentary canal, can no
longer be maintained. On the contrary, so far from considering it
an intestinal disease, pure and simple, we should rather look upon it
as a modified form of septicaemia. It is septicaemia in that always,
and in all cases, the bacilli pass into the blood, and then into the
various organs, and in that the symptoms, excepting so far as they
are intestinal, are referable to the poisons there produced. It is a
modified form, however, in that in nearly all cases there is a definite
local and primary disease, whence1 the secondary dissemination of
the micro-organism takes place."*

Whilst it has been held that typhoid infection can pass out of
the infected person by means of the Bweat, the expectoration, the
faeces and the urine, it is only the latter two which need be considered
as a rule. Typhoid stools should always be considered infective,
both in the early and late stages, and the bacilli have even been found
in the stools fifteen days after the temperature lias become normal.
Further, it is possible that the typhoid bacillus may be distributed

* Brit. Med. Jour., 1900, i., pp. 827-31- (Gulstonian Lectures, 1900), P. Horton-
Smith.

300 BACTERIA AND DISEASE

by persons not suffering from the disease. It is believed that the
virus of typhoid fever is chiefly distributed by the contents of the
alimentary canal, and this view is so universally held that it is
unnecessary to elaborate it.

The urine is the other chief excretion by which the bacilli of
typhoid fever are voided from the body. Horton-Smith has demon-
strated that the urine of typhoid patients contains the bacilli of the
disease in the proportion of one in every four cases. He has also
shown, that, as a rule, it is towards the end of the disease, or during
convalescence, that this condition occurs. Further, whilst it is
always difficult to find the bacilli in the stools, in the urine it is
generally easy, for when they are present they are nearly always in
pure culture, and not uncommonly they are present in such extra-
ordinary numbers that one cubic centimetre may contain many
thousands of micro-organisms (Horton-Smith). Cammidge found 37
per cent, of all the typhoid urines examined contained the bacillus.
In one case the organism was found eight months after convalescence.
In London, Horton-Smith found typhoid bacilli present in the
urine in 25 per cent, of all cases examined. Working in Boston,
Eichardson obtained a positive result in 22*5 per cent, of the
cases examined. Both the last-named investigators found the
bacilli present, in certain cases, in such large numbers that the
urine was rendered turbid by their presence. Nor are such
cases rare. Out of the cases in which the specific bacillus was
present in the urine, in as many as twelve it was present to the
degree of turbidity, and in only two was the urine described as
"clear" (Horton-Smith). Eeferring to the stage of the disease in
which the bacilli appear in the urine, they have been found as early
as the thirteenth day from the commencement, and as late as the
fourteenth day of convalescence (Horton-Smith). Speaking gener-
ally, the condition is rare before the third week of the disease. The
duration of this specific bacilluria also varies. The shortest duration
recorded by Horton-Smith was eight days, but in four other cases it
had not disappeared until after the lapse of twenty-one days, twenty-
five days, thirty days, and seventy days. The phenomenon of typhoid
bacilli in the urine probably occurs because one or more bacilli find
their way into the bladder, and there commence rapid growth in the
urine within the bladder, which medium is by no means unfavour-
able to the multiplication of the bacillus (Horton-Smith).

From these facts there are two broad deductions which concern
the bacteriologist and epidemiologist : — First, that enteric fever
occurs as a result of infection by the typhoid bacillus ; and secondly,
that the typhoid bacillus leaves the body of the infected person
through two chief channels, namely, the urinary and alimentary
systems. It has been shown further, that the typhoid bacillus is

TYPHOID FEVER 301

capable of a saprophytic existence in soil, dust, water, milk, and other
natural media. It can survive in ordinary earth for two months,
on sterilised linen for sixty days, on woollen cloth for eighty days, in
sterilised water one hundred and ninety-six days, in particular soils
four hundred days (Martin, Firth and Horrocks, etc.). Therefore it
follows that the organism may remain in the body for long periods,
may pass from the body in urine or in fasces, and find its way into
natural media, and from such media, sooner or later, back to man.
The line of infection may be direct or indirect ; but that it occurs
there can be no doubt.

The Bacillus of Typhoid Fever (Eberth-Gaffky).— The evidence
that Eberth's bacillus is the cause of typhoid fever consists in the
main of three parts : — (1) The bacillus is found with almost invariable
regularity in the spleen of persons dying of typhoid fever, when an
adequate bacteriological examination is made. (2) Eberth's bacillus
elaborates specific toxins. These toxins are for the most part intra-
cellular, contained within the bacillus itself, and are chiefly set free
when the latter is destroyed; and they are comparatively feeble
compared with those of other pathogenic organisms ; and to account
for the clinical conditions of the disease the number of bacilli
present in the infected body would have to be exceptionally large.
This fact, coupled with the varying virulence of the bacillus, is all
the more remarkable when it is remembered that not a few of the
epidemics of the disease have arisen from a dose of poison, so
excessively minute in itself, and so enormously diluted, as to appear
out of all proportion to the number of persons attacked. It is
possible that a few organisms introduced into the human body are
able, under certain conditions, to multiply rapidly, and so bring
about the same results as large dosage. Again and again it has been
shown that considerable epidemics have arisen from a pollution of
water so slight as to escape detection by any methods of chemical
or bacteriological analysis at present known, (o) The blood serum
of individuals suffering from typhoid fever lias a specific agglutinative
action upon the Eberth bacillus, similar to that observed in the
blood serum of animals, rendered immune to this germ (compare
also Pfeiffer's reaction). And whilst there is no evidence to suppose
that animals suffer from typhoid fever as fche disease occurs in man,
there is evidence to show that under certain conditions, a disease,
not unlike enteric fever, can be produced by inoculation of the B.
typhosus into guinea-pigs, mice, rabbits, etc. (Frankel and Simmonds).
Klein has also recently demonstrated by inoculation, that the bacillus
is able to multiply and develop in the lymph-glands of the calf.
For all practical purposes, therefore, the B. typhosus of Eberth is
now generally accepted as the causal agent in typhoid fever.

The channels of infection in typhoid fever are almost entirely

302 BACTERIA AND DISEASE

concerned with the alimentary tract. Water, milk, shell-fish, fried
fish, ice-cream, watercress, etc., have all been proved to be the
vehicle of infection in spreading the disease. Personal contact may
and does operate in spreading infection, and by this means food
also may become contaminated. Flies are held to have acted as
carrying agents in the Spanish-American War of 1898 and the
Boer War of 1900-1902. Corfield records some dozen outbreaks of
typhoid fever due to general insanitary conditions, 60 outbreaks
to infected water, and a large number to minor channels.* The
writer lias collected 160 records of milk-borne outbreaks of the
disease.f

In 1880-81 Eberth announced the discovery of the typhoid
bacillus in cases of clinical enteric fever. In 1884 it was first
cultivated outside the body by G-affky. Since then other organisms
have been held responsible for the causation of enteric (or typhoid)
fever. In 1885 the B. coli communis was recognised, and it has
been a matter of some debate among bacteriologists as to how far
these two organisms are the same species, and interchangeable.
There is evidence on both sides of the question, but bacteriologists
generally regard the Eberth-Gaffky bacillus as the specific cause of
typhoid fever, though complete proof is still wanting.

Under the microscope the bacilli appear as rods, 2-4 jm long,
*5 jul broad, having round ends. Sometimes threads are observable,
being 10 /ul in length. In the field of the microscope the bacilli
differ in length from each other, but are approximately of the same
thickness. Eound and oval cells constantly occur even in pure
culture, and many of these shorter forms of typhoid appear to be
identical in morphology with some of the many forms of B. coli.
There are no spores. Motility is marked ; indeed, in young culture
the typhoid bacillus is the most active pathogenic germ we know.
The small forms move about with extreme rapidity; the longer
forms move in a vermicular manner. Its powers of motility are
due to some five to twenty flagella of varying length, some of them
being much longer than the bacillus itself. The flagella are both
terminal and lateral, and are elastic and wavy.

The organism may be isolated from the ulcerated Peyer's patches
in the intestine, from the spleen, the mesenteric glands, and the
urine. Owing to the mixture of bacteria found elsewhere, it is
generally most readily isolated from the spleen. The whole spleen
is removed, and a portion of its capsule seared with a hot iron to
destroy superficial organisms. With a sterilised knife a small cut
is made into the substance of the organ, and by means of a sterilised
platinum wire a little of the pulp is removed and traced over the

* The Milroy Lectures on Typhoid Fever, 1902.

j Bacteriology of Milk, 1903 (Swithinbank and Newman).

PLATE 20.

- — T~r~~*z. -

-z-&$*

Bacillus typhosus. Film preparation from agar, 16
hours at 37° C. Stained with carbol fuchsin.
x 1000.

W *

' - *

Bacillus typhosus, showing fiagella. x 1000.

6

Widal-Gruber Reaction. Agglutination of /.'.
typhosus by blood serum of a typhoid
patient. X 400.

:v

*c

****,'/

*-■>$

Bacillus typhosus. From human mesenteric gland. Stained with
methylene blue. X 700.

[ To face page 302.

TYPHOID FEVER 303

surface of agar. The agar reveals a growth in about twenty-four
hours at 37° C, which is the favourable temperature. A greyish,
moist, irregular growth appears, but it is invariably attached to
the track of the inoculating needle. On agar plates the superficial
colonies appear as circular spots, dull white by reflected light and
bluish-grey by transmitted light. The deep colonies are opaque
and finely granular. On gelatine the growth is much the same, but
its irregular edge is, if anything, more apparent. There is no
liquefaction and no gas formation. On plates of gelatine the
colonies become large and spreading, with wavy margins. The whole
colony appears raised and almost limpet-shaped, with delicate lines
passing over its surface. When magnified, there is an appearance
of transparent iridescence. The growth on potato is termed
" invisible/'' and is of the nature of a potato-coloured pellicle, which
appears to be moist, and may at a later stage become light brown
in colour, particularly if the potato is fresh. Milk is a favourable
medium, and is rendered slightly acid. No coagulation takes place.
Broth is rendered turbid, and there is the formation of a sediment.
The organism is stained with carbol-thionin blue, carbolic fuchsin,
etc. It is decolorised by Gram's methods.

The problem of isolating the typhoid bacillus is greatly com-
plicated by the fact that B. coli communis is a normal inhabitant
of the alimentary canal, is widely distributed in nature, and is in
many respects similar to the typhoid bacillus. (For full account
of B. coli and its similarities to the typhoid bacillus, see pp. 46-51.)

We have pointed out elsewhere the relation between soil and
typhoid. In water, even though we know it is a vehicle of the
disease, the Bacillus typhosus has been only very rarely detected.
The difficulties in separating the bacillus from water (like that at
Maidstone, for example), which appears definitely to have been the
vehicle of the disease, are manifold. To begin with, the enormous
dilution must be borne in mind, a comparatively small amount of
contamination being introduced into large quantities of water.
Secondly, the group of the B. coli species considerably complicates
the search.

Further, we must boar in mind a point that is frequently neglected,
namely, that the bacteriological examination of a water which is
suspected of having conveyed the disease is from a variety of circum-
stances conducted too late to detect the causal bacteria. The in-
cubation period of typhoid we may take at fourteen days. Let us
■suppose a town water supply is polluted with some typhoid excreta
on the 1st of January. Until the 14th of January there may be
no knowledge whatever of the Btate of affairs. Two or three days
are required for notification of cases. Several more days elapse
generally before bacteriological evidence is demanded. Hence arises

304 BACTERIA AND DISEASE

the anomalous position of the bacteriologist who sets to work to
examine a water suspected of typhoid pollution three weeks
previously. There can be no doubt that these three difficulties are
very real ones. The solution to the problem will be found in the
dictum that " a water in which sewage organisms have been detected
in large numbers should be regarded with suspicion " as the vehicle of
typhoid, even though no typhoid bacilli are discoverable. The chief
of these sewage bacteria are believed to be Proteus vulgaris, Bacillus
coli, Proteus Zcnkcri, and Bacillus enteritidis sj)orogenes.

It may occur to the reader that, as the typhoid bacillus is, as far
as we know, comparatively common, drinking water may frequently
act as a vehicle to carry the disease to man. But, to appreciate the
position, it is desirable to bear in mind the following facts. The
typhoid bacillus is found, with other bacteria, in the excrement of
patients suffering from the disease ; it is short-lived ; in waters there
exist organisms which can exert an influence in diminishing its
vitality ; it is, so to speak, enormously diluted in waters ; exposure
to direct sunlight destroys it; and it has a tendency to be carried
down stream, or in still waters settle to the bottom by subsidence.
Even when all the conditions are fulfilled, it must not be forgotten
that a certain definite dose of the bacillus is required to be taken,
and that by a " susceptible " person.

Epidemic Diarrhoea

By "epidemic diarrhoea" {zymotic or epidemic enteritis) is meant
a specific disease, which may be defined as an acute infective
disease, affecting chiefly children under two years of age, occurring
during the summer months in epidemic form, and characterised by
the occurrence of diarrhoea, vomiting, and wasting, accompanied in
severe cases by toxaemia and collapse. The disease is a large
contributor to infant mortality, and in many urban districts it is
the most serious of all infant diseases, if measured by fatality.

The exact cause of epidemic diarrhoea is not at present known.
In 1887 Ballard formulated certain propositions which have
obtained general acceptance. They are as follow : —

" That the essential cause of diarrhoea resides ordinarily in the
superficial layers of the earth, where it is intimately associated in
the life-processes of some micro-organism not yet detected or
isolated.

" That the vital manifestations of such organism are dependent,
among other things, perhaps principally upon conditions of season
and on the presence of dead organic matter, which is its pabulum.

"That on occasion such micro-organism is capable of getting
abroad from its primary habitat, the earth, and having become air-

EPIDEMIC DIARRHCEA 305

borne, obtained an opportunity for fastening on non-living organic
material, and of using such organic material both as nidus and as
pabulum in undergoing various phases of its life-history.

" That in food, inside as well as outside the human body, such
micro-organism finds, especially at certain seasons, nidus and
pabulum convenient for its development, multiplication, or evolu-
tion.

" That from food, as also from the contained organic matter of
particular soils, such micro-organism can manufacture by the
chemical changes wrought therein through certain of its life-
processes a substance which is a virulent chemical poison; and
that this chemical poison is, in the human body, the material
cause of epidemic diarrhoea."*

Bacteriology of Diarrhoea. — The three causal agents which
Ballard mentions as playing a large part in the production of this
disease are the soil, season, and food — and the causa causans is
"some micro-organism not yet detected or isolated." It must be
said that we have not got much further than this during the last
fifteen years.

In 1885 Escherich published his classical researches on B. coli
communis. He pointed out that the meconium of the newly-born
infant is free from bacteria, but by the second day they are present
in large numbers, and in the ordinary excreta of healthy infants he
found chiefly two organisms, B. lactis wrogencs and B. coli communis.
Of these the former was the more abundant in the upper part of
the small intestine, and the latter in the lower part and in the
colon, so that in the excreta B. coli was abundant, and B. lactis
comparatively scarce. Booker, working in 1886 and onwards, found
that the constant bacteria of the healthy excreta of the infant
(B. coli and B. lactis mrogcncs) do not disappear in the excreta of
diarrhoea. B. coli, however, does not predominate in the same
degree, and B. lactis is present generally in greater numbers than
in the healthy excreta. Booker examined the excreta of 31
children, and isolated 33 different species of bacteria. Many
varieties of bacteria are sometimes found in individual cases of
diarrhoea. The greatest number were found in cases of cholera
infantum, and a larger number in catarrhal enteritis than in
dysenteric enteritis. The actual number of individual bacteria
was, he found, as great in the healthy excreta as in the diarrhceal
excreta. Proteus vulgaris was found very generally, and in the
most serious cases. No chromogenic bacteria were isolated, and
cultures from a large number of green stools failed to develop green
colonies. From these facts Booker concluded " that not one specific

* Supplement to the Report of the Medical Officer of the Local Government Board,

1887.

U

306 BACTERIA AND DISEASE

organism, but many different varieties of bacteria, are concerned in
the etiology of the summer diarrhoeas of children.

From 1889-1895 Booker continued his studies, isolating bacteria
from the rectum in 92 infants affected with epidemic diarrhoea, and
also from the organs of 33 infants who died from this disease. He
found the conditions for the development of bacteria in the
intestine of infants affected with summer diarrhoea different from
those in the healthy intestine of milk-fed infants, in that they
favoured more varied bacterial vegetation, a rich growth of the
inconstant species of intestinal bacteria, and a more uniform dis-
tribution through the intestine of the two constant varieties of
healthy excreta bacteria (B. coli communis and B. lactis cerogenes).
The first step in the pathological process, Booker believes to be a
direct injury to the epithelium from abnormal or excessive fermenta-
tion and from toxic products of bacteria ; and secondly, a general
intoxication may be brought about indirectly through the production
of soluble poisons. He holds that, bacteriologically and anatomically,
three principal forms of summer diarrhoea of infants may be
provisionally distinguished: — (i.) dyspeptic or non-inflammatory
diarrhoea ; (ii.) streptococcal gastro-enteritis ; and (iii.) bacillary
gastro-enteritis.* As a result of his extended researches, Booker
came to a general conclusion which he expressed as follows : — " No
single micro-organism is found to be the specific excitor of the
summer diarrhoea of infants, but the affection is generally to be
attributed to the activity of a number of varieties of bacteria, some
of which belong to well-known species, and are of ordinary occurrence
and wide distribution, the most important being the streptococcus
(enteritidis) and Proteus vulgaris." The streptococcus termed S.
cnteritidis varies in morphology, and seems to be associated with
two classes of cases, one of which simulates cholera, the other
typical enteric fever. "Micrococci are present in all cases, some-
times in enormous numbers." -j- It may be added that Cumston,
Hoist, Escherich, and Hirsch have also laid emphasis upon the
causal relationship of certain streptococci and diarrhoea.

Klein was one of the first workers to isolate an anaerobic
organism from cases of epidemic diarrhoea. This organism, which
he named B. enteritidis sjporogenes, was found in three successive
outbreaks of diarrhoea occurring among patients in St Bartholomew's
Hospital. In the first two outbreaks the milk was evidently the
channel of infection, in the third it was some rice pudding. The

* Johns Hopkins Hospital Reports, 1896, vol. vi., p. 253. See also a paper "On
the Growth of Bacteria in the Intestine," by Lorrain Smith and Tennant — Brit.
Med. Jour., 1902, vol. ii., p. 1941. Also Jeffries, Trans. American Pediatrics
Society, vol. i., 1889; Baginsky, Archiv. f. Kinderheilkunde, xii., Nos. 1 and 2;
Berliner klin. Woch., 1889 ; and Flexner and Holt's Rockefeller Inst. Rep., 1904.

f Johns Hopkins Hospital Reports, 1896, vol. vi., p. 251.

PLATE 21,

i

J0

% st'-m9

Sm-

/

*#\. '

/

Bacillus enteritidis sporogenes. (Klein.)
Film preparation from serum culture, showing spores, x 2000.

Anaerobic Milk Culture, showing the "Enteritidis Change.' (Klein.)
From left to right, tubes contain &, Ti„, X^S) t<to<5<j c-c- oi" Nottingham crude sewage.

[To face page 307.

EPIDEMIC DIARRHCEA 307

bacillus was carefully studied, and the main facts respecting it may
be stated briefly here : —

B. enteritidis sporoqenes (Klein) is an anaerobic bacillus: 1-6-4 -8 fi long, and
0-8 fi broad; stains by Gram's method and ordinary stains. Motile; spore forma-
tion present ; large, oval spores often situated near one end of bacillus ; grows well
on gelatine and agar. In the former gas is produced and the gelatine liquefies.
On agar there is also gas fermentation, and the colonies in the depth are round,
white by reflected light, brown and granular in transmitted light. On the surface
of agar plates flat, circular, moist, grey colonies appear in twenty-four to forty-
eight hours ; thicker in centre than at margin, and showing granularity. Bacillus
grows well in milk, producing the "enteritidis change." After thirty-six hours of
anaerobic incubation at 37° C. in milk, the cream is torn or disassociated by develop-
ment of gas, so that the surface of the medium is covered with stringy, pinkish-
white masses of coagulated casein enclosing a number of gas-bubbles. The main
portion of the tube of milk contains a colourless, thin, watery whey, with a few
casein lumps here and there adhering to the sides of the tube. The whey has a
smell of butyric acid, and is acid in reaction. It contains many bacilli. Patho-
genesis— If 1 c.c. of milk whey containing the bacillus be injected into a guinea-pig
(200 to 300 grammes), a swelling appears in six hours, extending over abdomen
and thigh, and death occurs in eighteen to twenty-four hours. Post mortem : sub-
cutaneous gangrene, with much sanguineous exudation, in which bacilli and spores
will be found. Klein considers this organism to be the cause of epidemic diarrhoea.

B. enteritidis sporogenes is a widely-distributed organism, and
occurs in normal and typhoid excreta, in sewage, manure, soil, dust,
and milk.* The etiological relationship between this bacillus and
epidemic diarrhoea has been called in question, and it is, of course,
not proved that the organism is the cause of the disease. On the
other hand, it has been very frequently found in the mucous flakes
of the dejecta in patients suffering from the disease, and in the
outbreak produced by the consumption of cooked rice pudding it
is difficult to understand how any organism except an anaerobe of
highly resistant qualities could have produced the condition. It
will be apparent, moreover, that B. enteritidis sporogenes fulfils in a
somewhat exceptional degree the requirements suggested by Ballard.

That epidemic diarrhoea is caused by the B. coli either alone or
in conjunction with other organisms, has been held by a number of
authorities. Cumston, who investigated 13 cases of the disease, con-
cluded that B. coli associated with Streptococcus pyogenes was the
chief pathogenic agent concerned, and he claims that the virulence of
B. coli is exalted by the association.! Lesage also formed the opinion
that the disease was due to B. coli, and investigate* I the agglutinative
properties of the serum of children suffering from epidemic diarrhoea
on B. coli isolated from the intestine. He obtained a positive result
in 40 out of 50 cases, and the serum of each of these 40 cases, also
agglutinated samples of B. coli from 39 oilier children seized with

* Reports of Medical Officer of Local Government Board. 1897-98, pp. 210-51 ;
10O-2, p/406. .

t international Medical Magazine, February 1897.

308 BACTERIA AND DISEASE

the same disease.* Some of the most recent work on the
relationship existing between B. coli and epidemic diarrhoea has
been done by Delepine, who examined milk in the outbreak of
epidemic diarrhoea which occurred in Manchester in 1894 (see
p. 224), and has also examined a large number of town and country
milks. His conclusion is that : —

" Epidemic diarrhoea of the common type occurring in this country
is apparently, in the great majority of instances, the result of
infection of food by bacilli belonging to the colon group of bacilli,
and which are present at times in f?ecal matter. It appears that
this infection of food does not generally lead to serious consequences,
unless the infection is massive from the first, or the food is kept
for a sufficient length of time, and under conditions of temperature
favouring the multiplication of these bacilli.

" Milk, which is the most common cause of epidemic diarrhoea
in infants, is usually infected at the farm, or (through vessels) in
transit. Of the bacilli of the colon group which are capable of
rendering the milk infectious, those which do not produce a large
amount of acid, and do not coagulate milk, are the most virulent,
and are probably the essential cause of epidemic diarrhoea." f

It is evident that our knowledge of the bacteriology of diarrhoea
is not sufficiently established to permit of any very definite con-
clusion on the matter. It may be that the whole group of choleraic,
enteric, and diarrhoeal diseases are caused by a group of micro-
organisms having many similarities and relationships to each other ;
or it may be that different forms of diarrhoea have their own specific
causal organism ; or, lastly, it may be a question of association of
organisms or of toxins which brings about the disease.^: In any
event, there is abundant evidence that epidemic diarrhoea is a
bacterial disease in the same sense as typhoid fever.

Conditions favourable to Epidemic Diarrhoea. — The provisional
results of Ballard's inquiry into the causation of epidemic diarrhoea
may be stated as follows : —

" The summer rise of diarrhoeal mortality does not commence
until the mean temperature recorded by the 4-foot earth thermometer
has attained somewhere about 56° F., no matter what may have been
the temperature previously attained by the atmosphere or recorded
by the 1-foot earth thermometer. The maximum diarrhoeal mortality
of the year is usually observed in the week in which the temperature
recorded by the 4-foot earth thermometer attains its mean weekly
maximum. The decline of the diarrhoeal mortality is in this con-

* La Semaine Med., October 1897.
t Jour, of Hygiene, 1903, vol. iii., No. 1, p. 90.

% See also Report of Medical Officer to Local Government Board, 1902, p. 395
(Martin), 404 et seq. (Klein).

EPIDEMIC DIARRHOEA 309

nection not less instructive, perhaps more so, than its rise. It
coincides with the decline of the temperature recorded by the 4-foot
earth thermometer, which temperature declines very much more
slowly than the atmospheric temperature, or than that recorded by
the 1-foot thermometer; so that the epidemic mortality may con-
tinue (although declining) long after the last-mentioned temperatures
have fallen greatly, and may extend some way into the fourth
quarter of the year. The atmospheric temperature and the tempera-
ture of the more superficial layers of the earth, is little if at all
apparent until the temperature recorded by the 4-foot earth ther-
mometer has risen as stated above ; then their influence is apparent,
but it is a subsidiary one."

In addition to these conditions of soil, Ballard and other workers
have concluded that insanitation in the widest sense of the term
favours epidemic diarrhoea. Density of population or houses upon
an area, unclean soil, dusty surfaces, bad light, absence of ventilation,
maternal neglect, etc., all have a share in creating an environment
favourable to the disease. As we have seen, Delcpine, like Ballard,
attributes the disease in large measure to milk. Ballard believed
that milk gained its infection by unsuitable storage and by the mode
in which it was used. He found that " infants fed solely from the
breast are remarkably exempt from fatal diarrhoea ; that infants fed
in whatever way with artificial food to the exclusion of breast milk
are those which suffer most heavily from fatal diarrhoea; that
children fed partially at the breast and partially with other kinds
of food, suffer to a considerable extent from fatal diarrhoea, but very
much less than those brought up altogether by hand ; and that, as
respects the use of 'the bottle/ it is decidedly more dangerous than
artificial feeding without the use of the bottle." This view has been
confirmed by Newsholme, Mven, Eichards, the writer, and others.

Dr Newsholme of Brighton lias published an interesting paper
on the causation of epidemic diarrhoea. Some of his chief conclusions,
which are now widely accepted, may be added : —

"(1) Epidemic diarrhoea is chiefly a disease of urban life. (2)
Epidemic diarrhoea as a fatal disease, is a disease of the artisan and
still more of the lower labouring classes to a preponderant extent.
This is probably largely a question of social status per sc ; that is,
it is due to neglect of infants, uncleanly storage of food, industrial
occupation of mothers, etc. (3) Towns which have adopted the
water-carriage system of sewerage have, as a rule, much less diarrhoea
than those retaining other methods of removal of excrement. (4)
Towns with the most perfect scavenging arrangements, including the
methods of removal of house refuse, have the least epidemic diarrhoea.
It has recently been suggested that epidemic diarrhoea is due to
surface pollution derived from street dust, particularly dried horse-

310 BACTERIA AND DISEASE

manure (Waldo). (5) The influence of the soil is a decided one.
Where the dwelling-houses of a place have as their foundation solid
rock, with little or no superincumbent loose material, the diarrhoeal
mortality is, notwithstanding many other unfavourable conditions
and surroundings, low. On the other hand, a loose soil is a soil on
which diarrhoeal mortality is apt to be high (Ballard). The pollution
of soil is probably the important element in the causation of diarrhoea
in towns on pervious soils. (6) Given two towns equally placed so
far as social and sanitary conditions are concerned, their relative
diarrhoeal mortality is proportional to the height of the temperature
and the deficiency of the rainfall in each town, particularly of the
third quarter of the year."

Dr Newsholme concludes that " the fundamental condition favour-
ing epidemic diarrhoea is an unclean soil, the particulate poison from
which infests the air and is swallowed, most commonly with food,
especially milk." In other words, epidemic diarrhoea is a so-called
" filth-disease," preventable by improved sanitation in the broadest
meaning of the term.*

From the facts and suggestions quoted above, and they are but
representative of many other similar views receiving the general
support of epidemiologists, it will be evident that at the present time
the cause of epidemic diarrhoea is to be found in four conditions,
which may be expressed shortly as two propositions, thus: (1)
Epidemic diarrhoea is a bacterial disease ; (2) its occurrence depends,
wholly or partly, upon surrounding temperature, deficiency of rain-
fall, and pollution of food, chiefly milk. The exact relationship
which these conditions have to each other is not known. Some
authorities hold that a certain temperature affects food, conducing
towards creating in it injurious properties. Others believe that it
is a question of pollution of milk by dust, which carries to the milk
the causal micro-organisms, and that deficient rainfall favours this
contamination, and increased temperature favours the growth and
multiplication of the bacteria thus conveyed to the milk. As Dr
Newsholme says, "Whatever be its mode of operation, a frequent
fall of rain during the summer weeks, even though its total amount
be not great, is one of the most effectual means of keeping down
the diarrhoeal death-rate "; f and whilst he considers temperature
conditions of great importance, "rainfall is more important than
temperature in relation to epidemic diarrhoea." Bain washes the air,
if the expression may be allowed, and carries to the surface aerial
dust. It, of course, also washes the surface of the soil and removes
surface pollution, and with it micro-organisms capable of infecting
infants, usually by food. Thus the relationship between these

* Public Health, 1899-1900, vol. xii., pp. 139-213.
f Annual Report on HealUi of Brighton, 1902, p. 48

SUPPURATION 311

meteorological conditions and milk, though an open question, may
be an essential one to the origin of the disease. Milk is probably
a common vehicle of infection (Ballard, Delepine, Newsholme), and
a number of outbreaks are now on record which appear to have been
due to the consumption of contaminated milk. In 1892 Gaffky
recorded an outbreak at Giessen,* in 1894 Mven reported on 160
cases of diarrhoea at Manchester,*)* in 1895 and 1898 three outbreaks
occurred at St Bartholomew's Hospital.j

The facts set forth above furnish sufficient indication of the
appropriate methods of prevention.

Suppuration and Abscess Formation

The term suppuration is used to designate that general breaking
down of cells which follows acute inflammation. An "abscess"
is a collection, greater or smaller, of the products of suppuration,
pus. Pus consists chiefly of two kinds of cells. First, leucocytes,
which have immigrated to the part affected; and secondly, broken
down and necrosed elements. Such an advanced inflammatory
condition may occur in any locality of the body, and it will
assume different characters according to its site. There are
connected with suppuration, as causal agents, a variety of bacteria.
Pus is not matter containing a pure culture of any specific species,
but, on the contrary, is generally filled with a large number of
different species, each playing a greater or lesser part in the process.
The most important are as follows : —

1. The Staphylococcus group. — This species consists of micrococci
arranged in groups, which have been likened to bunches of grapes
(Plate 30, p. 398). They are the common organisms found in pus, and
were, with other auxiliary bacteria, first distinguished as such by Pro-
fessor Ogston of Aberdeen. There are several forms of the same species,
differing from each other in certain respects. Thus we have the S.
pyogenes aureus (golden-yellow), alius (white), citrcus (lemon), and
others. They occur commonly in nature, in air, soil, water, as well as
on the surface of the skin, and in all suppurative conditions. The
aureus is the only one credited with pathogenic virulence. It occurs
in the blood in blood-poisoning (septicaemia, pyaemia), and is present
in all ulcerative conditions, including ulcerative disease of the valves
of the heart. The Staphylococcus cereus alius and S. cereus fiavus are
slightly modified forms of the S. pyogenes aureus, and are differenti-
ated from it by the fact of their being non-liquefying. They produce
a wax-like growth on gelatine.

* Deut. Med. Woch., vol. xviii. p. 14.

t Annual Report Medical Officer of Health, Manchester, 1894.
X Report of Medical Officer of Local Government Board, 1895-96, pp. 197-204 ;
ibid., 1897-98, p. 235 ; and ibid., 1898-99, p. 33(5.

312

BACTERIA AND DISEASE

Staphylococcus pyogenes aureus, the type of the species, is grown
in the laboratory on all ordinary media at room temperature, though
more rapidly at 37° C. Liquefaction sets in at a comparatively early
stage, and subsequently we have in gelatine test-tube cultures a
nocculent deposit of a bright yellow amorphous mass, and in gelatine
plates small depressions of liquefaction with a yellow deposit. The
organism renders all media acid, and coagulates milk. Its thermal
death-point in gelatine is 58° 0. for ten minutes, but when dry con-
siderably higher. Outside the body it may retain vitality for

Diagram of Types of Streptococci.

months. It stains by Gram's method. It is a non-motile and a
facultative anaerobe; but the presence of oxygen is necessary for
the production of much pigment. Its virulence readily declines.

2. Streptococcus pyogenes. — In this species of micrococcus the
elements are arranged in chains. Most of the streptococci in pus,
from different sources, are probably of one species, having approxi-
mately the same morphological and biological characters. Their
different effects are due to different degrees of toxic virulence ; they
are generally more virulent when associated with other bacteria, for
example, the Proteus family.

The chains vary in length, consisting of more elements when

SUPPURATION 313

cultured in fluid media (hence S. longus and S. brevis). They
multiply by direct division of the individual elements, and in old
cultures it has been observed that the cocci vary in form and size
(involution forms). This latter fact gave support to the theory that
streptococcus reproduced itself by arthrospores, or " mother-cells."

In culture upon the ordinary media, Streptococcus pyogenes is com-
paratively slow-growing, producing minute white colonies on or
about the sixth day. It does not liquefy gelatine, and remains
strictly localised to the track of the inoculating needle. Like the
staphylococcus, it readily loses virulence. The thermal death-point
is, however, lower, being 54° C. for ten minutes. Marmorek has
devised a method by which the virulence may be greatly increased,
and he holds that it depends upon the degree of virulence possessed
by any particular streptococcus as to what effects it will produce.
By the adoption of Marmorek's methods, attempts have been made to
prepare an antitoxin.

Streptococcus pyogenes has been isolated from the membrane in
cases of diphtheria, and from small-pox, scarlet fever, vaccinia, and
other diseases. In such cases it is probably not the causal agent, but
merely associated with the complications of these diseases. Suppura-
tion and erysipelas are the only pathological conditions in which the
causal agency of Streptococcus has been sufficiently established.

3. The Bacillus pyocyaneus occurs in green pus, and is the cause
of the coloration. G-essard was the first to prove its significance,
and he described two varieties. It is a minute, actively motile, non-
sporulating bacillus, which occasionally complicates suppuration and
produces blue-green pus. It stains with the ordinary aniline stains,
but is decolorised by Gram's method. Oxygen is necessary for
pigmentation, which is clue to two substances : pyocyanin, a greenish-
blue product extracted with chloroform, and pyoxaiithosc, a brown
substance derived from the oxidation of the former pigment. Botli
these colours are produced in cultivation outside the body. On
gelatine the colour produced is green, passing on to olive. There is
liquefaction. On potato we generally obtain a brown growth (com-
pare B. coli, B. mallei, and others). The ,.
organism grows rapidly on all the ordinary •••/ *7gJ 2:
media, which it has a tendency to colour •*,% **
throughout. It will be remembered that "v /. **
when speaking of the antagonism of organ- jf :f
isms, we referred to the inimical action of * **z*
B. pyocAjanens upon the bacillus of anthrax. ** #

4. Micrococcus tetragOUUS. — This species Fia. 26.— Diagram of Micrococcus

occurs in phthisical cavities, and in certain

suppurations in the region of the face. The micrococcus usually

occurs in the form of small tetrads. A capsule is generally present.

314 BACTERIA AND DISEASE

It is a non-liquefying organism, pathogenic for white mice (producing
septicaemia). It grows on ordinary laboratory media, producing a
viscid tenacious culture.

5. B. coli communis and many putrefactive germs commonly
occur in suppurative conditions, but they are not restricted to such
disorders (see p. 46).

6. Micrococcus gonorrhoeae (Neisser, 1879). — This organism is
more frequently spoken of as a diplococcus. It occurs at the acute
stage of the disease (and in the purulent secretion of gonorrheal
conjunctivitis), but is not readily differentiated from other similar
diplococci except by laboratory methods. Each element of the diplo-
coccus presents a straight or concave surface to its fellow. A very
marked concavity indicates commencing fission. The position which
these diplococci take up in pus is intracellular, and they are arranged
more or less definitely around the nucleus. In chronic gonorrhoea

the diplococci are diminished in number.
Difficulty has often been found in culti-
vating the organism in artificial media
outside the body. Wertheim and others
have suggested special formulae for the
preparation of suitable media, but it is a
comparatively simple matter to secure
cultures on agar plates smeared with
human blood from a pricked finger. The
plate is incubated at 37° C. At the end
of twenty-four hours small raised grey
colonies appear, which at about the end
of four days show adult growth. The
margin is uneven, and the centre more
opaque than the rest of the colony. This diplococcus is readily
killed, and sub-cultures must be frequently made to retain vitality
and virulence. Light, desiccation, and a temperature of 55° C.
all act germicidally. The organism stains readily with Loffler's blue,
but is decolorised by Gram's method. It is more or less strictly
parasitic to man, and has been definitely proved to be the cause
of gonorrhoea. A toxin has been separated. The shape, size,
character of growth, intracellular position, and staining pro-
perties of the gonococcus assist in differentiating it from various
similar diplococci.* An organism not greatly different from the
gonococcus is the diplococcus intracellular is meningitidis isolated by
Weichselbaum from cases of cerebro-spinal meningitis. It occurs in
the interior of leucocytes.

Such are the chief organisms associated with suppuration. In

the condition known as septicemia, these organisms multiply in the

* See Trans. J miner Inst. (First Series), A. G. R. Foulerton, F.R.C.S., pp. 40-81.

ANTHRAX 315

blood, and give rise to general poisoning without abscess formation;
in pycemia, however, multiple abscesses occur in various parts of the
body, including internal organs. From the results of experiment it is
now believed that suppuration in any form or degree is invariably
the result of bacterial infection. But it is not known in what way
bacteria exactly cause the condition ; it may be due to extracellular
toxins, or intracellular poisons, or to the bodies themselves setting up
primary irritation, or to all three conditions. Positive chcmiotaxis is
probably the explanation of the immigration of the leucocytes.

Anthrax

This disease was one of the first in which the causal agency of
bacteria was proved. In 1849 Pollencler found an innumerable
number of small rods in the blood of animals suffering from anthrax.
In 1863 Davaine described these, and attributed the disease to them.
But it was not till 1876 that Koch finally settled the matter by
isolating the bacilli in pure culture and describing their biological
characters.

It is owing in part to its interesting bacterial history, which
opened up so much new ground in this comparatively new science,
that anthrax has assumed such an important place in pathology.
But for other reasons, too, it claims attention. It appears to have
been known in the time of Moses, and was perhaps the disease
described by Homer in the First Book of the Iliad. Eome was
visited by it in 740 B.C.

Anthrax is an acute disease, affecting sheep, cattle, horses, goats,
deer, and man. Cats, white rats, and Algerian slice]) are immune.
Swine become infected by feeding on the offal of diseased cattle
(Crookshank).

Clinical Characters. — In most instances the first intimation of an outbreak of
anthrax is the discovery of a dead animal in the pasture or byre. The animal may
have been left a few hours earlier in apparent good health ; at least, there may have
been nothing to attract attention, or give warning of the near approach of death.
Occasionally there are, however, premonitory symptoms of an attack of anthrax
which can be recognised by an expert. The affected animal is dull, and disinclined
to move. If the case occurs in a herd at pasture the fact is sometimes indicated by
the separation of the sick animal from the rest. The affected animal will occasionally
cease to feed, and stand with its head bent towards the ground, and sometimes a
little blood is discharged from the nostrils and also with the faeces. Close attention
will enable the observer to detect an occasional shiver and trembling of the limbs,
which passes rapidly over the body, and then ceases. The shivering fits may then
become more frequent, and perhaps, while these signs are being noted, the animal
will suddenly roll over on its side, and, after a few violent struggles, expire. On
close inspection, especially in the case of swine, it will often be found that there is a
good deal of swelling under the throat extending down the neck; and the swollen
part will at first be hot and tender to the touch, but as the disease progresses
it becomes insensitive and cold.

316 BACTERIA AND DISEASE

The post-mortem signs are mainly three: The spleen is greatly
enlarged and congested, is dark red in colour, friable to the touch,
and contains enormous numbers of bacilli; the shin may show
exudations forming dark gelatinous tumours ; and the blood remains
fluid for some time after death, is black and tar-like, contains bubbles of
air, and shows other degenerative changes in the red corpuscles, whilst
the small blood-vessels contain such vast quantities of bacilli that they
may be ruptured by them. Particularly is this true in the peripheral
arteries. Many of the organs of the body show marked congestion.

The bacilli of anthrax are square-ended rods 1 /m broad and
4-5 jul long. In the tissues of the body they follow the lines of the
capillaries, and are irregularly situated. In places they are so
densely packed as to form obstructions to the onward flow of blood.
In cultures they occur in chains end to end, having, as a rule, equal

interbacillary spaces. But long filaments
and threads also occur. The exact shape
of the bacillus depends, however, upon
staining and spore formation. Both
these factors may very materially modify
the normal shape. The spores of anthrax
are oval endospores, produced only in
the presence of free oxygen, and at any
temperature between 18 and 41° C. On
account of requiring free oxygen, they
are formed only outside the body. The
homogeneous protoplasm of the bacillus
fig 28. -Diagram of Bacillus of becomes granular ; the granules coalesce,

Anthrax and Blood Corpuscles. . . . . 9 2, ,

constituting spores. Each spore pos-
sesses a thick capsule, which enables it to resist many physical
conditions which kill the bacillus. When the spore is ripe, or
has exhausted the parent bacillus, it may either take on a resting
stage, or under favourable circumstances commence germination,
very much after the manner of a seed. The spores may infect
a farm for many months; indeed, cases are on record which
appear to prove that the disease on a farm in the autumn may,
by means of the spores, be carried on by the hay of the follow-
ing summer into a second winter. Thus, by means of the spores,
the infection of anthrax may cling to the land for very long periods,
even for years. Spores of anthrax can withstand 5 per cent, carbolic
acid or 1-1000 corrosive sublimate for more than an hour; even
boilinQ- does not kill them at once, whilst the bacilli without their
spores are killed at 54° C. in ten minutes. When the spores are dry
they are much more resistant than when moist. The persistence of
the anthrax bacillus is due to its spores.

The bacillus is aerobic, non-motile, and liquefying. Broth

ANTHRAX 317

cultures become turbid in thirty-six hours, with nebulous masses of
threads matted together. The pellicle which forms on the surfaces
affords an ideal place for spore formation. Cultures in the depth
of gelatine show a most characteristic growth. From the line of
inoculation delicate threads and fibrillse extend outwards horizontally
into the medium. Liquefaction commences at the top, and eventually
extends throughout the tube. On gelatine plates small colonies
appear in thirty-six hours, and on the second or third day they
appear, under a low power of the microscope, not unlike matted
hair. The colonies after a time sink in the gelatine, owing to lique-
faction. On potato, agar, and blood serum the anthrax bacillus
grows well (Plates 17 and 22).

Channels of Infection. 1. The Alimentary Canal. — This is the
usual mode of infection in animals grazing on infected pasture land.
A soil suitable for the propagation of anthrax is one containing
abundance of air and proteid material. Feeding on bacilli alone might
possibly not produce the disease, owing to the germicidal effect of
the gastric juice. But spores can readily pass uninjured through
the stomach, and produce anthrax in the blood. Infected water, as
well as fodder, may convey the disease. Water becomes infected by
bodies of animals dead of anthrax, or, as was the case once at least
in the south-west of England, by a stream passing through the
washing-yard of an infected tannery. Manure on fields, litter in
stalls, and infected earth, may all contribute to the transmission of
the disease. Darwin pointed out the services which are performed
in superficial soils by earthworms bringing up casts ; Pasteur was of
opinion that in this way earthworms were responsible for continually
bringing up the spores of the anthrax bacillus from buried corpses
to the surface, where they would reinfect cattle. Koch disputed this,
but more recently Bollinger has demonstrated the correctness of
Pasteur's views by isolating anthrax contagium from 5 per cent, of
the worms sent him from an anthrax pasture. Bollinger also
maintains that flies and other insects may convey the disease from
discharges or carcases round which they congregate.

Alimentary infection in man is a rare form, and it reveals itself
in a primary diseased state known as mycosis intestinalis, an inflamed
condition of the intestine and mesenteric lymph-glands.

2. Through the Skin. — Cutaneous anthrax, when it occurs in the
human subject, goes by the name of malignant pustule, and is caused
by infective anthrax matter gaining entrance through abrasions or
ulcers in the skin. This local form is obviously mostly contracted by
those whose occupation leads them to handle hides or other anthrax
material (butchers and cleaners of hides), and it naturally affects the
skin of the hands, forearms, face, or back (as it occurs amongst hide-
porters). Two or three days after inoculation a red pimple appears,

318 BACTERIA AND DISEASE

which rapidly passes through a vesicular stage until it is a pustule.
Concomitantly, we have glandular enlargement (the pustule acting
as a centre of subcutaneous oedema), general malaise, and a high
temperature. Thus from a local sore a general infection may result.
Unless this does occur, the issue is not likely to be fatal, and the
bacilli will not gain entrance into the blood. The spleen is usually
not affected, and the organs generally contain few or no bacilli.
When a fatal issue occurs, it is due to the absorption of toxins.
Early excision of the pustule is usually followed by recovery.*

3. Respiratory Tract. — In man, this is perhaps the commonest
form of all, and is well known under the term " wool-sorters' disease,"
or pulmonary anthrax. This mode of infection occurs when dried
spores are inhaled in processes of skin-cleaning. It frequently com-
mences as a local lesion, affecting the mucous membrane of the
trachea or bronchi, but it rapidly spreads, affecting the neighbouring
glands, which become greatly enlarged, and extending to the pleura
and lung itself. The lung shows collapse and oedema leading to
pulmonary embarrassment. There is also fever. Such cases, as a
rule, rapidly end fatally. Even in wool-sorters' disease the bacilli
do not become widely distributed.

Preventive Methods. — As a rule, anthrax carcases are better
not opened and exposed to free oxygen. An extended post-
mortem examination is not necessary. A small prick, for example,
in the auricular vein will extract enough blood to examine for
the anthrax bacilli, which are driven by the force of the blood
current to the small surface capillaries. This occurs, of course,
only when the disease has become quite general, for in the early
stage the healthy blood limits the bacilli to the internal organs.
In such cases examination of the blood of the spleen is necessary.
The chief source of danger is the infection by anthrax blood or
discharges (containing sporulating bacilli) of the field, farm-yard,
byres, etc., and it is therefore necessary for thorough disinfection to
be carried out if infection has occurred. Burning the entire carcase
in a crematorium would be the ideal treatment. As such is not
generally feasible, the next best thing is to bury the carcase deeply
with lime below and above it, and rail in the area to prevent other
animals grazing off it.

In the German Special Eules relating to the establishment and
management of horse-hair spinning-mills, factories for hair and bristle
dressing, and brush factories of all kinds, j* it is laid down that
disinfection may be done in one of the three following ways : (1) by

* Accidental infection with anthrax has been held to be an accident to employes
under the Workmen's Compensation Act, 1897 (Courts of Appeal), Justice of the
Peace, May 7, 1904, vol. lxviii., p. 193.

f Order dated October 22, 1902, under Industrial Code {Gewerbeordnung), 120e.

PLATE 22.

j

1
' "1

1

Iv-V'':^il

11

*!»*

Bacillus imth rods.
Gelatine stab culture. 3 days' growth at 20" C x I', times.

i

~>y

Bacillus anthraeis.

Smear preparation from splenic blood of cow.

x 1000.

l'l; wkki.'s PNEUMOCOCCUS IN PNEUMONIC SPUTUM.

Stained with Neelsen and methylene blue,
x 1000.

[To face page 31S.

PNEUMONIA 319

letting a current of steam act on the material for not less than half
an hour, at a temperature of 218° F. ; (2) by boiling for not less than
one hour in a solution containing 2 per cent, potassium permanganate,
bleaching it afterwards with a solution containing 3 to 4 per cent, of
sulphurous acid; or (3) by boiling in water for not less than two
hours. A number of other regulations are included in the Order.*
Various experiments have been carried out in this country with a
view to determining the most effectual methods of disinfection^
Boiling does not appear to be always effective, and is, moreover,
frequently impracticable owing to the damage it causes. In steam
disinfection of horse hair and similar materials, a temperature of
230° F. for 30 minutes is as effective as higher temperatures, but the
hair must be loose and not closely packed in bundles. Probably one
of the most practical methods for disinfection of hair is to soak it
for twenty-four hours in a solution of one part of corrosive sublimate
in a thousand parts of warm water (Klein). But apart from actual
disinfection of the material, considerable protection is afforded by
(a) the avoidance of horse hair from Eussia, Siberia, and China, and
wool from Persia (from which sources most infection is derived), unless
it is guaranteed as thoroughly disinfected ; (b) by compelling employes
to wash with soap and hot water before leaving work or taking food,
the more general the washing, as a rule, the greater the security
obtained ; (c) the use of fans creating a down-draught to remove dust
when sorting; and (d) the exclusion of workpeople suffering from
cuts or scratches of the skin from processes in which they are likely
to come into contact with dust from horse hair.

Anthrax covers a wide geographical area all over the world, and
no country seems altogether exempt. In Germany as many as 3700
animals have been lost in a single year. In 1903 there were 761
outbreaks of anthrax in Great Britain, in which 1127 animals were
attacked. This is the largest return recorded since the passing of
the Anthrax Order in 188G.

Pneumonia

Some of the difficulty which has surrounded the bacteriology
of inflammation of the lungs is due to the confusion arising
from supposing that attacks of the disease differed only in
degree. Pneumonia, however, has various Tonus, arising now from
one cause, now Prom another. The lobar or croupous pneumonia is
associated with two organisms: Frankel's diplococcus and Fried-
lander's pneumo-hacillus. Acute catarrhal pneumonia generally
arises as a secondary complication to other disease, such as diphtheria,
influenza, bronchial affections, etc. Septic pneumonias arc also not

Annual Report of Chief Inspector of Factories and Workshops, 1902, p. 214.
I Ibid., 1900, 1902," and 1903.

320 BACTERIA AND DISEASE

specific, but secondary or mechanical. Other bacteria in addition
to the two named have from time to time been held responsible
for pneumonia, a streptococcus receiving, at one time, some support.
But whilst opinion is divided as to the role of various extraneous
and concomitant bacteria in lung disease, importance is attached to
Frankel's and Friedlander's organisms.

The diplococcus of Friinkel is a small oval diplococcus found in
the "rusty" sputum of croupous pneumonia. It is non-motile,
non-liquefying, aerobic, and facultatively anaerobic. When examined
from cultures the cliplococci are frequently seen in chains, not unlike
a streptococcus, and there is some reason to suppose that this form
gave rise to the belief that it was another species ; when examined
from the tissues, sputum, or pus, it possesses a capsule, like an
unstained halo (stained by MacConkey's method), but in culture this
is lost except in gelatine at 37° C. (Gordon). Involution forms occur.
The diplococcus is difficult to cultivate, but grows on glycerine
agar and blood serum at body temperature. On ordinary gelatine
at room temperature it does not grow, or if so, very slightly. The
ideal fluid is a slightly alkaline liquid medium, and in twenty-four
hours a powdery growth will occur in such broth. On potato there
is apparently no growth. The pneumococcus always requires a
temperature of about blood-heat for its maximum development.
It rapidly loses its virulence on solid media, and is said to be
non-virulent after three or four sub-culturings. A temperature of
54-58° C. for a few minutes kills the bacteria, but not the toxin.
This, however, is removed by filtration, and is therefore probably
intracellular. It is attenuated by heating to 70° C. This diplococcus
stains by Grain's method (see Plate 22, p. 318).

Frankel's diplococcus occurs, then, in the acute stage of true
croupous pneumonia, in company with streptococci and staphylococci.
It is by far the most frequently present organism in croupous
pneumonia. It also occurs in the blood in certain suppurative
conditions, in pleurisy and inflammation of the pericardium, and some-
times in diphtheria, and therefore it is not peculiar to pneumonia.

Frankel's organism is said to be frequently present in the saliva
of healthy persons. Inflammation depresses the resistant vitality
of the tissues, and thus affords to the diplococcus present in the
saliva an excellent nidus for its growth.*

Friedlander's Pneumo-bacillus is a capsulated oval coccus, assuming
the form of a small bacillus. It is inconstant in pneumonia, unequally
distributed, and scarce ; it is aerobic, and facultatively anaerobic ;

* For further particulars respecting the pneumococcus, see Practitioner \ March
1900, pp. 280-304 (J. W. Eyre); and Brit. Med. Jour., 1902, vol. ii., pp. 1585,
1646, 1704, 1765 (Croonian Lectures on Natural History and Pathology of
Pneumonia, by J. W. Washbourn).

ACTINOMYCOSIS 321

it occasionally occurs in long forms and filaments; it is non-motile,
non-liquefying, and has no spores; it does not stain by Gram's
method, which stain is therefore used for differential diagnosis ; it will
grow fairly well in ordinary gelatine at 20° C. ; and it is a denitrifying
organism, and also an actively fermentative one, even fermenting
glycerine. It is not unlike B. coli communis, and to distinguish it from
that organism it should be remembered that the B. coli is motile,
never has a capsule, produces indol, and does not ferment glycerine.

It is now generally held that Frankel's diplococcus is the chief
factor in the causation of croupous pneumonia, and probably plays
an important part in other forms of the disease. In the septic
pneumonias the different suppurative organisms are found, and some-
times in ordinary pneumonias these organisms may be the causal
agents.

Influenza

In 1892, during the pandemic of influenza, Pfeiffer dis-
covered a bacillus in the bronchial mucus of patients suffering
from the disease. It is one of the smallest bacilli known, and
frequently occurs in chains not unlike a streptococcus. Canon
obtained the same organism from the blood. In the bronchial
expectoration it can retain its virulence for as long as a fortnight,
but it is quickly destroyed by drying. The bacillus is aerobic, non-
motile, and up to the present spores have not been found. It is
non-motile, and does not stain by Gram's method. It has no
capsule. It grows somewhat feebly in artificial media, and readily
dies out. Blood serum, glycerine agar, blood agar, and gelatine have
all been used at blood-heat. It does not grow at room temperature.
On blood agar colonies appear in twenty-four hours in the form
of minute circular dots, almost transparent. The bacilli die out
quickly in cultures. Pfeiffer's bacillus appears most abundantly at
the height of the disease, and disappears with convalescence. It is
said not to appear in any other disease. It is chiefly found in the
respiratory passages in cases of influenza, and is usually isolated from
nasal secretion and the masses of greenish-yellow bronchial sputum.
The bacilli may persist after recovery of the patient.

Actinomycosis

This disease affects both animals and man. As Professor
Crookshank has pointed out, it has long been known in this
country, but its various manifestations have been mistaken for
other diseases or have received popular names.*

* Bacteriology and Infective Diseases (1896), pp. 413-447. Professor Crookshank's
Reports to the Agricultural Department of the Privy Council constitute a most
complete account of this disease. See also Trans. Jenner Institute (Second Series),
1899, p. 17.

X

322 BACTERIA AND DISEASE

Here mention can only be made of the most outstanding facts
concerning the disease. It is caused by the "ray fungus," or
Strcjptothrix actinomyces, one of the higher bacteria which, growing
on certain cereals, may gain entrance to the tissues of man and beast
by lacerations of the mucous membrane of the mouth, by wounds, or
by decayed teeth. Barley has been the cereal in question in some
cases. The result of the introduction of the parasite is an " infective
granuloma." This is, generally speaking, of the nature of an inflam-
matory tumour composed of round cells, epithelioid cells, giant cells,
and fibrous tissue, forming nodules of varying sizes. In some cases
they develop to large tumours, in others they soon break down.
Actinomycosis resembles tuberculosis in some of its tissue characters.

In the discharge or pus from human cases of the disease small
sulphur-yellow bodies may be detected, and these are tufts of " clubs "
which are the broken-down rays of the parasite ; for in the tissues
which are affected the parasite arranges itself in a radiate manner,
growing and extending at its outer margin and degenerating behind.
In cattle the centre of the old ray becomes caseated, or even calcified.
In the human disease abundant " threads " are formed as a tangled
mass in the middle of the colony. As clubs characterise the bovine
actinomycosis, so threads are the feature of the human form of the
disease. But in both there is a third element, namely, small round
cells, called by some spores, by others simply cocci. They are
probably formed from the filaments, but authorities are not yet
agreed as to the precise significance and role of these round cells.
The life-history of the micro-organism may be summed up thus :
"The spores sprout into excessively fine, straight or sinuous, and
sometimes distinctly spirilliform, threads, which branch irregularly
and sometimes dichotomously. The extremities of the branches
develop the club-shaped bodies. The clubs are closely packed
together, so that a more or less globular body is formed, with a central
core composed of a dense mass of threads " (Crookshank). (Plate 13,
p. 140.)

In man the disease manifests itself in various parts according
to the point of entrance. It has occurred in the mouth, vertebra?,
oesophagus, intestine, liver, kidneys, lungs, etc. When occurring in
the mouth, it attacks the lower jaw most frequently. In one recorded
case the disease was localised to the bronchi, and did not even extend
into the lungs. It was probably contracted by inhalation of the
parasite. The disease may spread to distant parts by means of the
blood stream (metastatic abscesses), and frequently the abscesses are
apt to burrow in various directions. The chronic inflammatory
change usually ends in suppuration.

In the ox the disease remains much more localised, is more
formative, and frequently occurs in the lower jaw, palate, or tongue.

GLANDERS 323

In the latter site it is known as " wooden tongue/' owing to the
hardness resulting. The skin and subcutaneous tissues are also a
favourite seat of the disease, producing the so-called wens or clyers
so commonly seen in the fen-country (Crookshank). Actinomycosis
in cattle is especially prevalent in river valleys, marshes, and on
land reclaimed from the sea. The disease occurs at all seasons, but
perhaps more commonly in autumn and winter. It is more frequently
met with in young animals. The disease is probably not hereditary
nor readily communicated from animal to animal.

The Streptothrix Actinomyces may be cultivated, like other
parasites, outside the body. Gelatine, blood serum, agar, glycerine
agar, and potato have been used for this purpose. After a few days on
glycerine agar at the temperature of the blood, small, white, shining
colonies appear, which increase and coalesce. In about ten days'
time the culture often turns a bright yellow, though it may remain
white or even take on a brown or olive tint. The entire mass of
growth is raised, dry, corrugated, and crinkled, and composed almost
exclusively of threads. In its early stage small bacillary forms
occur, and in its later stage coccal forms. True clubs never occur
in pure cultures, although the threads may occasionally show bulbous
endings.

Glanders

Glanders in the horse and ass, and sometimes by communication
in man also, is caused by a short, non-motile, aerobic bacillus, named,
after the old Eoman nomenclature (malleus), Bacillus until,;. It, was
discovered in 1882 by Loffler and Schutz. It is found in the nasal
discharge of glanclered animals. In appearance, the bacillus is not
unlike B. tuberculosis, except that it is shorter and thicker. The
beading of the bacillus of glanders, like that in tubercle, does not
denote spores. B. mallei can be cultivated on tin1 usual media,
especially on glycerine agar and potato. On the last-named medium
at blood-heat it forms a very characteristic honey-like growth, which
later becomes reddish-brown. High temperature is usually necessary.

In the horse glanders may affect the nasal mucous membrane,
forming nodules which degenerate and emit an offensive discharge.
From the nose, or nasal septum, as a centre, the disease may spread
to surrounding parts. It may also occur as nodules in and under
the skin, and involving the superficial lymph vessels and glands,
when it is known as "farcy." Persons attending a glandered animal
may contract the disease, often by direct inoculation'. Horned cattle
are immune.

In man glanders occurs in two forms, an acute and a chronic.
The site is, of course, usually on the hand or arm. The acute form
has the appearance of a "poisoned wound," locally, and there are

324 BACTERIA AND DISEASE

also the general symptoms of pyaemia, and an eruption on the
surface of the body. Such cases usually terminate fatally. The
chronic form results in local ulceration and involvement of the
lymphatics. It may at any time become acute.

The glanders bacillus is not quickly destroyed by drying, but it
possesses comparatively feeble resistance to heat (55° C. for ten
minutes), and antiseptics (5 per cent, carbolic in three minutes). It
differs widely from the tubercle bacillus in staining properties.
Gram's method and that of Ziehl-Neelsen are inapplicable. Carbol-
thionin blue is the best stain to use. (Plate 13, p. 140.)

Mallein is a substance analogous to tuberculin, and is made by
growing a pure culture of B. mallei in glycerine-veal broth in flat
flasks, with free access of calcined air. After a month's growth the
culture is sterilised, filtered, concentrated, and mixed with an equal
volume of a "5 per cent, solution of carbolic acid. The dose is 1 c.c,
and it is used, like tuberculin, for diagnostic purposes. If the
suspected animal reacts to the injection, it is suffering from glanders.
Eeaction is judged by three signs, (a) a rise of temperature 2-3° C,
(b) a large " soup-plate " swelling at the site of inoculation, and (c)
an enlargement of the lymphatic glands.

In 1903 there were in Great Britain as many as 1463 outbreaks
of glanders in which 2490 horses were attacked. This is the highest
number of outbreaks since 1892, when they numbered 1657. The
prevalence of the disease is localised often to certain counties and
districts. In 1903, 855 of the 1463 outbreaks occurred in the county
of London.

CHAPTEK X

TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

Pathology and Bacteriology of Tuberculosis — The Bacillus of Koch — Animal Tuber-
culosis, Bovine, Avian, etc. — Bovine and Human Tubercle Bacilli compared
— Intercommunicability — Diagnosis of Bovine Tubercle — The Prevention of
Tuberculosis— Pseudo-Tuberculosis— Acid-fast Bacteria Allied to the Tubercle
Bacillus : in Man, in Animals, in Butter and Milk, in Grass— Differential
Diagnosis — Streptothrix Group.

Tuberculosis is from several points of view the type of bacterial
disease which most concerns the public health. Its bacteriology is
perhaps more worked out than that of any other disease. Its pre-
valence in all parts of the world, and among animals as well as man,
makes it a disease of vital importance and interest to man. More-
over, the growth of our knowledge respecting it has led to the
introduction of methods of prevention, and the world is beginning
to understand that a scientific control of this disease is becoming
possible. For these reasons it is desirable 1<> treat somewhal fully
of the chief facts respecting it.

Pathology and Bacteriology *

As far back as 1794, Baillie drew attention to the grey miliary
nodules occurring in tuberculous tissue, which gave rise to the term
"tubercles." This observation was confirmed by Bayle in 1810.
In 1834 Laennec described all caseous deposits as "tubercles,"
insisting upon four varieties: —

* A detailed study of tuberculosis from its pathological and bacteriological aspect
will be found in La Tuberculose et son Bacille, part i., Straus, Professor a la Faculte
de Medecine de Paris; and in Tuberculosis, by Professor Comet, edited by W. B.
James and A. Stengel, 1904.

325

326 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

1. Miliary, which were about the size of millet seeds, and
generally occurring in groups.

2. Crude, miliary tubercles in yellow masses.

3. Granular, similar to the last, but scattered.

4. Encysted, a hard mass of crude tubercle with a fibrous or
semi-cartilaginous capsule.

The " tubercle " possesses a special structure, although it is not
always apparent, and certain cell-forms occur in it and give it a
more or less characteristic appearance.

The typical lesion is a nodule of granulation tissue, as small as
the size of a millet seed. The centre consists of one or more
multinucleated cells known as giant cells, immediately surrounded
by a zone of slightly elongated cells with a somewhat faintly-staining
nucleus, termed epithelioid cells, owing to their origin. These cells in
their turn are surrounded by another zone of small round cells which
have but little cell protoplasm, yet contain a deeply-staining nucleus,
and are known as lymphoid cells. They are apparently identical
with lymphocytes. The whole nodule is inflammatory tissue pro-
duced as a result of the action of a specific irritant, namely, the
tubercle bacillus.

It was not till 1865 that the specific nature of tuberculosis was
asserted by Villemin. Burden Sanderson (1868-9) in England con-
firmed his work, and it was extended by Cohnheim, who a few years
later laid down the principle that all is tubercular which by trans-
ference to susceptible animals is capable of inducing tuberculosis,
and nothing is tubercular unless it possesses this property.

Klebs (1877) and Max Schiller (1880) described masses of living
cells or micrococci in many tuberculous nodules in the diseased
synovial membrane of joints and in lupus skin. In 1881 Toussaint
declared that lie had cultivated from the blood of tubercular animals
and from tubercles an organism which was evidently a micrococcus,
and in the same year Aufrecht stated that the centre of a tubercle
contained small micrococci, cliplococci, and some rods. But it was
not till the following year, 1882, that Koch discovered and demon-
strated beyond question the specific Bacillus tuberculosis.

It is now held to be absolutely proved that the introduction of
this bacillus, or its spores, is the one and only essential agent in the
production of tuberculosis. Its recognised manifestations in the
body of man are as follows : — Tuberculosis in the lungs = acute or
chronic phthisis ; in the skin = lupus ; * in the mesenteric glands =
tabes mesenterica ; in the brain = hydrocephalus ; in lymphatic
glands = scrofula*

The disease may occur generally throughout the body, or it may

* There are, obviously, differences of virulence between these conditions and
pulmonary tubercle.

KOCH'S TUBERCLE BACILLUS 327

occur locally in the lungs, liver, glands, intestine, larynx, bones,
kidneys, spleen, and other parts.

We may summarise the history of the pathology of tubercle
thus : —

1794 Baillie drew attention to grey miliary nodules occurring
in tuberculosis, and called them " tubercles."

1834. Laennec described four varieties: miliary ; crude; granu-
lar; encysted.

1843. Klencke produced tuberculosis by intravenous injection
of tubercular giant cells.

1865. Villemin demonstrated infectivity of tubercular matter by
inoculation of discharges; Cohnheim, Armanni, Burdon
Sanderson, Wilson Fox, and others showed that nothing
but tubercular matter could produce tuberculosis.

1877. Living cells were found in tubercles, " micrococci " (Klebs,
Toussaint, Schiller).

1882. Koch isolated and described the specific bacillus, and
obtained pure cultivations (1884).

The Bacillus of Koch

Biology. — The B. tuberculosis of artificial culture is usually an
unbranched, slender, immotile rod, 1/5 to 4 /ul long and '4 /m broad,
often slightly bent. In sputum and tissues the bacillus may appear
branched and in thread forms. The protoplasm of the bacillus
consists of fat and wax (26 per cent.), protamin (24 per cent.), nucleo-
proteid (23 per cent.), nucleic acid (8 per cent.), and the remainder
of mineral and proteinoid (chitiu) substances. The protoplasm is
frequently vacuolated and irregularly segmented, and this becomes
particularly obvious after staining. As to staining, the bacillus is
acid-proof, and stains well with Ziehl-Neelsen or Gram. Klein and
Marmorek have shown that very young tubercle bacilli are not
resistant to acid and alcohol. Growth does not occur in the absence
of oxygen, is most favoured by a temperature varying from 29° C.
to 42 D C, and is at all times slow on artificial media. In sputum
and in tissues it will be found that many of the bacilli are straight
with rounded ends; others are slightly curved. They are usually
solitary, but may occur in pairs, lying side by side or in small
masses. They are chiefly found in fresh tubercles, more sparingly
in older ones. Some lie within the giant cells ; others lie outside.
When stained, they appear to be composed of irregular cubical or
spherical granules within a faintly-stained sheath. In recent lesions
the protoplasm appears more homogeneous, and only takes on the
segmented or beaded character in old lesions, pus, or sputum. As
a rule, the capsule stains. There are no flagella. So far as is known,

328 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

the bacillus tuberculosis discovered by Koch is the only immediate
cause of all forms of human tuberculosis. In the majority of cases
the micro-organism is met with in the form of slender rods, but under
certain conditions at present imperfectly understood, the micro-organism
may show filaments, true dichotomous branching and club forma-
tion, and, in the tissues, especially in experimental tuberculosis, it may
assume a radiate arrangement — characters which from a taxonomic
point of view bring it into close relation with a large group of micro-
organisms variously designated Streptothricese, Oospora, Noeardiaceae,
and more recently Actinomycetes (Lachner-Sandoval). According
to all experience, the tubercle bacillus is an aerobic facultative
parasite, which grows extremely slowly outside the body, and the fact
that for its growth it requires a relatively high temperature is against
the supposition that the tubercle bacillus multiplies extra-corporeally,
at least in temperate climates.

Morphological differences are found under different circumstances,
and within limits variation occurs according to the environment.
The filaments, threads, and true branching forms of old cultures have
been met with, though only occasionally, in sputum. Clubbed
actinomycotic forms have also been described. On these facts some
bacteriologists are disposed to look upon the tubercle bacillus as
belonging to the higher bacteria (Plate 18).

Cultivation on Various Media. — Koch inoculated solid blood serum
with tubercular matter from an infected lymphatic gland of a guinea-
pig, and noticed the first signs of growth in ten or twelve days in
the form of whitish, scaly patches. These enlarged and coalesced
with neighbouring patches, forming white, roughened, irregular masses.
The blood serum is not liquefied. Nocard and Eoux showed that by
adding 5 to 8 per cent, of glycerine to the media commonly used in the
laboratory, such as nutrient agar or broth, better growth is obtained.
In glycerine broth abundant growth appears at the end of seven
or eight days, and eventually cultures taken from glycerine broth
will be found to grow well in ordinary bouillon. A pellicle generally
forms. On glycerine agar, minute crumb-like colonies of whitish-
yellow colour appear in six to twelve days. Later, the whole growth
turns browner in colour, and is sometimes dry, sometimes moist, in
appearance depending on age of culture, and consistence of medium.
Ultimately, the discrete colonies coalesce and form a lichenous
growth. By continuous sub-culture on glycerine agar the virulence
of the bacillus is diminished. But in fifteen days after inoculation of
the medium the culture equals in extent a culture of several weeks' age
on blood serum. In alkaline broth to which a piece of boiled white
of egg was added, Klein obtained copious growth, and found that
continued sub-culturing upon this medium also lessens the virulence.
On potato the tubercle bacillus grows well in crumb-like masses.

PLATE 23.

Bacillus tuberculosis.

In sputum from a case of human phthisis.

Stained by Ziehl-Neelsen method.

x 1000.

Bacillus tuberculosis.

Giant cell. Bacilli in situ within the cell.

Stained by Ziehl-Neelsen method.

x 1000.

&-« ■- r%v

,*,

-•\/

*»v

-V

^

/^

- </ '

A

V

-^/

"7 A,

H

-£>S

Bacillus tuberculosis.

From edge of caseous patch in human lung.

X 750.

Bacillus tuberculosis.

mm preparation from glyeerine-glucose-agar culture

4 weeks at 87° C. stained with carbol fuel, sin.

X 1000.

[To fare pane 828.

KOCH'S TUBERCLE BACILLUS 329

Spore formation. — In very old cultivations spore-like bodies can
be observed both in stained and unstained preparations, but neither
the irregular granules within the capsule nor the unstained spaces
between the granules are spores (Babes and Crookshank). That the
bacilli probably possess spores is believed on account of their
behaviour under certain circumstances. For example, tubercular
sputum when thoroughly dried retains its virulent character. Even
cultures of tubercle artificially dried retain their virulence. Now,
no sporeless bacillus is known at present which can withstand thorough
desiccation. Again, non-spore-bearing bacilli are killed with a less
exposure to heat than that which is required to destroy tubercular
sputum. Koch, Lingard, Klein, and others long ago pointed out the
resistance of the bacilli of tubercle to solutions of perchloride of mercury
and to heating in suspension in salt solution, whilst sporeless bacilli
succumbed to the same treatment. So that it is commonly believed
that B. tuberculosis produces spores, even though such have not been
demonstrably proved.

Koch and other bacteriologists have declared the bacillus to be
a " true parasite." Koch based this view upon the belief which he
entertained that the bacillus can only grow between 30° C. and
41° C, and therefore in temperate zones is limited to the animal
body, and can only originate in an animal organism. " They are,"
he said, "true parasites, which cannot live without their hosts.
They pass through the whole cycle of their existence in the body."
But at length Koch and others overcame the difficulties and grew
the bacillus as a saprophyte. Schottelius * has observed that
tubercle bacilli taken from the lung of phthisical persons buried for
years still retains its virulence and capability of producing tuber-
culosis upon inoculation. He further showed that tubercular lung
kept in soil (enclosed in a box) revealed a marked rise in temperature.
Klein quotes these experiments as indications that "tubercle bacilli
are not true parasites, but belong to the ectogenic microbes which
can live and thrive independently of a living host."

It has now been abundantly proved that the tubercle bacillus
is capable of accommodating itself to circumstances much less
favourable than had been supposed, as regards temperature and
environment. For it is now known that it is possible to grow the
bacillus upon glycerine agar at 28° C. (82° F.), obtaining an ample
culture which develops somewhat more slowly than on blood serum,
and to a less extent than at 37° C. Sheridan, Delepine, Czaplewski,
Eansome, Beevor, and others have also been successful in obtaining
growths at room temperature both in Bummer and winter. Moeller
succeeded in growing the bacillus at 20° C, after passing it through
a blindworm.

Centralblatt. f. Bad. und Parasti,, vol. vii. , p. 9.

330 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

The Relation of the Bacillus to the Disease.— Having con-
sidered the structure of " tubercles " and the chief biological facts of
the tubercle bacillus, we may now ask : How does the bacillus set up
the changes in normal tissues which result in tubercular nodules ?
In arriving at a solution of this problem; we are materially aided if
we bear in mind the fact that when such an organism is present
in the tissues it has a double effect. First, there is an ordinary
inflammatory irritation ; and, secondly, there is a specific change set
up by the toxins of the bacillus. Many authorities believe that the
process is, generally speaking, as follows: — Directly the invading
bacilli find themselves in a favourable nidus they commence multi-
plication. In the course of a few days this acts as an irritant upon
the surrounding connective-tissue cells, which proliferate, and become
changed into the large cells known as epithelioid cells. At the periphery
of this collection of epithelioid cells, we have a congested area filled
with lymphocytes drawn thither by the process of inflammation and
constituting the zone of lymphoid cells. The production of the
bacillary poisons changes the epithelioid cells in the centre of the
nodule, some of which become fused together, whilst others expand
and undergo division of nucleus. By this means we obtain a series
of large multinucleated cells, giant cells. Thus is formed the typical
" tubercle." But if the disease is very active, this soon caseates and
breaks down in the centre. In a limb we get a discharge; in a
lung we get an expectoration. Both discharge and expectoration
arise from a breaking down of the new cell formation. Previously
to breaking down we have in a fully developed nodule commencing
at the periphery where the normal tissue is, healthy tissue, then the
inflammatory zone of lymphoid ceils, then epithelioid cells, and in
the centre giant cells, containing nuclei and bacilli. The sputum
or the discharge will, during the acute stage of the disease, at all
events, contain countless numbers of the bacilli, which may thus
be readily detected, and their presence used as evidence of the
disease. It is obvious that if the centre of the nodule degenerates
and comes away as a purulent discharge, a cavity will be left behind.
By degrees this small cavity becomes enlarged, as is frequently the
case in the lung, which particularly lends itself to such a condition.
Hence, though at the outset the affected part of a tubercular lung
becomes solid, ultimately the affected part becomes a cavity, unless
repair sets in, and by growth of fibrous tissue the commencing cavity
is obliterated.

The exact period of giant cell formation depends on the rapidity
of the formative inflammatory processes. Thus different conditions
occur. Giant cells are a constant feature of interstitial tubercles
in connective tissue, but in uncomplicated caseous tubercular
pneumonia there may not be found a single giant cell in

ACTION OF THE BACILLUS 331

a whole lung. Some authorities look upon giant cell formation as
a sign of chronicity of the process. Further, in some of the lower
animals, the giant cells become packed with tubercle bacilli, while
in man it frequently occurs that few or none at all are found.
When the giant cells do contain bacilli they are usually arranged
in one of four ways: (a) polar, (6) zonal, (c) mixed, or (d) at the
periphery of the giant cell. The breaking down of the nodule is
partly due to the bacterial poisons, and partly to the nodule being
non-vascular, owing to the fact that new capillaries cannot grow into
the dense nodule, and the old ones are occluded by the growth of
the nodule.

At first the disease is local, owing to the unfavourable action of
the blood, to phagocytic action, or to the fewness of the number
of bacilli absorbed. From the local foci of disease the tuberculous
process spreads chiefly by three channels : —

(a) By the lymphatics, affecting particularly the glands. Thus
we get tuberculosis set up in the bronchial, tracheal, mediastinal,
and mesenteric glands, and so frequently present as to be a
characteristic of the disease. This is the common method of
dissemination in the body, and by this channel the virus of
tuberculosis is carried along with the stream of lymph and infects
progressively the lymph vessels and glands. It may also be pro-
pagated along the lymphatics in an opposite direction to the lymph
stream.

(b) By the blood-vessels, by means of which bacilli may be
carried to distant organs. But this channel is comparatively
rare. Blood is not a favourable medium for the tubercle
bacillus.

(c) By continuity of tissues, that is by infective giant cell systems
encroaching upon neighbouring tissues, or discharge from lungs or
bronchial glands obtaining, for example, entrance to the gullet and
thus setting up intestinal disease.

It has been abundantly proved that the respiratory and digestive
systems are those principally affected by the tubercle bacillus.
Wherever the bacilli are arrested, they excite formation of granula-
tions or miliary tubercular nodules, which increase and eventually
coalesce. The lymphatic glands which collect the lymph from the
affected region are earliest affected, always the nearest first, and for
a time the disease may appear to be appreciably stopped on its
invading march. Each lymphatic gland acts as a temporary barrier
to progress until the disease has broken its structure down. It
remains "local," in spite of increase in number and importance of
the foci of disease, as long as the bacilli have not gained access to
the body generally.

Channels of Infection, — The common methods of invasion by

332 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

which the tubercle bacillus gains access to the human body are three,
namely, through the skin, and through the alimentary and respiratory
systems. A great variety of cases of skin infection are on record,
although the uninjured epidermis affords a fairly reliable protection, so
that simple contact with tuberculosis sputum does not suffice to pro-
duce infection if the skin be uninjured. The exact means and occasion
of entry are innumerable. Wounds play a great part in rendering
possible the invasion by tubercle bacilli. Infection by the alimentary
tract takes place in a variety of ways. The bacilli may be carried
in with air in mouth-inspiration, by dirty objects placed in the mouth
(in children), by kissing tuberculous persons, or by the ingestion of
infected food. Thus, we may have tuberculosis of the mouth and
tonsils, of the stomach, and of the intestine and other abdominal
organs, including the mesenteric glands. Elsewhere we remark upon
the comparative rarity of primary abdominal tuberculosis in man,
though the disease is more common in animals.

The chief channel of infection is, of course, the respiratory tract,
and the two means by which tubercle bacilli thus reach the body
are (a) inhalation of the dust of dried tuberculous sputum, and
(b) the inhalation of moist particles from the cough-spray of a
phthisical patient. Wherever tuberculous sputum is allowed to dry
the risks are great that the dust so produced may be inhaled in a
virulent form, and lodging at one or more points may set up varying
degrees of tuberculosis. This broad fact is based upon experimental
and clinical evidence. Tuberculosis has been produced experimentally
in animals in this way, and there is clinically the overwhelming
frequency of tuberculosis of the lungs among men exposed to just
such a manner of infection. But Koch, Flugge, and others have
shown that not only is sputum a source of infection when dried
and pulverised, but also when disseminated by coughing, shouting,
etc., in the form of minute moist particles of spray. Koch exposed
rabbits, guinea-pigs, rats, and mice to an infected spray for half an hour
on three successive days, and produced tuberculosis in every animal.
Heymann found that such spray particles from human beings
inoculated into guinea-pigs produced tuberculosis. Most of the
droplets are large and settle rapidly, but some may remain suspended
in the air for more than an hour, retaining, of course, their virulent
properties. Heymann found the duration of life of the bacilli in these
droplets was eighteen days in the dark, and three days when exposed
to light. Under ordinary circumstances and an absence of draughts,
the zone of danger from a coughing consumptive extends to a
distance of about three feet. It must be remembered that the
tubercle bacilli in the moist particles of cough-spray are probably of
higher virulence than those in dried sputum dust, and therefore it
seems reasonable to suppose that the cough-spray is the most

IN BOVINES 333

dangerous channel of infection in tuberculosis.* At the same time
experience shows that the degree of inf ectivity of phthisis is not a very
high one. It is a truly infective disease, but not an extremely
infectious disease. It may be rightly described as sub-infectious.^

Toxins of the Tubercle Bacillus. — Many investigators have isolated
products from pure cultures of the tubercle bacillus. These have
comprised chiefly albumoses, alkaloids, various extractives, and inorganic
salts. Koch isolated " tuberculin " from cultures of tubercle bacillus
upon glycerine broth by means of evaporation and precipitation with
alcohol. Buchner obtained by trituration and compression of fresh
tubercle bacilli a substance termed " tuberculo-plasmine." But of
the real nature of the toxins of the tubercle bacillus little is known.

Bovine Tuberculosis

Cattle come first amongst animals liable to tubercle. Horses may
be infected, but it is comparatively rare, and among small ruminants
the disease is rarer still. Dogs, cats, and kittens may be easily
infected. Amongst birds, fowls, pigeons, turkeys and pheasants
the disease assumes almost an epidemic character. Especially do
animals in confinement die of tubercle, as is illustrated in zoological
gardens.

Bovine Tuberculosis. — Kespecting the lesions of bovine tuberculosis,
it will be sufficient to say that nothing is more variable than the
localisation or form of its attacks. The lungs and lymphatic glands
come first in order of frequency, next the scions membranes, then
the liver and intestine, and lastly the spleen, joints, and udder
(Nocarcl). The anatomical changes in bovine tubercle are mostly
found in the lungs and their membranes, the pleurae. It also affects
the abdomen and its chief organs, the peritoneum, and the lymphatic
glands. In both of these localities a characteristic condition is set
up by small grey nodules appearing on the pleura and peritoneum,
the nodules, increasing in size, giving an appearance of "grapes."
Hence the condition is called grape disease, or Perlsucht. The organs,
as we have said, are equally affected, and when we add the lymphatic
glands we have a fairly complete summary of the form of the disease
as it occurs in cattle. In about half of all cases the lungs and serous
membranes become simultaneously affected, in about one-third the
lungs alone; and in about one-fifth the serous membranes alone
(Friedberger and Frohner). As has been pointed out by Martin,

* For a discussion on the channels of infection in tuberculosis, see Carnet,
Tuberculosis, 1904, pp. 96-282; Fliigge, Zeitschrift far Hyg. u. Jufek,, Band
xxxviii., 1901.

f Koch, Etiology <\f Tuberculosis; in Brit. M<<1. Jour., 1903, i., p. 593 (Hillier),
will be found a useful summary of modern views on the question.

334 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

Woodhead, and others in their evidence before the Royal Commis-
sion, the organs, glands, and membranes are the common sites for
tubercle, not the muscles (or meat).

The following table records the findings of Geddes, who in 1901-
1902 was sent by the American Government to examine by means
of tuberculin some of the chief breeds of British dairy cattle.*

Breed.

No. Tested.

No. Rejected.

Percentage
of Rejections.

Jersey (in Great Britain) .
Aberdeen — Angus
Ayrshire ....
Shorthorn ....
Guernsey (in Great Britain)
Galloway ....
Highland ....
Red Polled

Hereford ....
Jersey (on island)
Dexter Kerry .
Guernsey (on island) .
Sussex ....

42

258

33

228

57

36

19

57

428

324

15

53

1

23
104

8
53
11

6

o

4
17
1
0
0
0

54'7b"

28-73

24-24

23-25

19-30

16-67

15-79

7 02

3-97

0-31

o-oo
o-oo
o-oo

Total.

1551

230

14-77

Eliminating the tests on Guernsey and Jersey, the proportions of
reactions among the tests made in Great Britain and Ireland were as
follow :— in 1901, 13*67 per cent. ; in 1902, 20'97 ; and for both years,
17'92. Hopkins examined 571 Shorthorns and found the percentage
of positive reaction was 23*0 as compared with Geddes's result of
23*25.f

Tuberculosis of the tedder is comparatively rare. Out of 100
tuberculous cows not more than 3 or 4 have tuberculosis of the
udder (Bang). The disease occurs as a diffuse, slightly hard, enlarge-
ment, generally unaccompanied by fever or tenderness of the organ.
Usually only one quarter is attacked, and that generally a posterior
quarter. The gland lobules become hypertrophied, and the larger
milk-ducts contain yellowish caseous masses full of bacilli. As the
condition advances, there is a considerable increase of the inter-
lobular connective tissue (interstitial mastitis) of the nature of a
sclerosis, and firm tubercles of various sizes begin to appear. Con-
sequent upon these changes the udder becomes nodular, and hard

* Nineteenth Annual Report of the Bureau of Animal Industry, 1902, p. 551.
f Report of Minister of Agriculture, Dominion of Canada, 1902, p. 134.

TUBERCULOUS MILK 335

and tough. Miliary tubercles appear in the walls of acini, and
enormous deposits of bacilli may be found in the udder. Simultane-
ously with these changes, the mammary lymphatic glands (pudic
glands) lying above the posterior region of the udder became
enlarged, indurated, and caseous. The disease may advance slowly or
with great rapidity. But finally the condition is such that the
glandular tissue of the udder is, as it were, smothered by the hyper-
trophy and fibrous transformation of the interstitial connective tissue.
The large excretory ducts become blocked by granulations or fibrous
growth outside them, or by caseous masses inside. This stage
inevitably leads to milk suppression (see also p. 203).

It should not be forgotten that tuberculosis of the udder is
associated with tuberculosis of the internal organs. It is almost
invariably secondary. It may exist with mild or advanced disease
of the internal organs. Its diagnosis is all the more difficult
owing to the fact that there may be no symptoms. Generally,
opinion must be guided by the local condition of the udder, coupled
with the condition of the milk. It may occur as a slow, painless
growth only evident when advanced, or it may increase with extra-
ordinary rapidity. This latter fact makes it desirable that every
animal suffering from tuberculosis of however mild a character should
be strictly eliminated from dairy stock. The three points usually
emphasised for diagnosis of tuberculous udder disease are — (a)
abnormal milk from one quarter, generally a posterior quarter ; (b)
some hardness, toughness, or irregularity of the udder; and (c)
enlargement of supra-mammary glands.* The best diagnostic of
general tuberculosis is the tuberculin reaction.

Changes in Milk from a Tuberculous Udder. — One of the first
signs of abnormality is the diminution in the yield. Previously to
this it is said there is an actual increase in the quantity of milk.
As soon as the disease begins to have effect, there is a definite decline
in the yield. For example, a cow which in health gave, say, fifteen
litres of milk, falls to one half or one quarter of that amount. The
milk also changes in consistence, becoming thin, watery, and Berous.
At the same time the colour may turn In yellow, and the flocculi and
Hakes which occur in milk from a healthy udder are present in
larger size. As the yield diminishes, the consistence of the fluid
becomes more and more irregular, the flocculi predominating. If
such milk be allowed to stand in a vessel, a deposil of solid matter,
composed of these fragments, settles down, leaving a superficial layer,
of thin fluid at the top. Finally, the consistence becomes sero-
purulent and then purulent. Hence, previously to suppression we
get a thick yellow purulent fluid, having ;in alkaline reaction,
coagulated casein, and diminution of lactose. As a, rule, tubercle

* See also Report of Royal Commission on Tuberculosis, 1S%, part iii., pp. 11, 42.

336 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

bacilli are readily found, and whether that be so or not the milk is
highly infective.*

The Entrance of the Bacillus into Milk. — There are two main
sources of the tubercle bacilli found in milk, namely, a bovine source
and a human source. The two common channels respectively are a
tuberculous udder and a phthisical lung. From the former, milk
may derive a direct and abundant supply of tubercle bacilli ; from
the latter, milk may become indirectly contaminated by the parti-
culate matter of dried sputum.

Tuberculosis may be introduced into healthy cows in a variety of
ways. The most common method is by means of a tuberculous
animal, from the excretions and discharges of which infection may
be conveyed to soil, water, air, fodder, and general surroundings. In
this way not only other animals cohabiting with a tuberculous
animal become infected, but premises, stables, and utensils may also
become infected. The milk of a tuberculous animal may also be
consumed by other animals on the farm, and so a vicious circle of
infection is completed. Eavenel has shown that by the cough of a
tuberculous cow tubercle bacilli may be distributed. Of thirty-four
examinations carried out on five tuberculous cows, tubercle bacilli
were detected on twenty occasions. One of the cows constantly
coughed up a tenacious mucus containing large numbers of tubercle
bacilli.f The saliva as well as the bronchial mucus of tuberculous
cows has been found to contain abundant bacilli, and by licking her
udder it is possible for a tuberculous cow to convey tubercle bacilli
to its exterior surface.

The excreta also are infective when lesions are located in the
alimentary canal. In tuberculosis affecting the alimentary canal of
the cow (1 per cent, of the cases), it is thus possible to get contam-
ination of the milk, indirectly, from the excreta. The mucous
membrane of the intestine, especially the colon, sometimes shows
tubercular ulcers, which are less frequently observed in the abomasum.
Tubercles may also develop under the mucous membrane, and serosa
of the stomach and intestines. In these ways arises a condition of
intestinal tuberculosis, which in its acute or ulcerating stage will
cause the excreta to be loaded with tubercle bacilli. Any one
familiar with a cowshed will at once recognise how readily milk
might become infected under such circumstances, which, though
undoubtedly exceptional, must not be overlooked.^ In these ways
stalls may become infected and transmit the disease to fresh herds
stabled in such premises. Nor are herds unstabled always free from

* See also Report of Royal Commission on Tuberculosis, 1896, part iii. , p. 142.
+ Commonwealth of Pennsylvania, Bulletin 75 (Pearson and Ravenel), 1901, p. 82.
% Trans. British Congress on Tuberculosis, 1901, vol. iii., p. 664 (Boinet and
Heron).

BOVINE AND HUMAN BACILLI 337

tuberculosis, as has been recently stated. A number of observers
have shown that whilst it is true that ill-ventilated, dark, damp cow-
sheds predispose to infection, milch cows living entirely in the open
do not, on that account, escape the disease.* It depends upon infec-
tion in the herd, that is, upon contagion. But it is probable that,
through more than any other channel, the udder is the most common
one for the conveyance of infection. When the udder is affected, the
milk invariably contains large numbers of bacilli, and it will be
understood when one cow in a herd is so diseased, the entire volume
of mixed milk from the herd may be contaminated. The presence of
the bacilli in the milk is not always proportionate to the extent of the
disease in the animal, especially when diagnosed clinically. The
reason of this is the difficulty of clinical diagnosis between chronic
interstitial mastitis and tuberculous udder. There can, however, be
little doubt that the chief source of tubercle bacilli in milk is the
tuberculous udder.

Finally, milkers affected with phthisis may readily infect the
milk, either by the repulsive habit of spitting on their hands prior to
milking, or by dried expectoration in cowshed, dairy, or milk-shop.
After distribution, milk is exposed in a variety of ways to dust, and
it cannot be doubted that such dust does at times contain particulate
matter derived from dried tubercular expectoration, and that there-
fore in this way also it is possible for milk to become infected.

The Bovine and Human Tubercle Bacillus Compared.— The
morphology of the bacilli in cultures of bovine origin is more
uniform and constant than in cultures from man. The bovine bacilli
are thick, straight, and short, seldom more than 2 ^ in length, and
averaging less (Theobald Smith). In the early generations many
individuals are seen which are oval, their length not more than
double their breadth. They are less granular than those from a
human source. They stain evenly and deeply with carbol-fuchsin,
beading being almost always absent from young cultures, and often
from old ones. In culture they have fairly constant and persistent
peculiarities of growth and morphology (Eavenel).

The human bacilli are, on the other hand, much longer, thinner,
and tend to increase in length in sub-cultures. They are generally
more or less curved, sometimes showing S-shaped forms. They
stain less intensely with carbol-fuchsin, but beading is generally
seen, even in early growths, and is often very well marked.

The above characteristics are most evident and persistent in
cultures grown on blood serum. On glycerine agar, glycerine
bouillon, and glycerine potato, bovine and human tubercle bacilli
approach each other in cultural features and morphology much more
closely, and by continued cultivation the differences tend to become
* Report on Bovine Tuberculosis, Government of New Zealand, 1900 (Gilruth).

Y

338 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

obliterated. Bovine cultures are more difficult to isolate than human,
are apt to grow as discrete colonies in the first culture, and for
several generations grow in a thin layer which somewhat resembles
ground glass. The optimum temperature and the thermal death-
point are practically the same in both forms.

The human bacillus, as a rule, grows somewhat more easily and
abundantly from the first, and will grow well on glycerine agar in
sub-cultures made directly from the original growth on blood serum.
All attempts to obtain a like result with the bovine organisms have
failed. In artificial culture the human bacillus rapidly loses viru-
lence. The bovine bacillus grows as a film on blood serum, whereas
the human bacillus produces warty growths.

The morphological distinctions tend to disappear also in the
tissues of susceptible animals. We may inoculate a typical bovine
culture, and in a short time obtain from the various organs long and
much beaded bacilli simulating the human variety (Hueppe).

The most striking dissimilarity is, however, seen in the action of
the bacilli from the two sources on animals. By whatever method
of inoculation, the bovine bacillus, as a rule, possesses a much
greater pathogenic power than the human bacillus for all animals on
which it has been tried (Villemin, Eavenel, and others), the only
exceptions being possibly those animals, like guinea-pigs, which are
so extremely susceptible to both types that it is difficult to draw very
much distinction between them. Dorset and other workers hold
that in bovine and human tuberculosis we have to do with organisms
differing usually in virulence, but between which there is no other
essential distinction.*

Intercommunicability of Human and Bovine Tuberculosis

Since the discovery by Koch in 1882 of the tubercle bacillus, it
has generally been held that tuberculosis in man and animals is one
and the same disease.f Villemin (1865) was the first to main-
tain this identity on the results of inoculation of bovine and
human tubercular matter into small animals. Chauveau (1868)
carried out similar experiments upon cattle.^; Both workers
were successful in transmitting the disease, which produced similar
effects in the inoculated animals. Many other workers have
obtained like results, which were more or less uniformly in support
of the view that the identity of bovine and human tuberculosis was

* Trans. British Congress on Tuberculosis, 1901, vol. iii., pp. 553-81. See also
experiment of Kossel and others, to which reference is made on p. 344, and
U.S. Dep. of Agriculture, 1904, Bull. 52 (Dorset).

f Kruse, Pansini, Fischel, Johne, etc. See also Twelfth and Thirteenth Annual
Reports of the Bureau of Animal Industry, Washington, 1895-96 (Theobald Smith).

% Congres pour Vttude de la Tuberculose, Paris, 1888.

INTERCOMMUNICABILITY 339

a tiling to be accepted as a proved and fundamental proposition.
Not only have various workers separately arrived at that conclusion,
but the conclusions of the Eoyal Commission on Tuberculosis, 1895,
included the following words :— " We find the present to be a con-
venient occasion for stating explicitly that we regard the disease as
being the same disease in man and the food animals, no matter though
there are differences in the one and the other in their manifesta-
tions of the disease ; and that we consider the bacilli of tubercle to
form an integral part of the disease in each, and (whatever be its
origin) to be transmissible from man to animals, and from animals
to animals. Of such transmission there exists a quantity of
evidence, altogether conclusive, derived from experiment."*

Whilst there was up to 1901 almost entire unanimity of opinion
amongst various workers in respect to this identity, it should not be
supposed that there was unanimity in respect to the degree of
pathogenicity. It was, in fact, conceded on all hands that
tuberculosis was a more virulent disease in animals than in man,
and that the bacillus in the two species differed in various respects
as to morphological, biological, and pathological properties (Theobald
Smith, Dinwicldie, Frothingham). In 1901, however, Dr Koch
expressed the opinion that, "human tuberculosis differs from bovine,
and cannot be transmitted to cattle," f and that bovine tuberculosis
was scarcely, if at all, transmissible to man. On the same occasion
counter-evidence was produced by MacFadyean,J Ravenel,§ Crook-
shank,! | and many others.

As a result of experiment, Koch felt "justified in maintaining that human
tuberculosis differs from bovine, and cannot be transmitted to cattle." He further
concluded that bovine tuberculosis was scarcely, if at all, transmissible to man. It
will be at once obvious that these two conclusions, that human tuberculosis is not
transmissible to cattle, and that bovine tuberculosis is not transmissible to man, are
of profound and far-reaching importance. Now if it were found on further
investigation that these conclusions were correct, the prevention of human
tuberculosis would be greatly simplified, and the precautionary measures hitherto
adopted for protecting human food from infection with animal tuberculosis need
not be enforced with the same stringency as at present, or, at least, would require
considerable modification. H

* Report of Royal Commission on Tuberculosis, 1895, part i., p. 10, par. 23.

I Trans. Brit. Cong, on Tuberculosis, 1901, vol. i., p. 29.

X Ibid., vol. i. , p. 79.

§ Ibid., vol. i., p. 91, and vol. hi., p. 553. || Ibid., vol. i., p. 92.

If It would not necessarily be justifiable to say that in this event such pre-
cautionary measures might be "altogether withdrawn," as has been suggested, for
it will be understood that tuberculous meat and milk from animals might still be
unwholesome and unfit for the food of man, even though there was evidence
to show that the exact specific disease was incommunicable. Presumption would
always be against the consumption of meat or milk plus disease products, whether
tubercle bacilli or not, for such food is not of the quality and nature reasonably
expected by the purchaser. Various non-specific diseases of animals cause meat to
be unfit for the food of man.

340 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

The evidence furnished by Dr Koch for the conclusion that human tuberculosis is
not communicable to animals is briefly this : — Nineteen young cattle which had
stood the tuberculin test (and were therefore presumably free from tuberculosis)
were treated as follows : — Six were fed with tubercular human sputum almost daily
for seven or eight months. Four repeatedly inhaled great quantities of bacilli
which were distributed in water and scattered with it in the form of spray. The
remainder (9) were infected in various ways with pure cultures of tubercle bacilli
taken from human tuberculosis, or tubercular sputum direct from consumptive
patients. In some cases the bacilli or sputum were injected under the skin, in
others into the peritoneal cavity, and in others into the jugular vein. None of
these 19 cattle showed any symptoms of disease. After six to eight months they
were killed, and in their internal organs not a trace of tuberculosis was found. The
result was utterly different, however, when the same experiment was made on
cattle free from tuberculosis with tubercle bacilli from bovine sources. In this
case virulent tuberculosis rapidly supervened. Further, an almost equally striking
distinction between human and bovine tuberculosis was brought to light by a
feeding experiment with swine. Six young swine were fed daily for three months
with the tubercular sputum of consumptive patients. Six other swine received
bacilli of bovine tuberculosis with their food daily for the same period. The
animals that were fed with sputum remained healthy and grew lustily, whereas
those that were fed with the bacilli of bovine tuberculosis soon became sickly, were
stunted in their growth, and half of them died. After three months and a half the
surviving swine were all killed and examined. Among the animals that had been
fed with sputum no trace of tuberculosis was found, except here and there little
nodules in the lymphatic glands of the neck, and in one case a few gray nodules in
the lungs. The animals, on the other hand, which had eaten bacilli of bovine
tuberculosis had, without exception (just as in the cattle experiment), severe
tubercular diseases, especially tubercular infiltration of the greatly enlarged
lymphatic glands of the neck and of the mesenteric glands, and also extensive
tuberculosis of the lungs and the spleen. The difference between human and
bovine tuberculosis appeared not less strikingly in a similar experiment with asses,
sheep, and goats, into whose vascular systems the two kinds of tubercle bacilli
►were injected. Dr Koch also stated that other experiments in former times, and
recently in America, have led to the same result.

In support of his second contention, namely, that bovine tuberculosis is not trans-
missible to man, Dr Koch points out that the direct experiment upon human beings
is, of course, out of the question, and hence it is necessary to rely upon indirect
evidence. Dr Koch, therefore, reasons as follows : Tuberculosis, caused by meat
or milk, can be assumed with certainty only when the intestine suffers first, i.e.,
when a so-called " primary tuberculosis " of the intestine is found. If bovine
tubercle bacilli are capable of causing disease in man there are abundant oppor-
tunities for the transference of the bacilli from one species to the other, and cases
of primary intestinal tuberculosis from consumption of tuberculous milk ought
therefore to be of common occurrence. " But such cases," he maintains, " are
extremely rare." In support of this view Dr Koch stated that he had only seen
2 cases; that only 10 cases had been met with in the Charite Hospital in
Berlin ; and that out of 3104 post-mortems of tubercular children, Biedert observed
only 16 cases. Reference was also made to other similar evidence.

Finally, Dr Koch maintained that " though the important question whether man
is susceptible to bovine tuberculosis at all is not yet absolutely decided, and will not
admit of absolute decision to-day or to-morrow, one is, nevertheless, already at
liberty to say that if such a susceptibility really exists the infection of human beings
is but a very rare occurrence. "

Such, then, was the position of the question at the end of 1901. It may be con-
venient here to add the chief reasons for supposing that bovine and human
tuberculosis are one and the same disease, and intercommunicable : —

1. That the tubercle bacillus of bovine tuberculosis possesses characteristics of
shape, size, staining, and cultivation on artificial media similar to, and in the
opinion of many authorities almost identical with, the tubercle bacillus of human
origin.

INTKRCOMMUNICABILITY 341

2. That in specially prepared and suitable media artificial cultures of the tubercle
bacillus from bovine and human sources have produced indistinguishable effects
when they have been employed to infect a variety of animals, which would seem
to indicate that the conditions produced are only variations of one and the same

3. That tuberculin * produces a specific reaction in tuberculous cattle, whether
human or bovine tubercle bacilli have been employed in its preparation.— (Mae-
Fadyean. )

[It will be seen that these three reasons have relation to the theory of the identity
of bovine and human tuberculosis. ]

4. That because the tubercle bacillus derived from bovine sources is, either by
inoculation or ingestion as food, admittedly very virulent and dangerous for such
diverse species of animals as the rabbit, horse, dog, pig, sheep, and cow, it is
highly probable that it is also dangerous to man.f For il is well known that the
majority of disease-producing bacteria are harmful to only one or two species of
animals, but those disease-producing bacteria that are common to all the domesticated
animals are also able to produce disease in man.

5. That the statistics and percentages set forth by Dr Koch with regard to
primary intestinal tuberculosis cannot be accepted as representing universal
experience. For example, in two separate reports from two children's hospitals in
London and Edinburgh dealing with 547 cases of death from tuberculosis in children,
it appears that 29*1 per cent, and 28 '1 percent, of the cases respectively primary
infection appeared to have taken place through the intestine. But quite apart from
statistics, the whole question of such primary intestinal tuberculosis (which Dr
Koch held as the only acceptable evidence of tuberculous infection through milk
and meat) is fraught with many difficulties and fallacies, and is at present sub judice.
It has been shown by Professor Sidney Martin and others that primary intestinal
tuberculosis may not be, by any means, an invariable criterion of tubercular infection
by means of food {vide infra).

6. That there are on record a number of cases in which there appeared to be
substantial evidence to show that persons had contracted tuberculosis, directly or
indirectly, by means of milk or meat. It is obvious that such cases, unless occurring
with extraordinary frequency, are only of relative value. Moreover, there arc other
channels of infection to eliminate, and this it is often impossible to do.

7. That the results obtained from the inoculation of human tubercle into animals
by Dr Koch cannot be accepted as in complete accord with universal experience.
In England alone somewhat similar experiments have been performed, having positive
results. Several years ago Professor Crookshank carried out such an experiment.
He obtained sputum containing numerous tubercle bacilli from an advanced case of

* Tuberculin is a product of the artificial cultivation of the tubercle bacillus (human
or bovine) now used as a diagnostic injection test into cattle. If such cattle are
suffering from tuberculosis they " react" (giving high temperature, swelling at the
point of inoculation, etc.) ; if not so suffering, they do not react.

t See the researches of Villemin (1865), Klebs, Chauveau(1867), Gerlach, Giinther
and Harms (1870-1873), Bollinger, and others. Further, Frfedberger and Frohner
state in their Veterinary Pathology that Wesener compiled reports up to 1884 of 369
feeding experiments, the positive and negative results of which were about equal in
number. From this compilation it appears that («) 71 animals, among which guinea-
pigs and swine proved most susceptible, were experimented upon with human tuber-
cular matter ; (6) 180 experiments were made with tubercular matter from cattle ;
(c) the flesh of tuberculous cattle was given on 32 occasions as food, with the result
that pigs were found to be more susceptible than other animals, and that dogs were
unaffected; and (d) the milk of tuberculous cows was given as food in 86 cases.
From these experiments it was found that in the scale of comparative racial sus-
ceptibility the herbivora (cattle, sheep, goats) proved highest, then swine, and after
these guinea-pigs and rabbits. Carnivorous animals were little affected. Bovine
tubercular matter was found to possess the greatest power of infection, then came
the sputum of tuberculous men, then the milk of tuberculous animals, and lastly,
tuberculous flesh.

342 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

human consumption. This was injected into the peritoneal cavity of a healthy calf.
The animal became ill and died forty-two days after inoculation from pyaemia (blood-
poisoning). On post-mortem examination it was found that there were abundant
signs of generalised tuberculosis.* This calf was not tested with tuberculin pre-
viously to the experiment. Professor Sidney Martin carried out a number of
experiments for the Royal Commission on Tuberculosis, t amongst which three out
of four calves fed on human tuberculous sputum contracted the disease.

Iii 1902 Koch again emphasised the comparative rarity of primary
intestinal tuberculosis in the human being, and the local, as
distinguished from the general, infective nature of accidental bovine
inoculation of man {tuberculosis verrucosa cutis). In isolated cases
the nearest lymph glands might become affected, but the disease
remained nevertheless a local one. Dr Koch further expressed the
view that if bovine tuberculosis was transmissible to man by means
of the milk of cows with tuberculous udders, it would be reasonable
to suppose that "groups of illnesses" would occur, in a manner
analogous to other infective diseases, though the circumstances would
differ owing to the different length of the incubation periods. By
way of illustrating the non-infectivity of bovine tubercle bacilli
conveyed by milk, Koch points out (a) that bovine tubercle bacilli
must be taken into the human system very frequently, as 1 to 2
per cent, of all milch cows suffer from tuberculous udders ; (b) that
in addition to being drunk in considerable quantity and for long
periods, such milk is also widely distributed; (c) that domestic
sterilisation of milk does not occur to any appreciable extent; (d)
that the same may be said of the large dairies ; and finally (e) that
if milk under such circumstances is dangerous, the butter derived
from it will also be dangerous. For these reasons he maintained
that any resulting disease must be widespread. Yet Koch has found
"instead of the countless cases," which we ought to expect, "two
groups of illnesses and 28 isolated cases of illness." On examina-
tion he finds most of these recorded cases not free from objection.
To carry conviction as to milk-borne tuberculosis, Koch maintains,
that the following conditions must be fulfilled: — (i.) Certain proof
of tubercle in the person affected; (ii.) exclusion of other sources
of infection ; (iii.) the condition of all the consumers of the suspected
milk; (iv.) the exact source of the suspected milk, particularly in
respect to the disease of the udder of the cow yielding the milk.
Finally, he concludes that all that can be said at present is that the
injurious effects of milk infected with bovine tuberculosis and its
products are not proven.

On the other hand, many other workers have been investigating

* Bacteriology and Infective Diseases— Edgar M. Crookshank, 1896, pp. 389-391.

t Report of the Royal Commission appointed to inquire into the Effect of Food
derived from Tuberculous Animals on Human Health, 1895, part iii., Appendix,
pp. 18 and 19.

1NTERCOMMUNICABILITY 343

the matter in Europe and America, and Delepine,* Hamilton,! Orth,
and Behring J are amongst those who have obtained positive results.
Hamilton was able again to establish the truth of Martin's statement,
that not infrequently tuberculosis occurred in animals fed on
tubercular sputum without affecting the mesenteric and other
intestinal glands upon which Koch relied as indication of positive
results.

The fundamental feature of Behring's theory based upon his
experiments, the results of which are entirely opposed to those of
Koch, is that tuberculosis in animals and in human beings
represents different varieties of the same disease, and that it is
transferable, especially by the agency of tuberculous milk. He
distinguishes in this respect between adults and infants, and main-
tains that while the former, except under special conditions of the
digestive organs, may safely partake of unsterilised milk, infants
are particularly liable to infection from this source. Experiments
made on newborn foals, calves, guinea-pigs, and other animals show
that the mucous membrane of the intestines at that stage of their
development is like " a filter with very large pores," and that the
bacilli of infection pass through it into the blood precisely as if the
animals had been inoculated with the poison. In subsequent stages
of their development these animals are provided by nature with a
mucous membrane which tends to exclude the danger of infection.
Behring is convinced that the same holds true of infants and adults,
and that a large portion of mankind is infected in infancy with the
germs of tuberculosis derived from cows' milk. In support of his
assertion he adduces statistics both of anatomical and of pathological
investigation.

This latter evidence tending to show the transmission of tuber-
culosis to man by means of milk and meat, is of the same character
as that upon which the Eoyal Commission relied when it
reported : — " We cannot refuse to apply, and we do not hesitate to
apply, to the case of the human subject, the evidence (of trans-
mission of the disease) thus obtained from a variety of animals
that differ widely in their habits of feeding — herbivora, carnivora,
omnivora. As regards man, we must believe that any person
who takes tuberculous matter into the body as food incurs some
risk of acquiring tuberculous disease" § And again, " Wc have
obtained ample evidence that food derived from tuberculous animals
can produce tuberculosis in healthy animals. In the absence of

";; Brit. Med. Jour., 1901, ii., p. 1224.

f Trans, of Highland and Agricult. Soc. of Scotland, 1903, and Public Health,
1903, p. 689.

J Deut. Med. Woch., 1903.

§ Report of Royal Commission, 1895, part i., p. 10, par. 22.

344 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

direct experiments on human subjects we infer that man also can
acquire tuberculosis by feeding upon materials derived from tuber-
culous food-animals." *

Viewing all the facts, there can be little doubt but that this con-
clusion is the right one from the point of view of the public health.
Various circumstances have in all probability contributed to render
unsuccessful or irregular in result the numerous feeding experiments
which have been made. The tissues of animals differ greatly in
susceptibility to tuberculosis; the infective material is exposed to
the digestive juices which are, in measure, germicidal, and yet not
equally so ; the virulence of the infective material itself varies
enormously, as does the virulence between different generations or
races of tubercle bacilli. Hence it comes about that one animal
may eat with its food a certain amount of tuberculous material, and
yet not develop tuberculosis, whilst another animal of the same
species might quickly develop the disease, which would in all
probability show itself at the animal's weakest point, and not
always necessarily in the intestine. Further, there is another point
which should not be overlooked, namely, the subsequent treatment
of the inoculated animal. Whilst it is essential to prove that the
animal to be inoculated is free from tuberculosis, it should be
remembered that in taking very healthy animals for experiment,
and in subsequently treating them in what may be termed an
"ideal" fashion, some of the very conditions essential to the pro-
duction of the disease in ordinary life are removed. As in men, so
in cattle and other animals, it may be presumed that abundance of
good food and fresh air, and, in general, an ideal environment, tend
to counteract the effect of the inoculated or communicated virus.
Thus such experiments as those stated above may not always fairly
represent the modes of transmission of the disease as they occur
in ordinary life. It is not the "very healthy" animal of a herd,
well housed and fed, which contracts tuberculosis.

As a result of the wide differences of opinion revealed by the
pronouncement of Koch's views, special Commissions of Inquiry were
instituted in Germany, Great Britain, and other countries, in addition
to the individual research work to which reference has been made.
As this book has been passing through the press, reports of these
inquiries have been made by the German Imperial Health Office and
the Eoyal Commission on Tuberculosis, appointed in 1901 by the
British Government. The conclusions are briefly as follows : —

Kossel, Weber, and Heuss, who carried out a comparative research

upon tubercle bacilli of different origins, made a number of experiments

on calves by injecting some forty different strains of human bacilli

and fifteen strains from bovine, fowl, and swine sources. They con-

* Report of Royal Commission, 1895, part i., p. 20, par. 77.

INTEUCOMMUNICABILITY 345

elude as a result of these experiments that in a preponderating
number of cases of human tuberculosis, tubercle bacilli were found
distinguishable from the bovine bacilli of Perlsucht in cows
morphologically, culturally, and in pathogenic properties, but that
exceptionally in man tubercle bacilli occur which cannot be distin-
guished. They hold, nevertheless, that the possibility of infection in
man, under certain circumstances, by milk from tuberculous udders is
proved. They found that generalised tuberculosis was produced in
animals by injecting strains of tubercle bacilli obtained from tuber-
culous diseases in children.*

The Eoyal Commission appointed in this country has also issued
an interim report (1904), signed by Sir M. Foster and Professors Sims
Woodhead, Sidney Martin, MacFadyean, and Boyce. The Commission
was appointed to inquire and report with respect to tuberculosis : —

(1) Whether the disease in animals and man is one and the same ;

(2) whether animals and man can be reciprocally infected with it;
and (3) under what conditions, if at all, the transmission of the
disease from animals to man takes place, and what are the circum-
stances favourable or unfavourable to such transmission.

The first line of inquiry upon which they entered may be stated
in their own words, as follows : —

What are the effects produced by introducing into the body of the bovine animal
(calf, heifer, cow), either through the alimentary canal as food, or directly into the
tissues by subcutaneous or other injection, tuberculous material of human origin,
i.e., material containing living tubercle bacilli obtained from various cases of tuber-
culous disease in human beings, and how far do these effects resemble or differ from
the effects produced by introducing into the bovine animal, under conditions as similar
as possible, tuberculous material of bovine origin, i.e., material containing living
tubercle bacilli obtained from cases of tuberculous disease in the cow, calf, or ox ?

We have up to the present made use, in the above inquiry, of more than twenty
different " strains " of tuberculous material of human origin, that is to say, of material
taken from more than twenty cases of tuberculous disease in human beings, including
sputum from phthisical patients, and the diseased parts of the lungs in pulmonary
tuberculosis, mesenteric glands in primary abdominal tuberculosis, tuberculous
bronchial and cervical glands, and tuberculous joints. We have compared the
effects produced by these with the effects produced by several different strains of
tuberculous material of bovine origin.

In the case of seven of the above strains of human origin, the introduction of the
human tuberculous material into cattle gave rise at once to acute tuberculosis, with
the development of widespread disease in various organs of the body, such as the
lungs, spleen, liver, lymphatic glands, etc. In some instances the disease was of
remarkable severity.

In the case of the remaining strains, the bovine animal into which the tuberculous
material was first introduced was affected to a less extent. The tuberculous disease
was either limited to the spot where the material was introduced (this occurred, how-
ever, in two instances only, and these at the very beginning of our inquiry), or spread
to a variable extent from the seat of inoculation along the lymphatic glands, with, at
most, the appearance of a very small amount of tubercle in such organs as the lungs
and spleen. Yet tuberculous material taken from the bovine animal thus affected,
and introduced successively into other bovine animals, or into guinea-pigs from which

* Tuberkulose-Arbeiten aus liem Kaiserlichen Gesundheitsamte, Heft i., 1904, p. 34.

346 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

bovine animals were subsequently inoculated, has, up to the present, in the case of
five of these remaining strains, ultimately given rise in the bovine animal to general
tuberculosis of an intense character ; and we are still carrying out observations in
this direction.

We have very carefully compared the disease thus set up in the bovine animal by
material of human origin with that set up in the bovine animal by material of bovine
origin, and so far we have found the one, both in its broad general features and in
its finer histological details, to be identical with the other. We have so far failed to
discover any character by which we could distinguish the one from the other ; and
our records contain accounts of the post-mortem examinations of bovine animals
infected with tuberculous material of human origin, which might be used as typical
descriptions of ordinary bovine tuberculosis.

The results which we have thus obtained are so striking, that we have felt it our
duty to make them known, without further delay, in the present interim report.

The Commission defer to a further report all narration of the
details of their experiments, as well as all discussions, including those
dealing with the influence of dose and of individual as well as racial
susceptibility, with questions of the specific virulence of the different
strains of bacilli, with the relative activity of cultures of bacilli and
of emulsions of tuberculous organs and tissues, and with other points.

Meanwhile they have thought it their duty to make this short
interim report, for the reason that the result at which they have
arrived, namely, that tubercle of human origin can give rise in the
bovine animal to tuberculosis identical with ordinary bovine tubercu-
losis, seems to them to show quite clearly that it would be most unwise
to frame or modify legislative measures in accordance with the view
that human and bovine tubercle bacilli are specifically different from
each other, and that the disease caused by the one is a wholly different
thing from the disease caused by the other.

In this final conclusion as to administrative measures both German
and British Commissions agree. They also agree as to the inter-
communicability of bovine and human tuberculosis.*

Diagnosis of Bovine Tuberculosis. — There are three methods
of diagnosis — clinical, bacteriological, and by means of tuberculin.

(a) Clinical. — When tuberculosis affects the lungs and respiratory
organs generally, it is accompanied by a frequent cough, but no fever.
There is disturbance of the respiration, the breathing being quickened
by slight exertion or excitement, and the cough stimulated by
changes of temperature. The departure from the normal in the
relative length of the inspiratory and expiratory movement (the
expiration being markedly prolonged) can be readily seen as a rule,
and not uncommonly in these cases a rough harsh sound may be
heard in the throat during respiration. By auscultation it is possible
sometimes to detect dull portions of the lung surrounded by areas of
increased resonance. The vesicular murmur is louder and harsher

* This subject is fully discussed in Bull. 53 (1904), of U.S. Dept. of Agriculture
(Salmon).

DIAGNOSIS OF BOVINE TUBERCULOSIS 347

than normal over the upper half of the chest, and is particularly
marked during expiration. Usually the superficial glands in the throat,
those between the jaws, and under the ear or of the udder are swollen
and hard. The animal may continue for months in an apparently
healthy condition. When the disease is abdominal and the glands
and organs in the belly are chiefly affected, the symptoms of defective
nutrition are early evident, namely, emaciation, lessened milk secre-
tion, indigestion, breathlessness, and more or less rapid failure in
general health. In these cases the udder should be especially
examined.

It should be noted that clinical diagnosis of tuberculosis, especi-
ally of udder tuberculosis, does not enable us to judge whether
tubercle bacilli are secreted along with the milk of the cow in
question. The bacteriological test and tuberculin are necessary.
The former is sometimes tedious, and the latter remains at present
our only quick and sure method.

(b) Bacteriological Examination. — This method of diagnosis can
be applied at once in suspected udder disease by examination of the
milk ; or, as recommended by Nocard, a trocher may be used by
which a small fragment of tissue from the indurated portion of the
udder may be obtained for examination. Mucus or discharges from
throat, wounds, and ulcers may also be examined and assist in
diagnosis. The only sure method of bacteriological examination is
by inoculation of animals. Microscopical and cultural tests arc
unreliable.

(c) Tuberculin. — In recent years the method of testing herds for
tuberculosis by means of tuberculin has come into vogue, and it will
be necessary to refer briefly to this subject. The discovery by Koch,
in 1890, of the production of fever, indicated by a rise in tempera-
ture, in tuberculous animals into which he injected a sterilised
glycerine extract of pure cultures of tubercle bacilli, while it
produced no effect whatever when the animals were free from that
disease, furnished us with a simple but fairly reliable diagnostic
agent.

Tuberculin is a soluble product of cultures of tubercle bacilli, of
which a glycerine extract is made, which is sterilised by licit and
filtered through porcelain, so that it contains no living germs, and
therefore cannot produce tuberculosis in animals injected witli it.
It has, therefore, no effect on healthy animals ; in some cases the
disease is aggravated by it when it exists, but it cannot be produced
by it. The lymph must not be exposed to sunlight ; it must not be
frozen, and must be kept well corked to exclude air.

Koch's "old tuberculin" is made from glycerine-veal broth
cultures of B. tuberculosis by means of evaporation and precipitation
with alcohol. The liquid cultures arc thus concentrated to one-tenth

348 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

of their original bulk, and then passed through a Chamberland filter.
The brown and viscid filtrate is the tuberculin. Buchner and
Bonier pointed out that the proteins of other bacteria have a similar
effect upon tuberculosis, that is, cause a reaction with rise of
temperature. In 1897, Koch was able to improve his tuberculin,
and under the name "Tuberculin T. B." recommended a new
preparation. In point of fact, the new preparation takes three
forms, distinguished by the letters T. A. (alkaline tuberculin), T. 0.
(upper tuberculin, Germ, obcr), and T. E. (residual tuberculin). T. A.
is extracted from a young and virulent culture of B. tuberculosis by
means of a one-tenth normal solution of caustic soda, and the
solution is filtered. The reaction on inoculation is intense, and may
be accompanied with abscesses. Accordingly, its clinical use is
open to objection. T. 0. and T. E. are prepared by vigorously
pounding in a mortar dried cultures of the tubercle bacilli and then
adding distilled water. The emulsion is thoroughly centrifugalised.
The clear, opalescent fluid collecting at the upper part of the tube
contains no tubercle bacilli, and constitutes in the first centrifugal-
isation T. 0. The ddbris or residuum of tubercle bacilli remaining at
the bottom of the tube is used for the production of T. E. This
residue is dried, triturated with distilled water, and centrifugalised
repeatedly until hardly any residue remains. Twenty per cent, of
glycerine is then added to both preparations for purposes of pre-
servation. T. E alone is used clinically.

The method of use is as follows : The animals are kept a day or
two in their byres, and the temperature is taken to standardise the
normal, which is generally about 1022° F. The tuberculin is then
injected (30-40 centigrammes), and if the animal be tuberculous,
there is a rise in temperature of 1J° to 3°. The fever usually begins
between the twelfth and fifteenth hour after injection, and lasts
several hours. The more nearly the temperature approaches 104° F.,
the more reason is there to suspect tuberculosis (Bang). The dura-
tion and intensity of the reaction, however, has not a direct relation
to the number or gravity of the lesions, but the same dose in healthy
cattle causes no appreciable febrile reaction. The tuberculous calf
reacts just as well as the adult, but the dose used is generally
smaller.

Tuberculin injection has no bad effects on the secretion of milk,
either in quantity or in quality. The consensus of opinion of those
most experienced is that it does not lessen the secretion of milk in
dairy cattle, consequently they may be tested even when in full
milk without disturbing its secretion, unless it be during the few
hours of its absorption. It does not cause abortion in cows, or
sterility in bulls.

It is the quickest and most reliable method of diagnosis of

IN PIG AND SHEEP 349

bovine tuberculosis which we possess. Schiitz maintains that
failures in diagnosis by tuberculin injection only amount to 29 per
cent. Obviously, it may fail in animals which are highly tuberculous,
owing to the fact that their tissue already contains so much tuberculin
that they are unable to respond any longer to the tuberculin test.
If there is any lesion whatever in the udder, and there is a reaction
to tuberculin, the milk from that cow should not be used.*

Tuberculosis of Other Animals

Tuberculosis of the Pig is less common than that of cattle, but
not so rare as that of the calf (Nbcard). In nine out of ten cases
the pig is infected by ingestion, particularly when fed on the refuse
from dairies and cheese factories. The disease follows the same
course as in cattle, but generalisation is more common and more
rapid. Lesions of the abdominal organs occur in almost every case.
The glands, particularly those of the throat, are markedly affected.
The finding of the tubercle bacillus is difficult, and the only safe
test is inoculation. The massive lesions are often thinly scattered,
rich in giant cells, and containing few bacilli. The disease usually
assumes the acute or "galloping" form, and not infrequently emacia-
tion is absent and the dead pork meat possesses a healthy and fat
appearance. The internal organs and glands are the chief sites of
disease. The whole carcase should be condemned. In a pork carcase
seized by the writer in Finsbury in 1904, the retro-pharyngeal,
submaxillary, cervical, and mediastinal glands were enormously
enlarged and caseous, the deep glands of the body and the udder
and its glands were also affected, and the joints of the right fore
foot were ankylosed owing to tuberculous infiltration. The internal
organs in this case were a mass of tubercle.

Sheep are very rarely affected with tuberculosis, though there
is evidence which goes to prove that very long confinement in limited
space with tuberculous cattle might result in transmitting tuber-
culosis to sheep. One of the few cases on record in tin's country
was met with, by the writer, and has been described by Fuulerton.f
This was a half-bred, emaciated ewe having both lungs extensively
consolidated, and containing numerous tubercles which were also
present on the pleurae. The liver, spleen, both kidneys, and lymphatic
glands were affected. An emulsion, made from one of the affected
glands, inoculated into the guinea-pig, caused death from generalised
tuberculosis. The sheep, though rarely attacked, is not naturally
immune.

* For a discussion as to the practical use of the tuberculin test, see Trans. Brit.
Cong, on Tuberculosis, 1901, vol. ii., pp. 235-78 (Delepine).

I Transactions of the Pathological Society of London, 1902, vol. liii., pt. hi., p. 428.

350 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

Tuberculosis in the Horse is relatively very rare. It attacks
the organs of the abdominal cavity, especially the glands ; it affects
the lung secondarily as a rule. The cases are generally isolated ones,
even though the animal belongs to a stud. Nocard holds that the
bacillus obtained from the pulmonary variety is like the human
type, whilst the abdominal variety is more like the avian bacillus.

Dog1. — Nocard says : " If the dog can become tuberculous from
contact with man, the converse is equally true. Infection is at any
rate possible when a house-dog scatters on to the floor, carpet, or
bed, during its fit of coughing, virulent material, which is rendered
extremely dangerous by drying, especially for children, its habitual
playmates. The most elementary prudence would recommend the
banishment from a room of every dog which coughs frequently, even
though it only seems to be suffering from some common affection of
the bronchi or lung." *

Birds. — Tuberculosis is a common disease among birds of the
poultry-yard : poultry, pigeons, turkeys, pea-fowl, guinea-fowl, etc.
They are infected almost exclusively through the digestive tract,
generally by devouring infected secretions or organs of previous
tubercular fowls, and though very susceptible in this way, birds can
consume large quantities of phthisical sputum without becoming
tubercular. Whatever the position or form of avian tuberculosis, the
bacilli are present in enormous numbers, and are often much shorter
but sometimes longer than those met with in tuberculous mammalia,
and grow outside the body at a higher temperature (43° C). They are
said also to be more resistant and of quicker growth. The species
is probably identical with Koch's bacillus, though there are differences
(Maffucci). In the nodule, which is larger than in human tuber-
culosis, there are few or no giant cells, and it does not so readily
break down. Guinea-pigs and other animals are not so readily
infected with avian tubercle as mammalian. The writer has prepared
a number of histological specimens illustrating the comparative
pathology of tuberculosis, particularly in birds which have died of
the disease in the Zoological Gardens, London, including guan, quail,
ostrich, rhea, currasow, swan, cuckoo, vulture, goose, eagle, fowl,
pheasant, parrot, etc. In most cases the disease affects the organs
of the alimentary canal, especially the liver. The lungs are rarely
affected except secondarily. The disease frequently develops rapidly
like an acute infective disease, and the bacilli may often be found in the
tissues arranged in large colonies, as in leprosy in the human tissues.

The bacillus of avian tubercle differs from the organism of
tuberculosis in mammals in the appearance of the cultures and
the temperature conditions. Fischel, who worked under Huppe's
direction, says both micro-organisms are of one and the same kind

* Animal Tuberculosis, p. 129.

n.ATE 24.

s *

IN BIRDS AND COLD-BLOODED ANIMALS 351

as regards nutritive media. In consequence of the different physio-
logical nutritive media of the colder mammalian bodies on the one
hand, and the warmer avian bodies on the other, a distinction has
arisen between the two kinds. By artificial cultivation Fischel
succeeded in bringing about approximation in outer qualities between
the two bacilli ; he succeeded in getting the organism of tuberculosis
in mammals accustomed to a higher temperature, and in given
nutritive media he obtained a resemblance in the appearance of the
cultures, but as regards pathogenesis he could not transfer one to
the other. Fischel states that he was able, with the organism of
avian tuberculosis, to bring about a general tuberculosis in a guinea-
pig, but the cultures started out of the organs of this animal were
not identical with those of avian tuberculosis.

Tuberculosis in cold-blooded animals is in the same way to be
regarded as a modification of the tuberculosis in mammals. Bataillon
and Terre succeeded in cultivating tuberculosis in mammals and
birds by means of passing it through the body of a frog at room
temperature. Lubarsh was able to modify tuberculosis in mammals
by passage through the body of a frog, so that cultures taken from
the spleen of a frog grew at a temperature of as much as 28-30°.
Dubard also cultivated cultures taken from fish inoculated with
mammalian tuberculosis which also thrived at room temperature.
Moeller was able to produce cultures from the spleen of a slow-worm
inoculated with sputum containing tubercle bacilli, which flourished
at 20°; but which ceased to grow at a temperature of 30° and over.
The cultures resemble in appearance those of avian tuberculosis, grow
at room temperature, and are moist and thick.

It should be added that certain abnormal and tubercle-like
conditions have been met with in the carp, and from such conditions
a bacillus morphologically and tinctorially similar to the tubercle
bacillus has been isolated.

Most bacteriologists maintain that the B. tuberculosis of Koch is
the common denominator in all tubercular disease, whatever and
wherever its manifestations, in all animals. The bacillus, they
hold, may, however, experience profound modifications owing to
successive passages through the bodies of divers species of animals.
But if the modifications which it undergoes as a result of trans-
mission through birds, for example, are profound enough to make
the bacillus of avian tubercle a peculiar variety, though not a distinct
and separate species, of Koch's bacillus, they are not enough, it is
generally believed, to make these bacilli two distinct species. We
may, therefore, take it for granted that tuberculosis is one and the
same disease, generically, with various manifestations, common to
man and animals, intercommunicable, and having but one vera causa,
the B. tuberculosis of Koch.

352 TUBERCULOSIS AS A TYPE OF -BACTERIAL DISEASE

The Prevention of Tuberculosis

At the present time much attention is being directed to the
administrative and personal control of tuberculosis. How greatly
this is needed in so preventable a disease is evident from a perusal
of the following table from the Begistrar-General's reports : —

ENGLAND AND WALES
ANNUAL DEATH-RATES FROM ALL TUBERCULAR DISEASES

The following is a table of death-rates to a million living (England
and Wales), 1880-1901 {Reg. -Gen. Annual Reports) : —

1880

1881

1882

1883

1884

1885

1886

1887

1888

1889

1890

Tabes Mesenterica .

370

284

313

289

310

251

300

253

240

269

265

Tubercular Meningitis

330

276

264

262

264

253

257

236

239

234

240

Phthisis ....

1869

1825

1850

1880

1827

1770

1739

1615

1568

1573

1682

Other Forms

129

145

153

160

170

157

177

179

174

183

189

Total ....

2698

2530

2580

2591

2571

2431

2473

2283

2221

2259

2376

1891

1892

1893

1894

1895

1896
196

1897
201

1898

202

1899

198

1900
185

1901

188

Tabes Mesenterica

251

242

265

192

243

Tubercular Meningitis

247

227

226

211

222

210

213

213

203

198

183

Phthisis ....

1599

1468

1468

1385

1398

1307

1341

1317

1336

1333

1264

Other Forms

203

2300

199
2136

186
2145

185
1973

200
2063

179
1892

175
1930

184
1916

174
1911

186
1902

172

1807

Total ....

Tabes mesenterica is a term applied to tuberculosis of the alimentary canal and
mesenteric lymph glands.

Tubercular meningitis is the name of the same disease as it affects the mem-
branes of the brain (acute hydrocephalus).

Phthisis is the term applied to " consumption," or tubercle in the lungs.

These figures show a marked decline in the three worst forms of
the disease. But this decline is less marked in tabes mesenterica

PREVENTION OF TUBERCULOSIS 353

than in phthisis or tubercular meningitis, i.e. less in the kind of
tubercle located in the abdomen chiefly. Fortunately, the State is
beginning to realise its duty in regard to preventive measures. The
abolition of private slaughter-houses, the protection of meat and
milk supplies, the condemnation of tuberculous milch cows, and such-
like measures, fall obviously within the jurisdiction of the State
rather than the individual, and claim the earnest and urgent attention
of the public health departments of Government.*

Methods of Prevention. — A consideration of the various facts set
forth in this chapter will suggest the best means of the prevention of
consumption. These means depend upon broad principles of sanitation
and personal health rather than on bacteriological niceties or theore-
tical considerations. What is required has been stated briefly by
Sir Hermann Weber as: — (a) Purity and free circulation of air; (b)
sufficiency of good and pure food ; (c) well -constructed and ventilated
sunny rooms in houses situated on dry and pure soil; and (d) the
maintenance of the resisting power of the body and its different
organs.f These general desiderata are to be secured by practical
preventive methods, of which the following are some of the more
important : —

1. Personal hygiene and the maintenance of a high degree of
resistance in the human tissues is a matter which must rest with the
individual rather than the State, which can only exert its influence,
generally speaking, on the environment of the individual. Below will
be found a number of particulars as to the disease, stated in simple
form, and many of which bear a direct relation to the management
of personal conditions. The healthy life with sufficient food, exercise,
etc., is what is necessary to maintain healthy tissues. Mention may,
however, be made of the abuse of alcohol and the neglect of simple
illnesses as two most powerful factors in the creation of conditions in
the body favourable to tubercle infection.

2. Only second in importance is general sanitation and the
creation of a healthy environment for the individual and the com-
munity, and in this, and subsequent methods, much of the preventive
work in tuberculosis is centred. Such conditions as dust in the air,
overcrowding in houses, too great a density of houses on the area, ill-
ventilated and unclean rooms and workshops, etc., exert an indirect
influence of great force in the propagation of tuberculosis, and there-
fore the reduction and abatement of these conditions serves as a
means of prevention. The right enforcement of the Housing of the
Working Classes Acts, the Public Health Acts, the Building Acts,

* See the llarben Lectures, November, 1898, by Sir Richard Thorne Thorne,
Medical Officer to the Local Government Board ; also the Reports of the Royal Com-
missions on Tuberculosis, 1896, 1898, and 1903.

1 Tuberculosis, 1899, vol. i., p. 9.

354 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

and the Factory and Workshop Act are therefore matters of great
importance in the prevention of this disease.

3. Food has also been shown to be infected in a greater or less
degree with the virus of tuberculosis, and though the disease is not
spread so greatly through this channel as in other ways, it is never-
theless necessary to protect the public from tuberculous food, especially
meat and milk. The Public Health Acts (1875 and 1891) give
powers of seizure of diseased food, and the Dairy, Milkshops, and
Cowsheds Order of 1885, and its amendments, operate in the direction
of the control of the milk supply. These latter Orders should be
unified, and much more vigorously enforced than has been the case in
the past.

4. Lastly, there are certain measures of great importance which
concern the avoidance of infection from diseased persons. The con-
sumptive is the chief agent in the spread of consumption. Therefore,
anything which lessens the degree of his contagiousness is a means of
prevention. The first requirement is evidently knowledge of the
existence of cases of phthisis, and this may be obtained in various
ways, e.g., through hospitals or private practice, through poor-law
institutions, or by voluntary or compulsory notification. Voluntary
notification was first adopted in this country by the Local Authorities
in Brighton, Manchester, and Finsbury, and is now in vogue in many
districts. The results are not wholly satisfactory, but are better than
no information at all. Compulsory notification has recently been
instituted for an experimental period in Sheffield.* The cases of
phthisis being known, the next steps are supervision, disinfection of
sputum, house, and clothes, and, if jDracticable, isolation and treatment
of suitable cases. Sanatoria act partly as therajjeutic agencies, partly
as prophylactic agencies. Much is now being done in civilised
countries in these directions, and many sanitary authorities carry out
disinfection regularly, and make various efforts to prevent consump-
tives infecting their neighbours or fellow- workmen.

Still, after all, the prevention of phthisis is in no small degree a
matter of personal hygiene and precautions to be exercised by the
people themselves.

Hence we hail with satisfaction the recent endeavours to educate
public opinion. In order to simplify this matter, we have placed
in a footnote a series of statements embodying some of the chief
facts which every individual in our crowded communities should
know.f

* Sheffield Corporation Act, 1903, sect. 45.

t 1. Tuberculosis is a disease mainly affecting the lungs (consumption, decline,
phthisis) of young adults and the bowels of infants (tabes mesenterica). It may
affect any part of the body, and its manifestations are very various. It also affects
animals, particularly cattle, by whom it may be transmitted to man.

2. Its direct cause is a microscopic vegetable cell, known as the B. tuber-

PREVENTION OF TUBERCULOSIS 355

Pseudo-tuberculosis

In 1899 the Pathological Society of London urged that this term
should he discarded. It is used here not as concerned with diseases

culosis, discovered by Koch in 1882. This fungus requires to be magnified some
hundreds of times before it can even be seen. When it gains entrance to the
weakened body it sets up the disease, which is an infectious, or sub-infectious,
disease, though different in degree to the infectiousness of, say, measles.

3. Trade influence and occupation, in some cases, undoubtedly predisposes the
individual to tubercle. Cramped attitudes, exposure to dampness or cold, ill
ventilation, and exposure to inhalation of dust of various kinds, all act in this way.
In support of the evil effect of each of these three conditions much evidence could
be produced.

4. Overcrowding has a definite influence in propagating tubercular diseases.
The agricultural counties without large towns, like Worcestershire, Herefordshire,
Buckinghamshire, and Rutland, are the counties having the lowest mortality from
tuberculosis ; whilst the crowded populations in Northumberland, South Wales,
Lancashire, London, and the West Riding suffer most. Speaking more particularly,
the overcrowded areas of London, such as South wark, Shoreditch, Finsbury, Hol-
born, and Central London generally, show very high tubercular death-rates.

5. Tuberculosis is not increasing. — During the last thirty years it has shown, with
few exceptions, a steady decline in all parts of England. " Consumption " is most
fatal in comparatively young people (fifteen to forty-five years), whilst " tabes "
and other forms of tubercle are fatal chiefly to young children. These forms have
not declined so much as the lung form. The mortality in consumption of males has
since 1866 been in excess of that of females. The age of maximum fatality from
consumption is later than in the past, which is probably due to improved hygiene
and treatment.

6. This decline has been due not to any special repressive measures — for few have
been carried out — but to a general and extensive social improvement in the life of
the people, to an increase of knowledge respecting tuberculosis and hygiene, to an
enormous advance in sanitation, and to more efficient land drainage.

7. Not all persons are equally liable to consumption, some being much more
susceptible than others. We have mentioned the predisposing influence of certain
trades. There is also heredity, which acts, as we have said, in transmitting a
tubercular tendericy, rarely, if ever, the actual virus of the disease ; there is, thirdly,
the debilitating effect of previous illness or chronic alcoholism; there is, fourthly,
the habitual breathing of stagnant, polluted air ; and, fifthly, there are the condi-
tions of the environment, such as dampness and darkness of the dwelling. Such
influences as these weaken the resisting power of the tissues, and thus afford a
suitable nidus for the bacillus conveyed in milk, or by the inspiration of infected
dust or mucus.

8. Consumption may be arrested if taken in time. In cases where the lungs are
half gone, and consist of large cavities, it is obvious that curability is out of the
question. But if the disease can be properly treated in its earliest stages, there is
considerable likelihood of recovery, or at least of arrest of the disease.

9. The breath is not dangerous, as far as we know, but there is danger from
discharges of any kind from any infected parr, whether lungs or bowels ; for such
discharges, when dry, may readily pollute the air, and either the bacilli or spores be
inhaled into the lungs. The breath itself in coughing, shouting, etc., may emit a
fine spray of mucus or saliva, in which the bacilli may be suspended, and thus
convey infection.

10. The chief channels of -personal infection in the spread of the disease amongst a
community are two: (a) dried tubercular sputum or "cough spray" (or other
tubercular discharges); (6) infected milk or meat. As for milk and meat, boiling
the former and thoroughly cooking the latter will remove all danger. In any case,
there is evidence to show that infection from milk or meat is nothing like so
common as infection from sputum or mucous particles from a consumptive patient.

356 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

simulating tuberculosis (protozoal infections, parasitical disease, etc.),
but to designate a pathological condition set up by a special group

11. The expectoration is infective. — This is one of the commonest modes of
infection, and to it is held to be due the large amount of respiratory tuberculosis
(consumption, phthisis). The expectoration from the lungs must contain, from the
nature of the case, a very large number of bacilli. As a matter of fact, a single
consumptive individual can cough up in a day millions of tubercle bacilli. When
expectoration becomes dry, the least current of air will disseminate the infective
dust, which can by that means be readily reinspired. Infective saliva on pave-
ments and floors, as well as on handkerchiefs, or even in books, may thus become a
source of danger to others. The discharges from the bowels of infants suffering
from the disease also contain the infective material.

12. Milk, though a much more likely channel for conveyance of tubercle than
meat, is only or chiefly virulent when the udder is the seat of tuberculous lesions.
The consumption of such milk is only dangerous when it contains a great number of
bacilli and is ingested in considerable quantity. Practically, the danger from using
raw milk only exists for those persons who use it as their sole or principal food, e.g. ,
young children. All danger is avoided by boiling or pasteurising the milk.

At the same time there is an increasing amount of evidence forthcoming at the
present time which goes to prove that milk is not infrequently tainted with tubercle
(see pp. 204-206). The tuberculin test should be applied to all milch cows, and the
infected ones isolated from the herd. They need not necessarily be slaughtered.
Milk supplies should be more strictly inspected.

13. There are several methods by which meat infection can be prevented. In the
first place, herds should be kept healthy, and tubercular animals isolated. Cowsheds
and byres should be under sanitary supervision, especially as regards overcrowding,
dampness, lack of light, and uncleanliness. Public slaughter-houses under a
Sanitary Authority would undoubtedly be advantageous. Meat inspection should
also be more strictly attended to; efficient cooking, and avoidance of "roll" meat
which has not been thoroughly cooked in the middle, are also wise measures to adopt.

14. Consumptive patients may diminish their disease. — Dr Arthur Ransome" has
laid down five axioms of hygiene for phthisical patients which, if followed, would
materially improve the condition of such persons. At Davos, St Moritz, Nordrach,
Nordrach-on-Mendip, and many other places where they have been practised, the
beneficial change has been in many cases extraordinary.

(1) Abundance of light, nutritious, easily digested food, which must comprise

a large allowance of fat ; small meals, but frequent.

(2) An almost entirely open-air life, with as much sunshine as can be

obtained.

(3) Suitable clothing, mostly wool.

(4) Cleanliness, and bracing, cold-water treatment.

(5) Mild but regular exercise.

15. Consumptive patients may also assist in preventing the spread of the disease. — In
the first place, they should follow the hygienic directions just mentioned, because the
fulfilment of such conditions will materially lessen the contagiousness of the disease.
Next, the expectoration must never be allowed to get dry. A spitting-cup containing
a little disinfectant solution (one teaspoonful of strong carbolic acid to two table-
spoonfuls of water) should always be used, or the expectoration received into paper
handkerchiefs which can be burnt. Spoons, forks, cups, and all such articles should
be thoroughly cleaned before being used by other persons. The patient should not
sleep in company with another, and should occupy, if possible, a separate bedroom.
Isolation hospitals for consumptives, as for patients suffering from the ordinary
infectious diseases, are now being established.

16. House influence has some effect, both directly and indirectly, upon tubercular
diseases. Damp soils, darkness, and small cubic space in the dwelling-house exert a
very prejudicial effect upon tubercular patients. Sir Richard Thorne Throne t has

* Arthur Ransome, M.D., F.R.S., Treatment of Phthisis.
t Practitioner, vol. xlvi.

PREVENTION OF TUBERCULOSIS 357

of bacilli of which the chief is the B. pseudo-tuberculosis of Pfeiffer.*
Other workers have described very similar organisms.

described the favourable house for such persons as one built upon a soil which is dry
naturally or freed by artificial means from the injurious influences of dampness and
of the fluctuations of the ground water. The house itself should be so constructed as
to be protected against dampness of site, foundations, and walls. Upon at least two
opposite sides of the dwelling-house there should be enough open space to secure
ample movement of air about it, and free exposure to sunlight. Lastly, it should be
possible to have free movement of air by day and night through all habitable rooms
of the house. It is clear many inhabited houses could not stand these tests ; but
effort should be made to approach as near to such a standard as possible.

17. Tubercle-infected Houses. — Many authorities have demonstrated the fact that
dust in houses may contain the tubercle bacillus, and that thus, presumably, persons
may become infected. In 1904, Klein found living tubercle bacilli in the sweepings
of the floors of public-houses, and some fifteen years ago Cornet published the
result of his investigations into the infectivity of the dust found in the dwellings
of consumptives in Berlin, and some work of a similar character has been done in
England by Coates. *

These investigations consisted of bacteriological examinations of dust collected in
houses of three types : —

(a) Dirty houses in which a consumptive patient is living who takes no precau-
tions to dispose of his expectoration, but spits freely upon the floor and into his
pocket handkerchief. In 66 "6 per cent, of these houses virulent tubercle bacilli were
found showing the large amount of dangerous infective material present in an in-
fected house.

(b) Clean houses in which a patient is living who is not sufficiently careful as to
the disposal of his sputa. In 50 per cent, of these instances the bacillus was found.
It is evident that ordinary household cleanliness alone is insufficient to prevent the
accumulation of infective material in rooms occupied by a consumptive.

(c) Very dirty houses in which there had been no case of consumption for some
years. In this class of house no tubercle bacilli were present, showing that virulent
dust found in classes 1 and 2 must have been due to the presence of the consumptive
patient.

Taking the first two classes, the average of houses infected was 61 per cent.
Cornet's similar work resulted in finding 71 per cent, tubercle infected.

It was ascertained that the dust nearer the floor than the ceiling possessed the
greatest virulency. It was also shown that the infective dust was most virulent in
cases where the access of sunlight and free circulation of air was prevented, while,
conversely, the beneficial effect of light and air was demonstrated even in the dirtiest
houses. Instances were given of the dangers attaching to infected rooms, and the
risk to healthy occupants and their successors. According to Koch, " it is the over-
crowded dwellings of the poor that we have to regard as the real breeding-places of
tuberculosis."

18. Sunlight and fresh air are the greatest enemies to infection.

19. Disinfection «s necessary after death from phthisis, and should be as complete
as after any other infective disease. Compulsory notification of fatal cases and
compulsory disinfection have been officially ordered by the Prussian Government.
In this country, also, absolute disinfection should always be insisted upon after
phthisis. Walls, floors, carpets, curtains, etc., should be strictly disinfected.
Spraying with 1-100 solution of chloride of lime, or other similar disinfectant, is
the best method (see p. 444).

* The pathology and etiology of pseudo-tuberculosis is fully treated of by
Klein in the Supplement to the Twenty-Ninth Annual Report of the Local Government
Board, 1899-1900, pp. 355-384. See'also Annates de Vlnstitut Pasteur, 1894, No. 4,
and Jour, of Path, and Bad., 1898, pp. 160-181 (Muir).

Trans. Brit. Cony, on Tuberculosis, 1901, vol. ii., pp. 88-101.

358 TUBERCULOSIS AS A TYPE OE BACTERIAL DISEASE

The B. pseudo-tuberculosis (Pfeiffer) resembles B. coli, and occurs as short, small
bacilli, cylindrical, and with round ends. Its manner of grouping is singly, or in
couples or chains ; sometimes filamentous forms and long chains occur in
bouillon culture. It stains by alkaline Loffler's methyl-blue, and also by Gram's
method (Klein). It is non-motile. In bouillon, in twenty-four hours a well-marked
granular cloudiness appears, and small flocculi float through the liquid. Imperfect
pellicle after several days' growth. No general turbidity. On gelatine the growth
resembles B. coli, but the colonies are more circumscribed and granular, and later,
they become tuberculated. Growth is slow, and the colonies become more opaque,
whiter, and less spread out than B. coli. No gas is formed in gelatine shake cultures.
There is no liquefaction of gelatine. On agar minute grey-white flat colonies appear.
Stroke and stab cultures are similar to B. coli, but not so luxuriant. There is
limited growth on potato, which forms a thin layer with crenated thicker margin of
a whitish-yellow colour. It grows well in milk, but leaves it unaltered. Patho-
genesis— Guinea-pigs inoculated subcutaneously with a small quantity of culture die
in a few weeks. Their organs are found to be studded with yellow-white nodules
containing the bacillus in pure culture. These nodules develop more rapidly than
true tuberculosis, but do not contain any giant cells. If fed with food contaminated
with this organism, similar nodules develop in the walls of the intestine and mesenteric
glands. Klein believes that : " The presence of the B. pseudo-tuberculosis in milk may
probably play a part in causing pseudo-tuberculous disease in the human subject."

Klein found this bacillus present in 2 out of 5 samples of London
milk,* and in 8 out of 100 samples of country milk delivered in
London. f Delepine found that out of 450 samples of milk, lesions
produced by pseudo-tubercle bacilli were met with four times. It
seems not unlikely that this group of bacilli includes several varieties
bearing a close general resemblance to each other, but possessing
slightly different properties. They gain access to milk in all
probability by some accidental contamination. The milk itself
remains unaltered in appearance, though it becomes alkaline. As
regards differential diagnosis, it may be said that the pseudo-tubercle
bacillus is not acid-fast, nor is it similar to B. tuberculosis in morpho-
logical or cultural characters. The pathological changes set up by
it, and which form its chief claim to be considered as in any way
related to tuberculosis, differ from that disease in showing an absence
of giant cells in the nodules, absence of the true tubercle bacilli,
copious presence of the pseudo-tubercle bacilli, and a more rapid
development of disease.

Acid-fast Bacilli allied to the Tubercle Bacillus

We may here suitably consider the group of organisms
morphologically and tinctorially similar to the bacillus tuberculosis.
This group is known as that of the acid-fast bacilli, on account of
the fact that in staining by the Ziehl-Neelsen method (see p. 459)
these organisms possess, like the tubercle bacillus, the power of

* Report of Local Government Board, 1899-1900, p. 360, and 1900-1901, p. 332.
t Jour, of Hygiene, 1901, vol. i., p. 83.

ACID-FAST ORGANISMS 359

resisting decolorisation by the acid following the red stain.* In
England such bacilli are termed acid-fast, in Germany saurefeste,
and in France acidophils The group is one of great importance,
partly on account of the ease with which its members may be
mistaken for the "true" tubercle bacillus, and partly on account of
the relationship which appears to exist between them and the
tubercle bacillus. Some bacteriologists hold that possibly these
acid-fast bacilli represent a saprophytic stage in the life-history of
the true tubercle bacillus.

In his description of the tubercle bacillus, Koch foretold the
probability of other acid-fast organisms being discovered, and some
fourteen years after, in 1896, Koch and Petri actually demonstrated
the occurrence of such bacilli in the butter and milk of Berlin. In
the years immediately following, Eabinowitsch, Korn, Coggi, Tobler,
and others, found further organisms of this nature in such articles
of food. In 1898 Moeller showed that these acid-fast bacilli
occurred naturally in animals and plants. Dust, grass, hay, manure,
and similar substances yielded them, and now it is known that a
considerable family of these bacteria exists. It should, however, be
understood that the group is provisional only. Further knowledge
may reveal facts which would considerably modify present views.

Classification of Acid - fast Bacilli.— These bacilli may be
divided provisionally and for convenience into four chief sub-
divisions : —

(a) The acid-fast bacilli of other diseases or conditions affecting
man (e.g., B. leprae, B. smegmatis, B. of syphilis ai Lustgarten, etc.).
Other non-tubercular acid-fast bacilli have been found in lung
gangrene (Frankel), in the nasal cavities (Karlinski), in excreta,
and in certain chronic eye diseases, etc.

(b) The acid-fast bacilli occurring in other animals (e.g., B.
tuberculosis avium of Maffucci ; the B. tubert iclosis piscium of Dubard,
Bataillon, Terre ; B. tuberculosis rani cola of Lubarsch; B. tuberculosis
anguicola of Moeller, etc.).

(c) The acid-fast bacteria of butter and milk (e.g., B. laticola
planus, perrugosum, Friburgense, etc.), of Petri, Moeller, Eabinowitsch,
Binot, Markl, Coggi, Tobler (Nos. i.-v.), Grassberger, and Korn (Nos.
i. and ii).

(d) The acid-fast bacilli of grass (e.g., B. plilei or Timothy bacillus,
and Grass bacillus, No. ii., of Moeller), the "manure bacillus" of
Moeller, the urine bacillus of Marpmann.

All these various organisms are morphologically and in staining

* Acid-fastness is due, in all probability, not to fat in the bacillus, but to a
substance of the nature of wax, which can be extracted by acid-alcohol, ether, or
other wax solvents. For a discussion of this subject, see Trans. Jh-ilisJi Congress of
Tuberculosis, 1901, vol. iii., pp. 498-502 (Bulloch).'

360 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

properties allied to the B. tuberculosis. The chief characters of most
of them are referred to under Tuberculosis, Leprosy, etc., but it will
be necessary to describe briefly the character of others.

(a) Acid-fast Bacilli of Human Origin.— The leprosy bacillus will
be described subsequently (see p. 398).

The smegma bacillus was first discovered by Tavel and Alvarez,
in 1885, in the normal preputial smegma, and also in the secretion of
the outer skin, particularly where a collection of epithelium may
occur, as in the fold of the groin, between the toes, etc. The
discovery of the bacillus was incidentally made in investigating an
observation of Lustgarten, in 1884, on the syphilis bacillus.
Morphologically and tinctorially, the smegma bacillus closely
resembles the syphilis bacillus of Lustgarten. This fact discounted
the importance attached to Lustgarten's discovery, and subsequent
investigations show that Lustgarten's bacillus has not been found in
sufficient numbers, or with sufficient constancy, in the syphilitic tissue.

The smegma bacillus, according to Tavel and Alvarez, is morpho-
logically exceedingly like the tubercle bacillus, and can be stained
by the same methods. Inoculation experiments on animals were
without result, nor were the authors able to obtain a pure culture.
Laser and Czaplewski have cultivated (the former from the secretion of
syphilitic affections, the latter from gonorrheal pus) micro-organisms
resistant to acids, similar to diphtheria bacilli, which have
been declared by both authors to be identical with the smegma
bacillus. Frankel only calls those micro-organisms smegma bacillus
which first attracted attention by their great resemblance to tubercle
bacilli. This resemblance is wanting in the cultures of Czaplewski
and Laser, and in his own cultures, which he described later. In
form and other characters they are much more like a pseudo-
diphtheritic bacillus. Moeller agrees with Frankel. Moeller was
not able to get a pathogenic effect in guinea-pigs either with the
diphtheroid bacillus cultivated in pure culture from smegma, or with
the genuine smegma bacillus containing cutaneous secretion in
abundance. In this way the smegma bacillus differs from other
acid-fast bacilli. He found human serum the best culture medium
for smegma bacillus. The morphology differs according to media,
especially in milk cultures. Moeller found the bacillus to be
absolutely acid-fast and alcohol-fast, and this property is not much
diminished by age or media. The organism is strongly aerobic, and
grows slowly. On glycerine agar at 37° C. dull grey-white scales
of growth occur, and on potato dull white-grey colonies. Growth is
rapid in milk, and the milk is not coagulated.

With respect to differential diagnosis, especially in reference to
urogenital tuberculosis, the smegma bacillus is undoubtedly of great
importance.

PLATE 25.

Acid-fast bacillus from butter of Berlin

(Petri-Rabinowitsch).

Flask culture on glycerine-agar — .3 months at 22° C.

Single colony — actual size.

Acid-fast bacillus from milk of Belzig

(A. Moeller).

Flask culture on glycerine-agar — 3 months at 22° C.

Single colony — actual size.

To face page 360.

yixbH )

ACID-FAST BACILLI IN BUTTER AND MILK 361

Neufeld and others have obtained acid-fast organisms from
smegma, and Neufeld concludes that some of these may be described
as similar to B. diphtheria and others to the tubercle bacillus.

(b) Acid-fast Tubercle Bacilli in Animals.— The members of
this group are provisionally assumed to be related to the tubercle
bacillus. Eeference is made to these organisms under Tuberculosis
(see p. 349).

(c) Acid-fast Bacilli in Butter and Milk. — Several years after
Koch's discovery of the tubercle bacillus, species of bacteria were
found possessing acid-fast properties, but it was not until 1896 that
Koch and Petri isolated such organisms from the milk and butter of
Berlin, and in the following year Lydia Eabinowitsch carried out her
research on the same subject. In 1899, Korn discovered two bacilli
in the butter of Freiburg, Binot a bacillus in the butter of Paris,
and Coggi a bacillus in the butter of Milan, all four of which were
acid-fast. In the same year G-rassberger published a statement upon
acid-fast organisms occurring in butter and margarine. In 1900,
Beck and Santori met with similar organisms in milk; and in 1901,
Maria Tobler, Markl, and Moeller and Jong isolated acid-fast bacilli
from both butter and milk. All these organisms were bacilli, but
they showed much variation in form and polymorphism, sonic
appearing to be like B. diphtheria, and others like actinomyces.
The staining properties were, in all cases, those of the true bacillus
of Koch, except that the power of resistance to decoloration by acid
was rather less.* Swithinbank has cultured many of these organ-
isms upon different media, and has found them to show various
modifications in form, chromogenicity, vitality, polymorphism, etc.
By a series of cultures, he showed the various characters of ten of
these acid-fast species compared with the human and bovine tubercle
bacillus, all the cultures having been grown on the same media, for
approximately the same length of time, under precisely the same con-
ditions. The cultures were in each case composed of one colony only,
and admirably revealed the differences between the species. These
acid-fast bacilli live and develop on all ordinary media at room tem-
perature and blood-heat, preferably under aerobic conditions. They
do not form indol or liquefy gelatine, nor do they possess much patho-
genic action.-)-

The Butter Bacillus of Petri- ■Eabinowitsch. — Morphologically, this
organism is like the ordinary tubercle bacillus, though somewhat
shorter and thicker. It stains in the same manner, bul grows readily
at room temperature and rapidly at blood-heat. It is non-motile.

Report of Medical Officer of Local Government Board, 1900-1, pp. 331-3.
I Sec Bacteriology of Milk (Swithinbank and Newman) ; in The Journal of State
Medicine, 1903-4, will be found a useful summary of present knowledge of arid-fast
bacteria, by A. C. Coles ; Potefs work should also be consulted.

362 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

The cultures appear as moist, thick, creamy, wrinkled layers of growth
on the surface of the medium. The bacillus renders broth turbid and
acid. Indol is also produced, and a disagreeable odour. The growth
is slow on gelatine, occurring as small granular colonies. It is non-
liquefying. On glycerine agar, growth is abundant, rapid, and char-
acteristic. It occurs as a creamy film — of a light golden colour —
moist, thick, and much wrinkled (see Plate). It possesses a
glistening appearance, and an unpleasant odour These characters
disappear after much sub-culturing. Milk is not coagulated. A dull
dry growth generally occurs on potato. The organism possesses less
virulence when inoculated in pure culture. But when inoculated
with, or without, butter, it has clearly defined effects. Giant cells,
nests of epithelioid cells, and typical tuberculous caseation are,
according to Eabinowitsch, never to be found in the foci of disease
set up by this bacillus. None of the animals injected with this
bacillus reacted to tuberculin. The intra-peritoneal injection of pure
cultures often produces a formation of nodules in the abdominal
organs which frequently heal. If, however, the animals are killed in
three or four weeks, the following characteristics are found, namely,
slightly distended abdomen, more or less severe peritonitis, nodules
on mesentery and beneath the intestinal serosa, mesenteric glands
enlarged, and liver, spleen, and kidneys showing small nodules with
yellowish exudation. When the butter itself containing the organisms
is used, a fatal result often follows the injection after three to fifteen
days. Similar changes to the above have occurred, but of a more
intense degree. Eabinowitsch found rabbits insusceptible in contrast
to guinea-pigs. It is not known whether this bacillus is in any
degree pathogenic for man. But probably such is not the case. It
appears to be widely distributed in nature, as 60 per cent, of butter
samples in Berlin were found to contain it. The only satisfactory
way to differentiate this bacillus from the tubercle bacillus is by
inoculation of animals.

Moeller isolated a somewhat similar organism from milk,
which is generally known as Mocllcvs Milch Bacillus. It was found
in pasteurised milk at Belzig, and is almost identical in morphology
to the tubercle bacillus. It is acid-fast, non-motile, and grows at
room temperature as well as blood-heat. Broth becomes but little
turbid, and there is no deposit. Surface membrane of fatty aspect
and amber colour, adherent to tube walls, is sometimes formed. On
gelatine plates and tubes a white wrinkled culture of creamy nature
occurs, and on glycerine agar, after about three weeks, the growth is
white, uniform, and of a creamy nature, though at times slightly
wrinkled, and dry. In old cultures it is dry, or glazed, and of a
yellowish colour, which later turns to a reddish tint, the culture
itself becoming of a wrinkled appearance. Frequently the raised

PLATE 26.

Acid-fast bacillus from butter of Friburg

(Bacillus Friburgensis — Korn 1.)

Flask culture on glycerine-agar — 3 months at 22° C.

Single colony — actual size.

Acid-fast bacillus from butter of Friburg

(Korn II.)

Flask culture on glycerine-agar — :> months at 22° C.

Single colony — actual size.

To face paye 362-

"/JCJ

'

ACID-FAST BACILLI IN BUTTER AND MILK 363

"granular" centre is surrounded by a light blue zone, but slightly
raised above the medium (see Plate); on glycerine potato, there is
at first a white creamy growth very little raised above the surface of
the medium, but as it grows older the culture becomes wrinkled and
of a deep yellowish tint, almost red. The bacillus grows quickly and
luxuriantly in milk, forming an ochre-yellow ring round the surface
edge of the medium. The organism was found to produce nodules in
the organs of inoculated animals.

Korn isolated two acid-fast bacilli from Friburg butter. B. fri-
burgensis, JYo. 1, varies in morphology under different circumstances.
In preparations made from the organs of an inoculated animal, the
bacilli resemble in shape and size the B. tuberculosis of Koch. In
bouillon, the shape very much resembles the B. colt, but is a little
longer and slightly curved. On agar the bacilli are slightly thinner.
In old cultures upon agar and serum, they assume the aspect of
" Coccothrix." Upon potato they appear under form of cocci, diplo-
cocci, and specially of short, stout bacilli slightly curved. Upon
cooked beetroot at the end of three or four days the organisms
resemble staphylococci. They are shorter when grown at ordinary
temperature than when grown at 37° C. In culture media
the older growths generally exhibit some orange or red coloration,
though on almost all media the growth is at first white and
non-wrinkled. Upon glycerine-agar a thick, white, brilliant growth
occurs with deposit in the water of condensation, which is soon
covered with a veil adherent to the walls of the tube. Later
on the growth becomes slightly folded. At the age of about three
weeks the culture is creamy, presenting a few, light folds, and
of yellow-orange colour. In older cultures the growth is very
abundant, raised considerably above the surface, and irregularly
folded and convoluted, the colour varying in depth from light to
dark orange or even red brick (see Plate). Milk is not coagu-
lated, but at the end of three weeks possesses a yellowish-brow n
colour. Subcutaneous or intraperitoneal injection of pure cultures
produce only an abscess at site. If injected with butter itself in
white mice, granulations are produced in thoracic and abdominal
viscera, showing no giant cells, but commencing caseation.

B. friburg ensis, No. 2, consists of a small rod two or three times
longer than broad, often irregular and sometimes clubbed. It is
much less sensitive to decolorisation by acid than No. 1; grows
feebly on ordinary laboratory media, with the exception of glycerine
agar, in which growth appears in twenty-four hours, and eventually
becomes abundant. The culture is creamy, glistening, and a
yellowish colour (see Plate). Milk turns of a dirty red colour in
about three weeks time.

Markl, Tobler, Coggi, P>inot, and Grassberger have also isolated

364 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

acid-fast bacilli from butter. Brief reference may be made to the
organisms discovered by the two last-named workers.

Binot's Butter Bacillus is in morphological and tinctorial char-
acters similar to the tubercle bacillus. It differs in cultural features.
Binot describes it as producing a thick, viscous, creamy layer in
broth, from which " stalactites " grow down into the liquid. It grows
on gelatine (which it does not liquefy), producing thin creamy irregu-
lar surface colonies. On glycerine-agar plates and tubes white colonies
appear, becoming straw-coloured, and finally orange. They may
attain diameter of two-franc piece or larger; and have a bright
glistening surface, and are very adherent to medium. On the surface
they soon become wrinkled and irregular, and have scalloped edges.
Chromogenic characters are more marked if the growth is exposed to
air and light (see Plate). On potato, a scanty growth occurs, which
is at first moist, and of a clear yellow colour. On glycerine potato
an abundant homogeneous growth occurs, of an opaque yellow colour,
turning to orange. Irregular nodosities appear on the colony. The
bacillus produces tuberculous-like changes in animals in which it is
inoculated.

The Butter Bacillus of Grassbcrgcr is very similar to the other
members of this group, except that it produces, especially on gelatine,
but also on glycerine agar, a dry, much-wrinkled growth not unlike
some forms of mould, and of a deep rose colour. There is no liquefac-
tion. The wrinkles in the large colonies appear as characteristic
markings (see Plate). Milk is coloured throughout by the organism,
but not coagulated. There is also a surface growth and deposit, both
rose coloured. On potato the growth is similar to tubercle bacillus,
but of a deep rose tint.

(d) Acid-fast Bacilli in Grass, Hay, and Manure.— This second
group of acid-fast bacilli associated with milk (marked (d) in the
classification above) is often designated as that of the grass bacilli.
They were first cultured on Timothy grass (Phleum pratense), which
is much valued for feeding cattle.* Since then, however, this grass
bacillus has been found in various places, and it or its allies have
been isolated from cattle fodder, hay, hay-dust, manure, milk and
its derivatives. Morphologically, this bacillus (B. phlci) is similar to
the tubercle bacillus, slender and slightly curved. It contains
highly -stained granules and oval, clear spaces ; often grows in threads ;
and is branched, and sometimes has clubbed swellings at one end.
It is acid-fast in staining, and grows best at incubation temperatures
on the ordinary culture media. The colonies become visible in thirty-

* Cat's tail or Timothy grass (Phleum pratense). Although well known to the
British grower this grass is more extensively cultivated in the United States, where
it was introduced from Britain, nearly a century ago, by Mr Timothy Hanson, after
whom it is named Timothy grass.

PLATE 27.

Acid-fast bacillus from butter of Paris

(Binot).

Flask culture on glycerine-agar — 3 months at 22° C.

Single colony — actual size.

Acid-fast bacillus from butter of Vienna

(Grassbergcr).

Flask culture on glycerine-agar — S months at 22r C,

Single colony -actual size.

To face page 864

'

tS 3TAJS

) (j(J it; anfaoi iw-^wnyyn^ no -j-iwllh

i in C md Manur >

mmi i<

•('• u))

iditiom

•

ACID FAST GRASS BACILLI 365

six bums, are scale-like and greyish-white or yellow in colour. It
grows best on glycerine agar. It grows slowly in milk, but does not
coagulate it. Under certain conditions its growth in artificial media
is very similar to the tubercle bacillus, which, however, does not thrive
at room temperature. As regards pathogenic properties, the grass
bacillus is almost identical with the Petri-Eabinowitsch butter
bacillus in its effects on guinea-pigs. It has somewhat different effect
on rabbits, producing a condition difficult to distinguish from true
tuberculosis (Lubarsch). Giant cells, epithelioid cells, and caseation
are all said to occur. In all animals injected with the grass bacillus
a negative reaction to tuberculin is obtained. Moeller has isolated a
grass bacillus, No. 2 (from the dust of a hay-loft), which he considers
essentially different from the Timothy bacillus. The colonies are
moist and sticky, become confluent, and are yellow in colour. It
loses its acid-fast properties in old cultures. Its pathogenic pro-
perties are most marked when cultured in milk. It frequently
shows marked polymorphism. In culture it is like the butter bacilli.
Freymuth has shown that the changes this organism sets up in cold-
blooded animals are indistinguishable from true tuberculosis. An
acid-fast bacillus similar to grass bacillus, No. 2, has recently been
isolated by Moeller from milk. A variety of the grass bacillus has
also been found by Moeller in the excreta of animals, and is therefore
termed the manure bacillus (mist bacillus). This acid-fast bacillus
has been isolated from the excreta of cattle and other animals, and
bears a morphological and tinctorial resemblance to the Timothy
bacillus, whilst in cultures it is like grass bacillus, No. 2. On agar
these organisms grow in a similar manner, but bouillon does not
become turbid with the growth of the mist bacillus. It has certain
pathogenic properties. When injected into guinea-pigs, nodules
resulted; but they contained few epithelioid cells. The true radial
arrangement of the bacilli occurs, however. It is possible that most of
the acid-fast bacilli found in milk and butter have their origin in the
soil or vegetal ion.

Differential Diagnosis.— This brief record of the acid-fast bacilli
is enough to Bhow thai there exisl a large number of bacilli, which
on occasion may be present in milk or milk products, having char-
acters which ally them closely to the tubercle bacillus. Moeller
holds that the primitive form is the grass bacillus, and that the
butter bacillus, manure bacillus, etc., are varieties thereof. The
main points of distinction between this group and the true tubercle
bacillus are live.

First, the tubercle bacillus shows a fairly uniform manner of
growth.

Secondly, it requires intubation temperature.

Thirdly, it is unique with respect to its excessively slow growth.

366 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

Fourthly, it is as regards growth and propagation a parasite.

Fifthly, on inoculation it produces pathological cellular changes
distinct from the nodular new growths following inoculation of
acid-fast bacilli. In particular this is true, as far as is known at
present, in regard to the human organism.

In a sentence, the acid-fast bacilli differ from the tubercle bacillus
in three main particulars, viz. : morphology of culture, conditions of
development (chromogenicity, rapidity of growth, range of tempera-
ture within which they flourish), and their feebler pathogenic proper-
ties. From these facts it follows that however great the degree of
similarity between these various acid-fast bacilli, and however much
it is possible by artificial cultivation to modify the morphology of
the various forms, there is sufficient difference to enable a differ-
ential diagnosis to be made if all the biological characters are
ascertained, and most of all the pathogenic properties. Hence the
importance of the inoculation test being applied to acid-fast and
tubercle-like organisms detected in milk or butter.

As a simple method of differential diagnosis, Moeller suggests that
the smegma, sputum, or other secretion should be mixed with nutritive
bouillon, and kept at about 30° C. If in two or three days there is
a visible increase in the bacteria resistant to acids, it is certain that
it is not the genuine tubercle bacillus, which requires 37° C. Some-
times in sputum, mixed with certain nutritive media, the tubercle
bacillus increases at incubation temperature. This proliferation,
due in all probability to the importation of globulin-like substances
from the body, is, however, exceedingly small, and ceases altogether
after, at the latest, forty-eight hours ; whilst in the pseudo-tubercle
bacilli a persistent further proliferation takes place at 30°.

The pathological differences from Koch's bacillus are that inocu-
lation with acid-fast bacilli gives rise to no " giant cells," no epithe-
lioid cell clusters, and no tuberculous caseation. Nodular lesions
occur suggestive of tubercle, but according to Potet,* and Abbot and
Gildersleive f : {a) they constitute a localised lesion only, having no
tendency to dissemination, metastasis, or progressive destruction of
tissue by caseation ; (b) they tend to terminate in suppuration like
ordinary abscesses; (c) when occurring as result of intravenous
inoculation they appear in the kidney, rarely in the lung and
other organs ; and (d) the form of granuloma set up is similar to
actinomyces.

This group of organisms is one of considerable importance to the
milk bacteriologist, and in all investigations dealing with the tubercle
bacillus, or with milk and its products, it is essential that the bacilli
met with should be clearly differentiated from the tubercle bacillus.

* Etude sur les Bacteries elites Acidophilus (Potet, Paris, 1902), pp. 188-194.
f The University of Pennsylvania Medical Bulletin, June, 1902.

PLATE 28.

Comparative Cultures or Acid-Fast Bacteria on Glycerine Agar.

(Timothy Grass Bacillus, Moeller'a Grass Bacillus, the Mist Bacillus).

(Grown and photographed by Nwithinbank).

f To face page 366.

STREPTOTHRIX GROUP 367

Insufficient care has been taken in this respect up to the present,
Any such organism found should be compared in cultural and
pathogenic properties with the human tubercle bacillus, the bovine
bacillus of pseudo-tuberculosis, and the various acid-fast organisms,
and not simply accepted on tinctorial properties as a tubercle
bacillus.

Acid-fast Streptothrix Group. — Kecently, considerable attention
has been given to the group of "higher bacteria" known as the
streptothricece. Seen in its mature form, a streptothrix appears to
consist of a tangled mass of mycelial threads, some short bacillary
forms, and spores. The life cycle is completed by the growth or
sprouting of spores by which a mycelium is developed. This branches
off in various degrees and directions, and probably sprouts itself, and
so produces, with the development arising from spores, a fresh
mycelial growth. The mycelium may undergo fragmentation, and
thus bacillary forms occur. Streptothrix is usually readily stained
by Gram's method, but several species are acid-fast. Foulerton and
Price Jones found this to be characteristic of S. Nocardii, S. caproc,
S. hominis (Sabrazes), and S. Eppingeri, in older culture, and then
only the mycelial threads and not the spores. Twenty-one other
species were not acid-fast.* The germs of streptothrix usually grow
better in culture media at 37° C. than at 22° C., and better ae'robically
than anaerobically. Pigment production occurs in some forms,
and certain species liquefy gelatine. One of the best media to use
is maltose-peptone agar, but potato, peptone broth, gelatine, or milk
are also used. A number of workers have now isolated species of
streptothrix from natural media, some of which are declared to be
acid-fast. Many of the group are pathogenic to man or animals.
S* actinomycis (actinomycosis), S. bovis communis, S. madurm
("Madura-foot"), and the disease known as "farcins des boeufs" are
examples. The organisms found in lachrymal concretions, alveolar
abscesses, and similar conditions, especially in relation to the jaws,
are probably frequently streptothrical in origin. These may gain
access to the tissues through carious teeth (by air or food). Infec-
tion may also occur through the tonsils, or cutaneously (as in
Madura-foot).

A description of Streptothrix antinomyces will be found on a
subsequent page. Here two forms of Streptothrix isolated by Fouler-
ton from cases of disease may be described.

1. Streptothrix luteola (Foulerton).

Isolated from a ease of sloughing keratitis in a girl of 12 years.

Staining Characteristics. — Stains well by Gram's method in young eultures ;

Trans, of the Path. Sue. of London, 1902, vol. liii., pt. i., p. 65.

368 TUBERCULOSIS AS A TYPE OF BACTERIAL DISEASE

cultures up to three months old show no acid-fast portions when stained by the Ziehl-
Neelsen method.

Cultural Characteristics. — Grows freely in presence of oxygen, very scanty
growth under anaerobic conditions ; on all media, except potato, growth is distinctly
more active at 37° C. than at 22° C. ; old cultures, especially those in peptone beef
broth, have a faintly feculent odour. On gelatine at 22° C. — Streak culture : obvious
growth is seen on third day, at first of an opaque white appearance, later may show
a very faint yellowish tinge. Growth sinks slowly into the medium, which gradually
liquefies. Stab culture : growth occurs along track of needle in the form of super-
imposed colonies, which have a somewhat flocculent appearance. On peptone-maltose
agar. — After 24 hours' incubation at 37a C. there is some indication of growth ; at
22° C. no obvious growth occurs within this period. After 72 hours' incubation at
37° C. there is a fair amount of growth, and rather less after incubation at 22° C.
The growth, at first of a faint drab or whitish tint, after longer incubation, becomes
usually of a faint yellowish colour. Cultures may yield a free formation of aerial
hyphae, giving a snowy appearance to the surface of the growth, or the surface may
assume a reticulated appearance, without any efflorescence. (For microscopic
appearances, see Plate 29.) On inspissated horse-serum. — Growth is comparatively
scanty on this medium, appearing after 72 hours' incubation at 37° C. At the end of
twenty-eight days' incubation at 37° C. there is a dry, wrinkled, drab-coloured growth,
which has sunk slightly into the medium; no liquefaction. On potato. — Growth
on this medium is equal at temperature 37° C. and 22° C. ; at the end of 48 hours'
growth appears as a brownish or faintly yellowish stain on the medium ; later,
growth usually assumes a ca/6 au lait colour ; there is no pigmentation or
erosion of the medium. Surface efflorescence is seen only in cultures incubated at
37° C, and not in cultures at 22° C. \n peptone beef broth.— After 48 hours' incu-
bation at 37° C. the appearance of some filmy growth is seen at the bottom of the
tube ; in older cultures growth appears as flocculent, more or less discoid colonies.
No pigmentation. (For microscopic appearances, see Plate 29.) In alkaline
litmus milk. — Medium at end of 72 hours' incubation at 37° C. is of a faint pink
colour ; no coagulation. The pink colour changes to a dirty white, and the milk
clears gradually from the surface downwards, becoming at last of a brown colour.
Growth on peptone agar is about equal to that on peptone-maltose agar ; growth on
glycerine agar (1 per cent, glycerine) is less free than it is on either of the two pre-
ceding media. Diastatic action. — No action of the sort is manifested within fourteen
days' incubation at 37° C. Resistance to heat. — Sporulating cultures resist exposure
to moist heat at 70° C. for 20 minutes, but are destroyed by an exposure to the same
temperature for 30 minutes. Pathogenicity for lower animals. — Not found patho-
genic for rabbits (intra-venous and intra-peritoneal inoculation), for guinea-pigs
(intra-peritoneal inoculation), or for tame mice (intra-peritoneal and subcutaneous
inoculation).

2. Streptothrix hominis (Foulerton).

Isolated from a case of pulmonary infection in a woman (especially from sub-
cutaneous abscesses).

Staining Characteristics.— Takes Gram's stain. A three-months' old culture from
glyoerine-peptone agar showed no acid-fast portions when stained by the Ziehl-
Neelsen method.

Cultural Characteristics.— Growth was obtained in peptone-beef broth and on
solid media under ordinary aerobic conditions ; no growth occurred on tubes of
peptone agar and glycerine-peptone agar incubated under anaerobic conditions.
Growth is more active at 37° C. than at 22° C. On peptone agar.— Growth very
slow and scanty ; after several weeks' incubation small whitish, heaped-up, dry-
looking colonies, resembling somewhat the growth of B. tuberculosis, are seen. On
glycerine agar and maltose-peptone agar.— The amount of growth is very much the
same as in the last case, and of much the same appearance. On inspissated horse-
serum and inspissated ox-serum.— Growth is very scanty ; the colonies sink slightly
into the medium. On potato.— No growth was obtained. In peptone-beef broth.—
Small globular colonies appear in the depth of the broth after about six days' incu-

PLATE 29.

*r^

%

.. <&

V «

X

ptothrix luteola (Foulerton). Film preparation from Streptothrix lutcola (Foulerton). Film preparation from
peptone-beef-brotli culture, 14 days at 37° C. x 1000. maltose agar culture, 0 weeks at 37° C. x 1000.

*i

H

t2*

'fm\

Streptothrix hominis (Foulerton). From pus of small abscess in chest-wall. Stained by Gram's method.

[To face page 308.

STREPTOTHRIX HOMINIS 369

bation at 37° C. These do not increase much in size, and after four weeks'
incubation are not much larger than a pin's head ; there is a tendency for the
colonies to cohere in flocculent masses. The growth is whitish in colour. Patho-
genicity for lower animals.— Two rabbits received intra-peritoneal injections of broth
cultures, but there were no obvious effects. *

Other acid-fast members of the pathogenic streptothrix group
have been isolated by Birt and Leishman from a case of pleural
effusion ; by Eppinger from an abscess of the brain ; by MacCallum
from peritoneal pus; by Nocard in "farcin du bceuf." These
organisms can, as a rule, be differentiated from the tubercle bacillus
by" morphological, cultural, and tinctorial properties.

General Note on Differentiation of Acid-fast Organisms

The acid-fast bacillus, pseudo-tubercle bacillus, and acid-fast
streptothrix may all be found to resemble the tubercle bacillus in
greater or less degree. It has been suggested that they are all of the
streptothrix genus. Whilst they are all acid-fast they are not equally
resistant, and this fact assists in their differentiation. Broadly, none
of these forms can resist decolorisation with 25 per cent, sulphuric
acid for more than sixteen hours, whereas the tubercle bacillus can
withstand decolorisation for seventy-two hours (Coles). Hence, if
him preparations be made in the usual way, stained for seven minutes
with hot carbol fuchsin, and then decolorised with 25 per cent, sul-
phuric acid for sixteen hours, the only bacilli remaining red are
tubercle bacilli. If this method fail, cultivation and inoculation tests
must be applied. The former by culturing in broth at 30° C. (shows
growth in three days in non-tubercle bacilli), the latter by inoculation
of guinea-pigs. And here, as elsewhere, it is necessary to observe all
the characters before forming a diagnosis.

* An excellent statement on the general characteristics and pathogenic action of
the genus Streptothrix will be found in the Trans, of the Path. Soc. of London,
1902, vol. 53, part i., pp. 56-127 (Foulerton and Price Jones).

2 A

CHAPTEE XI

THE ETIOLOGY OF TROPICAL DISEASES

Malaria : Forms of Malarial Fever, the Mosquito Theory, Prevention of Malaria —
Cholera: Methods of Diagnosis — Plague: Symptoms, Rats and Plague,
Bacteriology, Administrative Considerations — Leprosy — Yellow Fever —
Malta Fever — Sleeping Sickness — Beri-beri.

It is now generally accepted that the future prosperity of the Anglo-
Saxon race depends upon the measure to which it is able to control
the Tropics. For it is obvious that that great middle band of the globe
which we term the Tropics is increasingly one of the most valuable
and important areas for colonisation in the world. Yet, though the
rewards are great, the risks and penalties are also great. The
coloniser from temperate regions knows this to his cost. Malaria
and plague and cholera, to speak of no other tropical diseases, have
made irretrievable claims upon him. Eecently it has come to be
recognised that much of this great loss is preventable, and ought
therefore to be prevented. The establishment of Schools of Tropical
Medicine in London, Liverpool, and other places, and the practical
means adopted for preventing cholera epidemics and stamping out
plague and malaria, are examples of the new sense of responsibility
which is stimulating nations and governments to do their utmost
to bring under control those scourges of pestilence, which have made
the Tropics so often the grave of the white man.

The channels of infection in tropical diseases are various.
Unhealthy surroundings, diet, the soil, bad water, and parasite
hosts (as, for example, rats in plague and mosquitoes in malaria),
seem to be the chief. But there is much yet to be done in the
investigation of the causes of certain tropical diseases.

We may now enter upon the consideration of five typical diseases
mostly limited to the Tropics : (1) Malaria ; (2) Cholera ; (3) Plague ;
(4) Leprosy ; (5) Yellow Fever. It is apparent that many bacterial

370

MALARIA 371

diseases are "cosmopolitan." Tuberculosis, for example, may occur
in all parts of the world; so may pneumonia or typhoid. But the
five diseases named above are in a greater or lesser degree endemic in
tropical regions.*

1. Malaria

The term malaria {lit., bad air) is often applied rather to a
group of fevers than to one specific affection. Such fevers have
certain points in common. One common feature is that, with few
exceptions, the disease originates in the blood. A second feature is
the elaboration of a black or brown pigment from the haemoglobin
present in the blood corpuscles. And a third common character is
that these diseases are produced not by bacteria but by hoematozoa,
that is to say, protozoa which can live and perform their functions
in the blood. The term " malaria " should, however, be reserved for
the specific disease caused by the malarial parasite.

For many years the group of diseases represented by malaria
were designated miasmatic, owing to the belief that they were caused
by some damp and unhealthy condition of the soil, from which
emanated a miasm or soil ferment. Thus was explained their
prevalence on and around marshy tracts of land, and their prevention
by land drainage. Whilst these two latter features remain true, a
new interpretation has been placed upon them.

In 1880 Laveran first discovered and described parasites in
the blood cells of malarial patients, and on further investigation
it was soon found that these assumed many different forms.
These differences depend upon the kind of fever and the stage of
fever.

The reasons for believing that Laveran's bodies — though they
have not yet been cultivated outside the human body — are the
specific cause of malaria are briefly these : — (1) The parasites found
in the blood of malarial patients of all countries are the same. (2)
Such parasites are not found in healthy persons. (3) Their develop-
ment fully accounts for the production of the inelancemia and
malarial pigmentations of viscera owing to the melanin-forming
property of the parasite. (4) The phases in the development of
such parasites corresponds with the clinical course of malaria (Golgi).
(5) Quinine, which cures malaria, kills the parasite. (6) Malaria
can be conveyed by the introduction of this parasite into the blood

:< The term endemic indicates that a disease affects people within a certain
geographical limit, and which seems therefore to arise from local or particular
causes. Epidemic indicates that a disease attacks a large number of people at the
same time and approximately in the same place. Whereas a pandemic is the same
with an indefinitely wide distribution.

372 THE ETIOLOGY OF TROPICAL DISEASES

of man, and the parasite reappears in the blood of the individual so
inoculated. It is interesting in this connection to observe the
negative results of the recent attempts of Koch to inoculate the
higher apes with malaria in Batavia, as reported by the German
Colonial Office.* Laveran's bodies have been variously classified as
knowledge of them has grown. It is now agreed that these parasites
belong to the Sporozoa, to the order of Rcemoeytozoa, and to the
genus of Hozmamwba.

Now if we examine a sample of human blood from, say, the
benign tertian form of malaria, we shall find not different parasites
as in three forms named below, but different stages in the evolution
of one parasite. These different stages are normal and regular, and
not accidental or chance forms, and for the sake of convenience we
may summarise them seriatim thus : —

1. Early Form of Parasite. — Looking through the microscope, we
shall readily observe large numbers of blood corpuscles, and in some
of these, and possibly many of them, there will be apparent certain
irregularities. In the first place, the protoplasm of the affected
corpuscles is paler than that of the healthy cells. Next, within the
protoplasm will be seen the parasite (amwoula), containing possibly a
few specks of black pigment, and of more or less irregular outline,
sometimes nearly filling the whole blood corpuscle. This body is
motile, and moves about like an amoeba inside the corpuscle, in the
tertian fever with great rapidity. As it increases in size, the
corpuscle becomes paler. The largest of these spherical forms are
outside the cells (extra-corpuscular spores, enhaemospores), and move
about free in the blood plasm. But the smaller ones are generally
inside the blood cells (intra-corpuscular amoebula). They live at the
expense of the haemoglobin in the corpuscle, and ultimately change
it into pigment (melanin).

2. Concentration of the Pigment. — After the parasite has gained
its mature form as regards size, an increase and concentration of the
pigment occurs. The body of the parasite now fills the corpuscle,
and the pigment which before existed in specks, or diffusely, becomes
gathered together towards the centre of the parasite.

3. The third change in the evolution of the amoebula is segmenta-
tion. By this process the parasite splits up into segments; the
tertian fever forming much smaller and more numerous segments
than the quartan. This segmentation gives rise to what is known
as the "rosette body."

4. In reality these segments are sporocytes, new amoeboid bodies,
which, by the rupture of the eaten-out corpuscle, become diffused
freely in the blood. Many of these " spores " are supposed to pass
to the spleen, some are absorbed by phagocytes or scavenging cells,

* Deutsche Medicinische Woehmsehrift, February 1900.

FORMS OF MALARIAL FEVER 373

but, in a few hours, many others reappear in the blood and inaugurate
another stage.

5. The Infection of the Corpuscles. — The spores now attach them-
selves to healthy blood corpuscles, slowly pass into their interior,
and set up a precisely similar series of changes; the actively
amoeboid stage, the increase of size and pigmentation, the con-
centration of the pigment, the mature form, and the segmenta-
tion resulting in the rosette body, and eventual escape of
sporocytes. In this way the multiplication of the parasite is
carried on in the human host. This is known as the Cycle of
Golgi. Each paroxysm of malaria is related to the evolution cycle
of a generation of these parasites — probably many millions in
number — the commencement of each paroxysm coinciding with the
maturation of a generation of parasites. The severity of a paroxysm
in a given type of fever is also in direct relation to the number of
parasites in the blood. It does not necessarily follow that the
gravity of the case is in proportion to the intensity of the paroxysms.

Malaria is characterised by marked intermittency, which is usually
divided clinically into three leading forms : —

(a) Quartan, depending upon a parasite which takes seventy-two
hours to pass through its cycle of development, and produces fever
every third day. The corpuscles invaded do not become so much
decolorised, hypertrophied, or altered in shape as in other forms.
The parasite shows distinctly less amoeboid movement, and is not
so delicate in structure or definition as in the Tertian varieties,

^8

Fig. 29.— Quartan Malaria Parasite.

though it carries a large amount of dark brown, pigmented material,
which is coarse in grain. The developed sporocyte has what is
described as a " daisy-head " appearance. The six to fourteen spores
are rounded in form, and possess a well-defined nucleus. Quartan
fever is relatively much more common in temperate and subtropical
latitudes than in the tropics.

(b) Benign or Mild Tertian. — In this fever the parasite takes
forty-eight hours to complete its cycle. The amoebula is actively
motile inside the corpuscle, giving rise to great and rapidly-changing

374

THE ETIOLOGY OF TROPICAL DISEASES

irregularities in the condition of the corpuscle, which becomes swollen,
pale in colour, and may show deeply-stained " spots." The pigment
granules are finer than in the quartan parasite. The final

@

Fig. 30.— Tertian Malaria Parasite.

decolorisation of the corpuscle is very marked. Tho "rosette
body" or sporocyte in this species is composed of some fifteen to
twenty-five spores, small, smooth, and oval; the gametes are
spherical. The benign tertian parasite is probably the commonest
form found in malaria, and is widely distributed over the globe.

(c) The Malignant Infections (aestivo - autumnal, malignant
quotidian, malignant tertian). The amaibulm in these conditions
are much smaller than in the benign types, but may occur in pro-
digious numbers, and their movements are very active. The organism
causes considerable modifications in the shape and size of the corpuscle,
which has a tendency to shrivel. It is not filled by the parasite in the
same degree as in the other forms. Sporulation occurs in the spleen

Fig. 31.— Malignant Malaria Parasite.

and other internal organs, and not in the blood, and, therefore, the
sporocytes in this form are not found in the blood in the usual way.
The most distinctive feature of all is that the malignant parasites
(gametes) form "crescents," and attack a larger proportion of red
corpuscles than in the other forms. The classifications, minor

FORMS OF MALARIAL FEVER 375

subdivisions; and clinical nomenclature have passed through a variety
of changes. The three old divisions have been retained here for
convenience. Sir Patrick Manson has suggested the following
classification : —

t r. . f Quartan \ _ In which the parasites do

f' ^enign • \ Tertian J ~ not form crescents.

( Quotidian — with pigmentation "j In which the para-

B. Malignant -J Quotidian— without pigmentation J- = sites do form
t Sub-tertian J crescents.

Now in some forms of malarial fever, namely, the malignant
infections, the spherical bodies or mature form immediately prior to
segmentation into rosette bodies, do not actually show segmentation,
but assume the form of crescents lying inside the blood corpuscle, the
haemoglobin of which has been absorbed. Between the poles of the
crescent may be seen the membrane of the blood corpuscle, the
crescent being folded somewhat on itself. These crescents represent
the form of the parasite which requires to enter the body of the
mosquito in order to attain development. The crescents do not, as
a rule, appear in the blood until about one week from the onset of
the fever, and are the first stage of the extra-corporeal phase of the
parasite. They are termed the gametocytcs, and are of two kinds,
male and female. It will be necessary to follow the development of
each kind separately.

The microgametoeyte, or male gamete, is the parasite in crescent
form, with the delicate membrane of the containing blood corpuscle
at first investing it, as described above. These crescents are hyaline
in appearance, and the motionless pigment is loosely arranged. The
crescent eventually absorbs or exhausts the blood corpuscle, and
becomes a free body in the blood serum. Next, it changes by
becoming kidney-shaped, then round at the poles, thicker, more
ellipsoidal, and eventually spherical. The pigment granules now
become mobile, and eventually assume most active movement and
become diffused throughout the whole sphere, and almost immedi-
ately thereafter the sphere itself becomes agitated, and from its
circumference shoot out long flagella (microgametes). This flagellated
parasite is a weird-looking, " octopus-like " body, with long lashing
tentacle-like flagella, and containing in its centre a mass of moving
pigment granules. The flagella or microgametes are of the nature
of spermatozoa, and fulfil a similar function, and are in length some
three or four times the diameter of the microgametocyte. They are
unpigmented, and may bear at their extremities bulbous swellings.
They break off and become free in the blood, continuing their
active movements.

The flagellated body may be produced from free spheres of the

376 THE ETIOLOGY OF TROPICAL DISEASES

quartan and benign tertian parasite (which clo not produce crescents)
as well as from crescents. But it is never seen in fresh blood
immediately after being drawn from the body. It only appears after
the blood has left the body for twenty minutes or half an hour. Such
a striking transformation of a free sphere or a crescent is evidently
a stage of great importance, and two different explanatory theories
have been advanced to account for it. Some have held it to be a
degenerative change in the parasite — that the coldness of the
outside air has killed it, that its contortions and wriggling flagella
are but its death struggles, and that the active movement of its
pigment particles are but Brownian movements of dead granules.
Other authorities, and particularly Sir P. Manson, have declared the
flagellated body to be a vital evolutionary change — a normal step in
the life of the parasite, the first stage in its life-history outside the
human body, the extra-corporeal homologue of the intra-corporeal
sporulating body. It is now agreed that these microgametes or
flagella are the essential sporulating bodies of an extra-corporeal
phase, and that their function is the impregnation of the female
gametocyte.

The macrogfametocyte, or female gamete, is the second kind of
gamete, and starts its course in much the same way as the male cell.
It also is a crescent inside the blood cell, which it eventually
breaks down, and thus becomes free in the blood serum.
Instead of being hyaline it is granular, and the pigment is
situated more centrally. It eventually becomes ellipsoidal, and
then spherical. The protoplasm of the female crescents stains
more deeply than that of the male crescents; the pigment is
more closely grouped together, generally in ring form, in the
centre of which will be seen one or occasionally two large
masses of chromatin. At its maturity as a macrogametocyte, two
small polar bodies or excrescences or papilla are seen on its circum-
ference, and it is at this site that impregnation by the free nagellum
(male cell or microgamete) is effected. The result is the zygote, or
travelling vermicule. In 1897, MacCallum observed this impregna-
tion actually taking place in a case of human malaria, and others
have observed it in one of the malaria-like organisms of birds, the
halteridiuin. After its entry into the female cell the nagellum
became quiescent, and the pigment became collected at the posterior
end of the cell, which then assumed the shape of a spear head, and
became the actively-motile zygote.

The Mosquito Phase. — Just as the segmentation body eventually
splits up into spores for the further propagation of the parasite in
the blood of the malarial patient, so the flagellated body provides
for the propagation of the parasite in some living host outside the
human body ; for, as is well known, parasites pass from one host to

THE MOSQUITO THEORY

37

another. There is now complete evidence to show that the host is
the Mosquito. It is therefore necessary to consider briefly the
outstanding features of the Mosquito Phase.

The Mosquito is widely distributed, especially in tropical

Fjo. 32.— Anopheles maculipennis 9 (Meigen).

countries, and is generally classified under the genus Culiciclce, of
which probably a large number of varieties exist. The Culcx is the
commonest species. Its larvae live almost everywhere in warm
countries, inhabiting any pot, tub, well, cistern, broken bottle, or,

378

THE ETIOLOGY OF TROPICAL DISEASES

indeed, anything in which a little water can lodge. They are
almost domestic animals. The palpi are short, the wings unspotted,
the proboscis thin, the thorax large, and the larvae have breathing
tubes. They prefer to lie in artificial collections of water. When at
rest, for example on a wall, the body of Gulex is found parallel to the
wall (see fig. 33). The rarer species, and that which has been proved
to be the host of the malaria parasite, is the Anopheles. This differs in
various essential particulars from the Culex. In Anopheles the body
is slim, but the proboscis long and thick, the palpi are long, and the

Fig. 33.— Diagram of Culex and Anopheles on Wall (Ross).

wing is "dappled" with dark spots on its anterior margin. The
larvae have no breathing tubes, and lie horizontally (not vertically
as in Culex) in the water of puddles, looking like bits of brown
stick or thorns floating on the surface. When at rest on a wall, the
axis of its body is almost at right angles to the wall, so that the
head of the Anopheles is directed towards the wall, whilst the body
projects out into the room. Culex has been briefly designated a
" pot-breeding " mosquito, whilst one of the features of Anopheles is
that it is " puddle-breeding." Its favourite haunts are slow, small
streamlets containing algae ; small, shallow, natural puddles with
confervoid growths in the water ; or stagnant and fairly permanent

THE MOSQUITO THEORY 379

collections of rain-water. The larvse cannot live when the puddles
in which they breed dry up. Nor does Anopheles appear to favour
swamps containing deep water. The puddles they select are in
immediate proximity to houses, where the adult mosquitoes may
frequently pass to find human beings or cattle, from whom they
may derive their nourishment. As is well known, it is the female
mosquito which is the blood-sucker. After she has filled herself
with blood, she retires to some dark, sheltered spot near a stagnant
puddle, and after a few days deposits her eggs (from 200-400 in
number) in a mass on the surface of the water. Then she dies and
falls into the water beside her eggs. The eggs give rise (sometimes
in sixteen hours time) to the tiny swimming larvse, which feed
greedily and grow rapidly, shed their skins, and become nymphce or
jpujoce. Eventually, the shell of the nyrrvpha cracks along its dorsal
surface, and the young mosquito emerges. Standing on the raft of
its empty pelt, it dries its wings and flies away.* Yery soon it also
lays eggs. The entire cycle from egg to egg is about fifty days.
The three conditions necessary for the multiplication of malarious
mosquitoes are (a) high atmospheric temperature, 75° to 104° E. ;
(b) collections of water, fresh or brackish ; and (c) the presence in
the breeding pools of low forms of animal and vegetable life.

Having now gathered the outstanding facts concerning the
mosquito, we may return to the part which it plays in the propaga-
tion of malaria. The method which nature elects for the liberation
of any given parasite is generally one that is reasonably regular and
frequent in its operations. Hence, it occurred to Laveran and
Manson that as the malarial organism is a passive blood parasite,
its escape from the human body might be effected on the same
principle that the escape of the passive muscle parasites is effected.
As the latter obtain their opportunity by being swallowed by some
flesh eater, Dr Manson reasoned that the blood parasite might obtain
similar liberty by being swallowed by some blood-eater, some
suctorial insect such as the sand-fly or mosquito. He was still
further led to the mosquito theory owing to the parallel conditions
which he had already found to exist in the case of the analogous
mosquito phase of Filaria sanguinis.^ The mosquito phase then is
the extra-corporeal stage of the life-history of the malaria parasite
which takes place within the body of the mosquito, and we may
now briefly follow the course of events in the further development
of the flagellated body inside the mosquito.

Whilst the mosquito theory was largely the suggestion of Sir

* For further particulars as to mosquito Anopheles, see Brit. Med. Jour. 1901,
i., p. 195; Jour, of Hygiene, 1901, i., pp. 4-77, 269, and 451, also vols. 1902 and
1903.

I Goulstonmn Lectures, 1896. Sir Patrick Manson. Brit. Med. Jour., 1896, i.,
641 et seq., and 1900, i., 328.

380

THE ETIOLOGY OF TROPICAL DISEASES

P. Manson, the practical investigation and final elucidation of it
were in great measure due to the patient observance and skill of
Ronald Ross, at that time a Surgeon-Major in the Indian Medical

Fig. 34.— Scheme showing Human and Mosquito Cycles of the Malaria
Parasite (Manson).

1-8, the human or endogenous or asexual cycle; 9-24, the mosquito or exogenous or sexual cycle. 1,
Normal red blood corpuscle ; 2-5, blood corpuscles containing an amoebula ; (3-8, sporocytes ;
9, young gametocyte ; 11, 13, 15, 17, microgametocytes or male gametes ; 10, 12, 14, 16, macro-
gametocytes or female gametes ; 18, female gametocyte being impregnated by microgametes ;
19, travelling vermicule ; 20, young zygote; 21, 22, zygotomeres ; 23, blastophore ; 24, mature
zygote.

Service. His chief contribution to the solution of the problem was
the discovery of the malarial parasite in the tissues of the mosquito.
He found that in the body of mosquitoes fed upon malarial human
blood 70 per cent, of the crescent forms of the parasite, as we have
seen one of the latent forms of the flagellated body, were transformed

THE MOSQUITO THEORY 381

into the flagellated body. This transformation occurred in a pro-
portion of instances very much greater than occurs in malarial
blood spread in the ordinary way on a slide and exposed to the air.
He found, also, that the flagella broke away from the flagellated
body, yet he was unable to trace what became of the free flagella. But
in a "dapple-winged" mosquito (anopheles) fed in the same way
with malarial blood containing crescents (summer-autumn fever),
he discovered, embedded in the tissues of the stomach wall, certain
peculiar oval cells containing the same pigment as the malarial
parasite in the human blood. This was, in fact, the extra-corporeal
form of the human parasite after impregnation, namely, the zygote.
Next he discovered pigmented cells in the body tissues of mosquitoes
fed upon sparrows' blood, affected with a similar parasitical condition
to malaria (protcosoma) ; and from one step to another he demon-
strated the evolution of the mosquito phase of these parasites.
Eoss made it evident that the cycle of extra-corporeal development
of the parasite, as we have already seen, is carried on inside the
body of the mosquito. After malarial blood is shed or swallowed
by the mosquito, the changes already described take place. Practi-
cally, all the crescents become spheres within a few minutes of being
taken into the stomach of the mosquito, then the male gametes
become flagellated, and the female gametes become impregnated.
According to Eoss, the condition which brings about the transfor-
mation of crescents into flagellated bodies is not low temperature,
nor exposure to air, nor contact with the wall of the mosquito's
stomach, but abstraction of the water from the serum of the blood.

However that may be, the changes resulting from impregnation
result in the mosquito's stomach in the production of the zygote or
fertilised cell. This body is a travelling vermicide, and on or about
the second day it penetrates the stomach wall and becomes encysted
between the muscle fibres. The number of zygotes produced in the
mosquito after its feed on malarial blood varies widely, sometimes
being few, sometimes many. In the enlarging encysted cell there
now come to be developed a number of cells known as zygotomeres,
which as development proceeds become blastophores filled with
filiform spore cells (sporozoits or germinal rods or zygotoblasts).
Ultimately, the zygote becomes thus transformed into a cyst (sporo-
cyst) packed full of zygotoblasts. When fully developed, at about
the eighth or ninth day after the mosquito ingested the malarial blood,
the sporocyst measures about 60 micromillimetres in diameter.
About the twelfth day it bursts, discharging the zygotoblasts, which
are, of course, " spores " or reproductive elements, into the body cavity
and fluid of the mosquito, and spreading from thence they become
accumulated in the large veneno-salivary gland, and are thus in a
position to be injected along with its secretion into the human

382 THE ETIOLOGY OF TROPICAL DISEASES

subject next bitten. These zygotoblasts are the actual source then
of infection of man, and on arriving in the human blood the parasite
(as spores) attacks the blood cells, and thus commences the intra-
corporeal or human phase described above. The mosquito phase
occupies a time varying between six to sixteen days or longer,
depending on temperature and other factors.

Such, in outline, is the mosquito theory of malaria. No one
supposes that the last word has been said. But sufficient is now
known to make it certain that the mosquito phase is a fact of
essential importance in the conveyance of the disease to man. In
the first place, the malarial parasite has been found repeatedly in
the body of the mosquito, and in the second place the crucial
experiment of inoculation has been performed, and has yielded a
positive result. Infected mosquitoes were brought from the Eoman
Campagna, and Dr Thorburn Mason and Mr George Warren con-
sented to be bitten, and thus contracted malaria. There yet remain
gaps to be filled up in our knowledge of the disease, but there can
be little doubt that future work will further establish and elaborate
the principles of the mosquito theory, and the lines of prevention
will of necessity follow the new facts now proved.*

As concerns preventive medicine, the new facts may be sum-
marised in three propositions : — (1) Malaria is caused by a number
of microscopical parasites which live and propagate themselves in
the blood. (2) These parasites are carried from infected persons to
healthy ones by the agency of the genus of mosquitoes termed
Anopheles. (3) These mosquitoes breed chiefly in shallow and
stagnant terrestrial waters.

Examination of Malarial Blood (see Appendix, p. 485).

The Prevention of Malaria

The new knowledge respecting malaria indicates the only
adequate preventive methods. The malarial parasite gains access
to the human subject by means of mosquito bites, and, as far as
is known, in no other way. Hence the methods of prevention must
be directed mainly against the mosquito : —

1. The prevention of mosquito -breeding.

2. The destruction of mosquitoes.

3. Avoidance of being bitten by mosquitoes.

4. The use of quinine.

1. The Prevention of Mosquito-breeding1. -~In order to prevent

the breeding of mosquitoes, it is necessary to eradicate all possible

breeding-places. As we have seen, such places are tanks, cisterns,

vessels of stagnant water, ditches, small pools, buckets, cocoa-nut

* For full account of malaria, see Tropical Diseases (Manson), 1903, pp. 1-173.

PREVENTION OF MALARIA 383

shells, tins, cans, pots, etc., wherever stagnant water readily collects,
especially near houses. The larvae of Culex float when at rest on the
surface of the water, suspended by their tails and with their heads
hanging downward; when disturbed, they wriggle to the bottom.
The larvae of Anopheles float flat on the surface like small sticks, and
when disturbed they wriggle on the surface with a backward skating
movement. The former are usually present in artificial collections
of water, such as pots, broken bottles, cans, etc., whilst Anopheles
prefer natural collections of water, chiefly rain-water puddles which
do not dry up quickly, or which contain green water-weed. Such
being the breeding-places, prevention is simple. Collections of
stagnant water must not be permitted. Search must be made for
them, vessels must be emptied and puddles brushed out with a
broom, and small pools drained and filled in. Water must not be
allowed to collect anywhere near the house. Land drainage is an
obvious preventive of the first importance.

Cisterns and similar necessary collections of water may be
protected by paraffin or petroleum, for these by lying on the
surface of the water prevent the pupae of the mosquito from reaching
the surface at the time of transformation. When water is required,
it may be drawn off from the bottom of the cistern. The surface of
paraffin may be renewed once a fortnight or oftener if necessary.
Paraffin (kerosene oil) also acts as a culicicide, destroying the larvae
by choking their air tubes. The essential condition in any scheme
for the sanitary improvement of a malarious region is that the eggs,
larvae, and nymphae of the mosquito should be exterminated in that
region. Covering small collections of water with healthy soil, accom-
panied by thorough drainage, inasmuch as they remove at the same
time both the water and the atmospheric air, the two indispensable
elements of mosquito life, are the best preventive methods.*

2. The Destruction of Mosquitoes. — Smoke from a wood fire or
damp tobacco leaves, or sulphur, or other gaseous disinfectants may
be used for this purpose. Kerosene may be used as recommended
above for killing the larvae. Individual mosquitoes should be killed
whenever possible.

3. Avoidance of being- Bitten by Mosquitoes —For the protec-
tion of the person from attack by mosquitoes, there are a variety of
contrivances, from mosquito nets to mosquito-proof houses. Mosquito
nettings on the bed should invariably be used in malarious countries.
There are several points as to the effectual use of such nets. In the
first place, the net should be square, should be hung inside a frame-
work, tucked carefully under the mattress all round and not allowed
to hang down, and stretched tight so as to allow air to pass in easily.

* See also Brit. Med. Jour. 1900, vol. i., pp. 300-300 (Celli) ; and 1901, vol. i.,
pp. 193-203 and p. 240.

384 THE ETIOLOGY OF TROPICAL DISEASES

In the second place, the roof should be made of netting similar to the
other parts of the net, and not of cloth. Thirdly, during the day,
when the net is not being used, it should be hung up in such a way
as to prevent mosquitoes entering it. Fourthly, the mesh of the net
should be sufficiently fine to effect its purpose, and should contain
no rents, holes, or other apertures. Where punkahs are available,
they should swing above the mosquito net.

Again, habits of life, especially temperateness and moderation,
not sleeping in the open air, living as far as possible in healthy
houses, not frequenting native quarters after sunset, and not
associating with native children, are methods by which to avoid being
bitten. It is now well known that in malarious towns and districts
the great majority of native children harbour the malarious parasites
in their blood, and therefore segregation is a necessary preventive
method. Europeans' houses should bo built at a distance from the
native quarters.

The experiments of Sambon and Low, of Celli, of Grassi, of Fermi,
and Tonsini, and of the Eed Cross Society of Italy, have demonstrated
beyond all question that it is practicable to construct habitable houses
which shall be mosquito-proof.*

Lastly, probably some protection is obtained by means of
perfumes, washes, pomades, soaps, etc., though these should not be
relied upon. Certainly flannel clothing is a great advantage.

4. Quinine. — When it is too late for preventive measures the
time has come for isolation, disinfection of rooms containing infected
mosquitoes, and treatment. A person with malaria is always a risk
to other persons, and should be isolated as far as practicable. Infected
rooms should be disinfected with gaseous disinfectants, such as
sulphur or formic aldehyde. Quinine is the specific remedy in
treatment. It acts not only as an antipyretic but as a specific drug,
destroying the parasite in the blood. For an ordinary intermittent
fever the dose of quinine may be 10 grains, given when the sweating
stage commences, followed by 5 grains every six or eight hours for a
week, and with a view of preventing relapse 5 grains three times
every fifth, sixth, or seventh day for two or three months. Five to
ten grains of quinine twice a week or oftener, tends to prevent or
abate incipient malarial infection.

2. Cholera

This word is used to denote a group of diseases rather than one
specific well-restricted disease. In recent years it has become
customary to speak of Asiatic cholera and British cholera, as if,
indeed, they were two quite different diseases. But, as a matter

* Practitioner, March 1901, p. 262.

CHOLERA 385

of fact, wc know too little as yet concerning either form to dogmatise
on the matter. Until 1884 practically nothing was known about
the etiology of cholera. In that year, however, Koch greatly added
to our knowledge by isolating a spirillum from the intestine, and in
the dejecta of persons suffering from the disease.

Cholera has its home in the delta of the Ganges. From this
endemic area it spreads in epidemics to various parts of the world,
often following lines of communication. Cholera is generally
conveyed by means of water. It is a disease which is characterised
by acute intestinal irritation, manifesting itself by profuse diarrhoea
and general systemic disturbance accompanied by collapse, cramps,
cardiac depression, subnormal temperature, and suppression of urine.
The incubation period varies from only a few hours to several days.
In the intestine, and setting up its pathological condition, are the
specific bacteria, in the general circula-
tion their toxic products bringing about
the systemic changes.

The spirillum of Asiatic cholera
(Koch, 1884) generally appears in the
body and in artificial culture, broken
into bacillary elements known as
" commas." These are curved rods
with round ends, showing an almost
equal diameter throughout, and some-
times united in pairs or even in chains
(spirillum). The latter rarely occurs in
the intestine, but may be seen in fluid
cultures. The common site for Koch's Ro" 8^tST(S^?oama
comma is in the intestinal wall, crowd-
ing the tubules of the intestinal glands situated ltd ween the epi-
thelium and the basement membrane, abundant in the detached
Hakes of mucous membrane, and free in the contents of the intes-
tine. The bacilli are present in enormous numbers, and lie usually
with their long axes in the same direction, giving the "fish in
stream" appearance (Koch). The bacilli do not occur in the blood,
nor are they distributed in the organs of the body. They occur
mostly in the lower intestine.

The bacillus is actively motile, and possesses at least one terminal
flagellum. The organism is aerobic, and liquefies gelatine. It
stains readily with the ordinary aniline dyes, but does not stain by
Gram's method. It does not produce spores, though certain refractile
bodies inside the protoplasm of the bacillus in old cultures have
been regarded as such. The virulence of the bacillus is readily
attenuated, and both the virulence and morphology appear to show
in different localities and under different conditions of artificial

2 L

386 THE ETIOLOGY OF TROPICAL DISEASES

cultivation a large variety of involution forms. Unless the organism
is constantly being sub-cultured, it readily dies. Acid, even the *2
per cent, present in the gastric juice, readily kills it. Prolonged
drying, or heating to 55° C. for sixty minutes, or treatment with
weak chemicals has the same effect. The bacilli, however, have
comparatively high powers of resistance to cold. .Unless examined
by the microscope in a fresh and young stage, it is difficult to
differentiate Koch's comma from many other curved bacilli.

Its characters in culture are not always distinctive. Microscopic-
ally, the young colonies in gelatine appear as cream-coloured, irregu-
larly round, and granular. Liquefaction sets in on the second day,
producing a somewhat marked " pitting " of the medium, which soon
becomes reduced to fluid. In the depth of gelatine, the growth
is very characteristic. An abundant, white, thick growth exactly
follows the track of the needle, here and there often showing a break
in continuity. Liquefaction sets in on the second day, and produces
a distinctive " bubble " at the surface. The process proceeds steadily,
at first a funnel-shaped liquefaction resulting, and then in the course
of a week or two all the gelatine may be reduced to fluid. On agar
Koch's comma bacillus produces within twenty-four hours a thick
greyish irregular growth. On potato, especially if slightly alkaline
and incubated at 37° C, an abundant brownish layer is formed.
Broth and peptone water are favourable media, and at 37° C. a general
turbidity occurs with the formation on the surface of a pellicle,
containing spirilla in active motility. In milk it rapidly multiplies,
curdling the medium, with production of acid. Unlike B. coli, it
does not form gas, but, like B. coli, it produces large quantities of
indol, and a reduction of nitrates to nitrites. Hence the indol test
may be applied by simply adding to a peptone culture several drops
of strong sulphuric acid, when in the course of several hours, if not
at once, there will be produced a pink colour, "the cholera red
reaction," due to the formation of nitroso-indol. Although the
bacillus readily loses virulence, and its resistance is little, it
retains its vitality for considerable periods in moist cultures,
upon moist linen, or in moist soil. In cholera stools kept at
ordinary room temperature the cholera bacillus will soon be out-
grown by the putrefactive bacteria. The same is true of sewage
water.

The lower animals do not suffer from any disease exactly similar
to Asiatic cholera, and hence it is impossible to fulfil the postulate
of Koch dealing with animal inoculation. In this respect it is like
the typhoid bacillus. It is, however, provisionally accepted that
Koch's bacillus is the cause of the disease. The four or five other
bacteria which have from time to time been put forward as the cause
of cholera have comparatively little evidence in their support. It

DIAGNOSIS OF CHOLERA 387

is less from these and more from several spirilla occurring in natural
waters that difficulties of diagnosis arise.

The reasons for believing Koch's bacillus to be the cause of
cholera are four: (a) its constant presence in cases of the
disease; (b) the results of accidental infection with this bacillus;
(c) the agglutinative and protective properties of the serum of cholera
patients ; and (d) the result of Haffkine's preventive inoculation.

There appears to be evidence to show that comma bacilli may be
introduced into the alimentary canal without producing the disease,
unless there be some injury or disease of the wall of the intestine
(Peiffer). Desquamation of the intestinal epithelium seems an essential
factor in the production of the disease in man. It need hardly be
added that the bacillus acts, like other pathogenic bacteria, by the
production of toxins (Peiffer), which appear to be intracellular. At
present very little is known of their chemical nature. Brieger
separated cadaverin and putrescin and other bodies from cholera
cultures, and other workers have separated a toxalbumin.

Methods of Diagnosis of Cholera : —

1. The nature of the evacuations and the appearance of the
mucous membrane of the intestine afford striking evidence in
favour of a positive diagnosis. Nevertheless, it is upon a minute
examination of the flakes and pieces of detached epithelium that
reliance must be placed. In these flakes will be found abundance
of bacilli having the size, shape, and distribution, of the specific
comma of cholera. The size and shape have been already referred to.
The distribution of comma bacilli ("fish in stream") in the
flakes of watery stools is, when present, somewhat characteristic of
Asiatic cholera, and may greatly aid in a correct diagnosis. But
unfortunately, it is not always present, and then search for other
characters must be made.

2. The appearance of cultivation on gelatine, to which reference
has been made, is of diagnostic value, and the growth on agar and in
peptone solution.

3. The " cholera red reaction." It is necessary that the culture
and the sulphuric acid be pure for successful reaction.

4. The intra-peritoneal injection in guinea-pigs is followed by
abdominal distention, subnormal temperature, and other characteristic
symptoms.

5. Isolation from water is, according to Dr Klein, best accom-
plished as follows: A large volume of water (100-500 c.c.) is placed
in a sterile flask, and to it is added so much of a sterile stock fluid
containing 10 per cent, peptone, 5 per cent, sodium chloride, as will
make the total water in the flask contain 1 per cent, peptone and
•5 per cent. salt. Then the flask is incubated at 37° C. If cholera
vibrios are present in the water, however few, it will be found

388 THE ETIOLOGY OF TROPICAL DISEASES

after twenty-four hours' incubation that the top layer contains
actively motile vibrios, which can now be isolated readily by gelatine-
plate culture.

6. To demonstrate in a rapid manner the presence of cholera
bacilli in evacuations, even when present in small numbers, a small
quantity must be taken up by means of a platinum wire and placed
in solution containing 1 per cent, of pure peptone and #5 per cent,
sodium chloride (Dunham's solution). This is incubated as in the
case of the water, and in twelve hours is filled with a turbid growth,
which when examined by means of the hanging drop shows
characteristic bacilli.

7. Pfeiffer's Test. — Take a loopful of six hours' agar culture of
suspected cholera bacilli, and add it to 1 c.c. of ordinary broth con-
taining '001 c.c. of anti-cholera serum (see p. 423). The mixture
is injected intra-peritoneally into a guinea-pig of 250 grammes.
In 20-30 minutes a drop of peritoneal fluid is withdrawn and
examined microscopically for comma bacilli, when, if the reaction is
positive, it will be found that the spirilla have broken down into
granules.

3. Plague

This disease, like anthrax and leprosy, has a long historical
record behind it. As the Black Death, it decimated the population
of England in the fourteenth century, and visited the country
again in epidemic form in the middle of the seventeenth century,
when it was called the Great Plague. It is highly probable that
these two scourges and the recent epidemic in the East are all
forms of one and the same disease. As a matter of fact, it is
difficult to be sure what was the exact pathology of a number of the
grievous ailments which troubled our country in the Middle Ages,
but from all accounts bubonic plague and true leprosy were amongst
them. The former came and went spasmodically, as is its habit;
the latter dragged through the length of several centuries.

There are four chief varieties of plague : first, the bubonic form,
the most common and typical ; the lymph glands are chiefly affected
in the groin, the axilla, or the neck ; secondly, the septiccem,ic form
in which the bacillus reaches the blood ; thirdly, the pneumonic, in
which the lungs are mainly affected ; and fourthly, pestis minor, in
which the affection of the glands stops short of the septicemic stage,
and even the local symptoms are slight. There are certain symptoms
common to all forms of plague, when at all severe.

Symptoms of Plague.— An ordinary attack of plague usually
begins three to five days after exposure to infection. Such attack may
develop gradually, but, generally, there is sudden onset with much
fever, as indicated by a high temperature, rapid pulse, headache, hot

SYMPTOMS OF PLAGUE 389

skin, and thirst. The eyes are injected as if inflamed ; the expression,
at first haggard, anxious and frightened, becomes subsequently vacant,
listless, and dull ; the utterance is thick, and the gait unsteady as in
one under the influence of drink. Mental aberration develops quickly.
There is frequently a marked tendency to faint. The tongue is at first
covered with a moist white fur except at the edges, which are red,
but later on it becomes dry and of a mahogany colour. Vomiting and
nausea are present from the onset. Sleeplessness is a characteristic
symptom.

The most distinctive sign of plague is the presence of swellings,
or " buboes " as they are called, in the groin, armpit, or neck. These
" buboes," which? led to the disease being called " bubonic plague,"
and which have no relation to venereal complaints, appear as a rule
on the second or third day of the disease. They occur as large,
smooth, tense swellings. They are usually painful and tender on
pressure, and in size they vary from that of an almond to that of
an orange. Later on they may "gather" and burst like an ordinary
abscess. There may appear about the body purple spots, and even
" carbuncles."

But buboes are not an essential feature of plague. Cases occur
in which these manifestations of the disease are greatly delayed or
even absent, as, for instance, in " pneumonic," " gastric," and " septi-
cemic" plague; forms of the malady which may be mistaken re-
spectively for inflammation of the lungs, typhoid fever, or acute
blood poisoning. Plague in these forms is always grave; not only
because of the fatality of the cases but for the reason that they,
especially the " pneumonic," are highly infectious to other persons.
It is important, therefore, that in localities where plague is present
or is threatened, cases of anomalous illness of the above sorts lie
without loss of time brought under medical supervision.

Besides the forms of plague already referred to, there is yet
another, namely, the so-called "ambulant form." In plague of this
description the affected person is hardly ill at all, presenting no definite
symptoms perhaps beyond indolent, though painful, swellings in
groin or armpit. Such plague cases may nevertheless be instrumental
in spreading the disease, and any persons therefore who, having
been possibly exposed to plague, exhibit these symptoms, should be
isolated and watched medically until the nature of their malady has
been definitely ascertained.

The sudden onset, the marked prostration, the mental aberration,
the splitting headache, vomiting and nausea, backache, the rise in
temperature, the furred tongue, when taken in conjunction with
tenderness and pain in some one of the groups of glands, are sufficient
to arouse suspicion as to the case being one of plague.

The distribution of plaque at the present time is fortunately a

390 THE ETIOLOGY OF TROPICAL DISEASES

somewhat limited one, namely, a definite area in Asia known as the
" Plague Belt." From Mesopotamia, as a sort of focus, the disease
spreads northwards to the Caspian Sea, westwards to the Eed Sea,
southwards as far as Central India, and eastwards as far as the
China Sea. This constitutes the "belt," but the disease may take
an epidemic form, and is readily, though very slowly, conveyed by
infection or contagion. It appears to be infectious by means of
infective dust, and contagious by prolonged and intimate contact
with the plague-stricken.

Rats and Plague. — Eats have been shown to be the agents
for conveying the disease from port to port, and even infecting
man. It is probable that rats are not the only agency acting
in this way.* Nevertheless, it is true that rats contract the
disease more readily than any other animals, and that when
suffering from it they may spread the infection. How it is
thus spread is not known. Cantlie and Yersin have pointed
out that previously to an epidemic of plague rats die in enormous
numbers, and Manson has declared that rats supply " the best and
most probably the initial opportunity" for the bacillus of plague.
" Were I asked," he continues, " how I would protect a state from
plague, I would certainly answer, exterminate the rats as a first and
most important measure." But, to be effective, this measure must be
employed in anticipation of the advent of the disease. " When the rats
are tumbling about drunk with plague it is too late." We may quote
Sir P. Manson's simile of the position of the rat in epidemic plague.
" I would compare a plague-threatened, but as yet not invaded, city,"
he says, " to a grate in which the fire is laid all ready for lighting.
There is the refractory though combustible coal on top, there is the
greasy paper and dry, resinous, inflammable wood underneath, and
there is the lighted match ready to be employed. Drop the match
on the top of the coal ; it flickers for a second and goes out — the
coals do not catch fire. But apply it to the paper and sticks under-
neath, and in a moment there is a blaze : the sticks are consumed,
the coals catch, and in a little while the fire burns merrily. The
coals will now burn by themselves, or, if they threaten to go out,
another stick or two will quickly revive the fire. In my simile the
coals stand for the human inhabitants, the sticks for the rodent
inhabitants, and the lighted match for the plague germ that has
dodged the quarantine intended to protect that threatened city. No
sticks, no fire ; no rats, no plague epidemic." f

Dr Doriga, the Principal Medical Officer of Health for Venice,
has set forth a brief rhumt of the chief facts relating to the

* See Indian Plaque Commission Report, 1902, and Report on Plague at Sydney,
1903 (Ashbnrton Thompson).

I Brit. Med. Jour., 1899, vol. ii., p. 924.

RATS AND PLAGUE 391

agency of rats and mice in the spread of plague, which is as
follows : — *

" 1. Kitasato and Yersin, and many others after them, have found
the specific bacillus of plague in the dead bodies of rats and mice
collected in houses in which cases of the disease subsequently broke
out among the occupants, or in the streets of infected towns. They
have also placed beyond question the great susceptibility of these
rodents to the bacillus.

"2. In all the towns of India manifest examples of contagion
from mice to men have been observed. At Bombay, in certain
establishments where the dead bodies of rats were found, it has been
noticed that the persons who collected them alone contracted plague,
although many other work-people were engaged at the same place.

" 3. The first cases of the disease have sometimes appeared in
warehouses where wheat, cotton seed, or other substances likely to
attract rats were stored. At Kurachee, where the warehouses are
situated in streets without dwelling-houses, the first sufferers were
the caretakers.

" 4. Well-constructed and well-maintained houses, i.e. where rats
cannot find harbour, nearly always remain free from plague. This
same immunity was demonstrated by Eennie at Canton, in 1894,
among the occupants of boats anchored in the river. On the other
hand, is to be observed the permanence of infection in the houses of
poor natives, notwithstanding the removal of the residents and
furniture and the most rigorous disinfection, because of reinfection
by means of mice.

" 5. The epidemics at Bombay, Kurachee, and Karad were
chiefly localised in quarters where the disease had broken out
amongst rats. The spread of infection in other parts of these same
towns was regularly preceded by the immigration and death of rats,
and its diffusion always corresponded to the route of travel taken by
these rodents in their migrations.

" 6. In healthy countries adjoining infected, the disease broke out
amongst the inhabitants without the importation of a single (human)
case, but was preceded by the immigration of rats from an infected
place.

"7. In many countries and towns the development of the
epidemic among the inhabitants followed a month after the importa-
tation of the first cases, or after the death of fugitives arriving from
infected localities. During the interval the plague had been propa-
gated by mice.

" 8. Lastly, the mode of infection and propagation of plague on
certain ships proved that the rats on board had been the vehicles of

* "The Prevention of Plague through Suppression of Rats," Revue, d'Hygiene,

August IS 99.

392 THE ETIOLOGY OF TROPICAL DISEASES

contagion." * It should be added that Doriga's views are not universally
held, and were not fully accepted by the Indian Plague Commission.

Quite recently, attempts have been made in Paris, with the con-
sent of the Prefect of the Seine, to exterminate rats wholesale, in
order to protect the city from an epidemic of plague. ■[•

The Bacteriology of Plag'ue is one of the latest additions to the
science. During the Hong Kong epidemic in 1894, Kitasato, of
Tokio, demonstrated the cause of plague to be a bacillus. This was
immediately confirmed by Yersin, and further proved by the isola-
tion in artificial media of a pure culture of a bacillus able, by means
of inoculation, to produce the specific disease of bubonic plague.

The bacillus was first detected in the blood of patients suffering
from the disease. It takes the form of a small, round -ended, oval
cell (0*7 juL broad by 1*5 jul long), with marked polar staining, and
hence having an appearance not unlike a diplococcus. In the middle
there is a clear interspace, and the whole is surrounded with a thick
capsule, stained only with difficulty. The organisms are often
linked together in pairs or even chains (especially in fluid cultures),
and exhibit polymorphic forms. In culture the bacillus may be
coccal or bacillary in form. Involution forms occur in old cultures,
and also, more rapidly, when 2-5 per cent, of sodium chloride is
added to the medium. On such salt-agar the involution forms are
very marked. The bacillus is non-motile (Plate 30, p. 398).

The plague bacillus grows readily on the ordinary media at blood-
heat, producing smooth, shining, circular cream-coloured colonies,
with a wavy outline, which eventually coalesce to form a greyish
film. The colonies slip about on the agar when touched with the
platinum wire. If melted butter (or ghee) or oil be added to
bouillon, this bacillus grows in " stalactite " form, that is, the growth
starts on the under surface of the fat globules, and extends down-
wards in the form of pendulous string-like masses which readily
break off if the tube is slightly shaken. The following negative
characters help to distinguish the bacillus : There is no growth on
potato ; milk is not coagulated ; gelatine is not liquefied ; Gram's
method does not stain the bacillus ; and there are no spores. The
bacillus is readily killed by heat or by desiccation over sulphuric
acid at 30° C. Both in cultures and outside the body the bacillus
loses virulence. To this may be attributed possibly the variety of
forms of the plague bacillus which differ in virulence. But it has
great powers of resistance against cold.J

On gaining entrance to the human body the bacillus affects in

* For a general discussion of the subject of plague in the lower animals, see Brit.
Med. Jour., 1900, i. pp. 1141 and 1216.

1 Brit. Med. Jour., 1900, vol. i., p. 722.

j For further particulars as to cultured characters of B. pesiis, see Brit. Med Jour. ,
1902, ii., 956,

THE PLAGUE BACILLUS 393

particular two organs, the spleen and the lymph glands (buhonic
plague). The latter become inflamed in groups, commencing gener-
ally with the inguinal (60 per cent.) followed by the axillary (17 per
cent.). The buboes consist usually of masses of inflamed and enlarged
lymph glands, attended with haemorrhage and often with necrotic
softening. The spleen suffers from inflammatory swelling, which
may affect other organs also. In both places the bacilli occur in
enormous numbers. In the pulmonary form the lung is affected with
broncho-pneumonia. This form of plague is said to be always fatal.
Kitasato considers that the bacillus may enter the body by the three
channels adopted by anthrax, namely, (a) the skin, (b) alimentary
canal, and (c) respiratory tract. But the vast majority of cases arise
from infection through the skin. Infection through the alimentary
canal is still doubtful. Soil, clothes, and contaminated articles gener-
ally are the agencies of infection. As stated already, rats play an
important part in the propagation of the disease. The Indian Com-
mission hold that suctorial insects are practically of no importance
as transmitters of infection.

Haffkine has prepared a vaccine to be used as a prophylactic (see
p. 424), and the Indian Plague Commissioners have recently reported
on its effects. Inoculation with this vaccine appears sensibly to
diminish the incidence of the plague attacks on the inoculated popu-
lation, although the degree of protection is not perfect. The disease
is four times more numerous among the uninoculated than among
the inoculated. The fatality of the attacks is also diminished in the
inoculated. Protection does not begin till a few days after the
inoculation, but it lasts many weeks and even months. It may here
be added that the means of stamping out plague are the ordinary
methods of notification, isolation, and disinfection. The latter should
include destruction of the patient's clothes, and the scraping of the
walls, and, in India, burning of the earthen floor of his dwelling.
The soil and dwellings are among the chief sources of infection,
and therefore require most attention.

As to the infectivity of plague, it is now generally held that the
bubonic form is, as a rule, dangerous from the excretions, and only
in the last stages of the disease; that the primary pneumonic form
is highly infective ; that houses in which plague patients or plague
rats have died, and in which clothing has been soiled by excretions,
are infective ; and that there is much more danger from living in an
infected house than from coming into contact with a plague patient.
Plague is a disease which is specially favoured by insanitation within
the walls of houses as contrasted with insanitation outside houses,
relating, for example, to drainage, removal of refuse, etc. Eats,
merchandise, clothing, etc., may each and all play a part in the con-
veyance of plague from one village to another, or one country to

394 THE ETIOLOGY OF TROPICAL DISEASES

another. But the chief agency for spreading the disease to unin-
fected places consists of travellers. The lines of human communica-
tion are followed by the infection in a marked degree, especially lines
of steamship and railway.

Plague is essentially a " filth disease/' and it is frequently pre-
ceded by famine. Temperature and overcrowding exert an influence
upon it. The areas affected by the disease in the Middle Ages in the
seventeenth century, and in 1894-96, are alike in being characterised
by filth and overcrowding. There is little fear, speaking generally,
of the plague ever flourishing under Western civilisation, where the
conditions are such that even when it appears there is little to
encourage or favour its development.*

Administrative Considerations. — Plague will not readily
fasten on any section of a population which is properly housed,
cleanly, and generally, in a sanitary sense, well-to-do ; rather it will
especially affect, if it obtains foothold in a district, insanitary areas
such as are peopled by the poorest class, and where overcrowding of
persons in houses and dirt and squalor of dwellings and of inhabi-
tants tend to prevail.

In these circumstances, and from an administrative point of
view, the following facts respecting plague should be borne in
mind : —

(1) Plague has an incubation period of three to five (in excep-
tional cases of perhaps eight to ten) days.

(2) Plague is wont, especially in its earlier manifestations, to
assume a mild form, or even to present anomalous symptoms,
tending to confound it with other and more innocent diseases.

(3) Plague in all its forms must needs be regarded as personally
infective.

(4) Plague affects rats as well as the human subject ; it may,
indeed, be found, causing mortality among these lower animals ante-
cedent to its definite invasion of the population. There can be no
doubt that the rat and man are, as regards plague, reciprocally
infective.

Although Local Authorities should be on their guard against
plague, when cases occur at the ports or elsewhere in these islands,
it is not intended to suggest that there exists, under these circum-
stances, cause for alarm. There can be no doubt that, in this
country, hygienic conditions and methods of dealing with infectious
diseases are far in advance of those of former centuries wherein
plague was repeatedly epidemic in our populations; they are in
advance, too, of those in localities abroad, where plague has shown
itself formidable in recent years. During the past fifty years there

* The most complete account of plague hitherto published is The Report of the
Indian Playue Commission, 1902, vols, i.-viii. (see in particular vol. v.).

THE CONTROL OF PLAGUE 395

has occurred in England and Wales a large diminution in the
mortality from most diseases of the infectious class, and in the same
period typhus fever has declined almost to extinction. This latter
disease is that which, as regards the conditions under which it
becomes prevalent, most closely resembles plague. Wherefore it
may be confidently anticipated that the measures of sanitary
improvement, of isolation and of disinfection, which have been
found effectual against indigenous disease such as typhus, will, if
promptly and thoroughly brought to bear, be equally effectual
against plague.

First among measures requisite for control of plague is prompt •
notification to the local authority of all cases of the disease
occurring in their district. As a rule, the first cases of an outbreak
will require bacteriological diagnosis in addition to or auxiliary to
clinical diagnosis.

Secondly, in the event of plague being detected in any district,
the measures to be taken to prevent its spread are, generally
speaking, those which are available against the more ordinary
epidemic diseases. These measures include prompt removal of the
sick persons to hospital and their isolation therein ; the destruction
or thorough disinfection of all infected articles, with the effectual
disinfection also of the invaded dwelling-place ; the keeping under
observation during ten days after detection of each plague case all
persons who have been in contact with the patient ; house to house
visitation for the discovery of unreported or suspicious cases; the
abatement as speedily as possible of all insanitary conditions in the
locality which may tend to the spread of the disease; and, in the
case of death, the prompt disposal of the body, with all due precau-
tions against its becoming a source of infection.

Thirdly, an essential measure of precaution, in view of the
observed relation between plague in rats and plague in the human
subject, will be the prompt destruction of all rats in districts
threatened or invaded by plague, care being taken that their
carcases are collected and burnt without being unduly handled.*

When treated in a well-appointed hospital, with plentiful fresh
air and proper attention to cleanliness and disinfection, plague, except
in its pneumonic and septicemic forms, shows but small infective
power ; and that therefore doctors and nurses in attendance on the
sick run but little risk of contracting the disease. Nevertheless, these
and other persons brought into close relation with plague may be
afforded protection against infection by submitting themselves to
protective inoculation ten days before contact with plague cases.

* Danysz has suggested the killing of rats by infecting them with an organism
fatal to them, Brit. Mad, Jour., 1904, vol. i., p. 947.

396 THE ETIOLOGY OF TROPICAL DISEASES

Divert ions for Obtaining and Forwarding for Bacterioscopic Examina-
tion, Material from Suspected Plague Cases

A. — From the Living Person.

1. Clean with soap and water, and then with alcohol, the last phalanx of either
the second or third finger. When dry, or after mopping with a clean cloth, put a
piece of tape round the proximal end of the last phalanx, so as to cause venous con-
gestion. Prick the palmar surface of this phalanx with a sterile needle, and
immediately take up the exuding blood in two sterile capillary tubes such as are
used for collecting vaccine lymph. These tubes when charged should be sealed at
both ends.

2. When there is a discharging bubo, collect fluid therefrom in capillary tubes, as
in the case of blood. When this discharge is not of a sufficiently fluid character for
collection in this way, place some of it in a small glass-stoppered phial, previously
well washed out with alcohol, care being taken that no alcohol remains in the
phial.

3. If expectoration be obtainable, collect some in a phial in the manner pre-
scribed in section 2.

[In blood, discharge, or expectoration, cover glass preparations should be made
and stained by simple stains, and by Gram's method. The plague bacillus does not
stain by Gram's method. Cocci, streptococci, and diplococcus pneumonia? do stain
by Gram's method. Cultivations and inoculations must also be made.]

B. — From the Dead Body.

1. Cut out any inflamed lymph gland, together with some of its surrounding
tissue, and place the whole in a wide-mouthed glass-stoppered bottle previously
well washed out with alcohol, care being taken that no alcohol remains in the bottle.
The bottle should have the stopper well secured and sealed.

2. Obtain also a piece of the spleen, dealing with it in the same manner.

All suspected plague material should be carefully packed, so as to avoid risk of
breakage.

4. Leprosy

This ancient disease is said to have existed in Egypt 3500
B.C., and was comparatively common in India, China, and even in
parts of Europe 500 B.C. We know it has existed in many parts
of the world in the past, in which regions it is now extinct. Some
of the earliest notices we have of it in this country come from
Ireland, and date back to the fifth and sixth centuries. Even at
that period of time also various classical descriptions of the disease
had been written and various decrees made by Church councils to
protect lepers and prevent the spread of the disease, which was
often looked upon as a divine visitation. In the tenth century
leprosy was prevalent in England; it reached its zenith in the
thirteenth century, or possibly a little earlier, and declined from
that date to its extinction in the sixteenth. But even two hundred
years later leprosy was endemic in the Shetlands, and it is recorded
that in 1742 there was held a public thanksgiving in Shetland on
account of the disappearance of leprosy. The last leper living in the
Shetlands was admitted to the Edinburgh Infirmary in 1798.

LEPROSY 397

At one time or another there were as many as 200 institutions
in the British Isles for the more or less exclusive use of lepers.
Many of these establishments were of an ecclesiastical or municipal
character, and owing to the fact that diagnosis was not accurately or
carefully made, it is certain that these institutions frequently housed
persons suffering from diseases other than leprosy. Bury St
Edmunds, Bristol, Canterbury, London, Lynn, Norwich, Thetford,
and York were centres for lepers. Burton Lazars and Sherburn,
in Durham, were two of the more famous leper institutions.

Elsewhere, the writer has furnished the evidence obtainable in
support of the view that true leprosy {elephantiasis grcccorum) was
prevalent in England in the Middle Ages.* It existed there anterior
to the crusades, which per se had little or no effect in spreading the
disease in England. It was generally supposed from the eleventh
century that leprosy was a contagious and hereditary disease, and
that it depended upon these two characters for its extension in
England. But probably such was not the case, for it is fairly certain
that strict segregation was never carried out. The disease as an
endemic disease reached its zenith in the thirteenth century or
earlier, and declined till final extinction in the eighteenth. In
England itself it disappeared approximately in the sixteenth century.
Probably the famine of 1315, and the Black Death of 1349, materi-
ally assisted in the extermination of lepers. The disease being
diffused neither by contagion nor heredity has under favourable
hygienic circumstances a tendency to die out. Hence the decline
and final extinction of leprosy in Great Britain was due to this
general tendency under favouring circumstances, viz., to an extensive
social improvement in the life of the people, to a complete change in
the poor and insufficient diet, and to general sanitary advancement.

At the present time the distribution of the disease is mostly
Asiatic. Norway contains about 1200 lepers, Spain a smaller
number. Scattered through Europe are perhaps another 2000 to
3000, in India 100,000, and a number in Japan. The Cape
possesses a famous leper hospital on Eobben Island, with a
number of patients. The disease is also endemic in the Sandwich
Islands.

Descriptions of the pathological varieties of leprosy have been
very diverse. The classification now generally adopted includes
three forms : the tuberculated, the anaesthetic, or (maculo-aiuesthetic),
and the mixed. Lepra tuberculosa is that form of the disease
affecting chiefly the skin, and resulting in a nodular tuberculated
growth or a diffuse infiltration. It causes great disfigurement. The
anmsthetie form causes a destruction of the nerve fibres, and so

* The Decline and Extinction of Endemic Leprosy in the British Islands, 1895,
p. 108.

398 THE ETIOLOGY OF TROPICAL DISEASES

produces anaesthesia, paralysis, and what are called " trophic "
changes. Not infrequently patches occur on the skin, which appear
like parchment, owing to this trophic change. Bullse may arise.
When the tissue change is radical or far advanced, considerable
distortion may result. The mixed variety of leprosy, as its name
implies, is a mixture of the two other forms.

The Bacteriology of Leprosy. — The B. lejorce was discovered
by Hansen in 1874. He found it in the lepra cells in the skin,
lymph glands, liver, spleen, and thickened parts of the nerves. It
is common in the discharge from the wounds of lepers. It is
conveyed in the body by the lymph stream, and has rarely been
isolated from the blood (Kobner).

The bacillus is present in enormous numbers in the skin and
tissues, and has a form very similar indeed to B. tuberculosis. It is
a straight rod, and showing with some staining methods marked
beading, but with others no beading at all. It measures 4 ja long
and 1 ju broad. Young leprosy bacilli are said to be motile, but old
ones are not. Neisser has maintained that the bacillus possesses a
capsule and spores. The latter have not been seen, but Neisser holds
that this is the form in which the bacillus gains entrance to the
body. There is a characteristic which fortunately aids us in the
diagnosis of this disease in the tissues, and that is the arrangement
of the bacilli, which are rarely scattered or isolated, as in tubercle,
but gathered together in clumps and colonies. The bacilli occur
for the most part inside the round cells, but they are also found free
in the lymphatics, inside connective tissue cells, and in the walls of
blood-vessels. A few may often be found in the hair follicles or
glands of the skin, or even in the epithelium. The bacilli also occur
in the lymphatic glands and in the internal organs. The brain and
spinal cord are almost always exempt. But recent research has
made it evident that the distribution in the tissues may be more
widespread than was formerly supposed. Bordoni - Uffreduzzi,
Carrasquilla (1899) and Campana, claim to have isolated the bacillus
and grown it on artificial media, the two former aerobically on
peptone-glycerine blood serum, at 37° C, the latter anaerobically.
Other workers have been unable to obtain successful results. Culti-
vated bacteria from the organs of lepers, described rather later by
Babes, and still more recently by Czaplewski, differ from the genuine
bacillus of leprosy in their incomplete resistance to acids. Both
authors maintain that the bacteria cultivated by them resemble the
bacilli of diphtheria. In any case it is very doubtful whether these
bacteria cultivated from leprosy are the genuine Bacillus lejorce.
Hence it is not possible to study the bacteriology of leprosy at all
completely ; inoculation experiments also have not proved successful.
Nevertheless there is little doubt that leprosy is a bacterial disease

PLATE 30.

v

t

,*

s

Bacillus lepra',.

Discharge from sore of leper. Stained by

Ziehl-Neelsen method.

x 1000.

gar

Bacillus lepra*.

,epra cells containing bacilli. From lobule of ear of leper.

Stained by Ziehl-Neelsen method.

x louo.

►•

Bacillus of Plague (/;. pestis bubontcce).

From liver of plague-stricken rat.

x 1000.

Staphylococcus pyogt nes aurt as.

x 1000.

[To face page 398.

LEPROSY 399

produced by the bacillus of Hansen. Bordoni-Uffreduzzi maintains
that the parasitic existence of the B. lejprce may alternate with a
saprophytic stage. This may be of importance in the spread of the
disease. There is evidence in support of the non-communicability of
the disease by heredity or contagion. Segregation does not appear
always to result in a decline of the disease, as we should expect if
it were purely contagious. Ehlers, of Copenhagen, has, however,
as recently as 1897 reaffirmed his belief in the contagiousness of
leprosy ; Yirchow, on the other hand, declared that it was not highly
contagious. There is evidence to show that persons far advanced
in the disease may live in a healthy community, and yet not infect
their immediate neighbours. Indeed, the transmission of the disease
is still an unsolved problem. Mr Hutchinson maintains that diet,
particularly uncooked or putrid fish, is a likely channel. Deficiency
of salt, telluric and climatic conditions, racial tendencies, social
status, poverty, insanitation, drinking-water, even vaccination, have
all secured support from various seekers after the true channel by
which the bacillus gains entrance to the human body. The real
mode of transmission is, however, still unknown. The decline and
final extinction of leprosy in the British Islands was, as we have
stated, probably due in part to the natural tendency of the disease
to die out, and in part to a general and extensive social improvement
in the life of the people, to a complete change in the poor and
insufficient diet, and to general sanitation.

At the Leprosy Congress held in Berlin in 1897, Hansen again
emphasised his belief that segregation was the cause of the decline
of leprosy wherever it had occurred. But there appears to be
evidence to show that leprosy has declined where there has been no
segregation whatever, and therefore, however favourable to decline
such isolation may be, it would seem not to be an actually necessary
condition. At the same Congress Besnier declared in favour of the
infective virus being widely propagated by means of the nasal
secretion. Sticker states that the nasal secretion contains myriads
of lepra bacilli, especially in the acute stages of the disease, and
Besnier and Sticker have pointed out how frequently and severely
the septum nasi and skin over the nose is affected in leprosy. Several
leprologists in India have recorded similar observations. These facts
appear to support Besnier's contention, that the disease is spread by
nasal secretion.

We may add here the conclusions arrived at by the English
Leprosy Commission * in India : —

" 1. Leprosy is a disease sui generis ; it is not a form of syphilis

; Dated 1890-91. The Commissioners were the late Beaven Rake, M.D., G. A.
Buckmaster, M.D., the late Prof. Kanthack, of Cambridge, the late Surgeon-Major
Arthur Barclay, and Surgeon-Major S. J. Thomson.

400 THE ETIOLOGY OF TROPICAL DISEASES

or tuberculosis, but has striking etiological analogies with the
latter.

" 2. Leprosy is not diffused by hereditary transmission, and, for
this reason and the established amount of sterility among lepers, the
disease has a natural tendency to die out.

" 3. Though in a scientific classification of diseases, leprosy must
be regarded as contagious, and also inoculable, yet the extent to
which it is propagated by these means is exceedingly small.

" 4. Leprosy is not directly originated by the use of any particular
article of food, nor by any climatic or telluric conditions, nor by
insanitary surroundings, neither does it peculiarly affect any race
or caste.

" 5. Leprosy is indirectly influenced by insanitary surroundings,
such as poverty, bad food, or deficient drainage or ventilation, for
these by causing a predisposition increase the susceptibility of the
individual to the disease.

" 6. Leprosy, in the great majority of cases, originates de novo,
that is, from a sequence or concurrence of causes and conditions dealt
with in the Eeport, and which are related to each other in ways at
present imperfectly known."

The practical suggestions of the Commission for preventive
treatment included voluntary isolation, prohibition of the sale of
articles of food by lepers, leper farms, orphanages, and "improved
sanitation and good dietetic conditions " generally. Serum-therapy
has been attempted on behalf of the French Academy of Medicine,
but without success. Many forms of treatment ameliorate the
miserable condition of the leper, but up to the present no curative
agent has been found.

5. Yellow Fever

This disease is admitted to be one of the most terrible of tropical
diseases. Fortunately, its area of endemicity is comparatively
limited. When, however, it breaks out, especially on board ship,
its high percentage of fatality is well known.

A number of investigators, from the beginning of last century
down to the present time, have been at work on the cause of yellow
fever. Sanarelli, the Director of the Institute of Hygiene, in the
University of Montevideo, in South America, is one of the more
recent workers, and he has isolated a bacillus which he believes is
the causal agent of the disease. The bodies of those who die of
yellow fever are, however, either so free from organisms, or so entirely
invaded by organisms, that the B. icteroides is difficult to discover.
Moreover, led by the clinical signs of the disease — " black vomit " and
other gastro-intestinal phenomena — investigators have, a priori,
supposed that the digestive canal was the seat of the disease, and

YELLOW FEVER 401

therefore the prohable locality of the causal bacillus; whereas, as
Sanarelli pointed out, the B. icteroides must be sought for in the blood
and tissues, and not in the alimentary canal. But even thus the
difficulties are not wholly removed. For it happens that this
organism may only be found in comparatively small numbers, and
certainly at the beginning of the disease multiplies very little in
the human body. Its influence upon the body, too, appears to be
such that the tissues of a yellow fever patient become the hunting
ground of vast numbers of secondary infective bacteria.

This bacillus (B. icteroides) may be obtained from the small
capillaries — in, say, the liver — by incubation at favourable tempera-
ture (37° C). It is a short bacillus with round ends, like B. coli. It is
motile, and possesses 4-8 flagella. It develops sufficiently well for all
practical purposes on the ordinary media. On agar at blood-heat it
grows well — a grey, iridescent, smooth layer, with regular margins ;
and on the same medium, at the temperature of the room, it
produces in twenty-four hours characteristic colonies not unlike drops
of milk. It grows on gelatine without liquefaction. The organism
is a facultative anaerobe, decolorised by Gram's method; ferments
sugar, but does not coagulate milk until after some weeks. It
appears strongly to resist drying. Direct sunlight kills it in seven
hours ; but it is said to be able to live for some time in sea water.
The organism can be isolated from the living patient as well as the
dead body.*

Sanarelli has maintained that atmospheric transmission is the
common channel of infection in yellow fever. As everyone knows,
it is a disease which, when once installed on board ship, seems to
cling to it tenaciously, more particularly in the hold, magazines,
merchandise, and in all close and restricted quarters. Humidity,
heat, and want of light and ventilation have been, until recently, the
supposed conditions necessary to the conveyance or harbouring of
yellow fever. Sanarelli has further suggested that moulds must be
considered "the natural protectors of the specific agent of yellow
fever."-)- By a series of interesting experiments, he demonstrated
the stimulating effect which moulds have upon gelatine-plate cultures
of this bacillus in the laboratory. Outside the laboratory, in houses
and on ships, the conditions favouring the growth of moulds appeared
also to be the conditions favouring yellow fever. For instance,
humidity, heat, and scanty aeration are highly favourable to mould
growth, and thus, according to Sanarelli, to yellow fever. To these
factors, also, is supposed to be due the unhealthiness of Eio Janeiro.
During the yellow fever epidemic in Montevideo in 1872, the

* Brit. Med. Jour., 1897, vol. ii., p. 7 (Prof. G. Sanarelli), 1900, vol. i., p. 334,
and The Medical News (New York), 9th December 1899.
t Brit. Med. Jour., 1897, vol. ii., p. 11.

2 C

402 THE ETIOLOGY OF TROPICAL DISEASES

inhabitants of the houses facing north were attacked much more
than others, and it was found that both these houses and the streets
in which they stood were distinguished by an exceptional degree of
humidity.

In 1901 the United States Army Commission reported, after
extensive investigations into the etiology of yellow fever, that whilst
B. icteroides was not always present in cases of yellow fever, the
blood of the patient appeared to contain the virus, whatever it was,
and retained it after being passed through a Berkefeld filter. The
Commission further reported that the disease was not communicable
by direct contact with those suffering from the disease, but was
probably communicated by mosquitoes in a similar way to malaria.
The species of mosquito found capable of carrying the infection in
this way is the Stegomyia fasciata. Though the matter was not
proved, nor the nature of the virus determined, preventive mea-
sures were adopted in Havana on the mosquito hypothesis, with
the remarkable result that the disease was stamped out.
Guiteras of Havana has carried out further experiments which
confirm many of the Commission's findings, and, in particular, the
transmission of the disease by mosquitoes.*

In 1902 a United States Army Expedition was appointed to
reinvestigate the subject, and it reported in 1903. The chief
conclusions reached were as follows : (1) Bacteriological examination
of the blood of persons with uncomplicated yellow fever during life,
as well as of organs and blood immediately after death, is negative.
(2) The mosquito known as Stegomyia fasciata, when allowed to suck
the blood of a yellow fever patient after the lapse of forty-one hours
after the onset of the disease, and subsequently fed on sugar and
water for twenty-two days can, if permitted to bite a non-immune
person, produce a severe attack of the disease. (3) Stegomyia fasciata,
contaminated by sucking the blood of a yellow fever patient, and
then killed, cut into sections and appropriately stained, presents with
regularity a protozoan parasite, Myxo-coccidium stegomyia?, which
can be traced through a cycle of developments from the gamete to
the sporozoite. (4) Stegomyia fasciata, fed on the blood of a person
with malarial fever, on normal blood, or artificially, does not harbour
the myxo-coccidium.

The etiology of yellow fever, therefore, remains at the present
time sub judice, but the probabilities are that the disease is mosquito-
borne and due not to bacteria but to sporozoal parasites.

There are other tropical diseases to which brief reference must be made.
Malta Fever (Mediterranean fever) is common along the coast of the Mediter-
ranean and on its islands. It also occurs elsewhere. In 1886 Bruce cultivated from

* American Medicine, 23rd November 1901, p. 809.

MALTA FEVER, SLEEPING SICKNESS, ETC. 403

the spleen of patients dead of the disease an organism now known as the Micrococcus
melitensis. Clinically, Malta fever is a disease of long duration and variable
symptoms, including remitting fever. Perspiration, pains, swelling of joints, en-
largement of the spleen, etc., are among the common signs. Micrococcus melitensis
is a small, round, or slightly oval coccus, singly, in pairs, or chains. Does not stain
by Gram's method. Can be cultivated on agar at 37° C. from the spleen ; colonies
appear about third day as small, round, slightly raised growths, old cultures assume
a buff tint. Addition of nutrose hastens growth of culture. On gelatine growth is
very slow ; there is no liquefaction. In broth there is a turbid growth, without
pellicle formation.

Cultures kept at 22° C. retain their vitality for fourteen months, but the organism
dies in about five days in sterile, fresh, and sea water and urine, but remains active
for longer in sterile milk, and for sixty-nine days in dust. The organism is present
in the peripheral blood in all cases during the early stages, and in severe pyrexial
relapses. It has recently been isolated from the urine of patients. The disease
appears to be inoculable in animals.

Sleeping Sickness. — Investigations point to the conclusion that sleeping-sickness
is caused by the entrance into the blood, and thence into the cerebro-spinal fluid, of
a species of trypanosoma (probably the Trypanosoma gambiene, discovered by Forde
and described by Dutton), which is transmitted from the sick to the healthy by a
species of tsetse fly (Glossina palpalis), and by it alone ; that, in short, sleeping-
sickness is a human tsetse-fly disease. From a series of carefully controlled and
minutely observed experiments, carried out by Bruce, Nabarro, and Grieg, it was
discovered that monkeys inoculated with cerebro-spinal fluid from sleeping sickness
patients, or with blood from natives not as yet showing symptoms of sleeping sick-
ness, but containing a similar parasite, sickened and died with all the symptoms of
sleeping sickness.

From the analogy of the closely related disease in cattle, the nagana or tsetse fly
disease of South Africa, it was suspected that in sleeping sickness a like method of
infection took place. It has been demonstrated by experiment that not only were
these flies, fed on sleeping sickness cases, capable of conveying the disease to healthy
monkeys, but that the freshly caught flies from an infected area, without any arti-
ficial feeding, were also capable of conveying the disease.

It was further discovered by a carefully-organised investigation that this fly, like
its congener the tsetse fly of South Africa, is confined to well-defined areas, and that
these areas correspond absolutely with the distribution of sleeping sickness ; whereas,
in regions where no Glossina palpalis is found, although other biting flies abound,
there is no sleeping sickness. Moreover, an examination of a large number of
individuals in the sleeping sickness areas and the non-sleeping sickness areas
respectively, revealed the fact that, while a large percentage (28) of the inhabitants of
the sleeping sickness areas have in their blood the trypanosoma already referred to,
in not a single case taken from inhabitants of non-sleeping sickness areas was this
parasite found.* The only other human trypanosome at present known is that
occurring in trypanosomiasis.

Dysentery is another tropical disease in which the etiology has not been finally
worked out. Endemic or tropical dysentery is possibly due to Amoeba coli. Epi-
demic dysentery is more probably due to Bacillus dysenteric?, and sporadic and
parasitic dysentery is due to various parasites, such as Balantidium coli and the
Bilharzia. B. dysenteric is a short rod, often occurring in pairs ; non-motile ; does not
stain by Gram's method ; does not curdle milk nor liquefy gelatine, t The researches

* See also Brit. Med. Jour., 1903, vol. i., p. 1431; and vol. ii., pp. 1343 and
1427 ; and Lancet, 1904 (July), p. 290, for a resume.

f For a full discussion of the subject, see Brit. Med. Jour., 1901, vol. ii., p. 786,
"A Comparative Study of the Bacilli of Dysentery"; and 1903, vol. i., p. 1315,
" Amoebic Dysentery in India " ; and Report of Royal Commission on Dysentery,
1903. For description of B. dysenteric, see Report of Medical Officer to Local Govt.
Bd., 1901-02, p. 396; Brit. Med. Jour., 1904, vol. i., p. 1002; Edin. Med. Jour.
(June), 1904, p. 489 (Eyre).

404 THE ETIOLOGY OF TROPICAL DISEASES

of Shiga, Kruse, Flexner, and others point to acute dysentery being caused by the
specific bacillus B. dysenteries, or some member of that group of organisms. Mott
holds that "asylum dysentery" is identical with tropical dysentery, and both
conditions are in all probability of bacillary origin."

Beri-beri. — The first medical writer to describe beri-beri, and by that name, was
Dr J. D. Malcolmson, F.R.S., of the Madras Medical Service, in a paper published
in 1835. Sir Joseph Fayrer, F.R.S., wrote on it, identifying one form of it with
the barbiers of the earlier European travellers.! The disease is endemic in Western
India, in the Indian Archipelago, and throughout the coasts of Further India and
Upper India, or China and Japan. It is practically confined to the labouring classes
where they are vegetarians. Dr Wallace Taylor traces it to a microscopic spore
infecting rice ; and other observers consider it a " place disease." The salient fact
is that it almost exclusively attacks those who are engaged in hard labour on insuffi-
cient nourishment, and it may be defined as the scurvy of the tropics. It is marked
by extreme weakness and dropsical distension of the abdomen, limbs, and face, both
symptoms developing so rapidly as to alarm alike the sufferer and those attending to
him. Hence its name beri, meaning " debility," and the reduplication of it, beri-beri
signifying " extreme," " alarming," " fatal," debility.

The disease is in all probability a germ disease but possibly not communicable
from man to man. It may be that the germ resides in soil or rice, or houses, and
surroundings of beri-beri localities, and produces a toxin which on being absorbed
produces a disease having many similarities with alcoholic neuritis (Manson). This
may be the explanation of the view that beri-beri is a " place disease. " Pekelharing
and Winkler hold that they have isolated a bacterium which is the cause of the
disease, but their views nave not been generally accepted.

* See also Bacteriological and Clinical Studies of the Diarrheal Diseases of Infancy \
by Flexner and Emmett Holt (Rockefeller Inst. Rep.), 1904.

t Practitioner (January), 1877 ; see also Report on Prison Administration in
Burma, 1878.

CHAPTEK XII

THE QUESTION OF IMMUNITY AND ANTITOXINS

Bacterial Products — Toxins — Question of Immunity — Kinds of Immunity — Theories
of Immunity — Applications of Immunity— Vaccination for Small-pox : Effect
of Vaccination — Pasteur's Treatment for Rabies — Inoculations for Cholera,
Typhoid, and Plague — Antitoxin Treatment of Diphtheria and its Effects.

The term natural immunity is used to denote natural resistance to
some particular specific disease. It may be due to species of animal,
or age, or individual idiosyncrasies. We not infrequently meet with
examples of this freedom from disease. Certain species of animals
do not, as a rule, take certain diseases. For example, cholera and
typhoid, which affect man, do not affect the lower animals. Swine
plague, which affects swine, does not affect man. The white rat
is immune to anthrax, which readily attacks cattle. Such examples
might easily be multiplied. Children, again, are susceptible to
certain diseases and insusceptible to certain others to which older
people are susceptible. The young of the lower animals also are
susceptible to diseases which do not attack adult animals. We
know, too, that some individuals have a marked protection against
certain diseases. Some persons coming in the way of infection at
once fall victims to the disease, whilst others appear to be proof
against it.

It is only in recent times that any intelligent explanations have
been offered to account for these phenomena. The most recent, and
that which appears to have most to substantiate it, is known as
immunity due to antitoxins. To understand the nature of antitoxins
it is necessary to consider briefly the products of bacterial activity.
They are chiefly seven : —

406 THE QUESTION OF IMMUNITY AND ANTITOXINS

1. Pigment. — We have already seen that many organisms exhibit
their energy in the formation of various pigments. These are,
as a rule, " innocent " bacteria. Oxygen is required for the pro-
duction of pigment by some of these species, absence of light by
others, and they all vary according to the medium upon which
they are growing. Eed milk, blue milk, and green pus are illus-
trations of materials owing their colour to pigment produced by
bacteria. Chromoporous bacteria are those in which the pigment is
diffused out into the surrounding medium; chromophorous bacteria
are those in which the pigment is stored in the cell protoplasm of
the organism.

2. Gas. — A large number of the common bacteria, like B. coli,
produce gas in their growth; hydrogen (H), carbonic acid (C02),
methane (CH4), sulphuretted hydrogen (H2S), and even nitrogen
(N) being formed by different bacteria.

3. Acids. — Lactic, acetic, butyric, etc., are common types of acids
resulting from the growth of bacteria.

4. Liquefying Ferment. — As we have seen, bacteria may also be
classified with regard to their behaviour in gelatine medium, as to
whether or not they produce a peptonising ferment which liquefies
the gelatine.

5. Phosphorescence. — Some species of bacteria, for example, certain
species in sea-water, possess the power of producing light (photogenic
bacteria).

6. Many organisms are capable of producing indol (a substance
formed by bacterial action from proteids by alimentary decomposi-
tion), or other metabolic substances as end-products.

7. Organic Chemical Products. — When a pathogenic bacillus grows
in the body, it produces as a result of its metabolism certain poisonous
substances termed toxins. These may occur in the blood as a direct
result of the life of the bacillus, or they may occur as the result of
a ferment produced by the bacillus. Toxins are of various kinds,
and by their effect upon the blood and body tissues they cause
the symptoms of the various diseases. We know, for instance,
that a characteristic symptom common to many diseases is fever,
which is produced by the action of the albumoses (bodies allied
to the albumins) upon the heat-regulating centres in the brain.
Whenever we have a bacillus growing in the body which has
the power of producing a toxin albumose, we obtain fever as a
result of that product acting upon the brain. Albumoses, as a
matter of fact, cause a number of symptoms and poisonous effects,
but the mention of one as an illustration will suffice. Toxins
act, broadly speaking, in two ways. They have a local effect

ACTION OF TOXINS 407

and a specific effect; as the two following illustrations will make
evident : —

(1) They have a local action, as, for example, in the formation
of an abscess. The presence of the causal bacteria in the tissue
brings about very marked changes. There is a multiplication of
connective tissue corpuscles, an emigration of leucocytic cells, a
congestion of blood corpuscles. These elements contribute towards
creating a swelling and redness, and pain results owing to the
subsequent pressure upon the nerve endings. We have, in short,
a state of inflammation. It is then that the toxin commences
its local action. The oldest cells in the mass of congestion will
break down, and necrosis or death will rapidly set in. The con-
nective tissue cells, leucocytes, blood corpuscles, etc., will thus
lose their form and function, and become pus. The local breaking
down of these gatherings of cells into fluid matter is believed
to be the work, not of the bacteria themselves, but of their
toxins.

(2) Toxins may be absorbed and distributed generally throughout
the body. When this occurs they produce degenerative changes in
muscles, in organs, and in the blood itself. Of such a change
diphtheria is an example. The bacillus occurs in a false membrane
in the throat, and occasionally other parts. It first causes the
inflammatory condition giving rise to the membrane, and then it
breaks it clown. In the body of the membrane the bacillus appears
to secrete a ferment which by its action and interaction with the
body cells and proteids, chiefly those of the spleen, produces albumoses
and an organic acid (Martin). These latter bodies are the toxins.
They are absorbed, and pass throughout the body. As a result, we
get the frequent pulse and high temperature of fever: the toxins
irritate the mucous membrane of the intestine, and cause various
fermentative changes in the contents of the intestines, therefore we
get the symptoms of diarrhoea : they penetrate the liver, spleen, and
kidney, setting up fatty degeneration and its results in these organs :
they finally affect many of the motor and sensory nerves, breaking
up their axis cylinders into globules, and producing the characteristic
paralysis. Loss of weight naturally follows many of these degenera-
tive or wasting changes. Thus, then, we have some of the chief
changes set up by the toxins, and these changes constitute the leading
symptoms in the disease as it is known clinically. In addition to
the presence of the specific bacillus in the membrane, we also have
a number of other organisms, like the B. coli, Streptococcus pyogenes,
and various staphylococci, diplococci, etc. Each of these produces,
or endeavours in the midst of keen competition and strife to produce,
its own specific effect. Thus we obtain the complications of
diphtheria, such as various suppurative and septic conditions. The

408 THE QUESTION OF IMMUNITY AND ANTITOXINS

whole of this compound process may be tabulated roughly as
follows : — *

Bacillus coli.
Staphylococci.
Diplococci.
Streptococci.

Toxins.

Suppurative glands,
septic poisoning,
etc.

Bacillus of diphtheria = primary infective agent.

Inflammatory changes and fibrinous exudation.

I
Ferment in membrane = secondary infective agent.

Passes through body, and )

u J- 4.- c V -J f ALBUMOSES ;

by digestion of proteids > '

produces . * . . j an organic acid.

1. Fever.

2. Diarrhoea.

3. Loss of body weight.

4. Fatty degeneration.

5. Degeneration of peri-

pheral nerves and
resulting paralysis.

Such is the specific effect of toxins in diphtheria. The same
principles apply with equal force in tetanus, typhoid, etc., the
differences being in degree of virulence, specificity, mode of onset,
and portions of the body affected.

Sidney Martin suggested a provisional classification of bacterial
toxins as follows : — f

1. Poisons secreted by the bacterium itself = (ferment ?*\

toxin ?) y = Extracellular

2. Products of digestive action of bacterium = albumoses ; I bacterial poisons.

3. Final non-proteid products = animal alkaloid ; J

4. Poisons present in the body of the baeillus { £*5a3SZL.
Such occur, for example, in the tubercle bacillus and the cholera vibrio.

The toxins of bacteria are of a kind which cannot be fully
expressed chemically, but only pathologically. They are probably
of a ferment nature in diphtheria and tetanus. The arguments in
support of that view are — (1) that they act in infinitesimal closes ;
(2) that they may act slowly and produce death after many days by
profoundly affecting the general nutrition ; and (3) that they are
sensitive to the action of heat in a way that no chemical poisons are
known to be. If they are considered as ferments, they must be

* It should be distinctly understood that this table is merely schematic and
provisional. The details of toxin production and its effect are of course still open
to revision and amendment.

t Sidney Martin, M.D., F.R.S., F.R.C.P. , Croonian Lectures delivered before the
Royal College of Physicians, June 1898.

ACTION OF TOXINS 409

substances which have a peculiar affinity for certain tissues of the
body on which they produce their special toxic effect. Hitherto, all
attempts at the separation of such bacterial ferments have been
without success, and for other reasons also the whole question of
such ferments must be left open at present. Sidney Martin and
others have demonstrated that many of the extra-cellular toxins are
albumoses or bodies of a similar nature. They are non-crystallisable,
soluble in water, precipitated along with the proteids by concentrated
alcohol, relatively unstable, having their toxicity diminished or
destroyed by heat, light, or certain chemical agents. As for the
products of digestion, they are formed either by the bacillus ingest-
ing the proteid and discharging it as albumose, or the digestion
occurs by means of a ferment secreted by the bacillus in the body of
the individual or animal suffering from the disease.

It is now held by some that the virus of anthrax produces
albumoses and an alkaloidal substance (Martin), the former producing
fever, the latter oedema, congestion, and local irritation. Hankin
arrived at the view that the bacillus first produces a ferment and
then elaborates albumoses. In tetanus the bacillus produces a
secretion of non-proteid toxin which causes the convulsions. The
albumoses present in this disease are probably due to the secretory
toxin. Ehrlich has isolated a spasm-producing toxin {tctant^asmin),
and a crude poison capable of destroying reel blood cells (tetanolysin).
The nature of the tetanus toxin is not determined, but it is known
that it is a most powerful poison, probably less than -oxoth of a grain
being poisonous to man. In diphtheria, too, we have a secretory
poison in the membrane and in the tissues, and an albumose which
is possibly the result of the secretion. But the true chemical nature
of the diphtheria toxin is also still unknown. In t^hoid fever
intra-cellular bacillary poisons exist, and a toxalbumin has been
obtained which has pathogenic effects of an indefinite character.
The toxins of the typhoid bacillus appear to have little digestive
effect.

Summary of Toxic Effects.— The action of bacteria as disease
producers depends (1) upon the effects of the presence of the bacteria
themselves, and (2) upon their power of forming, directly or indirectly,
certain chemical organic products known as toxins. The effects of
the bacteria, though very diverse, may be classified generally as of a
necrotic or a, separative character, leading to increased functional
activity at first (such as phagocytosis), and subsequently to in-
creased formative activity (such as cell growth and subdivision).
In most diseases the lesion has a special Bite (as in typhoid fever),
and the body generally is only affected indirectly. This locali-
sation may be due to specific action, or to point of entrance of
the bacillus (as in malignant pustule). Secondarily to tissue

410 THE QUESTION OF IMMUNITY AND ANTITOXINS

changes, the body metabolism is affected owing to the distribution
of toxins, and it is to this cause that the chief symptoms of disease
are due.

The Question of Immunity

However the details of the modus operandi of the formation of
toxins are finally settled, we know that there comes a time when
the disease symptoms vanish, the disease declines, and the patient
recovers. In past times this was explained by saying that the
disease had exhausted itself, having gone "through" the body. In
a sense that idea is probably true ; but recently a number of
investigators have applied themselves to this problem, and with
some promising results. And it is now known that, as a result
of the action of the toxins in the body tissues, powers of
resistance are stimulated or conferred in or upon the body cells
affected. What has been found to be true of lower animals by
experimentation is now known to be true of the human body. It
has, therefore, become possible to inoculate resistant blood serum
into toxic blood with the result of opposing the toxins, and bringing
about a condition of resistance, and ultimately, recovery. Or, in
other words, one of the means of defence against the invasion of
such organisms which is possessed by the animal body is the
capacity to manufacture, and set free in the blood stream, sub-
stances which combine with the toxins and so render them inert.
By habituating a large animal, such as the horse, to the action of
toxin in increasing quantities, cells or fluids of its body can be
thereby so stimulated to produce and throw into the blood stream
antitoxins in excessive quantity, that the serum of the animals may
contain sufficient excess for its useful employment as a remedy for
the disease in man or animals. From such results it is but a step
to protective inoculation.

Various protective inoculations against anthrax, for instance,
were practised as early as 1881, and the protected animals remained
healthy. In 1887 Wooldridge succeeded in protecting rabbits from
anthrax by a new method, by which he showed that the growth of
the anthrax bacillus in special culture fluids gave rise to a substance
which, when inoculated, conferred immunity. In 1889 and 1890
Hankin and Ogata worked at the subject, and announced the
discovery in the blood of animals which had died of anthrax of
substances which appeared to have an antagonistic and neutralising
effect upon the toxins of anthrax and upon the anthrax bacilli
themselves. These substances, they afterwards found, were products
of the anthrax bacillus. Behring and Kitasato arrived at much the
same results in tetanus and diphtheria. In 1890 they showed that
the blood serum of an animal which had been immunised against

PRODUCTION OF ANTITOXIN 411

tetanus was capable, when injected into other animals, of protecting
them not only against poisoning with tetanus toxin but also against
infection with living tetanus bacilli. They also proved that, under
certain conditions, a curative action could be demonstrated in animals
which already presented symptoms of tetanus infection. Similar,
though less striking, results were described in the case of diphtheria.
The next step was to isolate these substances, and separating
them from the blood, investigate still further their constitution.
A number of workers were soon occupied at this task, and
Buchner, Hankin, the Klemperers, Eoux, Sidney Martin, and
others have added to our knowledge respecting these toxin-
opposing bodies now known as antitoxins. Some believed these
bodies were a kind of ultratoxin — substances of which an early
form was a toxin ; others held that, as the toxins were products of
the bacteria invading the tissues, the antitoxins were of the nature
of ferments produced by the resisting tissues. A third view is that
possibly antitoxins may be the result of an increased formation of
molecules normally present in the tissues. Finally, antitoxins came
to be looked upon as protective substances produced in the body cells
as a result of toxic action, and held in solution in the blood, and
there and elsewhere exerting their influence in opposition to the
toxins. These antitoxic bodies gradually increase in the blood and
tissues, and their action falls into two groups : (a) antitoxic, which
counteract the effects of the poison itself; and (b) antimicrobic,
which counteract the effects of the bacillus itself. " In one and
the same animal the blood may contain a substance or substances
which are both antitoxic and antimicrobic, such, for example, as
occurs in the process of the formation of the diphtheria and
tetanus antitoxic serums " (Sidney Martin). Antitoxin must,
therefore, be looked upon as a normal constituent of the living
cells which is produced in increased quantity. Of the chemical
nature of toxins and antitoxins, very little is known. Martin
and Cherry have come to the conclusion that toxins are prob-
ably of the nature of albumoses, and antitoxins probably have
a molecule of greater size, and may be allied to the globulins.
Antitoxin has been shown to appear in the various secretions
of the body as well as in the blood, though in a less concentrated
state.

The relation of the antitoxin to the toxin, and its mode of
antagonism, is probably one analogous to chemical union. The two
bodies unite to form an inert compound possessing no toxic or
pathogenic effect. It is found that a definite period of time elapses
before the effect of the toxin is neutralised, and that it takes place
more rapidly in strong solutions than in weak, and in warm
temperature than in cold, which all goes to confirm the view that

412 THE QUESTION OF IMMUNITY AND ANTITOXINS

such union is the mode of antagonism.* The progress of disease
is, therefore, a struggle between the toxins and the antitoxins:
when the toxins are in the ascendency we have an increase of
the disease ; when the antitoxins are in the ascendency we have a
diminution of disease. If the toxins triumph, the result is death ;
if the antitoxins and resistance of the tissues triumph, the result
is recovery.

Different Kinds of Immunity. — We have gathered, then, that
whenever bacteria, introduced into the blood and tissues, fail to
multiply or produce infection (as in saprophytic bacteria, or in
immunity of a particular animal from a specific microbe), this
inability to perform their role is brought about by some property
in the living blood serum which opposes their life and action ;
and further we have seen that this protective property is ex-
haustible according to the number of bacteria, and differs with
various species of bacteria, and in different animals. Buchner
designated these protective bodies, held in solution in the blood,
alexines, and regarded them as belonging to the albuminous bodies
of the lymph and plasma. Alexines are naturally produced anti-
toxins; ordinary antitoxins are acquired alexines. Hence we have
the well-known terms "natural" and "acquired" immunity. Of the
former we have already spoken. The latter, acquired immunity, is
a protection not belonging to the tissues of individuals naturally and
as part of their constitution, but it is acquired during their lives as
a further protection of the tissues. This may happen in one, or
both, of two ways. Either it may be an involuntary acquired
immunity, or a voluntary acquired immunity, a natural attack of
disease, or an artificial attack due to inoculation. Small-pox, typhoid
fever, even scarlet fever, are diseases which rarely attack the same
individual twice. That is because each of these diseases leaves
behind it, so to speak, its antitoxic influence. Hence the individual
has involuntarily acquired immunity against these diseases. An
example of voluntary acquired immunity is also at hand in the old
method of preventive inoculation for small-pox, or variolation. This
was clearly an inoculation setting up an artificial and mild attack
of small-pox, by which the antitoxins of that disease were produced,
and protected the individual against further infection of small-pox ;
that is to say, it was a voluntary acquired immunity. This form of
artificial production of protection is artificial immunity. It may be

* Ehrlich has shown that the antitoxic power of these anti-bodies varies widely,
and is not uniform. Moreover, antitoxins are specific in their action. He suggests
that the ultimate toxin molecule contains two unsatisfied affinities, one of which can
combine with antitoxin {haptophorous)* and the others having a toxic action
{toxophorous). These groups under certain conditions can lose none of their combin-
ing power, the toxophorous being more readily weakened than the haptophorous.
The weakened toxins are termed toxoids or toxones.

THEORIES OF IMMUNITY 413

convenient to marshal together these various terms in a table as
follows : —

, .. . f = a condition of protection or insusceptibility to certain

Immwmty in manj diseases.

1. Natural immunity = constitutional protection produced by at

Acquired naturally (involuntary) produced by anti-
toxins formed by an attack of the disease.
Acquired artificially (voluntary) =

(a) Active immunity, produced by direct inocula-
tion of the weakened bacteria or weakened
toxins of the disease, e.g. vaccination, or
Pasteur's treatment of rabies, or Haffkine's
inoculation for cholera.
(/,) Passive immunity, produced by inoculation,
not of the disease or of its toxins, but of the
antitoxins produced in the body of an animal
suffering from the specific disease. These
antitoxins combine in some way with the
toxins, and so avert their harmful effects.
An example of passive immunity occurs in
diphtheria antitoxin.

Theories of Immunity

We may now consider shortly how these new facts were received,
and what theories of explanation were put forward to explain con-
tinued insusceptibility to disease. It had, of course, been known for
a long time that one attack of small-pox, for example, in some
degree protected the individual from a subsequent attack of the
same disease. To that experience it was now necessary to add a
large mass of experimental evidence with regard to toxins and
antitoxins. The chief theories of immunity which have been pro-
pounded are as follows : —

1. The Exhaustion Theory. — The supporters of this view argued
that bacteria of disease circulating in the body exhausted the body of
the supply of some pabulum or condition necessary for the growth
and development of their own species (Pasteur).

2. The Retention Theory. — This theory, on the contrary, was based
upon the view that there were certain products of micro-organisms
of disease retained in the body after an attack which acted antagon-
istically to the further growth in the body of that same species, as
occurs in a test-tube culture.

3. The Aequired Tolerance Theory. — Some have advanced the
theory that, after a certain time, the human tissues acquired such
a degree of tolerance to the specific bacteria or their specific products,
that no result followed their action in the body. The tissues became
acclimatised to the disease.

4. The Phagocyte Theory. — This theory, which gained so many
adherents when first promulgated by Metclmiicoff, attributes to

414 THE QUESTION OF IMMUNITY AND ANTITOXINS

certain cells in the tissues the powers of " scavenging," overtaking
germs of disease, and absorbing them into their own protoplasm.
This, indeed, may be actually witnessed, and had been observed before
the time of Metchnikoff. But he it was who applied the observation
to the destruction of pathogenic organisms. He came to the con-
clusion that the successful resistance which an animal offered to
bacteria depended upon the activity of these scavenging cells, or
phagocytes. These cells are derived from various cellular elements
normally present in the body : leucocytes, endothelial cells, connective
tissue corpuscles, and any and all cells in the body which possess
the power of ingesting bacteria. If they were present in large
numbers and active, it was argued, the animal was insusceptible to
certain diseases; if they were few and inactive, the animal was
susceptible. It appears that the bacteria or other foreign bodies in
the blood which are attacked by the phagocyte become assimilated
until they are a part of the phagocyte itself. Metchnikoff explained
how the phagocyte is able to encounter bacteria when both are
circulating through the blood. It is guided, he holds, in this attack
on the organisms by the power of chemiotaxis. The bacteria elaborate
a chemical substance which attracts the phagocyte, and this is
termed " positive chemiotaxis." But it may occur that the chemical
substance produced by the bacteria may have an opposite, or repellent,
effect upon the leucocytes, in which case we have " negative chemio-
taxis." Metchnikoff distinguishes two chief varieties of phagocytes
which become active in disease : (a) the microphages, which are the
polynuclear leucocytes of the blood, and (b) the macrophages, which
include the larger hyaline leucocytes, connective tissue cells, etc. It
is now known that blood serum, from which all leucocytes (phagocytes)
have been removed, possesses immunising effects as before, it is
therefore clear that such effect is a property of the serum per se, and
not wholly or only due to the scavenging power of certain cells in it.
Metchnikoff explains this fact by stating that the phagocytes possess
digestive ferments (cytases) which may be set free in the blood
serum, giving it its bactericidal properties. Metchnikoff admits that
antitoxin and toxin form a neutral compound, but holds also that
acquired resistance of body cells is of importance in toxin immunity.
5. Ehrlich's Side Chain Theory. — Ehrlich looks upon a molecule
of protoplasm as composed of a central atom cell with a large
number of side chains of atom groups. The central cell is the
mother cell, the side chains are receptors, that is, cells having com-
bining affinity with food stuffs by which nutriment is brought to
the mother cell. These receptors are of two kinds, those having
power of combining with molecules of simple constitution, and those
having power of breaking up compound bodies by ferment action for
the purposes of assimilation. Now if toxins be introduced into the

VACCINATION 415

system they are fixed to the receptors by their haptophorcms
elements, and their toxophorous elements are therefore free, and if
in sufficient numbers or amount produce the toxic changes. If the
dose of toxin molecules is small, the mother cell is able to throw off
the receptor plus the toxin (E + T), which thereby becomes free in
the blood. The central atom group, however, is able to produce new
receptors, which in their turn come to be free in the blood. As a
result of repeated loss, the regeneration of receptors becomes an over-
regeneration, and the excess of unfixed receptors become free in the
blood, constituting antitoxin molecules. When forming part of the
mother cell the receptors anchor the toxin which is thus able to set
up toxic effects in the body cells and tissues, but when the receptors
are free in the blood (E + T), we have an inert compound, and
therefore no toxic effect. This ingenious theory of Ehrlich explains
the facts of antitoxic effect better than any other, and though not
established, and still requiring much more elucidation, is the theory
which mostly holds the field at the present time.

The Application of the Principles of Immunity

We propose now to consider in some detail four illustrations of
the application of the facts concerning immunity to the prevention
or treatment of disease, viz., vaccination, Pasteur's treatment of
rabies, antityphoid and antiplague inoculation, and antitoxin inocu-
lation for diphtheria. The vaccination in small-pox is an inoculation
of the virus of an attenuated form of the disease ; the rabies inocula-
tion is a transmission of the vital products of the attenuated disease ;
the typhoid and plague inoculations are of pure cultures of living
virus from outside the body; and the diphtheria inoculation is the
introduction of antitoxins (passive immunity).*

Vaccination for Small-pox

In 1717, Lady Mary Wortley Montagu f described the inoculation
of small-pox as she had seen it practised in Constantinople. So
greatly was she impressed with the efficacy of this process, that she
had her own son inoculated there, and in 1721, Mr Maitland, a
surgeon, inoculated her daughter in London. This was the first time
inoculation was openly practised in England. \ For one hundred and
twenty years small-pox inoculation (or variolation, as it is more

* See also Serums, Vaccines, and Toxines, W. C. Bosanquet, 1904.

t The friend of Addison and Pope, who married Mr Edward Wortley Montagu
in 1712, and on his appointment to the ambassadorship of the Porte in 1716 went
with him to Constantinople. They remained abroad for two years, during which
time Lady Wortley Montagu wrote her well-known Letters to her sister the Countess
of Mar, Pope, and others.

$ Crookshank, History and Pathology of Vaccination.

416 THE QUESTION OF IMMUNITY AND ANTITOXINS

correctly termed) was practised in this country, until by Act of
Parliament in 1840 it was prohibited. There were different methods
of performing variolation, but the most approved was similar to the
modern system of arm-to-arm vaccination, the arm being inoculated,
by a lancet in one or more places, with small-pox lymph instead of,
as now, with vaccine lymph. As a rule, only local results or a
mild attack of small-pox followed, which prevented an attack of
natural small-pox. But its disadvantage is apparent : it was in fact
inoculating small-pox, and it was a means of breeding small-pox,
for the inoculated cases were liable to create fresh centres of infection.

In 1796, Edward Jenner, who was a country practitioner in
Gloucestershire, observed that those persons affected with coiv-jjox,
contracted in the discharge of their duty as milkers, did not contract
small-pox, even when placed in risk of infection. Hence he inferred
that inoculation of this mild and non-infectious disease would be
protective against small-pox, and would be preferable to the process
of variolation then so widely adopted in England. Jenner therefore
suggested the substitution of cow-pox lymph (vaccine) in place of
small-pox lymph, as used in ordinary variolation.

It should not be forgotten that variolation was thus the first
work done in this country in producing artificial immunity, and
was followed by vaccination, which was only partly understood.
Even to-day there is probably much to learn respecting it. Vaccina-
tion may be defined as active immunisation hy means of a weakened
form of the specific virus causing the disease. The nature of the
specific virus of both small-pox and cow-pox awaits discovery.
Burdon Sanderson, Crookshank, Klein, Copeman, and others have
demonstrated bacteria in cow-pox or vaccine lymph, and in 1898
Copeman announced that he had isolated a specific bacillus and
grown it upon artificial media.* Numerous statements have been
made to the effect that a specific bacillus has been found in small-pox
also. But neither in small-pox nor cow-pox is the nature of the
contagium really known.f

These facts, however, did not remove the suspicion which had
hitherto rested upon vaccine lymph as a vehicle for bacteria of other
diseases which by its inoculation might thus be contracted. A few
remarks are therefore called for at this juncture upon the work of
Copeman and Blaxall, in respect to what is known as glycerinated
calf lymph. Evidence has been forthcoming to substantiate in some
measure the distrust which many persons have from time to time

* An exhaustive account of vaccine may be found in the Milroy lectures, delivered
in 1898 at the Royal College of Physicians by S. Monckton Copeman, M.D., Brit.
Med. Jour., 1898, vol. i., pp. 1185, 1245, 1312 ; see also paper on the " Bacteriology
of Vaccinia and Variola," Brit. Med. Jour., 1902, vol. ii., pp. 52-67.

f Crookshank, Bacteriology and Infective Diseases ; Virchow, The Huxley Lecture,
1898.

EFFECT OF VACCINATION 417

felt in the vaccine commonly nsed in vaccination, hence the new
form as above designated. This retains the toxic qualities required
for immunity, but is so produced that it possesses in addition three
very important advantages : namely, it is entirely free from extran-
eous organisms, it is available for a large number of vaccinations, and
it retains full activity for eight months. It is prepared as follows :
— A calf, aged three to six months, is kept in quarantine for a week.
If then found upon examination to be quite healthy, it is removed
to the vaccinating station, and the lower part of its abdomen anti-
septically cleaned. The animal is now vaccinated upon this sterilised
area with glycerinatecl calf lymph. After five days the part is again
thoroughly washed, and the contents of the vesicles, which have of
course appeared in the interval, are removed with a sterilised sharp
spoon, and transferred to a sterilised bottle. This is now removed to
the laboratory, and the exact weight of the material ascertained. A
calf thus vaccinated will yield from 18 to 24 grains of vaccine
material. This is now thoroughly triturated and mixed with six
times its weight of a sterilised solution of 50 per cent, chemically
pure glycerine in distilled water. The resulting emulsion is asepti-
cally stored in sealed tubes in a cool place. At intervals during four
weeks it is carefully examined bacteriologically until by agar plates
it is demonstrably free from extraneous organisms, when it is ready
for distribution.

The Effect of Vaccination. — The Royal Commission on Vaccination, 1896,
concluded (p. 90) that the protection vaccination affords against small-pox may be
stated as follows : —

"(1) That it diminishes the liability to be attacked by the disease. (2) That it
modifies the character of the disease and renders it less fatal and of a less severe
type. (3) That the protection it affords against attacks of the disease is greatest
during the years immediately succeeding the operation of vaccination. It is
impossible to fix with precision the length of this period of highest protection.
Though not in all cases the same, if a period is to be fixed, it might, we think, fairly
be said to cover in general a period of nine or ten years. (4) That after the lapse of
the period of highest protective potency, the efficacy of vaccination to protect against
attack rapidly diminishes, but that it is still considerable in the next quinquennium,
and possibly never altogether ceases. (5) That its power to modify the character of
the disease is also greatest in the period in which its power to protect from attack is
greatest, but that its power thus to modify the disease does not diminish as rapidly
as its protective influence against attacks, and its efficacy during the later periods of
life to modify the disease it still very considerable. (6) That revaccination restores
the protection which lapse of time has diminished, but the evidence shows that this
protection again diminishes, and that to ensure the highest degree of protection which
vaccination can give the operation should be at intervals repeated. (7) That the
beneficial effects of vaccination are most experienced by those in whose case it has
been most thorough. We think it may be fairly concluded that where the vaccine
matter is inserted in three or four places it is more effectual than when introduced into
one or two places only, and that if the vaccination marks are of an area of half a
square inch they indicate a better state of protection than if their area be at all con-
siderably below this. "

These findings are well illustrated in the returns of the London Epidemic of
Small-pox which occurred in 1901-2. These returns are the most recent evidence as
to the protection afforded by vaccination. They are as follows : —

2 D

418 THE QUESTION OF IMMUNITY AND ANTITOXINS

Vaccinated.

Unknown.

Unvaccin

ated.

Total cases and
deaths.

Ages.

CD
(ft

o

03

p

w
w
O

+3

CD

Q

« o

O CD

o

,2

CO

n

3 ®
^ p.

GO

o

c«

CD

Q

is

Years.

Under 1 .

. . •

187

130

69-52

187

130

69-52

1 to 5

18

7

1

14-28

524

209

39-88

549

210

38-25

5 to 10

116

2

1-72

26

5

19-23

563

103

18-29

705

110

15-60

10 to 15

334

4

1-19

29

5

17-24

386

88

22-79

749

97

12-95

15 to20

829

19

2-29

44

10

22-73

233

62

26-61

1106

91

8*22

Total under 20

1297

25

1-93

106

21

19-81

1893

592

31-27

3296

638

19-35

20 to 25 .

1274

60

4-71

49

16

32-65

149

47

31-54

1472

123

8-35

25 to 30

1243

87

7-00

49

24

48-98

94

38

40-43

1386

149

10-75

30 to 35

997

111

11-13

45

21

46-67

52

22

42-31

1094

154

14-07

35 to 40

758

137

18*07

32

15

46-87

38

19

50-00

828

171

20-65

40 to 50

893

188

21-05

62

33

53-23

31

24

77-42

986

245

24-84

50 to 60

320

61

19-06

52

23

44-23

13

8

61-54

385

92

23-89

60 to 70

126

30

23*8

25

8

32-00

5

156

38

24-39

70 to 80

31

5

16-13

15

9

60-00

1

1

100-00

47

15

31-91

Over 80

6

1

16-67

1

1

100-00

1

1

100-00

8

3

37*50

Total between)

20 and 80 J

5648

680

12-04

330

150

45-45

384

160

41-66

6362

990

15-56

Grand Total .

6945

705

10-15

436

171

39-22

2277

752

33-03

9658

1628

16-85

From these figures it will be seen : —

(a) That under five years of age 18 cases of small-pox occurred in children who
had been vaccinated, whilst 711 cases occurred in children who had not been
vaccinated, (b) That under ten years of age the cases of small-pox in vaccinated
persons had a mortality percentage of 1-72, and the unvaccinated cases had a
mortality percentage of over 42 per cent, (c) That under twenty years of age the
cases of small-pox occurring in vaccinated persons had a mortality percentage of 1 -93,
and the unvaccinated cases had a mortality percentage of 31 *27. {d) That over
twenty years of age the cases of small-pox occurring in vaccinated persons numbered
5648, and there were 680 deaths, giving a mortality percentage of 12*04 ; whereas the
cases occurring in unvaccinated persons were 384, 160 of whom died, giving a
mortality percentageof 41 -66. The larger number of cases of small-pox in vaccinated
persons is, of course, due to the fact that by far the larger proportion of the popula-
tion at that age-period have, at some time or other in their lives, been vaccinated.
Three broad facts stand out with clearness : — (1) That small-pox among the
vaccinated is nowadays mainly a disease of adults, because children are protected
by primary vaccination and adults are not protected by revaccination. (Ninety-two
per cent, of the vaccinated cases were over fifteen years of age. ) (2) That among
the unvaccinated, small-pox is still, in great measure, a disease of the young as it was in
prevaccination days. (Seventy-three per cent, of the unvaccinated cases were under
fifteen years of age.) (3) That the mortality rate among the vaccinated is at all

EFFECT OF VACCINATION

419

ages much less than among the unvaccinated, and that this difference is very striking'
and complete in children because of their recent vaccination.

Those who advocate vaccination and revaccination as protective in a greater or
lesser degree against small-pox do so upon three main grounds. In the first place,
they claim that, other things being equal, persons who have been vaccinated (especi-
ally within ten years) are less liable to attack from small-pox. This is abundantly
established by the figures quoted above. In the second place, they claim that
persons who have been vaccinated, and yet, on account of their greater number in
the population, and, therefore, their consequent greater probability of infection, are
attacked by small-pox, do not die so readily from the disease as those who have not
been vaccinated. This claim also is more than proved in the returns quoted above.
In the third place, they claim that the protection afforded by vaccination depends
upon the efficiency of the vaccination. This may be measured, as is frequently done,
by the number of marks, but it is more satisfactorily measured by area of vaccination
mark (i. e. area of cicatrix). The return of the Metropolitan Asylums Board respect-
ing this point is given below, and a study of it will amply prove the claim made.

Admissions.

Deaths.

Mortality,
per cent.

Vaccinated Cases —
Area of Cicatrix :

Half and upwards of half
square inch
Area of Cicatrix :

One-third, but less than half
square inch
Area of Cicatrix :

Less than one-third square
inch ....
Area of Cicatrix :
Not recorded

5163
835

860

87

379

131

162
33

7-34

15-69

16-87
37-93

Totals of Vaccinated Class
Unknown and Doubtful Class .
Unvaccinated Class .

6945

436

2277

705
171
752

10-15
3922
33-06

Grand Totals

9658

1628

16-S7

Pasteur's Treatment of Rabies
Eabies is a disease affecting dosrs (in Western Europe)1

and

wolves (in Russia), and can be transmitted to other animals (chiefly
mammals and especially the Carnivore/) and man. Infection may
he conveyed from the rabid animal by biting (which is the most
frequent mode), by licking raw surfaces, by suckling, and possibly
by the ingestion by animals of the flesh of other animals which have
died from the disease.

• In the decennium 1894-1903, 1555 dogs in Great Britain were reported as suffer-
ing from rabies, of which only 29 eases occurred during the last five years. The
marked decline in recent years is attributed to the effects of the muzzling order and
stricter inspection of ownerless vagrant dogs. Sec also Y ear-Book of Department of
Agriculture, U.S.A., 190O.

420 THE QUESTION OF IMMUNITY AND ANTITOXINS

Although rabies was mentioned by Aristotle, and has been studied
by a large number of workers since, the contributions of Pasteur
have been greater than all the other additions to our knowledge of
the disease put together. Professor Eose Bradford has pointed out
that Pasteur's discoveries concerning rabies may be said to be four
in number : (a) that the virus was not only in the saliva, but also in
the central and peripheral nervous system, yet absent from the
blood; (b) that the disease was most readily inoculated in the
nervous system; (c) that by suitable means the virus could be
attenuated ; and (d) that by means of an attenuated virus preventive
and even curative methods might be adopted.

The disease takes two chief forms: (1) furious rabies, and (2)
paralytic or dumb rabies. The former is more common in dogs.
The animal becomes restless, has a high-toned bark, and snaps at
various objects; sometimes it exhibits depraved appetite. Briefly,
the animal passes from a melancholy to a maniacal and then a
paralytic state, ending in coma and death. In man, the incubation
period is fortunately a very long one, averaging about forty days.
Nervous irritability is the first sign ; spasms occur in the respiratory
and masticatory muscles, and the termination is similar to rabies in
the clog. The symptom of fear of water is a herald of coming
fatality.

Although a number of the workers at the Pasteur Institute and
elsewhere have addressed themselves to the detection of a specific
microbe, none has as yet been found, although, in the opinion
of Pasteur, such an agent may be suspected as the cause.

Pathologically, rabies and tetanus are closely allied diseases, and
the recent remarkable additions to our knowledge of the latter
disease only make the similarity more evident. There are in rabies
three chief sets of post-mortem signs. First, and by far the most
important, are the changes in the nervous system. Here we find
patches of congestion in the brain, and breaking down of the axis
cylinders of the nerves. The stomach, in the second place, exhibits
hemorrhagic changes, not unlike acute arsenical poisoning. Thirdly,
the salivary glands show a degenerative change in a breaking down
of their secretory cells. Eoux has pointed out that in life the
saliva of a mad dog becomes virulent three days before the appear-
ance of the symptoms of disease. The poison appears to be present
mainly in the nervous system and the saliva ; it is not present in
the blood.

The method of treatment by inoculation was introduced by
Pasteur. Before his time cauterisation of the wound was the only
method adopted. But if more than half an hour has elapsed since
the bite, cauterisation is of little or no avail. The basis of Pasteur's
treatment was the difference in virulence obtainable in spinal cords

PASTEUR'S TREATMENT FOR RABIES

421

infected with rabies. Pasteur found that drying the cord led to a
lessening of its virulence, just as certain other conditions increased
its virulence. Next he established the fact that subcutaneous
injection of a weak virus, followed up with doses of ever-increasingly
virulent cords, immunised dogs against infection or inoculation of
fully virulent material. From this he reasoned that if he could
establish a standard of weakened virulence he would have at hand
the necessary " vaccine " for the treatment of the disease.

Subsequent research and skilled technique resulted in a method
of securing this standard, which he found to be a spinal cord dried
for fourteen days. The exact details of preparation of this vaccine
are as follows: The spinal cords of two
rabbits dead of rabies are removed from
the spinal canal in their entirety by means
of snipping the transverse processes of the
vertebrae. Each cord is divided into three
more or less equal pieces, and each piece,
being snared by a thread of sterilised
silk, is carefully suspended in a steril-
ised glass jar. At the bottom of the jar
is a layer, about half an inch deep, of
sterilised calcium chloride. The jars are
then removed to a dark chamber, where
they are placed at a temperature of
20-22° C. in wooden cases. Here
are left to dry. Above each case
tube of broth, to which has been
added a small piece of the corre-
sponding cord, in order to test for
any micro-organism that may by
chance be included. In case of the
slightest turbidity in the broth, the
cord is rejected. Fourteen series of
cords are thus suspended on four-
teen consecutive days. The first, second, and third are found to be of
practically equal virulence, but from the third to the fourteenth the
virulence proportionally decreases, and on the fifteenth day the cord
would be practically innocuous and non-virulent. When treatment
is to be commenced, obviously the weakest — that is, the fourteenth
day — cord is used to make the "vaccine," and so on in steadily
increasing doses (as regards virulence) up to, and including, a third-
day cord. The fourteenth-day cord is therefore taken, and a small
piece cut off and extracted in 10 c.c. of sterile broth, which are
placed in a conical glass and covered with two layers of thick filter-
paper, the glass with its covering having been previously sterilised

Kn.. :sr..— Srsi'KNDED Spinal Cokd.
In drying, jar containing Calcium Chloride.

422 THE QUESTION OF IMMUNITY AND ANTITOXINS

by dry heat. When the patient bitten by the rabid animal is
prepared, 3 c.c. of this broth emulsion of spinal cord are inoculated by
means of a hypodermic needle (under aseptic precautions) into the
flanks or abdominal wall. On the following day the patient returns
for an inoculation of a cord of the thirteenth clay, and so on until a
rabid cord emulsion of the first three days has been inoculated. As a
matter of practice, the dosage depends upon the three recognised
classes of bites, viz. (1) bites through clothing (least severe); (2.)
bites on the bare skin of the hand ; (3) bites upon the face or head,
most severe owing to the vascularity of these parts. An example of
each, which the writer was permitted to take in the Pasteur Institute,
may be here added to illustrate the usual practice.

Inoculation Treatment for

affected with Babies.

1. For those Bitten through

2. For those Bitten

on

3. For those Bitten on

Face

Clothes.

Uncovered Skin of Hands, etc.

or Head.

bb

bt

bo

'H-C S

Ch a

«- s .

<!- 3

Days of

° 2 o
8£ o

CO t-l

Days of

OOq

SB

Days of

o .2 o

2- o

CO Ch

o

Treatment.

22 3 *->
o a o>

Treatment.

co 3 t_,

ogo)

=1

Treatment.

3h&

O

o

O

14

1 at 11 a.m. 3

14

1 at 11 a.m.

3

14

1 at 11 a. m.

3

1 „ „ 3

13

■f *i »»

3

13

1 „ „

3

13

2 „ „ 3

12

2 „ „

3

12

1 at 3 p. m.

3

12

2 „ „ 3

11

2 „ „

3

11

1 „ M

3

11

3 „ „ 3

10

3 „ „

3

10

2 at 11 a.m.

3

10

3 „ „ 3

9

3 „ „

3

9

2 „ „

3

9

4 „ „

3

8

4 „ „

3

8

2 at 3 p.m.

3

8

4 „ ,,

3

7

4 „ „

3

7

2 „ „

3

7

5 m ,,

3

6

5 ,, M

3

6

3 at 11a.m.

3

6

5 i , , ,

3

6

o »? >»

3

6

0 » » 5 »

3

6

6 „ „

3

5

6 ,, ,,

3

5

4 „ „

3

5

7 ,, ,,

3

5

7 „ „

3

5

•F> »i ?i

3

5

8 „

3

4

8 „ „

3

4

6 „ „

3

4

9 „

2

3

«> i» 95

2

3

7 „

2

3

10 „ „

3

5

10 „ „

3

5

8 „ „

3

4

11 M .,

3

5

11 „ „

3

5

9 „ „

3

3

12 „ „

3

4

12 „ „

3

4

10 „ „

3

5

13 „ „

3

4

13 „ „

3

4

11 ,, „

3

5

14 „ „

3

3

14 „ „

3

3

12 „ „

3

4

15 „ „ 3

3

15 „ „

3

3

13 ,, ,,

3

4

16 „ „

3

5

14 „ „

3

3

17 ,, ,,

3

4

15 „ „

3

3

...

18 „ „

3

3

16 „ „

17 „ „

18 ,, ,,

19 „ „

20 „ „

21 „ „

3
3
3
3
3

b

5

4
3
5

4
3

CHOLERA, TYPHOID, PLAGUE

423

Effect of Treatment.— It may be well to add the returns of inoculation made at
the Pasteur Institute, Rue Dutot, Paris, as above described, and the mortality rate
resulting. The record is as follows : —

Number of

Number of

Rate of

Year.

persons
Inoculated.

Deaths.

Mortality.

1886

2671

25

0*94

1887

1770

14

0*79

1888

1622

9

0-55

1889

1830

7

0-38

1890

1540

5

0-32

1891

1559

4

0*25

1892

1790

4

0'22

1893

1648

6

0-36

1894

1387

7

0-50

1895

1520

5

0-33

1896

1308

4

0-30

1897

1521

6

0-39

1898

1465

3

0-20

1899

1614

4

0-25

1900

1420

4

0-35

1901

1321

5

0-38

1902

1105

2

0*18

Of the 1105 persons under treatment in 1902, 9 were English, 2 Spaniards, 2
Russians, and one each Greek, Dutch, and Swiss— making 16 foreigners, 1089 French.
The diminution in the number of French patients, as compared with several preceding
years, is explained by the opening of anti-rabic institutes at Lille, Marseilles,
Montpellier, Lyons, and Bordeaux, at one or other of which persons residing in the
neighbourhood of those towns have been sent instead of going to Paris.

Pasteur's treatment of rabies by inoculation of emulsions of dried
spinal cord is, therefore, a "vaccination" of attenuated virus, result-
ing in antitoxin formation, to the further protection of the individual
against rabies.

Inoculations for Cholera, Typhoid, and Plague

Anti-Cholera Inoculation. — Inoculating cholera virus against
cholera has been made illegal, like variolation was in 1840. But
Haffkine has prepared two vaccines. The weak one is made from
pure cultures of Koch's spirillum of Asiatic cholera, attenuated
by growth to several generations on agar or broth at 39° C, or by
passing a current of sterile air over the surface of the cultures.
The strong one, virus exalte, is from similar culture the virulence of
which has been increased by passage through guinea-pigs. One
cubic centimetre of the first vaccine is injected hypodermically into
tin1 flank, and the second vaccine three or four days afterwards.
The immunisation is prophylactic, not remedial, and its action takes
ei'fect five or six days after the second vaccine has been injected.

424 THE QUESTION OF IMMUNITY AND ANTITOXINS

Anti-Typhoid Inoculation.— It is now known that the serum
of persons who have recovered from typhoid fever, and the serum
of animals artificially immunised against virulent typhoid bacilli,
protect against the typhoid bacillus. Animals have now been
immunised by injections of the toxins of the typhoid bacillus ; and
their serum aids in the destruction of the bacilli which produce the
toxins. Acting on these principles, Wright has prepared a vaccine
against typhoid fever. A virulent twenty-four hours culture is
emulsified in bouillon, and killed by heating for five minutes at 60°
C. For use, one-twentieth to one-fourth of the dead culture is
injected hypodermically, usually in the flank.* The effect of the
inoculation is some local tenderness and swelling with enlargement
of adjacent lymph glands. Within ten days the blood of the
inoculated person begins to show »a positive Widal reaction, owing
to its immunising properties, and it is also bactericidal in vitro.

Haffkine's Preventive Inoculation for Plague.— In plague
the same plan has been followed. Luxurious crops of Kitasato's
plague bacillus are grown on ordinary broth with the addition to
the surface of a film of oil or fat (" ghee "). Under the globules of
fat flakes of plague culture grow like stalactites, hanging down into
the clear broth. The culture is kept at 25° C. These are, every few
days, shaken to the bottom, and a second crop grows on the under-
surface of the fat. In the course of six weeks a number of such
crops are obtained and shaken down into the fluid, until the latter
assumes an opaque milky appearance. The purity of this culture
is controlled by transferring with a sterile pipette a small quantity
to a dry agar tube, and noting the appearance of the growth by
reflected light through the thickness of the agar. The culture is
now, unlike the cholera vaccine, exposed to a temperature of 65° C.
for one hour in a water-bath, and a small quantity of carbolic acid is
added ('5 per cent.), by which processes the bacilli are killed. The
dose is 5 to 10 c.c. This preparation has the advantage of being
easily prepared, obtainable in large quantities, and requires no animals
in its preparation. When inoculated, it produces local pain and
swelling at the site of inoculation, and general reactive symptoms such
as fever. From a careful analysis of the results of this inoculation, it
is shown that the efficacy of the prophylactic depends upon the
virulence of the bacillus culture from which the vaccine is prepared,
and upon its close and ability to produce a well-marked febrile
reaction. It appears to be more effective in the prevention of deaths
than of attacks.f

* For methods employed in preparation of the vaccine, see Brit. Med. Jour.,
1900, vol. i., p. 122 (Wright).

t Proc. Boy. Soc, 1900 ; Report of Medical Officer to Local Government Board,
1902, pp. 357-94.

DIPHTHERIA ANTITOXIN 425

The Indian Plague Commission concluded that (1) inoculation
sensibly diminishes the incidence of plague attacks on the inoculated
population, but the protection afforded is not absolute ; (2) inocula-
tion diminishes the death-rate among the inoculated population ; (3)
inoculation does not appear to establish protection until after some
days ; and (4) protection is conferred for a considerable number of
weeks and possibly for months. Finally, the Commission recom-
mend that under the safeguards and conditions of accurate
standardisation and complete sterilisation of the vaccine, and the
thorough sterilisation of the syringe in every case, inoculation should
be encouraged wherever possible, and in particular among disinfecting
stuffs, and the attendants of plague hospitals.

Antitoxin Inoculation fop Diphtheria

We may now consider an illustration of passive immunity. This,
it will be remembered, may be defined as a protection (against a
bacterial disease) produced by inoculation, not of the disease itself,
as in small-pox inoculation, nor yet of its weakened toxins, as in
rabies, but of the antitoxins produced in the body of an animal
suffering from that particular disease. Examples of this treatment
are increasing every year. The chief examples are to be found in
Diphtheria, Tetanus, Streptococcus, and Pneumococcus.

To be of value, antitoxins must be used as early as possible,
before tissue change has occurred and before the toxins have, so to
speak, got the upper hand. When the toxins are in the ascendency
the patient suffers more and more acutely, and may succumb before
there has been time for the formation in his own body of the neutral
compound of toxin and antitoxin. If he can be tided over the
"crisis," theoretically all will be well, because then his own anti-
toxin will eventually gain the upper hand. But in the meantime,
before that condition of affairs, the only way is to inject antitoxins
prepared in some animal's tissues whose disease began at an earlier
date, and thus add antitoxins to the blood of the patient, early in the
disease, and the earlier the better, for, however soon this is done, it
is obvious that the toxins begin their work earlier still. It should
not be necessary to add that general treatment must also be con-
tinued, and indeed local germicidal treatment, e.g. of the throat in
diphtheria and the poisoned wound in tetanus. Further, in a mixed
infection, as in glandular abscesses, with diphtheria, it must be borne
in mind that the antitoxin is specific, and may therefore probably
fail to reduce the complication which must be treated separately.

In the production of antitoxins, an animal is required from
whose body a considerable quantity of blood can be drawn without
injurious effect. Moreover, it must be an animal that can stand

426 THE QUESTION OF IMMUNITY AND ANTITOXINS

an attack of such diseases as diphtheria and tetanus. Such an
animal is the horse. Now, by injecting into the horse (a) living
organisms of the specific disease, but in non-fatal doses, or
(b) dead cultures, or (e) filtered cultures containing no bacteria
and only toxins, we are able to produce in the blood of the horse
first the toxins and then by natural processes the antitoxins of the
disease in question. The non-poisonous doses of living organisms
can be attenuated, by various means. Dead cultures have not
been much used to produce immunity except by Pfeiffer. In actual
practice the third method is much the most general, viz., filtering a
fluid culture free from bacteria, and then inoculating this in ever-
increasing doses. The preparation of diphtheria antitoxin may be
taken as an example, but what follows would be equally applicable
to other diseases, such as tetanus : —

1. To obtain the Toxin. — First grow a pure culture of the Klebs-
Loffler bacillus of diphtheria in large flasks containing " Loftier' s

medium," or a solution made by mixing three
parts of blood serum with one of beef broth,
and adding 1 per cent, of common salt
(NaCl) and 1 per cent, of peptone. An
alkaline medium is necessary, and a free
supply of oxygen and the presence of a
large proportion of peptone in the medium
favours a high degree of toxicity. The
bouillon must be glucose-free. The flask,
thoroughly sterilised before use, is now
plugged with sterile cotton-wool and incu-
bated at 37° C. for three weeks. Sterile air
may be passed over the culture periodically,
thereby aiding the growth. After the lapse
of about a month a scum of diphtheria
growth will have appeared over the surface of the fluid. This is
now filtered through a Chamberland filter into sterilised flasks, and
some favourable antiseptic added to ensure that nothing foreign to
the toxin shall flourish. The flasks are kept in a cool place in the
dark. Here, then, we have the product, the toxin, ready for injection
into the horse.

The power of the toxins is estimated by subcutaneous injection
of varying amounts into a number of guinea-pigs, and the minimum
lethal close (M.L.D.) is obtained. The standard M.L.D. is the
smallest amount which will kill a 250-gram guinea-pig in four days.
According to Behring, a normal M.L.D. (expressed as D.T.N.1) is
•01 c.c. ; a toxin of which '02 is M.L.D. will therefore be expressed
as D.T.N.5

2. Immunisation of the Horse. — The general principle is that

Fig. 37.— Flask used for Prepara-
tion of the Toxin of Diphtheria.

DIPHTHERIA ANTITOXIN 427

the animal is treated with increasing doses of the particular poison.
The toxins, which have been previously tested on small animals, such
as rabbits and guinea-pigs, are injected subcutaneously, intramus-
cularly, or intravenously. At first either very minute doses of weak
toxins, or toxins which had been modified by chemical agents, or in
other ways, are employed. In the case of tetanus, in the early stages
the toxin is usually modified by being treated with iodine. The
injection of the toxin may be followed by swelling at the site of
inoculation, loss of appetite, general malaise, and rise of temperature.
When these have passed off the animal receives a second, rather
larger injection, and in this way the quantity of toxin is increased
until within a few months the horse is capable of tolerating many
thousand multiples of what would be a lethal dose if given as a first
injection. When the serum has reached the strength suitable for
clinical use, blood is withdrawn from time to time by venesection.

It is evident that only healthy horses are of service in pro-
viding healthy antitoxin, even as healthy children are necessary
in arm-to-arm vaccination. To provide against any serious
taint, the horse is tested for glanders (with mallein) and for
tuberculosis (with tuberculin). The close of the injection of
toxin is at the commencement about TV c.c, or a little more.
The site of the inoculation is the apex of the shoulder, which
has been antiseptically cleaned. After the first injection there is
generally a definite febrile reaction and a slight local swelling.
From yV or \ c.c. the dose is steadily increased, until at the end of
two or three months * perhaps as much as 300 c.c. (or even half a
litre) may be injected without causing the reaction which the initial
injection of y1^- c.c. caused at the outset. This shows an acquired
tolerance of the tissues of the horse to the toxic material. After
injecting 500 c.c. into the horse without bad effect, the animal has
a rest of four or five days.

3. To obtain the Antitoxin. — During this period of rest the
interaction between the living body cells of the horse and the toxins
results in the production in the blood of an antitoxin. By means
of a small sterilised cannula, five, or eight, or even ten litres of blood
are drawn from the jugular vein of the horse into sterilised flasks or
jars. As used in Paris, the top of the jar is closed by two paper
coverings before it is sterilised. Then it is again covered with a
further loose one. Before use the loose one is removed and replaced
by a metal (zinc) lid, which has been separately sterilised. This
metal lid contains an aperture large enough for the tube which

■ To shorten this period Dr Cartwright Wood has adopted a plan by which time
may be saved, and 200 c.c. injected, say, within the first two or three weeks. This is
accomplished by using a " serum toxin M (containing albumoses, but not ferments)
previously to the broth toxin.

428 THE QUESTION OF IMMUNITY AND ANTITOXINS

conveys the blood from the cannula to pass through. The tube,
therefore, passes through the metal lid and two paper covers, which
it was made to pierce. When enough blood has passed into the
vessel the tube is withdrawn, and the metal lid slightly turned.
Thus the contained blood is protected from the air.* The jar con-
taining the blood (which contains the antitoxins) is next placed in
a dark, cool cellar, where it stands for separation of the clot. During
this time the blood naturally coagulates, the corpuscles falling as a
dense clot to the bottom, and the faintly yellow serum rising to the
top. The serum, or liquor sanguinis, averages about 50 per cent, of
the total blood taken. Sometimes antiseptic ('3 per cent, carbolic
acid) is added with a view to preservation. It is generally filtered
(through a Berkefeld) before bottling for therapeutic use, and
examined bacteriologically as a test of purity, for sterility and for
absence of toxicity, and for antitoxic value.

The latter step is the estimation of the antitoxic power of the
serum, or what is termed the " standardising " of the serum. This
is accomplished by testing the effect of various quantities upon a
certain amount of toxin. Ehrlich has adopted as the immunity unit
the amount of antitoxic serum which will neutralise a hundred times
the minimum lethal dose of toxin, the serum and toxin being mixed
together, diluted up to 4 c.c, and injected subcutaneously. A normal
antitoxic serum is one of which 1 c.c. contains an immunity unit.

Process of Standardisation of Antitoxins. — This matter will be
best illustrated by an illustration, as follows : —

Stage 1. — Varying amounts of a toxin are added to a definite amount of antitoxin,
i.e. to 1 Ehrlich unit, and injected into a guinea-pig of 250 grammes. That mixture,
which kills the guinea-pig in four days, is held to contain 1 M.L.D., over and above
the amount of toxin required to neutralise 1 antitoxin unit. The total amount of
toxin used to bring about the death of the guinea-pig in four days = the standard
toxin for that particular standardisation.

Stage 2. — Varying quantities of the antitoxin to be tested are added to the
standard toxin and injected into guinea-pigs. That mixture, which kills the guinea-
pig in four days as before, contains 1 M.L.D. of toxin over and above that
neutralised by the added antitoxin. The amount of toxin used in Stages 1 and 2
is the same, therefore the amount of antitoxin in Stages 1 and 2 must be equal. In
Stage 1, 1 Ehrlich unit was used, therefore the amount of antitoxin in Stage 2 which,
with the standard toxin killed the guinea-pig in four days, also contains 1 Ehrlich unit.

Example. Stage 1.

•25 c.c. Toxin + 1 Ehrlich Antitoxin Unit . Guinea-pig alive fifth day.

*_o c.c. ,, ,, ,, . ,, ,,

•27 c.c. „ ,, ,, . ,, died fourth day.

•28 c.c. ,, ,, ,, . ,, died first day.

.'. -27 c.c. Toxin = Standard toxin, containing 1 M.L.D. of toxin after neutral-
isation with 1 Antitoxin unit.

* At the conclusion of the operation the cannula is removed from the jugular vein
and the wound is closed by the valvular character of the slit in the skin and vein,
and the elasticity of the wall of the vein. No stitching or dressing is required.

DIPHTHERIA ANTITOXIN 429

Example. Stage 2.

Tfor c.c. Antitoxic Serum to be tested + *27 c.c. Toxin . Guinea-pig alive fifth day.

^inr c.c. »• >» »» • »» >'

-^ c.c. „ „ „ , ,, died fourth day.

*fo c.c. „ „ „ . „ died first day.

.-. The mixture of ^ac, of Antitoxic Serum +'27 c.c. Toxin, killing guinea-
pig in four days, contains 1 M.L.D.

•'• ttoit c-c- Antitoxic Serum = 1 Ehrlich unit.

.-. i c.c. = 300 Ehrlich units.

4. The Use of Antitoxin. — The antitoxin is now ready for injection
into the patient who has contracted diphtheria, and in whose blood
toxins are in the ascendency, and under which the individual may
succumb. They are injected in varying doses, as we have already
pointed out. As large a dose should be given as practicable. A
common first dose varies from 2000 to 5000 units. For prophylactic-
purposes a smaller dose is administered (500, and for children under
two years of age, 300 units). Early administration is of great
importance. The flank between the crest of the ilium and the last
rib and the lower part of the abdomen are generally selected as the
sites of injection, but any region with loose subcutaneous connective
tissue is suitable. The injections should be subcutaneous. In
performing the injection strict asepsis must be observed. The
syringe must be well washed and boiled before use. The skin must
be well cleansed with soap and water, and afterwards treated with an
antiseptic such as a 1 in 1000 corrosive sublimate solution, or 1 in
20 carbolic acid solution. The antitoxin of diphtheria has been used
on various recent occasions as a prophylactic in outbreaks of the
disease, and it is now considered as one of the practicable means for
controlling an epidemic. Antitoxin inoculation played a greater or
less part in the checking of diphtheria outbreaks at Cambridge,*
Golchester,f KempstonJ and other places. In the Cambridge
outbreak antitoxin was supplied free for prophylactic use in the case
of those who had come into contact with actual cases of diphtheria,
or where those who, not being ill, were known by bacteriological
examination of the throat to be harbouring the diphtheria bacillus.
Thus free bacteriological examination of the throat of suspected or
known "contacts" was first carried out. In the cases yielding
positive results antitoxin was injected. At Cambridge 500 units of
antitoxin were given in such contact persons; at Kempston 1000
units was the dose. The general result is that mortality has been
lessened, and that in fatal cases there has been a considerable
lengthening of the period of life. Moreover, the whole clinical
course of the disease is greatly modified, and its severity reduced.

■ Jour, of Hygiene, 1901, vol. i., pp. 228 and 487.

t Ibid., 1902,* vol. ii., p. 170.

X Report on cm Outbreak of Diphtheria at Kempston, p. 21.

430 THE QUESTION OF IMMUNITY AND ANTITOXINS

Effect of Diphtheria Antitoxin Inoculation

The following summary of the Antitoxin Treatment of all forms of Diphtheria at
the Hospitals of the Metropolitan Asylums Board, 1895-1903, compared with the
results obtained before the adoption of that treatment," affords striking evidence of
the efficacy of diphtheria antitoxin : —

Cases.

Deaths.

Mortality,
per cent.

1890-3 (before antitoxin) .
1894 (antitoxin occasion-
ally used) .

7111

3042

2161
902

30-39
29-65

1895 .
1896.
1897.
1898.
1899.
1900.
1901 .
1902.
1903 .

2182
2764
4381
5186
7038
7271
6499
6015
4839

615
717
896
906
1082
936
817
714
493

28-1
25-9
20*4
17-5
15-3
12-8
12*5
11-8
10-1

The value is particularly noticeable among children. Amongst cases in the first
year of life the rate has fallen from 61'8 to 37 "8; in the second year from 63*1 to
35*4 ; in the third year from 55*1 to 26*4 ; in the fourth year from 48 '3 to 22*9 ; and
in the fifth year from 39*6 to 20*7.f

At the Brook Hospital, Shooters Hill, Woolwich (Metropolitan Asylums Board),
Dr MacCombie has kept records since 1897 showing the results of the antitoxin
treatment on all the cases at the hospital, with special reference to the day of the
disease on which treatment began, in order to illustrate the effect of early adminis-
tration : —

Mortality per cent

in Cases Treated.

1897.

1898.

1899.

1900.

1901.

1902.

Cases treated

on 1st day of disease .
2nd ,,
3rd
4th
5th day and after

o-o

5-4
11-5
19-0
21-0

o-o

5-0
14-3
18-1
22-5

o-o

3-8
12-2
20-0
20*4

o-o

3-6

6-7

14*9

21*2

o-o

4-1
11-9
12-4
16*6

o-o

4-6
10-5
19-8
19*4

* Annual Report of Metropolitan Asylums Board, 1902, p. 172.

t For the most complete account of diphtheria antitoxin and its effects, see Report
on the Bacteriological Diagnosis and Antitoxic Serum Treatment of Cases admitted to
the Hospitals of the Metropolitan Asylums Board, 1895-6, by Professor Sims Wood-

DIPHTHERIA ANTITOXIN 431

" During the past six years," Dr MaeCombie reports, " the total number of cases
treated with antitoxin has been 4202. Not a single death has taken place among
the cases that came under treatment on the first day of disease, and among those
coming under treatment on the second day of disease, the mortality has not exceeded
5*4, and has been as low as 3 '6. While among those that came under treatment later
the average mortality is very much higher. Were it possible to secure the admission
to hospital of all cases on the first or second day of illness, the lives of a large number
of patients would thereby be saved.""" Dieudonne has collected similar returns to
the foregoing table from four or five different sources. The importance of early
administration is therefore widely established.

* Annual Report of Metropolitan Asylums Board, 1902, p. 208 ; see also Report,
1903, p. 218.

CHAPTEE XIII

DISINFECTION

General Principles— Means of Disinfection: by Heat; by Chemicals— Practical
Disinfection : Rooms, Walls, Bedding, Clothing, Excreta, Books, Linen,
Stables, etc. — Disinfection of Hands — Disinfection after Special Diseases :
Phthisis, Small-pox, Scarlet Fever, Diphtheria, Typhoid, Plague.

The object of modern bacteriology is not merely to accumulate
tested facts of knowledge, nor only to learn the truth respecting
the morphology and life-history of bacteria. These are most
important things from a scientific point of view. But they are
also a means to an end ; that end is the prevention of preventable
diseases and the treatment of any departure from health due to
micro-organisms. In a science not a quarter of a century old, much
has already been accomplished in this direction. The knowledge
acquired of, and the secrets learned from, these microscopic
vegetable cells which possess such potentiality for good or evil
have been, in some degree, successfully turned against them. When
we know what favours their vitality and virulence, we know
something of the physical conditions which are inimical to their
life ; when we know how to grow them, we also know how to kill
them.

We have previously made a brief examination of the methods
which are adopted for opposing bacteria and their products in the
tissues and body fluids. We must now turn to consider shortly the
modes which may be adopted in preventive medicine for opposing
bacteria outside the body.

It will be clear at once that we may have varying degrees of
opposition to bacteria. Some substances kill bacteria, and are thus

432

GENERAL PRINCIPLES 433

germicides ; other substances prevent their development and resulting
septic action, and are termed antiseptics. The word disinfectant is
used more or less indiscriminately to cover both these terms. A
deodorant is, of course, a substance removing the odour of evil-
smelling putrefactive processes. These are the four common
designations of substances able to act injuriously on bacteria and
their products outside, or upon the surface of, the body. But a
moment's reflection will bring to mind two facts not to be forgotten.
In the first place, an antiseptic applied in very strong dose, or for an
extended period, may act as a germicide; and, vice versa, a
germicide in too weak solution to act as such may perform only
the function of an antiseptic. Moreover, the action of these, dis-
infecting substances not only varies according to their own strength
and mode of application, but it varies also according to the specific
resistance of the protoplasm of the bacteria in question. Examples
of the latter are abundant ; for instance, between the bacillus of
typhoid fever and the spores of anthrax there is an enormous
difference in power of resistance. In the second place, there are
the physical conditions injurious to the development of bacteria.
At a low temperature bacteria do not multiply at the same rapidity
as at blood-heat. Within the limits of a moist perimeter the
air is, to all intents and purposes, germ-free. Direct sunlight has
a definitely germicidal effect in the course of time upon some of
the most virulent bacteria we know. In a certain sense these
three examples of physical conditions — low temperature, moist
perimeter, direct sunlight — may become first antiseptics and then
germicides. Yet for a limited period they have no injurious effect
upon bacteria. These would seem to be very simple points, and
calling for little comment, yet the pages of medical and sanitary
journals reveal not a few keen controversies upon the injurious
action of certain substances upon certain bacteria, owing to the
discrepancies of necessity arising between results of different
skilled observers who have been carrying out different experiments
with different solutions of the same substance upon different proto-
plasms of the same species of bacteria. We feel no doubt that in
these pioneering researches much labour has been to some extent
misspent, owing to the neglect of a common denominator. Only a
more accurate knowledge of bacteria or a recognised standard for
disinfecting experiments can ever supply such common denominator.
Species of bacteria for comparative-observation-experiments into the
action of chemical or physical agents must be not only the same
species, but cultured under the same conditions, and treated by the
agent in the same manner, otherwise the results cannot be compared
upon a common basis, or with any hope of arriving at comparable
conclusions.

2 E

434 DISINFECTION

In 1884 was issued from the English Local Government Board
one of the first adequate statements respecting the principles of
disinfection, as applied to the facts known respecting bacteria.*
In that report Dr Franklin Parsons arrived at the following
important conclusions : (a) that all infected articles which could
be treated by boiling water could not be so well disinfected in
any other way as by simply boiling for a few minutes; (b) that
for articles which could not be so treated, high pressure steam, with
complete penetration, was most satisfactory ; and (c) where articles
would be injured by either boiling or steam, dry heat at 240° F., if
sufficiently prolonged, would be effectual. He found that with the
exception of anthrax spores all the infected materials he experimented
upon were destroyed after an hour's exposure to dry heat at 220° F.,
or five minutes exposure to steam at 212° F. Anthrax spores
required four hours' dry heat at 220° F. Dry heat penetrates very
slowly into bulky and badly conducting articles, such as bedding.
Parsons also pointed out that at or above 250c F. "scorching"
occurred, and above 212° F. many kinds of stains were fixed in
fabrics, so that they could not be removed by washing. He advocated
that the standard of true disinfection should be the destruction of
the most stable infective matter known.

Previously to this period, experiments had shown the efficacy of
washing articles in boiling water, and Koch had shown the value of
corrosive sublimate. He had also shown the inefficacy of dry heat,
and of a number of chemical substances which it had been supposed
were disinfectants.

In 1887 the Committee on Disinfectants of the American Public
Health Association reported a number of findings, as the result of
experiment, which crystallised known facts. For infectious material
containing spores or sporulating bacilli they recommended burning,
steam under pressure 105° C. for ten minutes, boiling in water for
thirty minutes, chloride of lime 4 per cent., and mercuric chloride
1-500. If such material did not contain spores, or sporulating
bacilli, a 2 per cent, solution of chloride of lime sufficed, also mercuric
chloride 1-2000, carbolic acid 5 per cent., chlorinated soda 10 per
cent., and sulphur if 3 to 4 lbs. per 1000 cubic feet, and exposure
not less than twelve hours. For excreta the Committee advised
chloride of lime 4 per cent., for soiled underclothing, bed linen, etc.,
burning, boiling, or immersion for four hours in mercuric chloride
(1-2000), or carbolic acid (2 per cent.). For washing furniture
or hands the same solution of carbolic acid; for disinfecting the
bodies of the dead carbolic acid (5 per cent.), chloride of lime (4 per
cent.), or mercuric chloride (1-500); and for washing surfaces in

* Report of Medical Officer of Local Government Board, 1884.

MEANS OF DISINFECTION 435

»ick rooms and hospitals, 2 per cent, carbolic, or 1-1000 solution of
mercuric chloride.*

More recently a number of experiments have been carried out
in Europe and America as to the efficacy of certain chemical
substances, and reference will be made subsequently to some of the
results. Much of the evidence has been of a conflicting nature
which is clue, as we have said, to varying conditions, strengths of
disinfectants, and resistance of organisms.

Two or three years ago several workers at Leipzig f drew up
simple directions, the adoption of which would considerably assist
in securing a common standard for disinfectant research. They were
as follows : —

1. In all comparative observations it is imperative that mole-
cularly equivalent quantities of the reagents should be employed.

2. The bacteria serving as test objects should have equal powers
of resistance.

3. The number of bacteria used in comparative observation should
be approximately equal.

4. The disinfecting solution should always be used at the same
temperature in comparative experiments.

5. The bacteria should be brought into contact with the dis-
infectant with as little as possible of the nutrient material carried
over. (This obviously will depend upon the object of the research.)

6. After having been exposed to the disinfectant for a fixed time,
they should be freed from it as far as possible.

7. They should then be returned in equal numbers to the
respective culture medium most favourable to the development of
each, and kept at the same, preferably the optimum, temperature
for their growth.

8. The number of surviving bacteria. capable of giving rise to
colonies in solid media should be estimated after the lapse of equal
periods of time. J

Means of Disiufeetioii

We may now mention shortly some of the commoner methods
and substances adopted to secure efficient disinfection. They are all
divisible, according to Buchanan's standard, into two groups : —

1. Heat in various forms ;

2. Chemical bodies in various forms.

In practical disinfection it is necessary to inhibit or kill micro-
organisms without injury to, or destruction of, the substance harbour-

* Sternberg's Bacteriology, p. 201 el seq.

I Zeitschr. f. H;/(j. und Inf. Krank , xxv.

t See also "Standardisation of Disinfectants," by Kideal and Walker, Jour, of

Sauit. Inst., 1903.

436 DISINFECTION

ing the germs for the time being. If this latter is of no moment,
as in rags or carcases, cremation or burning is the simplest and most
thorough treatment. But with mattresses and beddings, bedclothes
and garments, as well as with the human body, it is obvious that as
a rule something short of burning is required.

Disinfection by Heat

From the earliest days of bacteriology heat has held a prominent
place as a means of disinfection. But it is only in comparatively
recent times that it has been fully established that moist heat is the
only really efficient form of heat disinfection. Boiling at atmo-
spheric pressure (100° C. or 212° F.) is the oldest form of moist heat
disinfection, and because of the simplicity of its application it has
gained a large degree of popularity. But it must not be forgotten
that mere boiling (100° C.) may not effectually remove the spores of
all bacilli, and obviously boiling is not applicable to furniture,
mattresses, and similar objects. For such objects hot-air ovens were
used in former days. But it was found that such dry heat disinfec-
tion (150° C. for an hour) injured articles of clothing, etc., and yet
left many organisms and spores untouched, as the degree of
temperature was rarely, if ever, uniform throughout the substance
being treated. The failures following in the track of these methods
were an indication of the need for some form of moist heat, viz.,
steam.

When water is heated certain molecular changes take place, and
at a certain temperature (100° C, 212° F.) the water becomes steam,
or vapour, and on very little cooling, or on coming into contact with
cooler bodies, will condense and give off its latent heat. But if the
vapour is heated, it will become practically a gas, and will not
condense until it has lost the whole of the heat, i.e. the heat of
making water into vapour plus the heat of making vapour into gas.
A gas proper is, then, the vapour of a liquid of which the boiling
point is substantially below the actual temperature of the gas.
But we know that the temperature at which it boils depends on the
pressure to which it is subjected (Eegnault's law). Hence in reality
"steam at any temperature whatever may be a vapour proper,
provided the pressure is such as prevents the liquid from boiling
below that temperature." In such a condition of vapour it is termed
saturated steam, or steam at or near its condensation point. Steam
at any pressure is "saturated," when it is at the boiling-point of
water for that pressure. But if it is at that same pressure further
heated, it becomes practically a gas, and is called superheated steam,
or steam heated above its natural condensation point. The former
can condense without cooling ; the latter cannot so condense at the

BY STEAM 437

same pressure. Saturated steam condenses immediately it meets the
object to be disinfected, and gives out its latent heat ; superheated
steam acts by conduction, and not uniformly throughout the object.
Its advantage is, that it dries moistened objects. It differs physically
from saturated steam, because it does not condense (and give out its
latent heat), until its temperature falls. Therefore, as a disinfecting
power, superheated steam is much less than saturated steam, it has
less heat in it, so to speak, and it has less penetrative power. One
further term must be defined, namely, current steam. This is steam
escaping from a clisinfector as fast as it is admitted, and may be at
atmospheric or higher pressures. The disinfecting temperature
which is now commonly used as a standard is an exposure to saturated
steam of 115° Q.for thirty minutes.

A number of different kinds of apparatus have been invented to
facilitate disinfection at this standard on a large scale. All the
larger Sanitary Authorities are now supplied with some form of
steam disinfector, though many are not furnished with high -pressure
disinfectors. Professor Delepine has pointed out that a current of
steam at low pressure may disinfect completely.* Whilst such simple
current-steam machines have thus been demonstrated as efficient
bactericides, for practical purposes it is important to have disin-
fectors, (a) capable of giving temperatures considerably above 100° C,
(b) of simple construction, (c) having a constant steam power of uni-
form temperature and rapid penetration, and (d) containing, when in
action, a minimum of superheated steam. In addition to these
characters of a first-rate steam disinfector, two other important
points in actual management should be borne in mind, namely, the
air must be completely ejected from the disinfection chamber before
the res alts due to steam are obtained, and some sort of automatic
indicator giving a record of eacli disinfection is indisjjensable.

The five chief types of steam disinfectors in common use are, the
Washington Lyon, the Goddard, Massey, and Warner, the Equifex
(Defries), the Thresh, and the Eeck.f

Washington Lyon's apparatus consists of an elongated boiler
having double walls, with a door at each end. The body of the
apparatus is in a "jacket" for the purpose of preventing loss of heat
and for " drying" disinfected articles after the process. The whole is
large enough to admit of bedding and mattresses, and generally is so
arranged that one end opens into one room, and the other end opens
into another room. This convenient position admits of inserting
infected articles from one room and receiving them disinfected into

Jour, of State Med., December 1897, p. 561.
I Full particulars of these various disinfectors may be obtained by communi-
cating with the makers. Elaborate catalogues are now issued with illustrations
and details of working.

438 DISINFECTION

the other room. Possible reinfection is thereby removed. Steam
is admitted into the jacket at a pressure of between 20 and 25
lbs., and its penetrating power may be increased by intermitting
the pressure during the disinfection. At the end of the operation
a partial vacuum is created, by which means much of the moisture
on the articles may be removed. In some cases a current of warm
air is admitted before disinfection in order to diminish the extent of
condensation.

The Equifax (Defiles) contains no steam jacket, but coils of pipes
are placed at the top and bottom of the apparatus, with the object of
imparting to the steam as much heat as is lost by radiation through
the walls of the disinfecting chamber, and at the same time of pre-
venting undue condensation, and to be available for drying. The air
is first removed by a preliminary current of steam, after which steam
at a pressure of 10 lbs. is intermittently introduced (for about
five minutes), and allowed to escape. The object of this proceeding
is to remove air from the pores of the articles to be disinfected by
the sudden expansion of the film of water previously condensed on
their surface.

The apparatus introduced by Thresh was constructed with a view
of overcoming the objection to some of the other machines, that
bulky articles retained a large percentage of moisture, thus necessi-
tating the use of some additional drying apparatus. A central
chamber receives the articles to be disinfected, and is surrounded by
a boiler containing a solution of calcium chloride (carbonate of potash
is now used) at a temperature of 225° F. This is heated by a small
furnace, and the steam given off is conducted into the central
chamber. Owing to the dissolved potash the temperature of the
steam given off when the solution boils is several degrees higher than
ordinary steam. The steam is not confined under any pressure
except that of the atmosphere. When the steam has passed for a
sufficient length of time, it is readily diverted into the open air. Hot
air is now introduced, and at the expiration of an hour the articles
may be taken out disinfected and as dry as they were when inserted.
The apparatus is comparatively inexpensive, and not of a complicated
nature. The current steam is saturated, and at a temperature a few
degrees above the boiling-point. The apparatus is now made in
various forms, portable or otherwise.

There are many other forms of steam disinfector, including the
apparatus by G-oddard, Massey, and Warner, the disinfector of Dele-
pine, and that of Keck and others, and each has its enthusiastic
supporters.

BY CHEMICALS 439

Disinfection by Chemical Substances

The effects of chemical substances as solutions, or in spray or
gaseous form, upon bacteria have been observed from the earliest
days of bacteriology. To some decomposing matter or solution a
disinfectant was added and sub-cultures made. If bacteria continued
to develop, the disinfection had not been efficient ; if, on the other
hand, the sub-culture remained sterile, it was assumed that disinfec-
tion had been complete. From such rough-and-ready methods large
deductions were drawn, and it is hardly too much to say that no
branch of bacteriology contains such a mass of unassimilated and un-
assimilable statements as that relating to research into disinfectants.
Most of the tabulated and recorded results are conspicuous in having
no standard as regards bacterial growth. Yet without such a
standard results are not comparable.

Silk threads, impregnated with anthrax spores, were placed in
bottles containing carbolic acid of various strengths, and at stated
periods threads were removed and placed in nutrient media, and
development or otherwise observed. But, as Professor Crookshank
pointed out, this method is fallacious, the thread being still wet with
the solution when transferred to the medium, and thus the culture
was modified or even inhibited altogether.* It is unnecessary for us
here to discuss every mode adopted by investigators in similar
researches. We may, however, point out that the most approved
methods at the present time are based more or less upon two simple
modes of exposure. In one a known volume of recent broth culture
of an organism grown under specified conditions is used, and to this
is added a measured quantity of the antiseptic. At stated periods
loopfuls of the broth and antiseptic mixture are sub-cultured in
fresh-sterilised broth, and resulting development or otherwise closely
observed. The other method is practised in dealing with volatile
bodies. In such cases a standard culture is made of the organism in
broth at a standard temperature. Into this are dipped small strips
of sterilised linen. When thoroughly impregnated, these are
removed from the broth and subsequently dried over sulphuric acid
in a vacuum at 38° C. These may now be exposed for a longer or
shorter period to the fumes of the antiseptic in question, and broth
cultures made at the end of the exposure. It is obvious that a very
large number of modifications are possible of these two simple
devices for testing the bactericidal power of chemical substances. It
should be remembered that here, perhaps, more than anywhere else
in bacteriological research, careful "control" experiments are
absolutely necessary.

Recently, Ainslie Walker suggested various conditions of experi-
* Bacteriology and Infective Diseases, p. 35.

440 DISINFECTION *

mentation with a view to obtaining comparable results. First,
he recommended the use of a well-known disinfectant giving
regular and consistent results, such as pure phenol. Secondly, the
source and age of the culture used is of importance. If the
culture be in broth, Walker suggests the following procedure :
to 5 c.c. of a twenty-four hours' blood-heat culture 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 dis-
tribute it evenly in sterilised water. The resulting emulsion may be
used in place of the broth culture. Thirdly, Walker emphasises the
importance of working with the same organism for comparable results.*
A substance, to be a satisfactory disinfectant, should, according
to Andrewes, possess five characters: (a) it should be germicidal
within a reasonable time-limit; (b) it should not possess chemical
properties which unfit it for ordinary use ; (c) it should be soluble in
water, or capable of giving rise to soluble products in contact with
the material to be disinfected ; (d) it should not produce injurious
effects on the human tissues ; and (c) it should not be too costly in
proportion to its germicidal value. f

Mineral acids (nitric, hydrochloric, sulphuric), especially concen-
trated, are all germicides, but owing to their corrosive action their
application is limited.

A number of bodies, such as chloroform and iodoform, have been
much advocated as antiseptics. The cost of the former and odour of
the latter have, however, greatly militated against their general
adoption.

Chloride of lime is a powerful disinfectant. Professor Sheridan
Delepine and Dr Arthur Ransome have demonstrated its germicidal
effect as a solution (1 per cent.) applied directly to the walls of rooms
inhabited by tuberculous patients, j Coates confirmed these results
in houses in Manchester infected by consumptives. Chlorinated
lime ought to be used which will yield not less than 33 per cent, of
available chlorine. It may also be used in solid form for decompos-
ing matter, excreta, etc.

Mercuric chloride (corrosive sublimate) has been an accepted
germicide for some time. But the experiments of Behring,
Crookshank, and others, have proved that the weaker solutions
(1-4000) cannot be relied upon. This is, in part, due to the fact that

* Practitioner, 1902, lxix., p. 523.

t Many useful hints and suggestions as to testing disinfectants, and on the whole
process of disinfection will be found in Lessons in Disinfection and Sterilisation, by
F. W. Andrewes, M.D., F.R.C.P., 1903, p. 81 et seq. ; see also Brit. Med. Jour.,
1904, ii., p. 13.

X Brit. Med. Jour., 1895, vol. i., p. 353.

BY SULPHUR 441

it forms in albuminous liquids an albuminate of mercury which is
inactive. Dilute solutions have the further disadvantage of being
unstable. Various authorities recommend a solution of 1-500 as
a germicide, and much weaker solutions are of course antiseptic.
An ounce each of corrosive sublimate and hydrochloric acid in 3
gallons of water makes an efficient disinfectant.

Potassium permanganate, is, of course, the chief substance in
Condy's fluid, as zinc chloride is in Burnett's disinfecting fluid. A 5
per cent, of the former and a 2 \ per cent, of the latter are germicidal.
Solutions are used for street-cleansing.

Boracic acid is used as an antiseptic with which to wash sore
eyes, or preserve tinned foods or milk. It is not a strong germicide
(it inhibits rather than kills), but an unirritating and effective wash.
Many cases of its addition to milk have found their way into the
law courts owing to cumulative poisoning, and as a rule its use as a
food perservative should be deprecated.

Carbolic acid has come much into prominence as an antiseptic
since its adoption by Lister in antiseptic surgery. It is cheap,
volatile, and effective. One part in 40 is antiseptic, and 1 in 20
germicidal. As a wash for the hands the former is used, and a
weaker solution for the body generally. Carbolic soap and similar
toilette combinations are now very common. At one time it
appeared as if corrosive sublimate would take the place of carbolic
acid as an antiseptic solution, but a large number of experiments
have confirmed opinion in favour of carbolic. Crookshank found
that carbolic acid, 1 in 40, acting for only one minute, was sufficient
to destroy Streptococcus pyogenes, S. erysipelatis, and Staphylococcus
pyogenes aureus, and in the strength of 1 in 20 carbolic acid completely
sterilised tubercular sputum when shaken up with it for one minute.
Klein, Houston, and Gordon, and other workers have found a 5 per cent.
solution of carbolic to be a reliable disinfectant for almost all bacteria.

Cresol, a member of the phenol series, is a good disinfectant and
the active clement in lysol, Jove's fluid, creolin, izal, and other
similar substances, which have been recently introduced and have
proved efficacious as disinfectants.

Sulphurous acid is one of the commonest disinfectants employed
for fumigation — the old orthodox mot hod of disinfecting a room in
which a case of infective disease had been nursed. It is evolved, of
course, by burning sulphur. For eacli thousand cubic feet from 1
to 5 lbs. of sulphur is used, and the walls may be washed with
carbolic acid. Dr Kenwood carried out some experiments in 1896
which appeared to support a belief in the disinfecting power of sulphur
fumes.* But he has since advocated formaldehyde as preferable.
He fdund that the B. diphtheria? was not killed by sulphur though

Brit. Med. Jour., 1896 (August), p. 439.

442 DISINFECTION

markedly inhibited, when the sulphurous gas (S02) did not
much exceed *25 per cent. But the bacillus was killed" where the
sulphur fumes exceeded "5 per cent. Both these results had
reference to the S02 in the air in the centre of the room at a height
of 4 feet, and after the lapse of four hours. There can be little
doubt that thoroughly fuming a sealed-up room with sulphur in a
moist atmosphere, and leaving it thus for twenty-fours, is generally
if not always, efficient disinfection. Moreover, its simplicity of
adoption is greatly in its favour. Anyone can readily apply it by
purchasing a few pounds weight of ordinary roll sulphur and burning
this in a saucer in the middle of a room which has had all its crevices
and cracks in windows and walls blocked up with pasted paper.*
But it is almost useless as a gaseous disinfectant unless used in a
particular way. The following seem to be the only lines upon which
anything like adequate disinfection can be secured by means of
sulphur : —

1. The room to be disinfected must be effectually sealed up.

2. Not less than 3 lbs. of sulphur should be used for every
1000 cubic feet.

3. Twenty-four hours should elapse between the time of lighting
the sulphur and the unsealing of the room.

4. The air in the room should be damp during the process, and
this may be achieved by steam, or spraying the walls with water, or
suspending wet blankets. By this means sulphurous acid is formed,
which is the essential part of the process.

5. At the end of the twenty-four hours the doors and windows
should be kept wide open for at least one, and if possible for two, days.

6. Furniture and fixtures should, as far as possible, be wiped
down with a damp cloth soaked in carbolic or some other disinfectant
solution. Dry dusting or sweeping should be strongly deprecated.
The walls may be stripped in cases where they are very dirty or
where there has been a recurrence of a disease. Sulphur fumigation
is not sufficient in disinfection after consumption.

The conclusions of Dr Novy respecting the efficacy of sulphur
fumes as a disinfectant may be added. He urges that "sulphur
fumes possessed little or no action on most bacteria when in a
dried state. If, however, the specimens are actually wet, they
will be destroyed except in the state of the resistant forms, such
as spore stage and tubercle bacilli. For tubercle bacilli or spore-
containing material, wet or dry, it is of no value. It can be used
for the disinfection of rooms which have been infected with
ordinary disease organisms. From 3 to 6 lbs. of sulphur must be
burned in each 1000 cubic feet of space. The walls, floors, and
articles should be sprayed with water. The room should be made

* See also Public Health, 1900, p. 438 et seq.

BY FORMALIN 443

perfectly tight, and should be kept closed at least twenty hours." *
Calmette states that sulphur vapour under pressure may he relied
upon for the disinfection of ships, etc.

Kecently, formalin has come much into favour as a room
disinfectant. Formalin is a 40 per cent, solution in water of
formaldehyde, a gas discovered by Hofmann in 1869. This gas is a
product of imperfect oxidation of methyl alcohol, and may be
obtained by passing vapour of methyl alcohol, mixed with air, over a
glowing platinum wire or other heated metals, such as copper and
silver. Its formula is CH20, and it is a colourless gas with a pungent
odour, and having penetrating and irritating properties particularly
affecting the nasal mucous membrane and the eyes of those working
with it. It is readily soluble in water, and in the air oxidises into
formic acid (CH202). This latter substance occurs in the stings of
bees, wasps, nettles, and various poisonous animal secretions.
Formalin is a strong bactericide even in dilute solutions, and, of
course, volatile. Its use should be restricted to disinfection of
articles injured by heat (furs, etc). A solution of 1-10,000 is said
to be able to destroy the bacilli of typhoid, cholera, and anthrax. A
teaspoonful to 10 gallons of milk is said to retard souring. When
formalin is evaporated down, a white residue is left known as
paraform. In lozenge form this latter body is used by combustion
of methylated spirit to produce the gas. Hence we have three
common forms of the same thing— formalin, formic aldehyde, para-
form— each of which yields formic acid, and thus disinfects. The
vapour cannot in practice be generated from the formalin as readily
as from the paraform.

By a variety of ingenious arrangements, formic aldehyde lias been
used by a large number of observers during the last two or three
years. We may refer to four modes of application: 1. The sprayer
produces a mixture of air and solution for spraying walls, ceilings,
floors, and sometimes garments. There are a number of different
forms of spraying apparatus such as the Equifex, the Mackenzie,
the Eobertson, etc. 2. The autoclave (Trillat's apparatus). In this
apparatus a mixture of a 30-40 per cent, watery solution of
formaldehyde and calcium chloride (4-5 per cent.) is heated under
a pressure of three or four atmospheres, and the almost pure, dry
gas is conducted through a tube passing through the keyhole of the
door into the sealed-up room. 3. The paraform larrup (the Alformant).
The principle of this lamp is that the hot, moist products from the
combustion of methylated spirit act upon the paraform tablets,
converting them into gas. 4. Lingners apparatus consists of a ring
boiler in which steam is generated and driven into a reservoir filled

• Tenth Report of State of Maine Board of ll<<tlih< 1898, p. 365. This report
contains a digest on the whole subjeet of disinfection.

444 DISINFECTION

with formalin or glyco -formal (30 per cent, formalin with 10 per
cent, glycerine), which is thus vaporised and ejected in the form
of a fine spray through four nozzles. A room is thereby speedily
filled with a dense formalin vapour. After four hours exposure,
Houston and the writer found that B. pyocyaneus, Staphylococcus
'pyogenes aureus, and various saprophytic organisms were killed.*
Klein and the writer found that B. anthracis and the tubercle
bacillus were killed by the same means. Eideal claims that the
resistant spores of anthrax may be killed when 7 '5 c.c. of formalin
per cubic metre (85 grammes of formaldehyde per 1000 cubic feet)
are vaporised with not less than four times its volume of water, and
that exposure need not exceed six hours.f Klein, Houston, and
Gordon found that B. typhosus, B. diphtherial, and certain suppu-
rative organisms were killed by means of the alformant lamp
method.^ It is agreed that the gas is harmless to colours, metals,
leather, and polished wood. The vapour acts best in a warm
atmosphere. As for its action on bacteria, it may be said that it
compares favourably with any other disinfectant.

Many observers have not recommended formaldehyde on account
of its professed lack of penetrating power. Professor Delepiue,
however, states that it possesses "penetration powers probably
greater than those of most other active gaseous disinfectants. B.
coli, B. tuberculosis, B. pyocyaneus, and Staphylococcus pyogenes aureus
were killed in dry or moist state, even when protected by three
layers of filter-paper." § In Professor Delepine's opinion, the
vapours of phenol, izal, dry chlorine, and sulphurous acid have,
under the same conditions, given inferior results. Since 1898 a
number of experimenters have confirmed these opinions. It is
extremely important that that disinfectant should be used which is
the most suitable one for the particular purpose at issue. A
germicidal substance which under certain conditions, and in relation
to one species of organism may be practically useless, may under
other conditions be most efficacious.

Practical Disinfection II

To disinfect a room, seal up cracks and crevices, spray the walls
with water, and burn, say, 3-6 lbs. of roll sulphur for every

* Practitioner, 1902, vol. Ixix., p. 328.

f Jour, of Sanitary Institute, 1903, vol. xxiii., part iv.

% Report of Medical Officer of London County Council, 1902.

§ Jour, of State Med., 1898 (November), p. 541.

|| For hints in the detail management of disinfection, the reader is recommended
to study A Practical Guide to Disinfection, by Rosenau and Allan, 1903 ; Lessons
in Disinfection, by F. W. Andrewes, 1903; the Practitioner, 1902, p. 300 (Houston);
RideaFs Disinfection and Disinfectants, 1904 ; and Public Health, 1904, pp. 558-570.

PRACTICAL DISINFECTION 445

1000 cubic feet of space.* Let the room remain sealed up for
twenty-four hours, then be freely opened. Formaldehyde gaseous
disinfection may be used as described above. But it would appear
that neither sulphur or formaldehyde are always reliable in dis-
infecting after tuberculosis. The most important point is to
cleanse surfaces, and probably the most efficient disinfection of a
room is by Lingner's apparatus (glyco-formal, 1 litre to every
1000 cubic feet), coupled with spraying or washing surfaces with
germicidal solution.

To disinfect walls, floors, etc., wash or spray with chloride of
lime solution (1-100), izal (1-100), formalin (2-100), or carbolic acid
(1-40). The last-named solution may be used to wipe down furniture.
These disinfectants may be used after sulphur fuming. Formic
aldehyde may also be used by autoclave or Lingner's apparatus.

To disinfect ocdding, etc., the steam sterilisation secured in an
efficient apparatus is the best (115° C. for thirty minutes). Bags and
infected clothing, unless valuable, should be burnt.

To disinfect garments and wearing ajpjparel. — If possible, steam in
an efficient steriliser ; if that be not available, such articles should
be washed in a disinfectant solution (5 per cent, carbolic), or fumed
with formic aldehyde (Lingner's glyco-formal apparatus).

To disinfect excreta or putrefying solutions, enough disinfectant
should be added to produce in the solution or matter being disinfected
the percentage of disinfectant necessary to act as such. Adding a
small quantity of antiseptic to a large volume of fluid or solid is as
useless as pouring a small quantity of antiseptic down a sewer with
the idea that such treatment will disinfect the sewage. The mixture
of the disinfectant with the matter to be disinfected must contain
the standard percentage for disinfection. Chloride of lime is a
common substance for use in this way (\ lb. to a gallon of water) or
in a 4 per cent, solution. Potassium permanganate (1-100), and
carbolic (5 per cent.), and many manufactured bodies containing
them, are also widely used. Corrosive sublimate (1-500), izal (1-100),
copper sulphate (1-20), lysol, cresol, or creolin (1-40), have all been
found efficacious (Houston). Drs Hill and Abram recommend that
the excreta and disinfectant be thoroughly mixed, and stand for at
least half an houf.f For various reasons they particularly advise
chinosol as the most convenient disinfectant for this specific purpose.
But subsequent experience has perhaps hardly supported this
recommendation.

Antiseptics for wounds. — Carbolic acid (1-40) or corrosive sub-
limate (1-1000) are commonly used in surgical practice. Boracic

• The measurement of cubic space is, of course, made by multiplying together in
feet the length, breadth, and height of a room.
I Brit. Med. Jour., 1898 (April), p. 1013.

446 DISINFECTION"

acid is one of the most unirritating antiseptics which is known. It
may be used in saturated watery solution (1-30) or dusted on
copiously as fine powder. It is especially applicable to open wounds,
and as an eye-wash.

Boots, books, leather '-covered articles, etc., should be disinfected by
dry heat or formalin (preferably Lingner's apparatus).

Infected linen should be steamed or boiled, but if that is not
available, immersion for one hour in corrosive sublimate (1-500) or
for twenty-four hours in the same solution 1-1000. Less powerful
germicides have, however, been found successful, e.g. izal (1-100),
carbolic acid (1-100).

Cups, saucers, plates, spoons, knives, forks, etc., should all be
disinfected in boiling water.

Bags in bales can only be disinfected by steam.

Pens, byres, stables, trucks, vans, markets, etc., are best treated
with some form of sprayer {e.g. Equifex hot-spray disinfector) or
distributor {e.g. the chloros distributor). Ships also may be treated
by this apparatus or by means of the Newcastle disinfecting hulk
(Goddard, Massey, and Warner).

Disinfection of the Hands. — To a surgeon, the disinfection of the
hands is a matter of vital importance. There are many opportunities
for conveying bacteria on the hands, which naturally come in the
way of dust and dirt, and so carry organisms in the cracks of the
skin surface, in the sebaceous glands, under the nails, and even in
the substance of the epithelium. This was demonstrated by
Lock wood in 1896, and again by Freeman in 1899. Subsequent
experiments confirmed the fact of the difficulty of completely freeing
the skin of the hands from micro-organisms. In 1902, Dr Schaeffer
of Berlin, whilst recognising that absolute asepticism of the hands
is not possible, showed by experiments that it is possible to render
the hands so free from organisms during a surgical operation that
the danger of wound contamination is exceedingly small. Collins
has pointed out (1904) that much depends upon vigorous scrubbing,
clean nail-brushes, and hot water. Soap, water, and carbolic acid
(1-20), permanganate of potash, corrosive sublimate (1-1000),
lysol, and many other similar antiseptic solutions have been
used with more or less satisfactory results. Schaeffer, how-
ever, decides in favour of the hot-water-alcohol method (96 per
cent, spirit), the chief advantage of which is that it removes
organisms from the skin rather than killing them on the skin.
Mikulicz advocates spirit-soap as cheaper than alcohol, but apparently
the difference in expense in this country is not great, and the spirit-
soap leaves the hands in a slippery condition. It may be pointed
out that washing in spirit rather than antiseptics preserves the
smooth surface of the skin and results in no roughness.

PRACTICAL DISINFECTION 447

Disinfection in or after Special Diseases

Disinfection after Phthisis.— The following statement was
drawn np in 1901, by Drs Newsholme, Mven, and the writer, for
the National Association for the Prevention of Consumption and
other forms of Tuberculosis. It may serve as a basis for practical
disinfection of rooms, etc., after phthisis : —

" The necessity for disinfection in consumption is based on well-
established facts. The essential cause of consumption and of all
other forms of tuberculosis is a living microbe, the B. tuberculosis,
though the condition of the bodily health of the individual greatly
influences the resistance to the disease and the prospect of recovery
from it.

" The disease is always contracted by taking into the system the
microbes causing it, which are derived solely from persons or
animals suffering from the same disease. These microbes may be
taken in infected milk or less commonly in infected flesh.

"The most frequent source of infection, however, is the dis-
charges and particularly the phlegm (spit or expectoration) of a
consumptive person. These discharges whilst moist are not likely
to be scattered, but if allowed to dry they become broken up into
dust, and are then extremely dangerous. There is little or no risk
of contracting consumption directly from the breath of a consumptive
person, but the phlegm infects everything upon which it falls —
handkerchiefs, books, papers, linen, floors, carpets, furniture, etc.,
and is then readily inhaled by healthy persons. This is the chief
means by which consumption is spread from person to person.

" On these facts rest the important question of disinfection. In
preventing a consumptive person from spreading the disease, two
sets of preventive measures are required: — 1st, the removal or
destruction of the infective matter already disseminated by the
patient's discharges, especially by his phlegm; and, 2nd, the
prevention of future dissemination. For the latter purpose the
main object is not to permit any discharge to become dry before
being destroyed. Before the consumptive person lias learned the
persona] precautions which must be taken, and up to the time when
he has been trained to carry them out carefully, he lias probably
distributed a considerable amount of infective matter. This is
especially liable to accumulate in a dangerous form at home, where
the space is small, and light and ventilation are defective. Infective
]) M holes will be found in greatest abundance on and near the floors,
on ledges, and in room-hangings. But the personal clothing and
bedclothes will also have become infected. Hence it is necessary
to disinfect the floor, walls, and ceiling of the rooms occupied by
the patient, as well as the furniture, carpet, bedclothes, etc.

448 DISINFECTION

" When this has been done, if the personal precautions advised
are carried out by the consumptive, further disinfection should not
be needed.

" It is, however, difficult to make sure that personal precautions
are fully carried out, and rooms should therefore be subsequently
cleaned at least once in six months, the floors being scrubbed with
soft soap, the furniture washed, the walls cleaned down with dough,
and the ceiling whitewashed.

"Confined workshops in which a consumptive has worked for
some time should be cleansed, and a notice in reference to spitting
should be suspended in all workshops. The latter precaution should
also be observed in all public-houses and common lodging-houses,
both of which require special attention to cleansing.

" Disinfection of rooms which have been occupied by con-
sumptive patients may be secured in various ways, but the following
are the practical rules which must underlie any methods adopted : —

" 1. Gaseous disinfection of rooms, or ' fumigation ' as it is
termed, by whatever method it is practised, is inefficient in such
cases.

" 2. In order to remove and destroy the dried infective discharges,
the disinfectant must be applied directly to the infected surfaces of
the room.

"3. The disinfectant may be applied by washing, brushing, or
spraying.

" 4. Amongst other chemical solutions used for this purpose, a
solution of chloride of lime (1 to 2 per cent.) has proved satisfactory
and efficient.

" 5. In view of the well-established fact that it is the dust from
dried discharges which is chiefly infective, emphasis must be laid
upon the importance of thorough and wet cleansing of infected
rooms.

"6. Bedding, carpets, curtains, wearing apparel, and all similar
articles belonging to or used by the patient, which cannot be
thoroughly washed, should be disinfected in an efficient steam
disinfector.

" 7. After all necessary measures of disinfection have been carried
out, the essential principle governing the subsequent control of a
case of consumption is that all discharges, of whatever kind (especi-
cially expectoration from the lungs), should under no circumstances
be allowed to become dry."

In Manchester and other places, where disinfection after phthisis
is regularly practised, a solution of chlorinated lime of the strength
of 1J ounces to the gallon is used. The wall-paper is thoroughly
saturated with this solution, applied with a soft brush or spray,

AFTER SPECIAL DISEASES 449

and is then, where necessary, stripped from the walls. The bare
walls, the ceiling, and floor are washed over several times with
the solution, and any articles of furniture which will admit of such
treatment are similarly washed over. Articles of clothing, bedding,
etc., are taken away to be disinfected in the steam disinfector.

In houses in Manchester which are in a clean condition, and
where it is certain that there has been no direct soiling of the walls
or floors with sputum, and where the infectious dust, if present, has
come from soiled pocket-handkerchiefs or articles of clothing, the
chlorinated lime method of disinfection is not considered necessary,
and the method of disinfection recommended by Esmarch is
practised: — The wall-paper is rubbed well with crumb of bread, or
with dough kneaded to a proper consistency. Floors, painted walls,
and woodwork are washed with soap and water, and ceilings are
limewashed. In addition, bedding, articles of clothing, etc., are
either disinfected by steam or washed with boiling water.

This method of disinfection, when properly carried out, was
found to remove practically all dust from a room, so that little or
no dust can be obtained by subsequently rubbing the wall-paper
with a sterilised sponge. The method, however, requires a certain
amount of care to make sure that all dust is removed from the walls,
especially from the angles and corners, and to properly rub down a
fair-sized room takes a considerable time. It is useless in cases
where the paper is directly soiled with sputum. Owing to the
mucus which it contains, the dried sputum sticks tenaciously to the
paper, in spite of repeated rubbing with dough.

This method of rubbing the walls with dough is an excellent
way of periodically cleaning a room, so as to keep it free from
dust.

After Small-Pox. — It is necessary that disinfection be very
thoroughly done. As a rule, the walls of the room used by the
patient must be " stripped and cleansed." Fumigation with formic
aldehyde and vigorous spraying of walls are usual. All bedding and
wearing apparel must be steamed, and if very unclean, burnt.

After Scarlet Fever.— The room used by the patient should be
disinfected in the ordinary way. Infection may be conveyed by
clothing, carpets, table-cloths, bell-ropes, etc., and such things must
receive attention. Infection is also probably conveyed by the peeling
skin, and even more so by the throat secretions. All discharges
from the mouth and nose, and also those from the ear when affected,
should be received on rags or thin paper handkerchiefs and burned.
The seat of infection may also be directly attacked by the use of
disinfectant gargles, of which chlorine water is one of the best.
During desquamation the skin may be oiled, and occasionally washed
in warm carbolic solution (1-40).

2 F

450 DISINFECTION

After Diphtheria. — The bacillus of diphtheria is non-sporulating,
and has comparatively little resistance against disinfectants. Ordi-
nary means of disinfection are therefore sufficient. Local disinfec-
tants should be used for the throat until bacteriological examination
is negative to the Klebs-Loffler bacillus, which may persist in the
throat for long periods. The throat may be painted with a solution
of perchloride of mercury (1-500); 15 to 20 minims of such a
solution would be a suitable amount to use for a single application.
Gargles or sprays may be employed, consisting of chlorine water, or
permanganate of potash (1-300). The throat and nose discharges
should be received on rags which can be burned.

After Typhoid Fever and Cholera.— Bedding and articles
which have come into contact with the patient require attention
in typhoid fever and cholera. The disinfection of the excreta
(faeces and urine) is the most important item. These discharges
should not be passed into the house drains until disinfected.
They should stand for some hours thoroughly mixed with the
disinfectant before being considered disinfected. Chloride of lime
(1-500 of the total mixture), izal (1-200 of the total mixture) and
carbolic acid (1-40 of the total mixture) are all used in this way.
If there is no house-drainage or water-carriage system, the excreta
should be treated as above, and deeply buried remote from any well
or water-course. The nurse's hands must be kept thoroughly
cleansed (thorough washing with hot water, soap, and perchloride
solution, 1-1000), especially before meals.

After Plague. — The detailed arrangements for the removal of
cases and disinfection of infected tenements after plague should be
under the personal supervision of the medical staff, and may be
detailed as follows : —

(a) Kemoval of patient to hospital.

(b) Kemoval of "contacts" to reception house, and kept under
medical observation for fourteen days.

(c') Fumigation of infected house by liquefied sulphur dioxide or
formic aldehyde from twelve to twenty-four hours, the disinfectant
being used in proportion to the cubic space dealt with.

(d) After the fumigation the house is entered ; all articles of
clothing, etc., to be removed are first of all thoroughly wetted with
2 per cent, solution of formalin (1 gallon 40-per-cent. solution
formaldehyde to 50 gallons water), or 2 per cent, chloride of lime,
then wrapped up in sheets soaked in the same fluid and removed to
the sanitary wash-house. There all articles which cannot be boiled
or steamed, or treated with formaldehyde, are burned.

(e) The walls, ceiling, flooring, woodwork, etc., and furniture of
the infected house are also sprayed with the formalin solution (1
gallon to 50 gallons water) or chloride of lime.

AFTER SPECIAL DISEASES 451

(/) All rooms in the infected dwelling are cleansed; the lobbies,
stairs, and landings being dealt with by formaldehyde or chloride of
lime solution.

Q/) Courts of such dwellings are watered with chloride of lime
solution.

(h) Ash-pits have contents watered with same, and then removed
and burned.

APPENDIX

NOTES ON TECHNIQUE

Synopsis of Technique :— General Methods of Examination; Staining Methods;
Flagella; Spores, etc. — Bacteriological Diagnosis — Examination of Water —
Examination of Milk — Bacteriological Diagnosis in Special Diseases —
Examination of Malaria Blood— Examination of Oysters — Examination of
Sewage — Miscellaneous.

General Elementary Methods of Examination

With the exception of pathological tissue and similar insoluble
substances, the common practice in bacteriology is to reduce as far as
possible the article to be examined to a fluid, that is to say, it is
chiefly fluids which can be systematically examined by the methods
of bacteriology. Water, milk, sewage, urine, blood, etc., are at once
in a condition to make examination available, but cheese, butter, foods,
soil, pus, dust, etc., require to be reduced to fluid, or washed in fluid
media, preparatory to examination. Thus soil particles may be washed
and macerated in sterilised broth, and the broth examined for contained
organisms. It will, on this account, be most convenient in the first
place to consider the application of bacteriological methods to the
examination of fluids.

The principle underlying the ordinary technique is the solidification
of fluid gelatine at or below room temperature. If a drop of con-
taminated water, for example, be added

to a tube of 10 c.c. of liquid gelatine, fC^ ^j

thoroughly mixed, and then the contents |^

of the tube poured out into a Petri plate C^______ _^J

(or other shallow glass dish) and allowed to ,,|( 88.— Petri Dish

solidify, we shall have scattered through

the solid film of gelatine the contained bacteria, in a favourable medium
for their growth and multiplication. Such a plate will be protected
from the air and incubated at a regulated temperature. This is the
principle of Koch's Plate Method. In the course of two or three days
the film of gelatine on the plate becomes covered with colonies of
germs, consisting of countless individual bacteria gathered round the

4r,3

454

APPENDIX

parent organism which found its way thither from the drop of con-
taminated water. The next step is to examine these quantitatively
and qualitatively.

1. Naked-Eye Observation of the Colonies. — By this means, at the very
outset certain facts may be obtained, viz., the size, elevation, configura-
tion, margin, colour, grouping, number, and kinds of colonies, all of
which facts are of importance, and assist in final determination as to the
quantity and character of the organisms present in the original drop of

Fio. 39.— A Diagram of Colonies of Bacteria on a Gelatine Hate.

water. Moreover, in the case of gelatine medium (owing to the fact
that it is liquefiable by ferments), one is able to observe whether or not
there is present what is termed liquefaction of the gelatine. Some
organisms produce in their development a peptonising ferment which
breaks down gelatine into a fluid condition. Many have not this power,
and hence the characteristic is used as a diagnostic feature.

2. The Microscopic Examination of Colonies (under low magnification,

x 60-100) confirms or corrects that which has been observed b}T the naked

eye. Micro-organisms, when growing in colonies, produce cultivation

features which are peculiar to themselves (especially is this so when

APPENDIX 455

growing in test-tube cultures), and in the early stages of such growths
a low power of the microscope or a magnifying glass facilitates
observation.

3. The Making of Cover-glass Preparations : (a) unstained — " the hang-
ing drop " ; (6) stained — single stains, e.g., gentian-violet, methylene-
blue, fuchsin, carbol-fuchsin, etc. ; or double stains by Gram's method,
by Ziehl-Neelsen's method, etc. This third part of the investigation is
obviously to prepare specimens for examination under the microscope.*
" The hanging drop " is a simple plan for securing the organisms for

microscopic examination in a more ^

or less natural condition. A hollow f\ \ CTTV^v "^^V

ground slide {i.e. a slide with a \f— >>t

shallow depression in it) is taken,
and a small ring of vaseline placed

round the edge of the depression. V7///////////>/}ih>^^n/)/////////friV7m
Upon the under-side of a clean fig. 40. -The Hanging Drop,

cover-glass is placed a drop of dis-
tilled water, and this is inoculated with the smallest possible particle
taken from one of the colonies of the gelatine plate on the end of a
sterilised platinum wire. The cover-glass is then placed upon the ring
of vaseline, and the drop hangs into the space of the depression. Thus
is obtained a view of the organisms in a freely moving condition, if
they happen to be motile bacteria. In ordinary practice the hollow
slide may be dispensed with, and an ordinary slide used.

With regard to staining, it will be undesirable here to dwell at length
upon the large number of methods which have been adopted. The

* A good microscope is essential. It should have objectives of 1 inch, \, and
TV (oil immersion). A white light, and proper adjustment of the substage condenser
and draw-tube are also necessary. A lens of Vjth inch focal depth is the usual
power required for the study of bacteria, although in some cases a lens of a focal
length of ,Vth inch, or even stronger, is desirable. Streptothrix actinomyces, which
belongs to the higher bacteria, is better seen with a power of J;th inch than with one of
X\X\\ inch. The principle of the immersion lens is the filling up of the space between
the lens and the cover-glass with a material whose refractive index is the same as
that of the lens, so that there will be no loss of illumination by the rays of light
passing through media of different powers of refraction, while proceeding from the
object to the lens. The power of a microscope varies not only according to that of
the lens, but also according to the power of the eye-piece. Thus the magnifying
power of a 1-inch objective in Swift's microscope varies, according to the strength
of the eye-piece and to the fact that the draw-tube is closed or extended, from 25
to 140 diameters; a Jth inch objective, from 175 to 690 diameters ; and a jVth inch
objective, from 385 to 1627 diameters. As a high-power lens gives a picture which
has comparatively very little "depth" of focus, it is necessary to place the object
under examination in as nearly the same plane as possible. Hence the material
to be investigated should be reduced to an extremely thin film.

The object should also be in the optical axis of the instrument, and secured in
position by means of the spring clips. In using the oil immersion lens, the body
tube of the microscope must be screwed down until the lens is in contact with the
oil and nearly touching the coverslip. The substage condenser must be screwed up
Hush with the stage. The best light must be obtained by adjustment of the mirror,
and fine focus must be used. A skilful use of the microscope depends, of course,
upon an understanding of its parts and upon practice.

F

456 APPENDIX *

"single stain" may be shortly mentioned. It is as follows. A clean
cover-glass or slide is taken (cleaned with nitric acid and alcohol, or
bichromate of potash and alcohol), and a drop of distilled water placed
upon it. This is inoculated with a particle of a colony on the end of a
platinum needle, and a scum is produced. The film is now " fixed " by
slowly drying it over a flame. When it is thus dried, a drop of the

selected stain (e.g. gentian-
violet) is placed over the film
and allowed to remain for a few
seconds. It is then washed off
with clean water, and the speci-
men dried, and mounted in
Canada balsam. The organisms
will now appear under the micro-
scope as violet in colour, and
fig. 4i.-Drying stage for Fixing Films. will thus be more clearly seen

than when unstained.
" Double staining " is adopted when it is necessary to stain the
organisms one colour and the tissue in which they are situated a contrast
colour. The chief methods will be mentioned subsequently.

4. Sub-culture of Colonies. — The plate method was introduced by Koch
in order to facilitate isolation of species. In a flask it is impossible to
isolate individual species, but when the growth is spread over a
comparatively large area, such as a plate, it is possible to obtain separate
detached colonies, and this being done, the colonies may be replanted,
by means of a platinum wire, in fresh media ; that is to say, a sub-culture
may be made, each organism cultivated on its favourable medium and its
manner of life closely watched. For example, a water may contain six
species of bacteria. On the plate these six species would reveal
themselves by their own peculiar growth. Each would then be isolated
and placed in a separate tube, on a favourable medium, and at a suitable
temperature. Thus each would be a pure culture ; i.e., one, and only one,
species would be present in each of the six tubes. By this simple means
an organism can be isolated and cultivated in the same sort of way as in
floriculture. From day to day the habits of each of these six species
may be observed, and probably at an early stage of their separate
existences it would be possible to determine to what species they
belonged. If not, further microscopic examination could be made, and,
if necessary, secondary or tertiary sub-cultures.

5. Inoculation of Animals. — It may be necessary to observe the action
of supposed pathogenic organisms upon animals. There is no means of
testing the pathogenic power of an organism, except by learning by
experiment, whether or not it produces disease. As a matter of fact, an
immense amount of bacteriological investigation can be carried on
without inoculating animals ; but, strictly speaking, as regards many
of the pathogenic bacteria, this test is the only reliable one. Nor
would any responsible bacteriologist be justified in certifying a water
as healthy for consumption by a large community if he were in

APPENDIX

457

doubt as to the disease-producing action of any of the contained
organisms.

By working through some such scheme as the above, it is possible to
detect what quantity and species of organisms, saprophytic or parasitic,
a water or similar fluid contains. For, observe what information has
been gained by following out these five steps in procedure. We have
learned the form (whether bacillus, micrococcus, or spirillum), size,
consistence, motility, method of grouping, and staining reactions of each
micro-organism ; the characters of its culture, colour, composition,
presence or absence of liquefaction or gas formation, its rate of growth,
odour, or reaction ; and, lastly, its effect upon living tissues. Here, then,

Fin. 42.— Types of Liquefaction of Gelatine.

are ample data for arriving at a satisfactory conclusion respecting the
qualitative estimation of the drop of water under examination.

As to the quantitative examination, that is fulfilled by counting the
number of colonies which appear, say by the third or fourth day, upon
the gelatine plates. Each colony has arisen, it is assumed, from one
individual, so that if the colonies be counted, though we do not thereby
know how many organisms there are upon the plate, it is known
approximately how many organisms there were when the plate was
first poured out, which are the figures we require, and which can at once
be multiplied up and returned as so many organisms per drop, or if the
quantity of water were measured, per c.c.

When counting colonies in a Petri's dish, it is sufficient to divide the
circle into eight equal divisions, and counting the colonies in the
average divisions, multiply up and reduce to the common denominator of
1 c.c. (or a Pakes' or Jeffer's Counting Disc may be used). For example,

458 APPENDIX

if the colonies of the plate appear to be distributed uniformly, we count
those in one of the divisions. They reach, we will suppose, the figure
of 60 ; 60 x 8 = 480 micro-organisms in the amount taken from the sus-
pected water and added to the melted gelatine from which the plate was
made. Let us suppose this amount was *25 c.c. Then the number of micro-
organisms in the suspected water is 60 x 8 = 480 x 4 = 1920 m.o. per c.c.

Double Staining1 Methods. — These are various, and are used when
it is desired to stain the bacteria one colour, and the matrix or ground
substance in which they are situated another colour. Two of the
common methods are those of Ziehl-Neelsen and Gram. They are as
follows : —

Gram's Method. — The primary stain in this method is a solution of
aniline gentian-violet (saturated alcoholic solution of gentian-violet 30
c.c, aniline water 100 c.c), or Nicolle's carbol-gentian-violet, which stains
both ground substance and bacteria in purple. The preparation is next
immersed in the following solution for thirty or forty seconds : —

Iodine ..... 1 part

Potassium iodide .... 2 parts

Distilled water . . . .300 parts

In this short space of time the iodine solution acts as a mordant by
chemical combination, fixing the purple colour in the bacteria, but not in
the ground substance. Hence, if the preparation be now (when it has
assumed a brown colour) washed in alcohol (methylated spirit), the
ground substance slowly loses its colour and becomes clear. But the
bacteria retain their colour, and thus stand out in a well-defined manner.
Cover-glass preparations decolorise more quickly than sections of
hardened tissue, and they should only be left in the methylated spirit
until no more colour comes away. The preparation may now be washed
in water, dried, and mounted for microscopic examination, or it may be
double-stained, that is, immersed in a contrast stain which will lightly
colour the ground substance. Eosin or Bismarck brown are commonly
used for this purpose. The former is applied for a minute or two, the
latter for five minutes, after which the specimen is passed through
methylated spirit (and preferably xylol also) and mounted. The result
is that the bacteria appear in a dark purple colour on a background of
faint pink or brown. Carbol thionine blue, picro-carmine, and other
stains, are occasionally used in place of the aniline gentian- violet, and
there are other slight modifications of the method. The application of
the method of Gram to coverslips and ordinary slide specimens for the
microscope may be shortly stated thus : —

1. Allow two or three drops of the gentian-violet stain to fall upon
slide and remain in contact with the film for Jive seconds. 2. Wash off
the stain wiih the iodine solution applied from a drop bottle for five or six
seconds. The film should then be black or dark brown. 3. Wash off
the iodine solution with a mixture of 1 part acetone and 2 parts alcohol
absolute, but allow to remain in contact for two or three seconds only.
4. Wash off with absolute alcohol, applied until no more stain comes

APPENDIX 459

away. 5. Wash in water, blot off superfluous water, and set aside to dry.
If thought desirable, the preparation may be counter-stained by the
application of a very weak solution of Ziehl-Neelsen.

The method of Gram enables us to classify bacteria into two great
groups. Certain organisms when coloured with a basic stain in aniline
or carbolic acid solution, and afterwards treated with a special mordant of
which iodine is the base, resist decolorisation by means of absolute
alcohol or other like reagent. Others, on the contrary, when treated in
the same fashion, readily give up their stain and decolorise when treated
with such reagents. The Bacillus anthracis may be taken as a type of the
former, the Bacillus typhosus of the latter.

Nicolle's Modification of Gram's Method, used in the staining of

diphtheria bacillus, consists in substituting carbolic acid for aniline water,
and in the use of a stronger iodine solution, and of acetone in the
decolorising fluid. Take 10 c.c. saturated alcoholic solution of gentian-
violet, and add 100 c.c. of a 1 per cent, solution of carbolic acid. The
iodine solution consists of iodine 1 gramme, potassium iodide 2 grammes,
and distilled water, 200 c.c. Place the film in the stain for five minutes,
then pass directly into iodine solution for five seconds, and decolorise by
passing rapidly through a mixture of one volume of acetone with four
volumes of absolute alcohol. This removes all unfixed stains at once. The
specimen is then dehydrated in xylol, allowed to dry, and mounted in
balsam.

Ziehl-Neelsen Method. — Here the primary stain is a solution of
carbol-fuchsin : —

Basic fuchsin . . . . . .1 gramme

Absolute alcohol . . . . . . 10 c.c.

Carbolic acid . . . . . . 5 grammes

Distilled water . . . . . . 100 c.c.

It is best to heat the dye in a sand bath, in order to distribute the
heat evenly. The various stages in the staining process are as
follows : — (a) The cover-glass with the dried film upon it is immersed
in the hot stain for one to three minutes. (A) Remove the cover-glass
from the carbol-fuchsin, and, after washing in water, place it in a capsule
containing a 33 per cent, solution of nitric acid to decolorise it. Here
its redness is changed into a slate-grey colour, (r) Wash in water, and
alternately in the acid and water, until it is of a faint pink colour, (r/) Now
place the cover-glass for a minute or two in a saturated aqueous solution
of methylene-blue, which will counter-stain the decolorised ground
substance blue, (e) Wash in water. (/') Dehydrate by rinsing in
methylated spirit, dry, and mount. A pure culture of bacteria will not
necessarily require the counter-stain (methylene-blue). Sections of tissue
may require twenty to thirty minutes in a primary stain (carbol-fuchsin).

This stain may be used for the bacillus of tubercle, with the
modification necessary to separate the bacillus of tubercle from other
organisms (leprosy, etc.) with similar staining properties. This modifica-
tion is to wash in absolute alcohol, after the carbol-fuchsin stain has been

460 APPENDIX

used, until all the colour has entirely disappeared. Then decolorise in
25 per cent, acid solution for a few seconds, wash in water and alcohol
and acid alternately, and counter-stain as usual. Honsell recommends
acid alcohol (absolute alcohol 97 per cent., HC1 3 per cent.) for ten
minutes before counter-staining. With a little practice, the staining of
the bacillus of tubercle when present in pus or sputum becomes a simple
and accurate method of diagnosis. A small particle of sputum or pus is
placed between two clean cover-glasses, and thus pressed between the
finger and thumb into a thin film. This is readily dried and stained as
above. But washing in alcohol and acid is not a reliable method of
differentiation between the tubercle bacillus and other acid-fast organisms.
Animal inoculation is the only reliable test.

Examination Of Moulds. — The examination of hyphomycetes or
mould fungi is, for differentiation purposes, best carried out on the Petri
dish itself, where the construction of the microscope will admit of this
being placed on the stage.

By the following method there is but little difficulty in recog-
nising the various species, and an excellent demonstration is given of
the hyphae with interstitial cells, and germinating conidia of the Oidium
lactis, the conidiophore and sclerotium of the Pencillium glaucum, or the
ramified mycelium, sporangia, and germinating zygospores of the various
species of Mucor, without disturbance of the growth. By means of a
finely drawn pipette allow to fall upon the centre of the mould colony a
small drop of aqueous solution (1 per cent.) of eosin. It is necessary to
exercise a little care in this, or the liquid will at once run off the colony
on to the surrounding medium. Place carefully upon the centre of the
drop a thin cover-glass, and press in order to obtain close contact.
Remove the Petri dish to the stage of the microscope, and examine the
margins of the growth with a sixth objective.

If the construction of the microscope will not allow examination on
the Petri dish, or if a permanent specimen is desired, the following
method can be recommended : — Detach by means of a pair of fine-
pointed forceps a portion of the young growth, holding it by the base,
and place it carefully on a slide. Place near it one drop of ammoniated
alcohol, and bring this in contact with the specimen by means of a finely
pointed needle. The absorption of the alcohol will allow the subsequent
penetration of the tissues by the liquids employed. Drop on to the
preparation a small quantity of Fleming's solution, and allow it to remain
for four or five minutes. Wash carefully with water, cover with a cover-
glass, and examine.

To make a permanent preparation, replace the water with glycerine
by placing a drop of the latter at one side of the cover-glass, and absorb
the water from the other by means of filter-paper. Dry carefully with
filter-paper damped with alcohol, and ring with paraffin.

Fleming's Solution

Chromic acid, 1 per cent. . . . .15 volumes

Osmic acid, 2 ,, . . . . 4 ,,

Glacial acetic acid ..... 1 volume

APPENDIX 461

Flagella Staining'

Successful staining of flagella is a matter of practice, and of careful
and exact technique. Whatever the method of staining adopted, the
preparation of the film is the same, and too much care cannot be
exercised at this stage. The slides should in no case have been
previously used, and they should be most carefully cleaned in the manner
described on p. 487. When taken out of the alcohol, the slide should be
carefully dried and wiped with a clean piece of old cambric, without
handling with the bare fingers. It should then be passed several times
through the flame, and set aside to cool.

Preparation of the Film. — 1. The cultures for examination should be
upon agar, and should not be less than six, or more than twelve hours
old, if incubation has taken place at 37° C. If incubated at 20°, slightly
older cultures may be employed (twelve to twenty hours).

2. Transfer from the young culture a small loopful by means of the
platinum needle, to a test-tube containing from 30 to 40 c.c. of sterile
water at room temperature. Or the emulsion may be made in a capsule
with a few c.c. of distilled water. Hold the loop in the water for a
few moments without shaking, until the water shows a slight turbidity.
Do not shake or handle the tube roughly. Incubate the emulsion for
five hours at 37° C. or for twelve to twenty-four hours at 20° C.

3. With a finely looped pipette, take up a small quantity of the
surface water from the inoculated tube, and distribute it in small droplets,
upon the slide.

4. Place aside to dry, carefully covered from chance of dust. When
dry, the staining can be proceeded with, according to the method
adopted. Do not fix the films in the flame ; the flagella are apt to be
injured thereby, and it will be found that the subsequent manipulations
will cause the organisms to adhere sufficiently to the slide.

Stabling the Film

The three ordinary methods practised in this country are : —

(1) PitJielcTs Method (Muir's modification).
The following solutions arc required : —

A. The Mordant.

Tannic acid, 10 per cent, aqueous solution . .10 c.c.

Corrosive sublimate, saturated aqueous solution . . 5 c.c.

Alum, saturated aqueous solution . . . .5 c.c.

Carbol-fuchsin (Ziehl) ..... 5 c.c.

The above must be thoroughly mixed and the precipitate which forms must be
allowed to deposit. The clear supernatant fluid is then drawn off with a pipette and
placed in a clean dropping bottle. The mordant will remain good for one or two
weeks, but not longer. It should be centrifugalised before use.

B. The Stain.

Alum, saturated aqueous solution . . . 25 c.c.

Gentian-violet, saturated alcoholic solution . . 5 c.c.

Filter twice. The stain must be freshly prepared.

The film is prepared as described above. The mordant is then dropped on to

462 APPENDIX

the slide and heated gently over the flame until the steam begins to rise. Allow to
steam for from one to two minutes ; wash well in running water and dry carefully.
When thoroughly dry, apply a sufficient quantity of the stain, and heat as before,
allowing to steam for two minutes. Wash in distilled water, dry, and examine.

(2) Van Ermengem's Method.

Three solutions are required in this method : —

A. Fixing Solution.

Osmic acid, 2 per cent, aqueous solution . . .10 c.c.

Tannin, 20 per cent, solution . . . .20 c.c.

To each 100 c.c. of this mixture add 4 to 5 drops of glacial acetic acid. The
colour of this solution should be violet, and the solution should be filtered before use.

B. Sensitising Solution.

Nitrate of silver . . . . 0 *5 aqueous solution

This solution should be kept in the dark, and filtered before use.

C. Reducing Solution.

Gallic acid ...... 5 grammes

Tannin . . . . . . 3 ,,

Fused acetate of soda (or potassium) . . 10 ,,

Distilled water . . . . . 350 ,,

(a) Cover the film with solution "A," and allow to act for five minutes at 37° C
or one hour at room temperature. Or heat gently until steam rises, and allow the
staining fluid to act for five minutes.

(6) Wash well with distilled water, then in absolute alcohol, and then again in
distilled water.

(c) Treat with solution " B," and allow it to act for thirty seconds, keeping the
fluid in movement on the slide.

(d) Allow the fluid to run off the slide, and without washing treat with "C" for
thirty seconds in the same manner.

{e) Allow fluid to run off, and again treat with " B " until the preparation begins
to turn black.

(/) Wash in distilled water, mount in water, and examine under the microscope.
The method is not wholly satisfactory.

A simple method is as follows : —

(3) Night-Blue Method (M'Crorie).

Place 2 or 3 drops of the emulsion on an absolutely clean slide, and dry at room
temperature. It is not necessary or desirable to fix by heat. The stain is made by
mixing 10 c.c. of night-blue, saturated alcoholic solution, 10 c.c. of a saturated
aqueous solution of potash alum, and 10 c.c. of a 10 per cent, aqueous solution of
tannin. The stain must be filtered before use. The slides, as prepared above, are
stained with this for two minutes in the hot incubator, and then washed gently in
running water. It may be found best to change the blue stain several times during
the two minutes. A counter-stain may be used if desired, and one of the best is
aniline gentian-violet. This should be applied for about a minute, after which a
cover-glass may be fixed over the film with Canada balsam. In such a preparation
the bacilli will be stained violet and the flagella blue. Better results may sometimes
be obtained by staining deeply with the blue (ten minutes), and then decolorising to
the necessary extent in dilute methylated spirit.

The Staining of Spores

The following are the methods commonly adopted : —

(1) Mbllers Method,

(a) Prepare the film as usual, fix and dry, observing the precautions taken in
preparing milk specimens.

APPENDIX 463

((>) Treat with alcohol for two minutes, and then with chloroform for two minutes ;
wash in water.

(c) Treat with chromic acid, 5 per cent, aqueous solution, for from one to two
minutes ; wash and dry.

(d) Pour on freshly filtered carbol-fuchsin and warm gently till it steams ; allow
it to act for ten minutes and wash off with water.

(e) Decolorise with sulphuric acid (5 per cent.) and water alternately, to remove
the carbol-fuchsin from the bacilli but not the spores.

(/) Dry and counter-stain with Loffler's blue until the film is of a faint bluish
tint. Wash off stain, dry and examine. The spores will be stained red and the
bacilli blue.

(2) Ziehl-Neelsen Method.

(a) Stain the film as for tubercle bacilli.

(/>) Decolorise with 1 per cent, aqueous solution of sulphuric acid, or alcohol 2
parts, acetic acid 1 per cent., 1 part.

(c) Counter-stain with Loffler's blue.

(d) Wash, dry, and examine.

(3) Abbott's Method.

Prepare films in usual way, and stain with Loffler's alkaline methylene-blue,
heating gently till steam rises (5 minutes). Then wash in water and decolorise with
nitric acid, 2 per cent, alcoholic (80 per cent.) solution, washing again in water.
Counter-stain with eosin, 1 per cent, aqueous solution. Wash, dry, and mount.
The spores are blue and the bacilli red.

Bacteriological Diagnosis. — The following points must be ascer-
tained in order to identify any particular micro-organism : —

(1) Its morphology : shape, size, etc. (bacillus, coccus, spirillum, etc.) ;
the presence or absence of involution forms ; motility, by the unstained
cover-glass preparation ("hanging drop ") ; note presence of Hagella ;
presence of spores, their appearance and position. Staining reaction ;
whether or not the organism stains by Gram's method.

(2) Cultural Characters. — The character of the growth upon various
media (gelatine, agar, milk, potato, blood serum, broth, and special
media) ; the presence or absence of liquefaction in the gelatine culture ;
its power of producing pigment, acid, gas, indol, ferments, phenol, etc.

(3) Biology : whether it is aerobic or anaerobic ; its powers of resist-
ance to external agencies; agglutination reaction, etc. ; pathogenesis, its
effect upon animal tissues and the course of the disease produced ; its
toxins, etc.

BACTERIOLOGICAL EXAMINATION OF WATEK

Collection of Samples. — Water from streams or wells should be
collected in glass bottles or flasks closed with glass stoppers previously
sterilised (at 150° C. for three hours), or washed out with pure sulphuric
acid. When the latter method is adopted, the bottle should be well rinsed
with the water which is to be examined before the sample is taken. In
taking the sample, the bottle should be held below the surface before the
stopper is removed, in order to obtain a sample of the main body of water
and not the surface water only. If it is an ordinary water supply
through pipes or from a cistern, the tap should be turned on and the
water allowed to run for a few minutes before taking the sample : and

464

APPENDIX

the same principle applies to a well not in regular use. Such a well
should be pumped for some time before taking the sample. For obtain-
ing samples from a considerable depth Miquel's apparatus may be used,
or, if that is not available, a weighted bottle.

After collection, the bottle should be at once stoppered, labelled, and
packed in ice and sawdust for transport to the laboratory, or placed in
one of the various ice cases now in use (Delepine's or Pakes'). Below
5° C, organisms do not multiply in water, and therefore it is important
to keep samples previous to examination at a low temperature. In all
cases where it is possible, the water should be examined at once after
collection.

Physical Examination. — The temperature and reaction of the water
should first be tested, and an examination made of any deposit or
suspended matter. Bubbles of gas, if present, should be noted. The
colour, character, and amount of particulate matter in suspension or
sediment should be observed and noted ; turbidity, odour, flavour and
taste, peatiness, etc., should all be noted. A record of the quantity of
the sample, its source, and the date and time of its collection is also
important. A microscopical examination of the matter obtained by
filtration followed by centrifugalisation may also yield important facts.

Bacteriological Examination. — This divides itself naturally into two
divisions — («) a quantitative examination, and (6) a qualitative examina-
tion.*

(a) Quantitative Examination

The sample should be gently mixed, and plate cultivations made.
Take five tubes of 10-15 c.c. of gelatine and five Petri dishes, and melt
the medium of the former in a water bath. The gelatine should be well
liquefied, but not overheated. The Petri dishes should be of even
surface, equal size, and properly sterilised. Take a 1 c.c. sterilised

Fig. 43.— Levelling Apparatus for Koch's Plate.

Fig. 44.— Moist Chamber for Koch's
Plate.

pipette accurately calibrated, and pass it into the bottle, removing the
necessary quantities of water. As a rule, 0*5 c.c, 0*2 c.c, 02 c.c,
0*1 c.c, and 0T c.c are suitable quantities for each of the five plates.
Add these quantities to the five tubes of liquefied gelatine, and gently

* An admirable illustration of how to examine a water is furnished in the Report
of Medical Officer to Local Government Board, 1901-2, pp. 494-547 (Houston).

APPENDIX 465

mix and pour into the Petri dishes. Allow the gelatine to set ; and
incubate at 22° C. for as long as possible before complete liquefaction
occurs. Count the colonies which appear after forty-eight hours incuba-
tion (agar), take the average at the period of maximum growth (gelatine
4-5 days), multiply up according to the fraction of a c.c. which has been
used, and return as so many organisms per cubic centimetre, stating
medium, period and temperature of incubation, etc. It is advisable that
each quantity of water from which the fractional part is added to the
gelatine should be taken up separately, and not that 1 c.c. of water should
be taken up and the fractional amounts, say of 05, 0*2, and 0*1 c.c, be
added to the gelatine. If Koch's plates are used they should be allowed

Fig. 45.— Wolfhiigel's Counter.

to set on the levelling apparatus, then placed in the moist chamber for
incubation at 22° C, and the colonies counted by means of Wolfhiigel's
counter (see Figs. 43, 44, and 45).

(6) Qualitative Examination

At the time of making the gelatine plates for quantitative examina-
tion, several agar, and litmus-lactose agar, plates may be made for quali-
tative purposes. The plates must be poured immediately after inoculation
of liquefied agar with small quantities of the water, as below 40° C. the
agar will resolidify. When poured, the agar plates should be placed on
cold stone or metal, and then incubated at blood-heat. On the second
or third day colonies will have appeared, and these should be studied
and sub-cultured (as pure cultures) on suitable media.

Valuable facts as to the quality of the water may also be obtained
from an examination of the five gelatine plates, particularly in respect
of the liquefying organisms, which should be counted as carefully as any
other colonies, and noted separately as well as in the total number of
colonies present. But in addition to the facts obtained from gelatine
and agar plates, other methods must be adopted in order to obtain
information respecting the quality of the water.

Take a sterilised Berkefeld porcelain filter, and pump or aspirate
through it 1000-2000 c.c. of the water under examination, and with a

2 G

466

APPENDIX

sterilised brush, transfer the particulate matter which has collected on
the candle into 10 c.c. of sterile water or broth. This is now a concen-
tration or emulsion of the organ ismal content of the
litre of water, and may be used for examination for
special organisms.

(a) B. enteritidis sporogenes.— Place 0-5 or

1 c.c. of the concentrated water in each of three
tubes of 10-15 c.c. of fresh sterilised milk. It is
important to use fresh milk, recently boiled, and
cooled down before inoculation. After inoculation
with the water to be examined, put the three tubes
into the water bath for fifteen minutes at 80° C, and
after allowing them to cool, place them in a Buchner's
tube or cylinder containing freshly-prepared pyro-
gallic solution (pyrogallic acid, 120 grains, strong
liquor potassae, 10 c.c). Accurately seal up the
Buchner, and place it, containing the tubes, in the
incubator at 37° C. The next day, or in forty-eight
hours, examine for B. enteritidis sporogenes. If that
organism is present, the following characteristic
appearances — the enteritidis change — will be ap-
parent (Klein). The cream of the
milk will be torn and altogether
dissociated by the development of
gas, so that the surface of the
medium becomes covered with
stringy white masses of coagulated
casein, enclosing a number of gas
bubbles. The main portion of the tube formerly occu-
pied by the milk will contain a colourless thin watery
whey, with a few lumps of casein adhering here and
there to the sides of the tube (see Plate 21, p. 307). If
the tube be opened, there will be found to be an odour of
butyric acid and an acid reaction. If some of the con-
tents of the tube are stained, as slide preparations, the
bacilli will be seen.

(b) B. COli Communis (p. 46). — Take from 0*1
to 0*5 of the concentrated or sample water, and
add to tubes of phenolated gelatine (-05 per cent,
phenol), or litmus-lactose agar, and make plates.
Colonies developing in these plates (red in latter medium) should be
suspected of being B. coli communis, and tested accordingly ;

Or, inoculate from the concentrated or sample water, three tubes of
Parietti's broth,* and incubate at 37° C, and those tubes which show

* PariettVs Formula consists of— phenol, 5 grams ; hydrochloric acid, 4 grams ;
distilled water, 100 c.c. To 10 c.c. of broth, 0*1-0*3 c.c. of this solution is added. The
tube is then incubated in order to test its sterility. If it be sterile, a few drops of the
suspected water are added, and the tube reincubated at 37° C. for twenty-four hours.
If the water contains the B. typhosus or B. coli, the tube will show a turbid growth.

Fig. 46.— Filter in posi-
tion for filter-brushing

method.

M

Fig. 47.— Buchner
Tube.

PLATE 31.

Apparatus fob Filtering Water ro paciutat] briolooical Examination.

(The lilter-brushing method).

f To fucr page 46G.

APPENDIX 467

growth in one to three days should be plated out on ordinary or pheno-
lated gelatine and colonies of B. coli examined for. Some authorities
recommend incubating Parietti's tubes at 42° C, a temperature favour-
able to B. coli but unfavourable to ordinary water organisms ;

Or, tubes of glucose-formate bouillon (meat infusion, 1 per cent,
peptone, 0*5 per cent, salt, 2 per cent, glucose, and 0'4 sodium formate)
may be inoculated with 0*1 to 0*5 c.c. of the water, and incubated in
Buchner's tube at 42° C, and the tubes which show turbidity in 24 hours
may be plated out on gelatine or glucose-litmus agar and B. coli, if
present, thus isolated (Pakes' method) ;

Or, inoculate tubes of M'Conkey's medium, bile-salt-glucose peptone,
with 1*5, or 10 c.c. of the water. "When this test yields negative results,
the absence of B. coli and of glucose fermenting coli-like microbes may
be accepted without reserve " (Houston) (see p. 484) ;

Or, place 1-20 c.c. of the water, or "01-'5 of the suspension, into tubes
of bile-salt solution, and incubate at 37° C. aerobically for twenty-four
hours, or at 42° C.anaerobically (in Buchner's tubes) for twenty-four hours;
and if, after this period, there is («) presence of growth, (6) formation
of acid, or (c) formation of gas, plate out on gelatine agar or bile-salt-
lactose-peptone agar, and sub-culture coli-like colonies on suitable media ;

Or, incubate at 38° C. 1 c.c. of the water in a Smith's fermentation
tube with glucose broth. If after twelve hours' growth gas collects, it
may be B. coli, and the species must be further tested. If there is no
gas there is probably no B coli.

In the examination for B. coli, the important media for sub-culturing
are as follows : gelatine shake cultures * (for gas production and lique-
faction), glucose gelatine or glucose agar (for gas production), milk or
litmus milk (for acidity and coagulation), peptone water (for indol), f and
potato. Eisner medium, J neutral-red agar, lactose and maltose media may
also be used.

"^Shake Cultures.'" — To 10 c.c. of melted gelatine, a small quantity of the
suspected organism is added. The test-tube is then shaken and incubated at 22° C.
In this medium the B. coli have opportunity for gas production.

\ The Indol Reaction. — Indol and skatol are amongst the final products of
digestion in the lower intestine. They are formed by the growth, or fermentation
set up by the growth, of certain organisms. Indol may be recognised on account of
the fact that with nitrous acid it produces a dull red colour. The method of testing
is as follows. The suspected organism is grown in pure culture in broth or peptone
water (or Dunham's solution), and incubated for forty-eight hours at 37° C. Two c.c.
of a *01 per cent, solution of potassium nitrite is added to the test-tube of broth
culture. Now 1 c.c. of concentrated sulphuric acid (unless quite pure, hydrochloric
should be used) is run down the side of the tube. A pale pink to dull red colour
appears almost at once, or in a few minutes, and may be accentuated by placing the
culture in the blood-heat incubator for half an hour. The presence of much dextrose
(derived from the meat of the broth) inhibits the reaction. B. typhosus does not
produce indol, and therefore does not react to the test ; B. coli and the bacillus of
Asiatic cholera do produce indol, and react accordingly.

t Eisner's Medium. — This special potassium-iodide-potato gelatine medium is used
for the examination of typhoid excreta. It is made as follows : 500 grams of potato
gratings are added to 1000 c.c. of water; stand in ice-chest for twelve hours, and
filter through muslin ; add 150 grams of gelatine ; sterilise and add enough deci-

468 APPENDIX '

Differential Diagnosis of B. coli. — The general characters of B. coli
will be found in the text of the present volume, but it may be stated
for diagnostic purposes that most reliance should be placed upon the
following characters. (But all characters must be taken into considera-
tion in forming an opinion.) B. coli produces a characteristic growth
on gelatine plate, smooth, milk-white colonies ; produces gas in
lactose, saccharose and glucose media ; is motile, non-liquefying (up
to the 14th day) and does not stain by Gram; produces acid
curdling of milk within four days at 37° C. The production of a
yellowish-green fluorescence in neutral-red agar shake culture and
the production of indol in peptone water or broth (but without
pellicle) are further tests relied upon by some. The Lawrence method
(State Board of Massachusetts) of testing for B. coli includes the follow-
ing seven tests : («) characteristic appearance on agar streak, (//) growth
on litmus-lactose agar, (c) gas production in dextrose broth, (d) coagula-
tion of milk, (e) production of nitrites in nitrate broth, (/') production
of indol in Dunham's solution, and (g) non-liquefaction of gelatine.*

B. lactis aerogenes is similar to B. coli, but coagulates milk much more
slowly and is non-motile.

B. Gaertner and its allies ferment glucose but not lactose in litmus
milk ; cultures are generally acid at first, and subsequently alkaline, and
there is no coagulation.

B. typhosus produces no gas in any media, does not coagulate milk,
stains by Gram, and serum diagnosis is also practicable.

Proteus group are similar to B. coli, except that they liquefy gelatine
and are slow in curdling milk.

(c) B. typhOSUS may be examined for by adopting exactly the same
methods as for B. coli. Its detection in, and isolation from, water
supplies is so difficult as to be well-nigh impossible. The condition of
a water is, however, ascertainable short of an absolute test for B. typhosus,
valuable though that would be.

(d) For the detection of the cholera Spirillum, add 10 c.c. of peptone
solution (10 per cent, peptone, 20 per cent, gelatine, and 5 per cent,
salt) to 90 c.c. of the water to be tested. Incubate at 37° C. After
twelve to twenty-four hours incubation, examine loopfuls from the
surface pellicle for spirilla ; or plate out loopfuls of the pellicle on
gelatine and agar ; or test for cholera red reaction and Pfeiffer's reaction
and agglutination test ; or culture emulsion from Berkefeld filter in
peptone water, and then plate out on gelatine and agar from tubes
showing pellicle.

(e) For the detection of Streptococci, plate out the emulsion obtained
from the Berkefeld filter on agar, and incubate at 37° C. After forty-
normal caustic soda until only faintly acid ; add white of egg ; sterilise and filter.
Before use add a gram of potassium iodide to every 100 c.c. Filter, and sterilise a
100° C. for twenty minutes on three successive days. Upon this acid medium
common water jbacteria will not grow, but B. typhosus and B. coli flourish — the
former like " small clear droplets," the latter as dark brown globular masses.

* Report of State Board of Health, Massachusetts, 1901, p. 400 ; ibid., 1902, pp. 262
and 280.

APPENDIX 469

eight hours examine the plate with a lens, and pick out the minute
colonies, streptococci, and sub-culture in broth, and incubate at 37° C.
Stain by Gram's method, and, if necessary, further sub-culture.

(f) Sewage Organisms and the organisms indicative of surface
pollution should also be examined for. If they be present in the water,
it may be taken as proved that such water has been recently polluted,
and should be condemned. Crude sewage generally contains in 1 c.c. :
(a) 1 to 10 million bacteria; (b) 100,000 B. coli (or closely allied forms) ;
(c) 100 spores of i). enteritidis sporo genes ; and (d) 1000 streptococci
(Houston). Further, so minute a quantity as ToVo °^ a c,c- °^ cru(le
sewage is usually sufficient to produce "gas" in a gelatine "shake"
culture in twenty-four hours at 20° C, and the inoculation of animals
with crude sewage always leads to a local reaction and not uncommonly
results in death. These three organisms, B. coli, B. entenlidis sporogenes,
and streptococci have been termed the " microbes of indication." These
bacteria are wholly, or relatively, absent from pure water, and their
presence, at all events in considerable numbers, must be taken as
indicating recent animal pollution* B. coli is a most accurate measure of
intestinal pollution, and far greater information as regards the sewage
pollution of water can be gathered by its estimation, than by simply
counting the total number of organisms present in water. It is an
intestinal parasite, and tends to perish in other media. f When it is
present in a small stream, contamination from houses can be traced.]:

Thresh has suggested the following scheme of examination of a
water as one furnishing the minimum amount of information which will
enable anyone to say positively that a water is faecally contaminated : —
(1) The detection of the presence of organisms of intestinal type ; (2)
the isolation and identification of B. coli ; and (3) the detection of the
presence of spores of B. enteritidis sporogenes. The process he recom-
mends is as follows : — (a) Make bile-salt broth cultures with 1, 5, 10,
and 20 c.c. of the water to be examined. After twenty-four hours the
tube containing the smallest quantity of water showing acid and gas
formation is selected for further examination. If after forty-eight hours
there is no such reaction, no further examination is made. (6) Two or
three loopfuls of the culture are added to 10 c.c. of sterilised water, and
a loopful of the solution is spread over a plate of bile-salt-lactose-
peptone agar containing neutral red and made faintly alkaline to litmus.
This plate is incubated for twenty-fours at 37° C, and the colonies pro-
duced carefully examined. As the B. coli communis ferments lactose
with the production of acid, any colonies of this organism will be of a
red colour and be surrounded by a haze, formed by the precipitation of
the bile acids. As this haze may not be apparent at the end of twenty-
four hours, types of all the red colonies are taken for the further
examination, (c) Each colony so selected is used to inoculate a tube of
lactose-peptone-bile-salt-litmus solution, and after twenty-four hours'
incubation, if acid and gas is produced, the growth is examined micro-

Second Report of Royal Commission on lSeica</e Disposal, 1902, pp. 2(5 and 27.
I Ibid., p. 99. X Ibid., p. 109.

470 APPENDIX

scopically to ascertain if the bacillus is motile or not (or spore-bearing),
and it is then treated by Gram's method to ascertain whether it retains
the stain. The results being recorded, (d) The turbid fluid is used to
inoculate — Litmus milk, for acid and clotting ; glucose-neutral-red agar,
for gas and fluoresence ; gelatin (stab or streak), for absence of liquefac-
tion ; and peptone solution, for indol. If the B. coli is present, all the
reactions indicated, with the possible exception of the production of
fluorescence, will be produced by one or more of the colonies selected.
The water is tested for the presence of B. enteritidis sporogcnes in the
ordinary way. It should be added that such an examination as the
above fails to obtain information upon the general constitution or the
bacteriological flora of a water which is obtained by the additional
means of the gelatine plate method, and whilst of value for rapid use,
should not be substituted for the systematic study of a water.

REPORT OF THE COMMITTEE APPOINTED BY THE ROYAL INSTI-
TUTE OF PUBLIC HEALTH TO CONSIDER THE STANDARDISA-
TION OF METHODS FOR THE BACTERIOSCOPIC EXAMINATION
OF WATER, 1904.

All the members of the committee are in agreement that the minimal number of
procedures should be :

(a) Enumeration of the bacteria present on a medium incubated at room

temperature (18°-22u C).
((>) Search for B. coli, and identification and enumeration of this organism if

present.
The committee regard these procedures as an irreducible minimum in the
bacterioscopic analysis of water. The majority of the committee recommend in
addition :

(c) Enumeration of the bacteria present on a medium incubated at blood-heat

(36°-38°C).

(d) Search for and enumeration of streptococci.

The committee do not think it necessary as a routine measure to search for the
B. enteritidis sporogenes, but are agreed that in special or exceptional instances it
may be advisable to look for this organism.

The Collection of the Sample. — No special precautions beyond those generally
recognised are suggested for taking the sample. The samples should be collected
in sterile stoppered glass bottles having a minimal capacity of 60 c.c. In special
instances it may be desirable to have much larger quantities.

Unless examined within three hours of collection the sample must be ice-packed.

(The committee recognise that under all circumstances the sooner the water is
examined after collection the more reliable are the results obtained.)

Media to be employed for Enumeration. — The choice of medium lies between
distilled-water gelatin, nutrient gelatin, distilled- water agar, gelatin agar, and
nutrient agar. The reaction of the medium is of importance.

For enumeration at room temperature, any of these media may be employed ;
but for enumeration at blood-heat, an agar or gelatin agar must be used.

The Americans seem to be using an agar medium only, and although on the
ground of simplicity it might be desirable to use a single medium for enumeration
under all circumstances — e.g. a distilled-water agar — it is felt by the committee
that gelatin media frequently give indications of value that are lacking with agar —
viz., liquefaction of the medium by many organisms and the more characteristic
appearance of the colonies in it ; gelatin is therefore recommended.

Since with a polluted water (detection of pollution being the ultimate aim in

APPENDIX 471

water examination) nutrient gelatin gives a relatively larger number of colonies than
distilled-water gelatin, nutrient gelatin should be used when one gelatin only is
employed. At the same time, it is recognised that cultures in distilled-water
gelatin compared with cultures in nutrient gelatin often give useful indications.
Thus with an unpolluted water the number of colonies is usually relatively larger
in distilled-water gelatin than in nutrient gelatin ; with a polluted water the con-
verse is the case. Therefore the use of both gelatins (distilled-water and nutrient)
is desirable, sets of plates being made with each medium.

Similarly, it was felt by many members of the committee that a comparison of
the ratio of the number of organisms developing at room temperature to those
developing at blood-heat gives useful indications. With a pure water this ratio is
generally considerably higher than 10 to 1, with a polluted water this ratio is
approached, and frequently becomes 10 to 2, 10 to 3, or even less. The actual
number of organisms growing at blood-heat is also of considerable value apart
from any question of ratio. Therefore it is suggested that plates of nutrient agar
should also be employed and incubated at blood-heat.

In certain instances it is true that this ratio may be unreliable. Thus with surface
waters, especially in tropical countries (as pointed out by Major Horrocks), varieties
of the B. fluoresceins liquefaciens and non liquefaciens and B. lujuefaciens may be
abundant and grow well at blood-heat.

Preparation and Reaction of Media for Enumeration

(a) Distilled- Water Gelatin. — Ten per cent, gelatin in distilled water, and brought

to a reaction of + 10 (Eyre's scale).
(6) Nutrient Gelatin. — Ten per cent, nutrient gelatin, preferably made with meat
(beef) infusion and Witte's peptone, and brought to a reaction of + 10
(Eyre's scale).
In hot weather it may be necessary to increase the percentage of gelatin.
Some members of the committee advocate the use of meat extracts in place of
meat infusion, on the score of convenience and uniformity of composition, Brand's
Essence being recommended as the best. It is the general opinion, however, that
Lieb'ufs Extract is less suitable for this purpose.

(c) For enumeration at blood-heat it is recommended that nutrient agar should be

employed, being prepared with the same constituents as nutrient gelatin,
but substituting H per cent, of powdered agar for the gelatin. Reaction
+ 10.

(d) Distilled- Water Agar. — Powdered agar U per cent., dissolved in distilled

water, and brought to a reaction of +10.

Owing to the changes which occur in the reaction of the medium on keeping, the
media employed should preferably be not more than three weeks old.

Amounts to be Plated, Size of Dishes, etc. — Gelatin. — For an ordinary water
amounts of 0*2, 0*3, and 0*5 c.c. may be plated in Petri dishes of not less than 10
centimetres diameter, preferably done in duplicate.

Agar. — Two plates may be made with 0"1 and 1*0 c.c, and are preferably
duplicated.

In dealing with an unknown water, and in all cases of doubt, additional sets of
plates should be prepared with a dilution of the water (made with sterilised tap-
water) often or hundred fold, according to circumstances.

The amount of the medium in a plate should be 10 c.c.

The sample must be thoroughly shaken and mixed in all cases before plating.

Temperature of Incubation.— (a) Room temperature = 18-22c C. ; (6) blood-heat =
36-38° C.

Counting. — Counting to be done with the naked eye, preferably in daylight, any
doubtful colony being determined with the aid of a lens or low-power objective.

Time of Counting.— Gelatin plates should be counted at the end of seventy-two
hours ; but in all cases the plates should be inspected daily, in order that the count
may be made earlier should liquefaction render this necessary.

The blood-heat agar plates should be counted at the end of forty to forty-eight
hours.

472 APPENDIX

Search for B. Coli

Method. — The committee recommend either —

(a) The glucose-formate broth method of Pakes.
(6) The bile-salt broth method of M'Conkey.

Incubation anaerobically at 42° C. increases the chances of success with either
medium, and is strongly recommended.

It has also been suggested that the neutral-red (Griibler's) glucose broth medium
may be employed.

The committee do not regard with favour the Parietti method, or the use of
carbolic acid media.

Quantity of Water to be Examined. — As a routine 50 c.c. should be the minimal
quantity examined for the presence of the B. coli, quantities from a minimum of
0*1 c.c. to a maximum of 25 c.c. being added to the tubes of culture media.

The committee are of opinion that it is preferable to add the water directly to the
tubes of culture medium, even with the larger amounts, rather than first to concen-
trate by filtration through a porcelain filter (the filter-brushing method). The
culture media recommended may be diluted with at least an equal volume of the
water without interfering with their cultural properties, and large tubes or small
flasks may be used for the larger amounts.

In the case of the bile-salt-lactose-peptone water, the medium may for the larger
amounts be prepared of double strength.

Isolation of B. coli, if Present. — If indications of the presence of the B. coli be
obtained in the preliminary cultivations, the organism must be isolated and identified.

This may be done by making surface cultures on plates of either (a) litmus-lactose
agar, reaction +10; (b) bile-salt agar ; (c) nutrose agar of Conradi and Drigalski ;
or (d ) ordinary nutrient gelatin.

The best medium of all is, probably, the nutrose agar of Conradi and Drigalski.
Agar media have the advantage of saving time.

Identification of and Tests for, the B. coli. — Having obtained coli-like colonies on
the plates made from the preliminary cultivations of the water, sub-cultures must be
made in order to identify the organism. The following, at least, should be made :

(a) Surface agar at 37° C. The abundant ' growth so obtained enables many sub-

cultures and preparations to be made if required.

(b) Stab and surface cultures in gelatine. This may be done in the same tube.

(c) Litmus milk incubated at 37° C.

(d) Glucose litmus medium.

(e) Lactose litmus medium.

(/) Peptone water for indol reaction.

Characters of the B. coli. — The B. coli is a small motile, non-sporing bacillus,
growing at 37° C. as well as at room temperature. The motility is well observed in
a young culture in a fluid glucose medium. It is decolorised by Gram's method of
staining. It never liquefies gelatin, and the gelatin cultures should be kept for at
least ten days in order to exclude a liquefying bacillus. It forms smooth, thin
surface growths and colonies on gelatin, not corrugated, growing well to the bottom
of the stab (facultative anaerobe).

It produces permanent acidity in milk, which is curdled within seven days at
37° C. It ferments glucose and lactose, with the production both of acid and of gas.

The typical B. coli must conform to the above description and tests.

It generally also forms indol (best obtained in peptone-water cultures), gives a
thick yellowish-brown growth on potato (greatly dependent on the character of the
potato), sometimes (about 50 per cent.) ferments saccharose, changes neutral-red
(Grii bier's), and reduces nitrates, and half the gas produced by it from glucose is
absorbable by KOH ; and these tests, if time and opportunity permit, may be per-
formed in addition to the foregoing.

The committee recognise that atypical B. coli are met with, but in the present
state of our knowledge hesitate to make any suggestion with regard to their
significance.

APPENDIX 473

Streptococci

The committee consider that it is a distinct advantage to search for streptococci.
They may be looked for by making hanging-drop preparations of the fluid media
employed for the preliminary cultivation of the B. coli (glucose-formate broth, etc.).
The presence or absence of streptococci in these tubes gives also a quantitative value
to the examination, just as in the case of B. coli, and the result obtained should be
stated. The streptococci should be isolated (best carried out on nutrose agar
plates), and their characters determined.

B. Bnteritidis Sporogenes

As already stated, the committee do not consider that it is essential as a routine
procedure to search for the B. enteritidis sporogenes, though in certain instances it
may be of advantage to do so. A negative result in such cases is probaby of more
value than a positive one.

This report is the outcome of prolonged deliberations, and every point has been
carefully considered and discussed by the members of the committee.

In conclusion, the committee suggest that if the above recommendations were to
be adopted by all engaged in the bacteriological examination of water it would
conduce to uniformity of results, and would render comparable the data obtained by
different observers. An addendum might be added to a report on an analysis
conducted on these lines, to the effect that the analysis had been carried out in con-
formity with the procedures recommended by the committee of The Royal Institute
of Public Health, 1904.

The committee beg to acknowledge their great indebtedness to Professor R.
Tanner Hewlett, M.D., D.P.H., upon whom the gnat burden of the work of the
committee has devolved.

RUPERT BOYCK, M.B., F.R.S.,
Chairman.

R. Tanner Hewlett, M.D., M.R.C.P.,
Hon. Secretary.

BACTERIOLOGICAL EXAMINATION OF MILK*

Physical examination (temperature, reaction, colour, cream, deposit,
specific gravity, etc.) of the milk should be made if necessary. The
microscopical examination of the milk before and after centrifugalisation
or sedimentation will likewise often yield useful results.

1. Plate Cultivation. -Dilute as required, and make plate cultivations in
Petri dishes or flat-bottomed Masks. Six or more gelatine plates should be
made and incubated at room temperature. Plates should also be made
with nutrient agar for incubation at :'>7 C. Other media may also be
used. The plates should be counted on the second, third, and fourth
days, and the necessary sub-cultures made. Agar plates incubated wholly
at 18° or 22 C, will in the long run show more colonies than when
incubated at 37° C. and then at 22 ('. or at 37° C. throughout.

2. Anaerobic Cultivation. — At the same time that the primary aerobic
plate cultivations are made, similar plates should be made on lactose
gelatine and lactose agar for anaerobic culture (see p. 117).

3. Primary Tube Cultivation. — Take ten tubes of 10 c.e. of the milk

For further particulars concerning the bacteriological technique in milk
examination, sec Bacteriology of Milk, by Swithinbank and Newman, 1903, pp.
30-115.

474 APPENDIX

under examination, and place three of them in the incubator at room
temperature and three of them at 37° C. Place four of them in a water
bath heated to 80° C. for fifteen minutes, and then enclose each of the
four tubes in a Buchner's tube. These primary cultures may be tested
in forty-eight hours for B. coli, the presence of indol, and B. enteritidis
sporogenes.

4. Secondary or Sub-cultures. — From the primary cultivations, make
sub-cultures on selected media for the isolation of organisms making
their appearance on the plates, or what is often preferable, make a set
of plates for qualitative examination only.

5. Examination for Special Micro-organisms. — The milk must be centri-
fugalised or the particulate matter allowed to gravitate by sedimentation.
It is, as a rule, useless to attempt examination microscopically or other-
wise without first using the centrifuge or sedimentation flask. The
deposit is then to be stained for the particular organism for which
search is being made (see p. 476).

For centrifugalisation, take two or three samples of the milk under
examination to the amount of about 40 c.c. each, and place it in the
sterilised tubes of the centrifuge. In these tubes the milk may be
centrifugalised for ten or fifteen minutes at 3000 revolutions a minute.
At the end of such a period the milk in each tube has separated into
three layers — at the top there is a dense layer of cream, at the bottom
there is the sediment or "slime" containing all the particulate matter,
between these two is the separated milk. Aspirate off the cream by
means of a sterile glass tube connected with an aspirator or vacuum
pump, and examine separately ; aspirate all the separated milk except
2 c.c. The remaining sediment is so compact and dense that the tube
may now be inclined and the sediment fully exposed without displace-
ment. By means of a sterilised platinum loop a small portion may be
taken up and spread on the surface of half a dozen slides, and stained.
The remainder of the sediment is well mixed with the 2 c.c. of milk and
used for inoculation of guinea-pigs.

For sedimentation, take two conical sedimentation glasses and fill
them with the milk under examination, allowing them to stand in the
refrigerator for twelve to fourteen hours. It is customary to add a few
small carbolic crystals to each flask. On the completion of sedimenta-
tion the milk has separated into three main strata : the cream at the top,
the sediment at the apex of the flask, and the separated milk in the
middle. The cream and milk may then be carefully decanted, and the
sediment will be available for examination.

Staining Methods in Milk Examination

The only difficulty which presents itself in the preparation of milk for
the microscope is the simultaneous staining of the casein and fat as well
as the organisms which may seriously confuse the issue. Hence the
removal of the two former substances is recommended, as follows :- —

(«) Staining after Clearing with 5 per cent. Acetic Acid. — The slides are

APPENDIX 475

thoroughly cleaned in the ordinary way, and immediately before use are
again washed with equal parts of alcohol and ether. Several loopfuls of
the milk to be examined are now placed on the slide and allowed to dry
at the temperature of the room, being protected from the air by means
of a small glass cover. When the film is dry it is fixed, preferably with
alcohol and ether, as described below. It is then washed alternately
with a 5 per cent, solution of acetic acid and distilled water until there
is but little apparent film left upon the slide, which is then dried
between layers of fine filter-paper. The specimen may now be stained
by means of any of the ordinary aniline dyes, washed in distilled water,
again dried, and examined under the microscope. Ether, chloroform,
various strengths of alcohol, and other clearing agents may be used if
preferred.

(6) Saponification. — If it be desired to retain the background of casein
and fat, it will be found best to saponify the milk in the following
manner : — Prepare the film of milk as before, but before drying it add
an equal number of loopfuls of a sodium carbonate or sodium hydrate
solution (5 per cent, to 50 per cent, dilution). The loopfuls of milk and
soda solution should be placed in immediate proximity to each other on
the slide, and thoroughly mixed by means of the platinum loop. By
this means an even distribution of the bacteria is obtained. The film is
then dried by gentle heating, stained, washed, and cleared with xylol.
The result will be that the organisms will be stained more deeply in
colour than the background of saponified matter.

(c) Clearing with Acetic Acid after Saponification. — The best ] repara-
tions are obtained by a combination of the above methods. For this
purpose the films are prepared exactly as in the ordinary saponification
method above described, but as soon as the films have become saponified,
instead of at once proceeding to stain with the desired dye, the film is
thoroughly cleared by several alternate washings with the 5 per cent,
solution of acetic acid and distilled water. The subsequent procedure
is as in (<-/).

Methods Of Fixation. — The object of fixing is to coagulate the
albuminous material, and cause perfect adhesion of the prepared film to
the slide. The following alternative nietlu ds are recommended : —

(a) Heat. — Holding the glass slide by one extremity between the
thumb and index finger of the right hand, pass it, film side upwards,
gently through the flame three times, allowing the under surface to rest
on the back of the left hand between each passage.

(()) Alcohol-ether. — Place one or two drops of a mixture of equal parts
of absolute alcohol and ether upon the dried film, and allow it to
evaporate.

(c) Formal-alcohol. — Formalin 1 part, absolute alcohol D parts. Leave
in contact for from three to four minutes, wash well in water, blot off
excess of moisture, and stain.

(jl) Perckloride of Mercury.- Saturated aqueous solution. Leave in
contact with the film for four or five minutes. Wash off with a stream
of water, and apply Gram's iodine solution in order to dissolve out any

476 APPENDIX

formed crystals of the salt. Wash again in water, blot off excess of
moisture, and apply stain. This fixing agent should be used on all
occasions when dealing with morbid material or cultures of a specially
virulent nature.

Methods of Examination for Special Micro-organisms in Milk

Bacillus pseudo-tuberculosis Of Pfeiffer (found in London milk
by Klein). — By the centrifuge or by sedimentation in an ice-chest for
twenty-four hours, obtain the particulate matter of the milk to be
examined. Inoculate 2 c.c. into a guinea-pig subcutaneously or intra-
peritoneally. In the course of three to four weeks caseo-purulent
nodules will occur in the inguinal glands (if subcutaneously inoculated),
or in the omentum and pancreas and other organs (if intra-peritoneally).
Cultures may be obtained best from glands, spleen, pancreas, or liver.
Examine the nodules by staining and culture. They will have the
following characters if the disease be pseudo-tuberculosis : (a) Absence
of giant cells ; (b) absence of the true tubercle bacillus ; (c) presence of
large numbers of B. pseudo-tuberculosis ; and (d) signs of a rapid and not
a slow development.

Method of Staining. — Make films in the ordinary way and stain with
Loffler's methylene-blue, heating the stain till it steams (Klein). Wash
in distilled water. Nodules may be hardened in Miiller's fluid and
spirit, and sections cut and stained by placing in Loffler's blue for
twenty -four hours and counter-staining in a mixture of eosin and
methylene-blue. Loffler's blue may also be used for staining the
bacillus in milk-films made from sediment. Gram's method is also
applicable, but the bacillus is not acid-fast, and will not hold the Ziehl-
Neelsen stain.

Bacillus diphtherise. — By centrifuge or sedimentation obtain the
particulate matter of the milk under examination and inoculate it into a
guinea-pig. Sub-culturing from the tissues of the guinea-pig, or, having
obtained sediment as above, inoculate six tubes or plates of Loffler's
medium (ox serum 3 parts, veal broth 1 part — the broth to contain
glucose 1 per cent., peptone 1 per cent., and sodium chloride 0*5 per
cent.). Upon this medium the Klebs-Loffler bacillus grows rapidly in
twelve to twenty hours, producing scattered nucleated, round, white
colonies which later become yellow.

Method of Staining. — Gram's method as modified by Nicolle (see
p. 459) will be found the most satisfactory, but the methylene-blue
solution of Loffler is often used. This consists of 30 c.c. of a saturated
alcoholic solution of methylene-blue added to 100 c.c. of a '01 per cent,
solution of caustic potash. By this stain the striped appearance of the
bacilli of older cultures on blood serum is obtained more readily than by
other methods.

Neissers Method for Differentiation of the Diphtheria Bacillus. — This
method consists in applying two stains as follows. Stain I. is made of
1 gramme of methylene-blue dissolved in 20 c.c. of a 95 per cent, alcohol,

APPENDIX 477

and 950 c.c. of distilled water. To this is added 50 c.c. of glacial acetic
acid. Stain II. consists of 2 grammes of vesuvin dissolved in 1000 c.c.
of boiling distilled water. Both stains are filtered before use. Prepare
films in usual way, and stain with No. I. for thirty seconds. Wash in
water and then stain with No. II. for thirty seconds. Wash, dry,
and mount. The bacilli are stained brown by vesuvin and the meta-
chromatin granules blue-black. Some bacteriologists place great
reliance upon the diagnostic value of Neisser's stain for the diphtheria
bacillus from blood serum cultures and from swabs. In the latter case
the stain is sometimes used as a " rapid method of diagnosis." It is not,
however, absolutely reliable.

StreptOCOCCUS in Milk. — By centrifuge or sedimentation obtain
the particulate matter of the milk. Take a sterilised platinum loop, dip
in the sediment, and remove a drop of it. Distribute this in a test-
tube containing 1 to 2 c.c. of sterile salt solution. Inoculate agar plates
with a drop of this dilution, and incubate at 37° C. When the colonies
appear, sub-culture those resembling streptococcus colonies in bouillon,
and on blood serum. Sub-culture from the bouillon in milk, gelatine,
and agar, carefully noting the characters of the growth, etc. Or
guinea-pigs may be inoculated in the subcutaneous tissue of the
groin or intra-peritoneally. An acute purulent inflammation will be
set up in the exudation of which streptococcus will occur in large
numbers.

Method of Staining. — Gram's method is the most satisfactory. Next to
Gram's stain the most useful is Loffler's blue. It may be noted that
most of the putrefactive organisms do not hold Gram's stain.

Bacillus COli Communis.— («) Dilute the milk to be examined 500
or 1000 times. Take a sterilised brush, dip it in the dilution, and brush
over the surface of six agar plates without recharging the brush.
Incubate at 42° C, and sub-culture the coliform colonies (bouillon, milk,
litmus milk, gelatine "shake" cultures, bile-salt-glucose-peptone, etc.).

(IS) Take six tubes of phenol bouillon (0'05 per cent, of carbolic acid),
and inoculate them with crude or diluted milk. Those which show
abundant turbidity after twenty-four to forty-eight hours at 37° C. may
be plated out on phenol gelatine, incubated at 20° C, and the coli
colonies sub-cultured ; or diluted milk may be at once plated out on
phenol gelatine, and colonies sub-cultured on such media as will show the
characteristics of the organisms.

o

The main characters of the B. coli group of organisms may be briefly
restated here, though particulars will be found elsewhere in the present
volume : — (1) They are non-sporing and non-liquefying ; (2) they rarely
stain by Gram's method ; (3) they are motile ; (4) they produce acid and
gas in glucose and lactose media ; (5) they produce acid in milk, and
usually coagulate it ; (6) they grow well at a temperature of 42° C.
Referring to the isolation of B. coli, Houston writes : " No test based on
observation of a change or changes produced in the nutrient medium,
and supposed to be characteristic of B. coli, can compare with isolation
from plate cultivations of the microbes suspected to be B. coli, and the

478

APPENDIX

subsequent attentive study of the biological characters of pure cultures
of these bacteria grown in various media."*

Bacillus enteritidis sporog-enes of Klein. —Take six tubes con-
taining 15 c.c. of fresh milk and sterilise them by boiling for half an
hour. Rapidly cool them by placing them in a beaker
of cold water, add to each tube 1 c.c. of a 1 in 500
dilution of the milk to be examined, or if it be pre-
ferred 0*1 c.c. of the crude milk. Heat the inocu-
lated tubes at 80° C. for fifteen minutes. Then remove
and cool, and place in Buchner tubes or cylinder con-
taining freshly prepared mixture of pyrogallic acid and
potassium hydrate solution. Seal the Buchner tubes
or cylinder with great care, making it absolute. Place
the Buchner apparatus, including the milk tubes, in the
incubator at 37° C. After forty-eight hours take out
the tubes, and examine them for the B. enteritidis
sporo genes. If necessary, inoculate guinea-pigs sub-
cutaneously with 1 c.c. of the whey, which in a few
hours causes swelling at the point of inoculation, and
extensive gangrene of the subcutaneous and muscular
tissues with sanguineous exudation ; the animal dies in
twenty-four or thirty hours. The B. butyricus of Botkin
may produce similar changes in milk tubes, but it has
no pathogenic action. Milk may be examined directly
by placing 20 c.c. in tubes and treating as above. For
the "enteritidis change" in the milk, see p. 307.

Bacillus tuberculosis. — Obtain the sediment of
the milk under examination and inoculate 2 c.c. of it
into the subcutaneous tissue of the guinea-pig. In
about four weeks' time, local if not general tubercu-
losis will have been set up. Take some of the dis-
charge and stain it after the Ziehl-Neelsen method.
The sediment of tuberculous milk may be stained
forthwith, without inoculation, by the same method,
and in some cases the tubercle bacillus may be thus
detected, but, generally speaking, the only sure test
is inoculation of animals. The pathological process is
slower than in pseudo-tuberculosis, and on exami-
nation the diseased tissues show giant cells and
numerous tubercle bacilli arranged within the giant cell
(see Plate 23, p. 328).
Method of Rabinoivitsch for Tubercle Bacillus in Butter. — The butter
is placed in sterile conical glass in the incubator at 37° C, where on
melting it will arrange itself in two layers. Three c.c. of the superna-
tant fatty liquid are injected into the peritoneal cavity of a guinea-
pig. A similar quantity of the deposit is treated in a like manner,

Fro. 48. — Another
form of Buchner Tube.

Second Report of Royal Commission on Sewage Disposal, 1902, p. 411.

APPENDIX 479

and finally, two or three other animals are inoculated intra-peritoneally
with the semi-liquid substance obtained on mixing together the two
layers into which the butter was formed. At the end of expiration of
seventy days the animals which have not already succumbed are
sacrificed, and a careful post-mortem examination is made. Microscopic
preparations and cultures are made from any organs affected. The
latter, taken together with the general aspect of the lesions, will, in the
majority of cases, be sufficient to enable a diagnosis to be made between
the true bacilli of tuberculosis, and other acid-fast organisms resembling
it. Bacilli which resists in a moderate or somewhat feeble manner
decolorisation by acids, which develop rapidly at a temperature of 37° C,
and grow feebly at ordinary room temperature, which exhibit chromo-
genic properties in culture, and give rise in the guinea-pig to lesions
which are not characteristically those of tuberculosis, must be regarded
as organisms of the acid-fast group, non-pathogenic for man, though
possibly related in some degree to the true bacillus of tuberculosis (see
also p. 358).

Another method is that indicated by Roth. Five grammes of butter
are vigorously shaken up in sterile water, and the whole is then centri-
fugalised. A fat-free deposit is thus obtained, and given quantities of
this are injected into animals in the ordinary manner.

Special Methods

Examination Of Colostrum. — Colostrum is the term applied to the
first milk yielded by the cow after parturition. It differs considerably
from ordinary milk, and generally appears as a thick, turbid, yellowish,
viscid fluid. When examined under the microscope, it is found to con-
tain, in addition to the ordinary milk corpuscles, peculiar conglomera-
tions of very minute fat granules which are hence known as colostrum
corpuscles. The chief chemical differences between colostrum (or
beastings) and milk are mainly three. First, colostrum is deficient in
casein. Secondly, it is proportionately rich in albumen. Thirdly, it con-
tains nearly three times more salts than milk. Probably it is this excess
of salts that usually causes it to exert a purgative effect upon the new-
born calf, and thus to remove the meconium which has accumulated in
the foetal intestine.

The difficulties of bacteriological examination of such a subject as
colostrum are considerable. At the outset, a fair sample is only obtain-
able by adopting the following precautions : (a) The teats and udder to
be cleansed ; (/>) milking to be carried out as soon after calving as pos-
sible, when the calf has sucked; (r) the first part of the "milking" to
be discarded, and the last part only to be examined. When the
colostrum reaches the laboratory, it must be diluted in precisely the
same manner as thick cream. After abundant dilution treat the solution
in the ordinary way, by staining preparations for the microscope, plating
out on various media, and sub-culturing.

Bacteriological Examination of Butter.— T ake a quarter of a

480 APPENDIX

pound of butter and place it in a sterilised flask with 150 c.e. of sterile
salt solution. Place the flask in the water bath at about 35° C, and
shake gently until the butter has melted. The contents of the flask now
appear as a milk-like emulsion. A small quantity of this mixture may
be used for plate cultivation on gelatine and agar, as in milk. The
remainder should be placed in a sedimentation flask in the refrigerator
for twenty-four hours. By this means the particulate matter of the
butter, including the contained organisms, are deposited. After remov-
ing the superficial solidified fat by means of a sterile spatula, the turbid
fluid may be decanted, and the sediment collected for microscopical
examination or the injection of guinea-pigs.

Examination Of Cheese. — With a knife previously sterilised by
p issing through the flame, cut off from the piece of cheese under
examination a thin slice parallel to the surface. Remove this, and
with a second sterile knife cut perpendicularly downward from the
bared surface. Pass down into the latter cut a coarse sterile platinum
needle, of which a small portion near the extremity has been slightly
roughened with a file.

Inoculate with this needle a sufficient number of tubes of bouillon
from which plate cultivations can subsequently be made for isolation
purposes, and placed under both aerobic and anaerobic conditions.

Examination of Milk for Pus Cells.— Place 10 cc. of the milk to be

examined in each tube of the centrifuge (Plate 5, p. 74) and centrifugalise
for two minutes. Pour off the supernatant fluid, and with a sterilised
needle or pipette take up a small quantity of the sediment remaining in
the tube. Spread the sediment evenly over the surface of an ordinary
glass slide, and dry over the flame of a Bunsen burner or on the drying
stage. Wash the fixed film with ether (or alternately with absolute
alcohol and ether) until all the superfluous fat is removed, and stain.
The preparation may be stained («) by one of the ordinary solutions
such as Lbffler's blue, etc. ; or (6) by Gram's method. Examine under
the microscope with a ^th oil immersion lens.

Inoculation of Guinea-pig's in Milk Examination.— It will

be sufficient to remark that the simplest forms of inoculation are all that
are usually required in milk investigation, namely, the intraperitoneal and
subcutaneous. In some cases it may be sufficient to inoculate a few cc. of
the original milk ; but, as a rule, it is advisable to centrifugalise, or use
the sedimentation flask containing about 250 cc. From the deposit or
sediment two guinea-pigs may be inoculated, the one subcutaneously in
the groin, the other intra-peritoneally. Particularly is this necessary in
making a reliable and exhaustive search for the B. tuberculosis. Micro-
scopic examination alone for this organism is not reliable (see p. 478).
The details of the process as carried out in practice are as follows : —

After centrifugalisation the deposit is mixed with the 2 cc. of milk
remaining in the tube after aspiration of that which is superfluous. Two
guinea-pigs (of say 250 grammes weight each) are taken and inoculated
with the deposit from about 40 cc of milk. The fluid is inoculated
subcutaneously on the inner side of the leg under strict aseptic precau-

APPENDIX 481

tions (the skin having been washed with 1-1000 corrosive sublimate, and
shaved). In less than a fortnight's time, if the inoculated milk contained
a considerable number of tubercle bacilli, typical infection of the
popliteal and inguinal glands can be detected. If the milk contained
very few bacilli the infection is much slower (fifth week). After the
animal has been killed the presence of the tubercle bacilli can be
detected in the inguinal glands and the spleen. Some workers make it a
rule to inoculate two guinea-pigs from the sediment of the milk, one
receiving half of the sediment subcutaneously in the groin, the other
receiving the other half intra-peritoneally.

BACTERIOLOGICAL DIAGNOSIS IN SPECIAL DISEASES

1. Diphtheria. — Obtain a piece of the membrane or a " swab " from
the throat. Take a piece of stout iron wire and twist a piece of cotton
wool round one end of it, and insert in a test-tube, and sterilise. By
means of such a swab obtain a rubbing of the suspected throat. Then
scraping off from the swab sufficient material for («) a microscopic
examination, (6) smear the swab over the surface of agar and blood
serum media, and finally (c) place in a tube of sterilised broth. Thus
we have material for a film preparation, for cultivation, and for animal
inoculation. Make the film in the usual way, and stain with Nicolle's
modification of Gram (see p. 458) or Neisser's stain (see p. 476).
Examine under the microscope. The value of examining such a prepara-
tion microscopically depends upon the experience of the bacteriologist.

Of culture media, blood serum is perhaps the best, but, if no serum
tubes can be had, an egg may be used. It should be boiled hard, the
shell chipped away from one end with a knife sterilised by heating, and
the inoculation made on the exposed white surface ; the egg is then
placed, inoculated end downwards, in a wine-glass of such a size that it
rests on the rim and does not touch the bottom. A few drops of water
may with advantage be put at the bottom of the glass to keep the egg
moist. The preparation is kept in a warm place for twenty-four to
forty-eight hours, and then examined. The examination, of course,
consists in staining and preparing specimens for the microscope, and
observing the form, arrangement, and characters of the organism or
organisms present. The same is clone for cultures on agar or blood serum.
On the latter the colonies show characteristic growth. A small piece of
the membrane may be detached, washed in water, and stained for
the bacilli.

To differentiate the true or Klebs-Loffler bacillus from the pseudo
or Hofmaim bacillus, note especially that Hofmann's bacillus is plumper,
shorter, and thicker in the middle than the true diphtheria bacillus. It
also stains more regularly, grows better on alkaline potato, and produces
an alkaline reaction in neutral litmus agar or bouillon incubated for two
days at 37u C. It is non-pathogenic for guinea-pigs, whereas the Klebs-
Loffler bacillus is pathogenic.

2. Tetanus. — The detection of the bacillus of tetanus in the dis-

2 H

482 APPENDIX '

charge of a tetanic wound is not always easy. Make preparations, and
stain with carbol-fuchsin. Drumstick-shaped, spore-bearing bacilli are
to be looked for. If a small piece of tissue is available, sections should
be prepared and double-stained. Cultivations should also be made from
the discharge in blood serum or glucose agar incubated at 37° C. for
forty-eight hours. Then keep the culture at 80° C. for twenty to thirty
minutes, to kill all non-sporing bacilli. Sub-culture in glucose gelatine
in hydrogen at 22° C, and examine in five days. Animal inoculation
(mice and guinea-pigs) is generally necessary.

To isolate the tetanus bacillus from soil, proceed as follows : — Make
an emulsion of the soil in sterilised water. Expose it to 80° C. for
twenty minutes. Add 1 c.c. of the emulsion to each of three tubes of
glucose-formate broth, and incubate anaerobically in Buchner's tubes at
37° C. After twenty-four hours' incubation, inoculate guinea-pigs sub-
cutaneously, using (M c.c, and observe results. Also make glucose-agar
plates from the same emulsion (after heating to 80° C), and incubate
anaerobically in Bulloch's apparatus.

3. Tuberculosis. — The tubercle bacillus is an acid-fast organism,
stained by Ziehl-Neelsen method. But several allied organisms possess
the same tinctorial properties, and therefore inoculation into a guinea-pig
is frequently necessary for diagnosis. Sputum, however, is generally
accepted as proved to be tubercular if bacilli having the morphology and
staining properties of the tubercle bacillus are present.

4. Typhoid. — Widal's Application of Grubers Reaction. This diagnostic
test depends upon the effect which the blood serum of a person suffering
from typhoid fever has upon the B. typhosus. The effect is twofold.
In the first place, the actively motile B. typhosus becomes immotile ; and
secondly, there is an agglutination, or grouping together in colonies, of
the B. typhosus. Neither of these features occur if healthy human blood
serum is brought into contact with a culture of the typhoid bacillus.

The method of using the test is as follows : —

(a) Collection of Serum. — Wash the lobe of the patient's ear with
antiseptic (2 per cent, lysol), and by rubbing render the ear hyperaemic.
Wash with methylated spirit and dry. Puncture the vein of the lobe
with a sterilised needle or lancet, and collect the issuing blood in a
pipette. Hold one end in contact with the bleeding point, and lower the
other end. By gravity the blood will enter the pipette ; if not, gentle
suction may be applied. When full to the shoulder, remove the pipette,
and placing the clean end to the lips, draw the blood gently but com-
pletely into the body of the pipette. Now seal the ends in a flame, and
let the pipette lie horizontally till the blood is coagulated.

(6) Dilution of Serum. — Place the pipette in the vertical position,
preferably in an ice-chamber, and in a few hours the clear serum, free
from corpuscles, will collect at the lower end, ready for dilution. If
necessary, centrifugalise to obtain corpuscle-free serum. There are
several methods of dilution used in practice, but broadly they are
divisible into two, a rough-and-ready dilution and an exact measured
dilution.

APPENDIX 483

The rough dilution is to take of the corpuscle-free serum to be
examined one drop. Dilute it with nine parts of neutral bouillon.
Mix on a slide or cover-glass a drop of this one-tenth dilutiou of
serum one or more drops of typhoid broth cultivation of eighteen to
twenty-four hours' growth. The serum and culture are thoroughly
mixed together in the trough of a hollow-ground slide, and a single drop
is taken, placed upon an ordinary clean slide, and a cover-glass super-
imposed ; or the mixture may be made on the cover-glass and super-
imposed on the slide.

The measurement method is to dilute the serum by exact quantities,
giving say, a 10 per cent., a 1 per cent., and a 01 per cent, dilution ; or
three mixtures containing respectively 50 per cent., 5 per cent., and
0*5 per cent, of serum. The 50 per cent, dilution is made by adding
equal loopfuls of serum and of a typhoid broth culture on a slide or
cover-slip. The 5 per cent, is made by diluting 10 cm. (measured by
graduated htematocytometer) of the serum, with 90 cm. of the broth
culture in a small sterilised test-tube. After thoroughly mixing, one
loopful of this dilution (now 10 per cent.) is mixed with one of cultivation.
The 0*5 per cent, is made by first diluting 10 cm. of the 10 per cent,
serum with 90 cm. of sterile broth in a small test-tube, and then mixing
equal loopfuls of this diluted serum and of the broth culture.

(c) The Typhoid Culture used should be one sub-cultured from a virulent
culture, and should be a broth or agar culture of about eighteen to twenty-
four hours; and, if preferred, may be filtered before use to remove any
normally agglutinated masses of bacilli before commencing the test.

(d) The Reaction. — The reaction is positive if the bacilli have become
grouped together tightly into clumps (agglutination), leaving the field
between the clumps free from bacilli. Immotility will also be present.
The reaction time is half-an-hour (see Plate 20). In his first experiments,
Widal used a test-tube in the following manner : — The blood to be tested
is diluted by one part of it being added to fifteen parts of broth in a
test-tube. The mixture is inoculated with a drop of a typical B. typhosus
culture. The tube is then incubated at 37" C. for twenty-four hours,
after which it is examined. If the reaction be positive, the broth appears
comparatively clear, but at the bottom of the test-tube a more or less
abundant sediment will be found. This is due to the clumps of bacilli
having fallen owing to gravity. If, on the other hand, the reaction is
negative, the broth will appear more or less uniformly turbid. This
method is not as satisfactory as the one described.

Some bacteriologists use two dilutions, 1 in 20 and 1 in 40, with a
time limit of one hour for each case. The reactions obtained are
interpreted as follows: — Where both dilutions show clumping and loss of
motility at the end of the hour a diagnosis of " enteric fever " is made ;
but if the reaction is present only in the 1 in 20 dilution, a guarded
opinion is given and the case stated to be " probably enteric fever " ; if
both preparations are unchanged, the case is reported as " probably not
enteric fever."

In the measured dilutions it may be said that if in half-an-hour there

484 APPENDIX *

is a positive result with the 50 per cent., 5 per cent., and 05 per cent.,
the case is undoubtedly one of typhoid fever, and if in half-an-hour there
is no reaction in all three, the result is definitely negative. Intervening
degrees of reaction must each be judged on its own merits, and a
subsequent examination made.

From the compilation of a large number of cases, the New York
Health Board concludes that Widal's reaction is present in typhoid
fever : —

From the fourth to seventh day in 70 per cent, of the cases.

From the eighth to fourteenth day in 80 per cent, of the cases.

During the third and fourth weeks in 90 per cent of the cases.

It is absent throughout in 5 to 10 per cent, of the cases.

Widal's reaction persists in the blood for months, or even years, but
after three or four months is usually feeble.

Differentiation of B. Typhosus

On p. 48 will be found some of the chief distinguishing tests for the
typhoid bacillus, which produces no gas in any media, does not coagu-
late milk, and stains by Gram's method. McConkey's test for B. coli
may also be used. The medium which he makes use of is bile-salt-
lactose agar, which is prepared as follows : To 1000 c.c. of tap- water in
a flask are added 2 per cent, of peptone, 0*5 per cent, of sodium tauro-
cholate, and 1*5 per cent, of agar. The flask is autoclaved at 105° to
110° C. for one and a half hours. The mixture is then cooled, mixed
with white of egg, and filtered ; then 1 per cent, of lactose is added.
The medium is distributed into test-tubes, 10 c.c. in each, which
are sterilised by steaming for 15-20 minutes on each of three successive
days. Plates are made and incubated at 42° C. for forty-eight hours.
There is a marked difference between the colonies of the organisms of
the typhoid group and those of the colon group. Of the typhoid group
the surface colonies are small, round, raised, and semi-transparent, the
deep one lens-shaped, white, and opaque, the medium remaining clear.
Of the colon group the surface colonies are roundish or irregular, with
flattened tops, opaque, white, with a yellow or orange spot in the centre ;
a few have a haze round them. The deep colonies all have a haze
round them, and are lens-shaped and orange-white. The haze is due
to precipitation of the sodium taurocholate by acid produced by fermen-
tation of the lactose. McConkey and Hill * have further modified this
method by the use of a bile-salt broth, composed as follows : Sodium
taurocholate, 0*5 per cent. ; glucose, 0*5 per cent. ; peptone, 2 per cent. ;
water, 100 c.c. The constituents are dissolved by heat, and the mixture
is filtered. After filtration, sufficient neutral litmus is added to give a
distinct colour, and the medium is then distributed into Durham's fer-
mentation tubes. These are ordinary test-tubes containing a piece of
light-glass tubing, about an inch in length, closed at the upper end.

* Thompson Yates Laboratories Report, 1901, vol. iv., part i., p. 151.

APPENDIX 485

This acts as a miniature gas-holder if fermentation of the medium
occurs. The tubes are finally steamed for twenty minutes for each of
three successive days. For the examination of water 1 c.c. is added to
each tube, and several are inoculated and incubated at 42° C. for forty-
eight hours. If the colon bacillus be present, the medium becomes
uniformly red, and is permeated with small gas bubbles, while the little
tube is filled with gas. Subsequently, plates may be made from the
tubes with the bile-salt agar medium.

Examination of Malarial Blood

1. Fresh Blood. — Thoroughly clean a cover-glass and wash a finger of
the patient. Then prick the finger and squeeze out a drop of blood.
This first drop of blood should be rejected. But when a second smaller
drop appears, just touch its surface with the clean cover-glass. Now
place the cover-glass on a clean slide, but do not exert any pressure
upon it. Under the weight of the cover-glass the blood will now spread
out into a very thin film. On examination under the microscope or by
the naked eye it will be seen that the blood corpuscles have, roughly
speaking, assumed the following zones : —

{a) A zone of scattered corpuscles immediately surrounding a central
portion empty of corpuscles and devoid of colour. This "scattered"
zone is composed of isolated, compressed, and much expanded
corpuscles.

(/;) Outside this first zone is one composed of corpuscles just touching
each other by the margin. This lias, therefore, been called the single
layer zone.

(c) The third zone lies still further outside, and is composed of
heaped-up corpuscles, overlapping each other and often in rouleaux.
Beyond them is the area of free haemoglobin, and valuable as enabling
the observer to see if there are pigment parasites present in the blood.

The ordinary pigmented amoeboid forms of the parasite will generally
be found in the single layer zone, whilst the flagellated bodies, if present,
will be seen chiefly in zone (e).

Pigmented leucocytes may appear anywhere in the field of the
microscope.

The intra-corpuscular parasites may generally be detected because
of their amoeboid movements, pigmentation, feeble definition, and effect
upon the corpuscle containing them.

2. Stained Preparation. — Whilst it is always best to examine malarial
blood in a fresh state if possible, it is generally desirable to make more
permanent preparations. This may be done as follows: — Make upon a
clean slide a very thin film of the malarial blood (by drawing a needle
or edge of cigarette paper over film). Allow it to dry in the air. Then
wash the slide containing the dried film with weak acetic acid (say two
or three drops of glacial acetic to an ounce of water) to clear the
haemoglobin. This may also be accomplished by dropping on the slide
a little alcohol, which may be dried up in several minutes' time with

486 APPENDIX

filter-paper. After either of these methods has been adopted, stain the
film for thirty seconds with a concentrated aqueous solution of methylene-
blue (or the following solution for the same period of time : Borax,
5 parts; methylene-blue, 2 parts; water, 100 parts). Wash in water,
dry with filter-paper, and mount in xylol-balsam under a cover-glass.*
Loffler's blue or carbol thionin may be used. For double staining,
Jenner's, Romanowsky's, or Leishman's stains may be used.

To demonstrate Jlagella, proceed as follows : — Take a piece of thick
blotting-paper, 3 x 1 J inches, with a round hole in the middle the size
of an ordinary cover-glass. Moisten the blotting-paper and place it on
a clean slide. Take a drop of the blood on another slide which has
been breathed upon, and invert it on the blotting-paper (moist cell).
In thirty minutes separate and dry the blood-film on both slides by
gentle warming over the lamp. Fix with absolute alcohol, which may
be allowed to evaporate or be dried with filter-paper. Wash with
acetic acid (15 per cent.) to dissolve out the haemoglobin, wash in water,
and dry as before. Stain the dried film with carbol-fuchsin (20 per cent.)
for six to eight hours. Wash and mount as before.

Bacteriological Examination of Oysters

Particular attention should be paid to the (a) washings of the shell,
(/>) the liquor in the pallial cavity, and (c) the contents of the alimentary
canal of the oyster. The two latter are the chief parts for examination
in the ordinary course, and to obtain knowledge of the contained
bacteria the method to adopt is as follows : —

Method. — Thoroughly cleanse the oyster shells by scrubbing with
soap and water, rinse under the tap, and again in sterile water. Also
the hands of the bacteriologist should be thoroughly cleansed and
rinsed in antiseptic (e.g. 1-1000 corrosive sublimate) and sterile water.
Now lay the oysters on the table with the flat shell uppermost, and
open with a sterile knife. Pour the pallial liquor into a sterilised flask
or capsule, and cut up the body of the oyster, adding the pieces to the
liquor or to another flask. Add to the flasks of liquor and of oyster
pieces, or to the one flask containing both sufficient sterile water (100
c.c. or 1000 c.c. as desired). The emulsion is now to be cultured as
follows, adding in each case suitable quantities of the emulsion, e.g. (10
c.c. or 5 c.c. or 1 c.c. or *5 c.c.) : — Three tubes of broth (for indol forma-
tion) ; three tubes of phenolated broth (for B. coli and its allies, and also
for secondary plate cultivation) ; three tubes of M'Conkey medium,
bile-salt-glucose peptone (for B. coli and its allies, coloration and gas) ;
three tubes of freshly sterilised milk, heated after inoculation to 80° C.
for 15 minutes, and cultured anaerobically (for B. enteritidis sporo genes) ;
three tubes of litmus milk (for acid and clotting) ; three gelatine
" shake " cultures (for gas production) ; three plates of phenolated
gelatine and three of ordinary gelatine ; and three plates of agar for
incubation at 37° C. For quantitative estimation of colonies on the

* See also Tropical Diseases (Manson), p. 40.

APPENDIX 487

plates, it will, of course, be necessary to multiply up according to degree
of dilution of the pallial liquor with sterile water in making the emulsion
in the first instance.*

Examination Of Urine. — Urine is examined in the same way as
water or sewage effluent. Plates (gelatine and agar) and sub-cultures
are made in the usual way. The urine should also be centrifugalised
and the sediment carefully examined by microscope and culture, and if
necessary inoculated into guinea-pigs. The organisms chiefly to be
looked for are B. typhosus (in cases of typhoid fever), B. tuberculosis,
septic organisms, and B. coli.

Examination Of Ice-cream. — Ice-cream usually contains vast
numbers of bacteria. It is examined in the same way as milk, and
requires high dilution before examination.

Examination Of Meat, Fish, etc. — Mince a portion of the unsound
meat or potted meat or fish by aid of sterile scissors and forceps, and
make an emulsion in broth in a flask at 42° C. (for thirty minutes).
Shake. Pipette off 10 c.c. of extract for inoculation of animals. Make
plates and further tube cultures of the emulsion. Incubate duplicates
anaerobically in Bulloch's apparatus. Feed animals on portions of the
samples.

Methods of Examination of Sewage and Sewage Effluents

The sample of sewage or effluent to be examined must be collected
in the same manner as in water.

1. Physical Examination. — Take note of quantity, colour, character
and amount of deposit and suspended matter, reaction, temperature,
bubbles of gas, etc.

2. Dilution. — This must be carried out as in the examination of milk,
500-1000 times.

3. Quantitative Examination. — Make plates on Petri dishes, gelatine
for incubation at 20° C, and agar at 37° C. Sewage is rich in intestinal
germs, most of which grow luxuriantly at blood-heat.

4. Qualitative Examination. — The three chief organisms of sewage
are: (a) B. coli (p. 4G), (b) B. enteriiidis sporogenes (pp. 156 and 307), and
(c-) savage streptococcus (p. 155). It is necessary, therefore, to examine
particularly for these organisms. It may also be necessary to estimate
quantitatively for B. coli and B. enteriiidis sporogenes.^

5. Subsidiary Differential Tests. — Inoculation of animals test ; produc-
tion of gas in gelatine "shake" cultures in twenty-four hours at 20° C. ;
acid clotting of litmus milk in twenty-four hours at 37° C. ; greenish-
yellow fluorescence in neutral-red broth cultures in twenty-four hours
at 37° C. ; the production of indol within five days at 37° C. ; and the
bile-salt broth test (growth, gas, and acid).

* A large number of methods and modes of experiment in the investigation of
Oysters will be found in the appendices of the Fourth Report of Royal Commission
on Sewage Disposal, 1904, vol. iii., pp. 191-309 (Houston).

t For methods, see Royal Commission on Sewage, Second Report, 1902, p. 140
(I louston).

488 APPENDIX

To Clean Glass Apparatus

Test-tubes and flasks may be washed in a bucket with hot water and
soap powder or soda, or boiled in the same. They should then be
cleaned with test-tube brushes and inverted for draining. Before use
they must be sterilised. Pipettes may be treated in the same way, and
then rinsed through with rectified spirit, and sterilised in the hot-air
oven. When test-tubes and pipettes are infected, they should be
treated in a similar manner, and also placed in strong disinfectant or
nitric acid (5 per cent.). Greasy slides should be placed in alcohol and
acid (5 per cent. HC1 or H2S04) for several hours, and then rinsed in
water. Greasy cover-slips may be treated in the same way, or boiled
in chromic acid (10 per cent.) and washed in acid alcohol and water.

Choice of Medium

This must be left very largely to individual experience and the
objects of the investigation. In a general way the constituents of the
various media described indicate the purposes to be obtained. The
general standard liquid media are bouillon and milk, the solid media are
gelatine (for room temperature cultivation) and agar (for blood-heat). In
tropical countries a combination of the two may be used. Further,
just as gelatine is a solid bouillon, so gelatinised milk may be used when
a solid milk medium is required. For anaerobes glucose and formate
media are commonly used. There are, of course, various media used for
different species of organisms. For the streptothrix group including B.
tuberculosis, glycerine media and potato are used. To isolate the B. typhosus,
carbolised media and Eisner are taken. Chromogenic bacteria nearly
always grow well on potato. The use of litmus milk, beer wort, icort
gelatine, milk agar, etc., is sufficiently designated in the names of the
media.

Preservation of Media. — Media may be kept in good condition for
months if a few simple precautions are borne in mind. The tubes or
flasks containing the medium must be effectually sealed, either with caps,
corks, or paraffin. The store of media must then be kept in a closed
metal box, and in a cool dark place.

INDEX

Abscess formation, 311
bacteria of, 312-313

Acetous fermentation, 102

Acid-fast bacteria, 358-369
classification of, 359
of human origin, 360
of butter and milk, 361
of grass and manure, 364
differential diagnosis, 365
streptothrix, 367

Actinomycosis, 321

Aerobic organisms, 23

Agar, 16

Air, bacteriology of, 76-91
dust and bacteria, 76
examination of, 73-75
moisture and bacteria, 79
of sewers, 82
currents and bacteria, 84
expired, 79-81
of workshops, 85
bacteria and gravity, 83
of bakehouses, 86
standard of bacteria in, 79, 91
of railway tubes, 88, 90
pathogenic bacteria in, 91
of House of Commons, 88
passages, bacteria in, 80

Alcohol, formation of, 96

Alcoholic fermentation, 96

Alexines, 412

Alformant lamp for disinfection, 443

Algae in water, 35

Ammoniacal fermentation, 110

Amylolytic ferments, 95

Anaerobic organisms, 23

methods of culture, 117-119
in hydrogen, 117
in glucose agar, 118
in Frankel's tube, 118
in Buchner's tube, 118

489

Andrewes on air of central railway tube,

90
Aniline dyes, 455, 458
Antagonism of organisms, 30
Antibiosis, 29
Anthrax, 315-319

clinical characters of, 315

pathology of, 316

spores of, 316

bacillus of, 316

in sewage, 177

channels of infection, 317
Antiseptics, 433

definition of, 433

some of the chief, 439-444
Antitoxins, 405

preparation of, 425

use of, 429

unit of, 428

effect of, 430
Appendix on technique, 453
Arthrospores, 12, 13
Artificial purification of water, 64-70
Ascospores, 14, 98
Asiatic cholera, 384
Association of organisms, 29
Attenuation of virulence, 31
Autoclave, 25

B At ii.i.us, definition of, 88
acetic 102

acidi lactici, 105, 106, 196
anthracis, 316
aquatilis, 45
bolulinus, 269
lutfj/ricus, 108-110
eapUlareus, 155
cloacce fluorescein ', 155
coli communis, 46-51, 56-60, 154, 466,

477
tests for, 48-51, 466

2 II 2

490

INDEX

Bacillus —

diphtheria, 288
enteritidis of Gaertner, 269
enteritidis sporogenes, 45, 154, 156,

307
fluoresceins liquefaciens, 45
fluoresceins nowAiquefaciens , 45
fluorescens stercoralis, 155
fusiformis, 155

friburgensis, Nos. 1 and 2, 363
of cholera, 385
of Binot, 364
of diarrhoea, 305
of dysentery, 403
of Grassberger, 364
of influenza, 321
lactis erythrogenes, 45, 200, 201
lactis pituitosi, 200
lactis viscosus, 200
liquefaciens, 45
mallei, 323

membranous patulus, 155
of leprosy, 398
phlei, 364

of glanders {mallei), 323
mes enter icus, 45
of Mceller, 362
mycoides, 45

of* malignant oedema, 144
No. 41, 245
of Rabinowitsch, 361
of symptomatic anthrax, 142
of plague, 392
prodigiosus, 200
pseudo- tuberculosis, 358
pyo-cyaneus, 157
pyogenes cloacinus, 155
radicicola, 134
saponacei, 200
smegmatis, 360
subtilis, 45
subtilissimus, 155
synxanthus, 201
of tetanus, 141, 481
of tubercle, 327, 337
typhosus, 48, 301
of yellow fever, 400

Bacteria, action of, 285
composition of, 9

Bacteria, in sewage, 151-158

and wheat supply, 131

and fixation of nitrogen, 131-139

in cheese-making, 241

in the dairy, 178-251

products of, 406

and disease, 280

the higher, 6, 8

in soil, 116
Bacterial action, 285

Bacterial action —

diseases of plants, 32

treatment of sewage, 162-177
Bacterio-purpurin, 10
Bacteroids, 136
Bakehouses, bacteria in, 86
Ballard on soil and disease, 145

on epidemic diarrhoea, 304, 308
Beer diseases, 110-113
Berkefeld filter, 71
Beri-beri, 404
Biogenesis, 2
Biology of bacteria, 1
Bitter fermentation, 112
Blood serum, 16
Blue milk, 201
Booker on bacteria of epidemic diarrhoea,

306
Boracic acid, 441
Boyce and others on bacteriological

examination of water, 470-473
Bread, bacteria in, 276-279

sour, 277

mouldy, 278

sticky, 278

red, 279
Broth, 16

Brownian movement, 11
Bubonic plague, 388-396
Buchner's tube, 118, 466, 478
Butter bacilli, the, 361-364
Butter, bacteria in, 241

making, 242-246

examination of, 479

bacterial flavouring of, 242
Butyric fermentation, 107

Carbol-fuchsin, 455, 459

Carbol-gelatine, 16

Carbolic acid as a germicide, 441

Carbonic acid gas and bacteria, 85-91

Caries, dental, 80

Carson's dairy farm, 232

Cellulose, 10

Chamber, moist, 464

Channels of infection in disease, 284

abnormal, 251
Cheese, bacteria in, 241

making, 246

examination of, 480

poisonous, 251
Chemical products of bacteria, 406

substances as disinfectants, 439-444

INDEX

491

Chemical products of bacteria —

and bacteriological examination of
water compared, 55

tests for nitrification, 128-132
Chemiotaxis, 11

Chloride of lime as a germicide, 440
Cholera, 384

bacillus of, 385

diagnosis of, 387

and filtration, 66

and milk, 223
Chromogenic bacteria, 406
Clams and bacterial infection, 266
Clark's process, 64
Classification, 5
Clowes on bacterial treatment of sewage,

164
Coccus, definition of, 6
Cockles and bacterial infection, 263-266
Collingridge on ice-cream poisoning, 274
Colon bacillus, see B. coli communis, 46
Comma bacillus, 385
Commensalism, 133
Commissions on food preservatives, 230

leprosy, 399

plague, 394, 425

sewage disposal, 49, 157, 161, 162,
172, 177, 261, 262, 469, 485

tuberculosis (1898) 270, (1895) 339,
343, (1901) 345

vaccination, 417
Composition of bacteria, 9
Conditions affecting bacteria in water,

60
Contact beds for sewage, 169
Contagion, 284

Contamination, organisms of, 56
Corrosive sublimate as disinfectant,

440
Counter (Wolfhiigel), 465
Cover-glass preparations, 455
Cream, bacteria in, 240
Crenothriv polyspora, 36
Cresol as a germicide, 441
Cultivation beds, 169
Culture media, 16
Cultures, anaerobic, 117

hanging drop, 455

plate, 453

pure, 456

shake, 467

sub-culture of, 456

Decomposition bacteria, 123

Delepine on bacteria in milk, 192-194

Denitrifying bacteria, 123

Dental caries, 80

Deodorants, 433

Desiccation, 18

Diagnosis, 463

Diarrhoea of infants, 304

and milk, 223

conditions favourable to, 304, 308

and soil, 308

bacteria of, 305
Diphtheria, 287-296

antitoxin of, 425-431

bacillus of, 288

bacillus in throat, 293

toxins of, 426

prevention of, 294

and milk supply, 211

and school influence, 292

pseudo-bacillus of, 295

diagnosis of, 290, 481

Diplococcus, definition of, 7
of gonorrhoea, 314
in pneumonia, 320

Directions for estimating disinfectants,

etc. , 434
Disease, production of, 280-287
Diseases of beer, 110-113

of plants, 32

conveyed by water, 53

and soil, 145
Disinfectants, 439
Disinfection, 432-451

means of, 435

by heat, 436

by chemicals, 439-444

of a room, 444

of walls, 445

of bedding, 445

of garments, 445

of excreta, 445

of wounds, 445

of hands, 446

of books, etc., 446

of stables, vans, etc., 446

after phthisis, 447

after small-pox, 449

after scarlet fever, 449

after diphtheria, 450

after typhoid, 450

after cholera, 450

after plague, 450

standards of, 434

Domestic purification of water, 70
Doriga on rats and plague infection, 390
Dunham's solution, 388

492

INDEX

Dysentery, 404

Earth temperatures and disease, 145,

308
Effluents, 158, 175
Eisner's medium, 467
Endospores, 12, 14
Enteric fever (see typhoid), 298-304
Enzymes, 94
Equifex disinfector, 438
Examination , bacteriological — technique
of, 453-463

air, 73

cholera, 387

diphtheria, 481

fish, 485

ice-cream, 485

leprosy, 398

malarial blood, 485

meat, 485

milk, 473-481

oysters, 484

sewage, 485

soil, 117

tetanus, 481

typhoid, 482

tubercle, 482

urine, 485

water, 463-473

yeasts, 97
Extracellular poisons, 408
External conditions, effect of, on bac-
teria, 15

Fermentation, 92-115

kinds of, 94

acetous, 102

alcoholic, 96-102, 198

ammoniacal, 110

butyric, 107, 197

lactic acid, 104, 196
Ferments, organised, 94

unorganised, 94

chromogenic, 200

curdling, 104, 197

bitter, 112-199

slimy, 199

soapy, 200
Films, 100
Filters, domestic, 71

sterilisation of, 72
Filtration of milk, 228-230

method of air-examination , 74
Filtration of water, 65-72
Filter-beds, 65

Firth and Horrocks on pathogenic
bacteria in soil, 148

Fission, 12

Fixing specimens, 475

Flagella, 11

staining, 461
Food, bacteria in, 178, 253
Formaldehyde and formalin, 443
Forms of bacteria, 6
Foulerton on pollution of water, 63

on bacteria in oysters, 260

on streptothrix group, 367
Fowler on bacterial treatment of sewage,

174
Fractional sterilisation, 24
Frankel's pneumococcus, 320
Frankland on bacteria in water, 37

on filtration of water, 65
Freezing, effect of, on bacteria, 18
Friedlander's pneumo-bacillus, 320

Gas, production of, 406
Gathering-ground, 34
Gelatine, 16

carbol, 466

liquefaction of, 457
Gemmation, 98
Gentian-violet, aniline, 455
Germicidal temperatures, 23-25
Germicides, 439-444
Gilbert on nitrification, 134
Ginger-beer plant, 137
Glanders, 323
Gonorrhoea, 314
Gram's method of staining, 458

Nicolle's modification of, 459
Gravity, influence on bacteria, 83
Gypsum block, 98

FLemocytozoa, 372

Hsemamoeba, 372

Haldane on ventilation of workshops,

85
Hanging-drop cultivations, 455
Hansen's method of dilution, 99
Heat as steriliser, 23-25
Heredity, 284

Hesse's method of air examination, 74
Hewlett and others on bacteriological

examination of water, 470-473
High yeasts, 101
Higher bacteria, 6, 8
Horrocks and Firth on soil and disease,

148

INDEX

493

Horrocks on classification of water
bacteria, 44

Hot-air steriliser, 24, 25

Houston on streptococci in water, 51
on pathogenic bacteria in soil, 148
on sewage bacteria, 153, 157,175, 177
on bacteria in oysters, 261

Hydrogen cultivation, 23

Hydrophobia, treatment of, 419

Ice, bacteria in, 274
Ice-cream, bacteria in, 272-274

manufacture of, 272

examination of, 485
Immunity, 405-431

acquired, 413

active, 412

Ehrlich's side chain theory, 414

artificial, 412

general principles, 405-412

natural, 413

passive, 413

theories of, 413

in small-pox, 415

in rabies, 419

in typhoid, 424

in plague, 424

in diphtheria, 425

in cholera, 423
Incubators, 17
Indol, formation of, 467

testing for, 467
Industries and bacteria, 113-115
Infection, channels of, 284
Influenza, 321
Inoculation, 456

Interpretations of bacteriology, 55
Intracellular poisons, 31
Inversive ferments, 95
Involution forms, 9

Jok dan's classification of water bacteria,
44

Kei>iiir, 136, 198

Kipp's apparatus for producing hydrogen,
23

Klebs-Lbffler bacillus, 288

Klein on bacteria in oysters, 259
on bacteria in cockles, 263
on bacillus entertiidu 9porogmtt% 156,
307

Koch's plate method, 453
steam steriliser, 24, 25
tubercle bacillus, '527 et seq.
postulates, 28]

Koch on filtration of water, 66

on inter-communicability of tuber-
culosis, 339 et seq.
comma bacillus, 385
bacillus of tubercle, 327, 337
views on tuberculosis, 338-346

Koumiss, 198

Lactic acid fermentation, 104, 196
Leguminosae, fixation of nitrogen by, 131
Leprosy, 396-400

history of, 396

forms of, 397

bacillus of, 398
Light, influence upon bacteria, 18-22
Lingner's apparatus for disinfection, 443
Liquefaction of gelatine, 457
Liquid hydrogen and bacteria, 18
Lloyd on Cheddar cheese-making, 249
Low yeasts, 101
Lymph, glycerinated calf, 416
Lyon's, Washington, disinfector, 4:i7

Macehatiox industries, 113
Malaria, 371-384

kinds of, 373, 375

cycle of Golgi, 380

parasites in, 372

microgametocytes of, 375

macrogametocytes of, 376

and mosquitoes, 376

anopheles of, 377

culex and, 378

examination of blood, 485

preventive measures, 382-384

mosquito breeding, 382

destruction of mosquitoes, 383

bites of mosquitoes, 383

quinine, 384
Malignant oedema, 144

bacillus of, 144
Mallein, 324
Malta fever, 402

Manchester sewage treatment, 170-174
Manson on malaria, 375 et seq.

on plague, 390
Martin on soil and disease, 147

on tuberculosis, 341-342
Mastitis, 184

M'Conkey's bile-salt method, 467, 484
Meat and bacterial infection, 267-272

examination of, 485

poisoning bacteria, 269

tuberculous, 270

decomposed, 270
Media, culture, 16

494

INDEX

Merismopedia, 8
Metabiosis, 29
Metachromatic granules, 10
Metchnikoff on phagocytosis, 414
Methods of examination, 453

Micrococcus, definition of, 6

aquatilis, Freudenreichii, 199
gonorrhoea?, 314
tetragonus, 313
viscosus, 199

Milk, bacteriology of, 178-251
composition of, 195
incubation period for bacteria in, 179
sources of pollution, 181-184
number of bacteria in, 184-194
influence of time and temperature

upon bacteria in, 185-194
fermentation bacteria in, 196-202
kinds of bacteria in, 194
disease-producing power of, 202
lactic acid fermentation of, 1P6
butyric fermentation of, 197
coagulation fermentation of, 197
alcoholic fermentation of, 198
anomalous fermentation of, 199
and tuberculosis, 203-207
and typhoid, 207-211
and cholera, 223
and epidemic diarrhoea, 223-226
and diphtheria, 211-214
and scarlet fever, 214-217
and sore-throat illnesses, 219-223
and thrush, 218

character of milk-borne disease, 218
prevention of milk-borne disease,

226
method of protection, 227
control of milk supply, 227-240
filtration of, 230
refrigeration of, 228
straining of, 228
sterilisation of, 231
of Liverpool, 205
pasteurisation of, 231
results of, 235
summary of control, 236
products, bacteria in, 240-251
examination of, 473-481
specialised milks, 237
and economic bacteria, 240-249
and municipal depots, 237
chromogenic fermentation of, 200

Miquel's method of air examination, 74

Modes of bacterial action, 25

Mohler on tuberculosis of the udder, 203

Moist chamber, 464

Moisture necessary for bacteria, 18

Morphology of bacteria, 6
Motility, 11

Mosquitoes and malaria, 376
Mycoderma aceti, 103
Mycoprotein, 9

Nasal passages, bacteria in, 80

Natural purification of water, 60-64

Needles, platinum, 17

New soil science, 138

Newsholme on conditions favourable to

diphtheria, 291

on causation of epidemic diarrhoea,
309

on disinfection after phthisis, 447
Nitric organism, 128
Nitrification, 125-131

chemistry of, 125

stages in, 129

bacteria of, 129
Nitrifying organisms, cultivation of, 127,

128
Nitrogen, fixation of, 131-139
Nitrogen-fixing bacteria, 131
Nitrous organism, 126
Niven on disinfection after phthisis, 447
Nodules on roots, bacteria in, 133

Oceax bacteria, 36

Oidium albicans, 218

Oxygen necessary for bacteria, 23

Oysters and typhoid fever, 253-263

poisoning, symptoms of, 256

infection, 257

and disease, prevention of, 262

examination of, 484

Pares' formate broth method, 467
Paraform for disinfection, 443
Parasitism, 25
Parietti's method, 466
Pasteur on fermentation, 93
Pasteur's treatment of rabies, 419
Pasteur filter, 71
Pasteurisation of milk, 231-236
Pathogenic bacteria, in soil, 140

in water, 53
Perlsucht, 333
Petri dishes, 453
Phagocytosis, 413-414
Phosphorescence, 22
Pigment, formation of, 406
Place of bacteria in nature, 4

INDEX

495

Plague, 388-396

varieties of, 388

symptoms of, 388

and rats, 390

distribution of, 389

bacillus of, 392

administrative control of, 394

vaccination for, 424

diagnosis of, 396
Plant diseases, 32
Plasmolysis, 9
Plate cultures, 453
Platinum needles, 17
Pleomorphism, 9
Pneumonia, 319

bacteria of, 320
Pneumo-bacillus, 320
Pneumococcus, 320
Polymorphism, 9
Postulates, Koch's, 281
Potato medium, 16
Pouchet's aeroscope, 74
Power on milk-borne scarlet fever, 214
Products of bacteria, 406
Proteolytic ferments, 95
Proteus family, 154, 45

cloacinus, 154

vulgaris, 154

Zenker i, 154

mirabilis, 154
Pseudo-diphtheria bacillus, 295
Pseudo-tuberculosis, 355-358
Purification of water, 60-72

natural, 60-64

domestic, 70

artificial, 64-70
Pus, 311
Pyocyanin, 313
Pyoxanthose, 313

Quantitative standard of water bacteria,

42
air bacteria, 77
milk bacteria, 191
soil bacteria, 116
Quarter-evil, 142

clinical characters, 143

Rabies, treatment of, 419

forms of, 420

pathology of, 420

results of treatment, 423
Red milk, 200

Reproduction of bacteria, modes of, 11
Retiin r, 113

Rivers, natural purification of, 38, 60-64

Robertson on soil and typhoid, 147

Ross on malaria, 380

Rotch's specialised milk, 239

Russel on butter-making, 244
on cheese-making, 247

Saccharomycetes, biology of, 97

methods of examination, 99

anomalous, 99

apiculalus, 102

aquifolii, 102

cerevisiw, 101

conylorneratus, 102

ellipsoideus 7., 101, 102

ellipsoideus //., 102

exiguus, 102

HansenU, 102

illicis, 102

Ludwigii, 99

tngcoderma, 102

pastorianus 1. , 102

pastorianus II., 102

pastorianus, 111., 102

piriformis, 102
Sand filtration of water, 65
Saprophytes, 25
Sarcina, 8

Savage on bacteria of made soil, 149
Scarlet fever, 296

milk and, -!1 i

bacteria of, 296

streptococcus in, 297
Sedgwick's method of air analysis, 75
Seed and soil, 26
Sedimentation, 62, 64
Septic processes, 311

tank, 165-169, 174
Sewage, organisms in, 1 51-158

bacterial treatment of, 1(52-177

constitution of, 151

examination of, 15 1. I v.">

organic matter in, 152

inorganic matter in, 152

number of bacteria in, 153

kinds of bacteria in, 1"' 1

spores in, 153

streptococcus of, 155

pathogenic bacteria in, 155

nitrification and denitrifieation in,
157, 158, 166

aerobic and anaerobic bacteria in,
157, 15S, 166

disposal of, 159

chemical treatment of, 160

biological treatment of, 158, 160-177

496

INDEX

Sewage —

irrigation of, 161

intermittent filtration, 160

effluents and pathogenic organisms,
175

London treatment of, 164

Manchester treatment of, 170

Sutton treatment of, 169

Exeter treatment of, 165

septic tank method of treating, 165-
169

contact bacteria beds method, 169

Leicester treatment of, 173

effect of bacterial treatment of,
175
Sewer air, 82

and toxicity of bacteria, 8.°>
Shake cultures, 467
Shell-fish and bacteria, 253-266
Sleeping sickness, 403
Small-pox, 416
Smith, Graham, on air of House of

Commons, 88
Smith, Horton, on typhoid urine, 300
Soil, bacteriology of, 116-150

bacteria in, 116

composition of, 119-122

denitrification in, 123

examination of, 1 1 7-1 1 9

and typhoid fever, 146-148

and tetanus, 140

polluted, 149

kinds of bacteria in, 119

nitrification in, 125-131

nitrogen-fixing bacteria in, 131-139

and its relation to disease, 1 15

pathogenic bacteria in, 140

classification of bacteria from, 119,
120

symbiosis in, 132, 136
Sorensen's dairy farm at York, 228
Species of bacteria, 28
Specificity of bacteria, 28
Spirillum, definition of, 88

of cholera, 385

of Obermeier, 8

Spontaneous generation, 3
Sponges, 114
Spores, kinds of, 12

resistance of, 12-15

staining of, 462

of yeasts, 98-99
Staining methods, 455-463
Standard of sterilisation, 24
Staphylococcus, 8, 311

cereus alius, 311

Staphylococcus —

pyogenes aureus, 312

pyogenes alius and c'ttreus, 311
Steatolytic ferments, 95
Steam, as a disinfector, 486-488

disinfectors, 437

steriliser, 24

saturated, 436

superheated, 436

current, 437
Sterilisation, 23-25

methods of, 28-25
Streptococcus, 7

in milk, 297

in water, 51

in sewage, 155

of scarlet fever, 297

pyogenes, 312

eonglomeratus, 297

Hollandicus, 199

Streptothrix group, 367

<ictinomyc.es, 821

liominis, 368

eppinger, etc., 369

luteofa, 367
Structure of bacteria, 6
Sub-cultures, 456

Sulphurous acid as a germicide, 441
Suppuration, 311
Swine fever, 272

Swithinbank and author on milk, 188-190
Symbiosis, 29, 132, 136
Symptomatic anthrax, 142

Table of economic bacteria in soil, 120
Temperature, influence of, on bacteria,

17
Tetanus, 140-142

toxin of, 141

bacillus of, 141, 481
Thermophilic bacteria, 17
Thresh's disinfector, 438
Thresh on bacteriological examination of

water, 469
Thrush, 218
Tobacco-curing, 114
Toxins, 406-412
Tropical diseases, 370-404
Tuberculin, 347
Tuberculosis, 325-358

pathology of, 325

varieties of, 326

history of, 326

conveyed by the air, 79, 331

and the milk supply, 335

INDEX

497

Tuberculosis —

of the udder, 334

giant cells in, 330

bacillus of, 327, 337

bovine, 333, 337

diagnosis of bovine, 346

cultivation of bacillus of, 328

spores of, 329

relation of bacillus to disease, 330

temperature for growth, 329

toxins, 333

of horse and dog, 350

of cold-blooded animals, 351

of animals, 349-351

of pig, 349

of birds, 350

of sheep, 349

prevention of, 352-357

disinfection in cases of, 447

decline of, 352

and overcrowding, 355

channels of infection in, 33]

inter-communicability of, 338-346

and house influence, 356

pseudo-, 355-358

Typhoid fever, 298-304
bacillus of, 301
effect of light on bacillus, 20
pathology, 298

bacillus compared with B. coli, 48
bacillus in sewage, 175
bacillus in drinking-water, 303
tests for bacillus of, 48, 468, 482, 484
and soil, 145

conveyed by the air, 78, 81
and milk supply, 207-211

Tyrotoxicon, 251

Udder tuberculosis, 203, 334
Unit of antitoxin, 428
Urea, 120

Vaccination, 415-419

effect of, 417
Vaccines, 415-425

plague, 424

Vaccines —

cholera, 423

small-pox, 415
Vaccinia, 415
Vacuolation, 9
Variolation, 415
Virulence, attenuation of, 31

Warrington on nitrification, 123 et seq.

Washington Lyon disinfector, 437

Water, bacteria, classification of, 44
bacteria in, 33, 44
collection of samples, 33
number of bacteria in, 34, 36, 42

56
examination of, 463-473
organisms of contamination, 56
pathogenic organisms in, 53, 56
multiplication of bacteria in, 34
natural purification of, 60-64
river, bacteria in, 36-42
artificial purification of, 64-70
B. coli in, 46-60
filtration of, 65
ordinary bacteria in, 45
domestic purification of, 70
London, bacteria in, 38-42
quantitative standard in, 42
quality of, 44-60
sewage bacteria in, 15
pollution of, 54-60
sea, bacteria in, 36

Watercress, bacteria of, 279

Wheat supply and bacteria, 131

Widal reaction, 482

Wolfhiigel's counter, 465

Wool-sorters' disease, 318

Yi wis, 14, 97-102
Yellow fever, 400-102

bacteria in. 100

and mosquitoes, 402
Yellow milk, 201

Zikiil-Nkli.sen stain, 459

PRINTED BY

OLIVER AND BCJ I)

EDINBURGH

2

UNIVERSITY OF CALIFORNIA
MEDICAL SCHOOL LIBRARY

THIS BOOK IS DUE ON THE LAST DATE
STAMPED BELOW

Books not returned on time are subject to a fine of 50c
per volume after the third day overdue, increasing to
$1.00 per volume after the sixth day. Books not in de-
mand may be renewed if application is made before expi-
ration of loan period.

6738747

3 1378 00673 8747

K ^