REESE LIBRARY

i >i TIII

UNIVERSITY OF CALIFORNIA.

AN INTRODUCTION

TO THE

BACTERIOLOGICAL
EXAMINATION OF WATER

AN INTRODUCTION

TO THE

BACTERIOLOGICAL "
EXAMINATION OF WATER

BY

W. H. HO R ROCKS

M.B.,MB.Sc., LOND.

ASSISTANT PROFESSOR OF MILITARY HYGIENE IN THE ARMY MEDICAL SCHOO1

NETLEY; ASSISTANT EXAMINER IN HYGIENE UNDER THE BOARD OF

EDUCATION, SOUTH KENSINGTON ; MAJOR, ROYAL ARMY

MEDICAL CORPS

LONDON
J. & A. CHURCHILL

7 GREAT MARYBOROUGH STREET
1901

PREFACE.

IN this volume I have endeavoured to present in a simple form
the methods and results of a bacteriological examination of
water. The object of such an examination being to discover
whether a water is likely to be prejudicial to health, it is
obviously necessary to have clear ideas as to the hygienic value
of the various micro-organisms which have been isolated from
water. Great difficulties have been introduced into the subject
owing to the extreme variability of water bacteria. The micro-
scopical appearances of these micro-organisms are largely
influenced by the media on which they are growing, and the
cultural characteristics are greatly dependent on whether the
micro-organisms are enfeebled or not. It is also well known
that bacteria, which normally inhabit the human body, find in
water very unfavourable conditions for their existence ; con-
sequently, after a comparatively brief existence in this medium,
they often exhibit weakened forms when isolated on ordinary
bacteriological media. In order to simplify matters as much as
possible, the bacteria which occur in water have been arranged
in three classes, viz., (1) bacteria which are found in pure water ;
(2) bacteria which are common in sewage but rarely met with in
pure water ; and (3) bacteria which give rise to specific disease
in human beings. The micro-organisms in the first class are
very numerous, but many of them have been so imperfectly
described that it has been very difficult to group them in a
satisfactory manner. As the easiest solution of the difficulty,
and to facilitate reference, certain well-defined types have been
selected, and under each type have been described all those
bacteria which are probably varieties, the variations being caused
by differences in food, habitat, £c. I have examined nearly all
the bacteria mentioned, and the descriptions are taken from

95829

vi PREFACE.

observations of pure cultures obtained either from Krai's
laboratory or from the original discoverers of the bacteria.

The subject of the bacterial flora of sewage is yet in its
infancy, but enough work has been done to enable certain
bacteria to be classed as the normal inhabitants of sewage, and
I hope this section will assist in the solution of the difficult
problems which are often placed before the water bacteriologist.

Of the bacteria which give rise to specific disease in human
beings, only the B. typhosus and Sp. Cholene Asiaticae occur
sufficiently often in water to merit description in a small work.
The differential diagnosis of the B. typhosus by the serum test
has lately received much attention, and an attempt has been
made to summarise the most recent work on the subject.

The plates have been drawn from my own sketches, and the
colonies depicted are believed to represent the forms most
commonly seen during routine examinations of " water-plates." I
have to thank Major Leishman, R. A.M.C., for kindly reading the
proofs, and for many valuable suggestions. I have further to
express my indebtedness to Dr. A. C. Houston, who has allowed
me to reproduce the descriptions of the sewage bacteria given in
his report to the London County Council ; and to Dr. Mervyn
Gordon for the use of his valuable paper on B. coli.

W. H.

SHOLING, SOUTHAMPTON.

September 1901.

CONTENTS.

CHAPTER 1.

PAGE

INTRODUCTORY . . ... . . . i

COLLECTION OF WATER FOR BACTERIOLOGICAL EXAMINATION . 4?

CHAPTER II.

QUANTITATIVE BACTERIOLOGICAL ANALYSIS . . . . 7
PREPARATION OF " WATER PLATES "..... 9

CHAPTER III.

THE MULTIPLICATION OF MICRO-ORGANISMS IN WATER . . 12
THE INFLUENCE OF 'LIGHT ON MICRO-ORGANISMS . . .14
THE INFLUENCE OF MOVEMENT AND REST ON MICRO-ORGANISMS 17
THE SEDIMENTATION OF MICRO-ORGANISMS . . . .18
THE DURATION OF LIFE OF MICRO-ORGANISMS . . .21
THE INFLUENCE OF THE CHEMICAL COMPOSITION OF A WATER

ON THE NUMBER OF MICRO-ORGANISMS .... 22
THE ACTION OF ELECTRICITY ON BACTERIA .... 24

CHAPTER IV.

THE BACTERIAL CONTENTS OF THE VARIOUS SOURCES OF SUPPLY 26*
THE RELATION OF THE RESULTS OF A QUANTITATIVE ANALYSIS

TO THE QUESTION OF POLLUTION . . . .30

CHAPTER V.

THE RELATION OF THE QUANTITATIVE ANALYSIS TO FILTRATION

OF WATER THROUGH SAND . ... 33

viii CONTENTS.

CHAPTER VI.

PAOK

QUALITATIVE BACTERIOLOGICAL ANALYSIS . . . 41

MICRO-ORGANISMS WHICH ARE FOUND IN PURE WATERS . 42

CHAPTER VII.
NITRIFYING AND DENITRIFYING MICRO-ORGANISMS , .81

CHAPTER VIII.

MICRO-ORGANISMS WHICH ARE FOUND IN SEWAGE AND POLLUTED

WATERS 88

CHARACTERISTICS OF THE BACILLUS COLI COMMUNIS GROUP . 88

METHODS EMPLOYED FOR THE ISOLATION OF THE B. COLI FROM

WATER SUPPLIES .... . . . -98

THE VALUE OF THE B. COLI COMMUNIS AS A SIGN OF SEWAGE

CONTAMINATION OF A WATER SUPPLY .... 103

CHAPTER IX.

THE BACILLUS ENTERITIDIS SPOROGENES . . . . . ltT8
VALUE OF BACILLUS ENTERITIDIS SPOROGENES AS A SIGN OF

SEWAGE CONTAMINATION . ... .112

CHAPTER X.

SPECIES OF STREPTOCOCCI AND STAPHYLOCOCCI. . . .115
VALUE OF STREPTOCOCCI AND STAPHYLOCOCCI AS A SIGN OF

SEWAGE CONTAMINATION . . . . . . 124

CHAPTER XI.

THE PROTEUS GROUP . . . . . . . .125

MICRO-ORGANISMS SUGGESTIVE OF SEWAGE CONTAMINATION 127 '

CONTENTS ix

CHAPTER XII.

PAGE

"SEWAGE BACTERIA" DESCRIBED BY JORDAN . . .135

39 ,) „ LAWS AND ANDREWES . 139

t> „ „ HOUSTON. . . .145

THERMOPHYLIC BACTERIA . . . . . . 155

CHAPTER XIII.

THE CHARACTERISTICS OF THE B. TYPHOSUS . . . .162
THE RELATION OF TYPHOID BACILLI TO OTHER PATHOGENIC

AND NON-PATHOGENIC BACTERIA . . . . .185
THE RELATION OF THE TYPHOID BACILLUS TO ACIDS AND

ALKALIES IN NUTRIENT MEDIA . . ... 187
MICRO-ORGANISMS W'HICH RESEMBLE THE B. TYPHOSUS . .188
METHODS OF ISOLATING THE B. TYPHOSUS FROM WATER . 1.97

EXPERIMENTAL RESEARCHES ON THE DURATION OF LIFE OF THE

B. TYPHOSUS IN WATER AND SEWAGE, ETC. . . . 209
CASES IN WHICH THE B. TYPHOSUS HAS BEEN ISOLATED FROM

WATER SUPPLIES . . . . . . . . 220

POSSIBILITY OF DIAGNOSING POLLUTION OF WTATER WITH DEJECTA

OF ENTERIC FEVER, APART FROM THE ISOLATION OF THE

TYPHOID BACILLUS . . . . . . . 226

CHAPTER XIV.

THE CHARACTERISTICS OF THE SPIRILLUM CHOLERA ASIATICS 236
MICRO-ORGANISMS WHICH RESEMBLE THE SPIRILLUM CHOLERA

ASIATICS 242

METHODS OF ISOLATING THE CHOLERA SPIRILLUM FROM WATER

SUPPLIES . . . . . . . 253

THE VITALITY OF THE CHOLERA SPIRILLUM IN WATER . . 254
THE RELATION OF CHOLERA VIBRIOS TO OTHER MICRO-
ORGANISMS ..... . 260

BEHAVIOUR OF SP. CHOLERA IN MEDIA CONTAINING ACIDS

AND ALKALIES . . . . . 263

x CONTENTS.

CHAPTER XV.

PAGE

THE MODE OF ACTION AND UTILITY OF THE PASTEUR-('HAM-

BERLAND AND BERKEFELD FlLTERS ..... 26,5

CHAPTER XVI.

SUMMARY OF THE PROCEDURE RECOMMENDED FOR THE BAC-
TERIOLOGICAL EXAMINATION OF WATER .... 282
PREPARATION OF CULTURE MEDIA . . . 284

BIBLIOGRAPHY ......... 289

INDEX .

V

PLATE I.

FIG.

1. B. FLUORESCENS LIQUEFACIENS. — Surface colony on a

gelatine-plate x 1 00 ; forty-eight hours growth at
22° C. Liquefaction has commenced.

2. B. FLUORESCENS LIQUEFACIENS. — Surface colony on a

gelatine-plate x 1 00 ; twenty-four hours growth at
22° C.

3. B. ARBORESCENS. — Surface colony on a gelatine-plate

x 100 ; forty-eight hours growth at 22° C.

4. B. ARBORESCENS. — Early stage of colonies ; of tenseen in

the depth of gelatine-plates.

5? B. FLUORESCENS NON-LIQUEFACIENS. — Surface colony on a
gelatine-plate x 1 00 ; forty-eight hours growth at
22° C.

6. B. AQUATILIS. — Surface colony on a gelatine-plate x 100;

forty-eight hours growth at 22° C.

7. B. ERYTHROSPORUS. — Surface colony on a gelatine-plate

x 100; seventy- two hours growth at 22° C.

5. B. MESENTERICUS VULGATUS. — Surface colony on a gela-

tine-plate x 100; forty-eight hours growth at 22° C.

Plate 1.

Fiy. 3.

Fig. 6.

. 5.

Ftff. 6'.

A S.HutK.LitKT London.

PLATE II.

PIG.

1. PROTEUS VULGARIS. — Colony in the depth of a gelatine-

plate x 100 ; thirty-six hours growth at 22° C.

2. PROTEUS VULGARIS. — Colony on the surface of a gelatine-

plate x 100; thirty-six hours growth at 22° C.

3. PROTEUS VULGARIS. — Young colony in a gelatine-plate

x 100 ; twenty-four hours growth at 22° C.

4. PROTEUS MIRABILIS. — Young colony in a gelatine-plate

x 100; twenty-four hours growth at 22° C.

5. PROTEUS ZOPFI. — Surface colony in a gelatine-plate x 1 00;

seventy- two hours growth at 22° C.

6. PROTEUS MIRABILIS. — Surface colony in a gelatine-plate

x 100 ; seventy-two hours growth at 22° C.

7. PROTEUS ZENKERI. — Surface colony in a gelatine-plate

x 100; seventy-two hours growth at 22° C.

8. PROTEUS MIRABILIS.— Zoogloea forms in the depth of a

gelatine-plate x 100.

Plate 11.

-"-I.. 3. HutK. Lithr London .

PLATE III.

TIG.

1. B. SUBTILIS. — Surface colony in a gelatine-plate x 100;

forty-eight hours growth at 22 J C.

2. B. MEGATERIUM. — Surface colony in a gelatine-plate

x 100; forty-eight hours growth at 22° C.

3. B. MYCOIDES. — Surface colony in a gelatine- plate x 100;

forty-eight hours growth at 22° C.

4. B. MEGATERIUM. — Surface colony in a gelatine-plate

x 100; seventy-twro hours growth at 22° C. A later
stage than fig. 2.

5. B. AUREUS. — Surface colony in a gelatine-plate x 100;

seventy-two hours growth at 22° C. The colony
has a golden-yellow colour.

6. B. OCHRACEUS. — Surface colony in a gelatine plate x 100;

thirty-six hours growth at 22° C. Liquefaction is
just commencing.

7. B. LIQUEFACIENS. — A young colony in a gelatine-plate

x 100. Liquefaction has not commenced.

8. B. Fuscus. — Surface colony in a gelatine-plate x 100.

The colony has a dark brown colour in the centre
and a thin filmy margin.

Plate 111.

. 2.

•r

A ,S. HutK. LitKr London

PLATE IV.

FIG.

1. B. ENTERITIDIS, GARTNER. — Surface colony in a gelatine-

plate x 100; seventy-two hours growth at 22° C.

2. B. COLI COMMUNIS. — Surface colony in a gelatine-plate

x 100; forty-eight hours growth at 22° C.

3. B. TYPHOSUS. — Surface colony in a gelatine-plate x 100;

four days growth at 22° C. "Ridges and valleys"
well marked.

4. B. ACIDI LACTICI. — Surface colony in a gelatine-plate

x 100; seventy-two hours growth at 22° C.

5. B. TYPHOSUS. — Surface colony in a gelatine-plate x 100;

four days growth at 22° C. " Glass pellicle " variety
of colony.

6. B. CAVICIDA. — Surface colony in a gelatine-plate x 100;

forty-eight hours growth at 22° C.

7. B. PYOCYANEUS. — Surface colony in a gelatine-plate

x 100; thirty-six hours growth at 22° C. Lique-
faction just commencing.

8. 13. VIOLACEUS. — Surface colony in a gelatine-plate x 100 ;

forty-eight hours growth at 22° C. The colony has
a violet colour. Liquefaction has not commenced.

Plate IV.

FUj.l.

Fiy. 3.

Fist. ',.

Fist.

Fi<f.

A.S.Hutk.LitKr London.

PLATE V.

PIG.

1. B. URE^E. — Surface colony on a gelatine -plate x 100;

seventy -two hours growth at 22° C.

2. B. AQUATILIS SULCATUS. — Surface colony in a gelatine-

plate x 100 ; forty-eight hours growth at 22° C.

3. B. ALBUS or B. UBIQITUS. — Surface colony 011 a gelatine-

plate x 100 ; forty-eight hours growth at 22° C.

4. B. MESENTERICUS RUBER. — An early stage of a colony in

a gelatine-plate x 100.

5. M. FLAVUS LIQUEFACIENS. — Colony in a gelatine-plate

x 100 ; forty-eight hours growth at 22° C.

6. B, CYANOGENUS. — Surface colony in a gelatine-plate

x 100; forty-eight hours growth at 22° C. The
gelatine surrounding the colony has a deep blue
colour.

7. B. DENITRIFICANS. — Surface colony on a gelatine-plate

x 100 ; seventy-two hours growth at 22° C.

8. SP. CHOLERA ASIATICS. — Surface colony in a gelatine-

plate x 100; seventy-two hours growth at 22° C.
Liquefaction well advanced.

Plate V.

Fty.l.

. 7.

Fig. 8.

A. S.HutK LitKr London.

CHAPTER I.

COLLECTION OF WATER.

OPINION as to the value of a bacteriological examination of
water has passed through many phases. The elaboration of
Koch's gelatine plate method gave a great impetus to the
bacteriological study of water supplies ; it enabled an estimate
to be made of the number of micro-organisms present in a
given quantity of water, and so permitted a comparative study
of various sources of supply. It was at first assumed that from
the number of bacteria found in a water distinct evidence of
purity or contamination could be at once deduced. Further
investigations, however, soon showed that water organisms were
capable of multiplying enormously within a very short time,
and waters of great organic purity, under certain conditions,
might be found teeming with bacterial life. It thus became
evident that, though a paucity of organisms might indicate a
condition of purity, it did not necessarily follow that a large
number of micro-organisms pointed to contamination. Realising
that the mere enumeration of the bacteria present in a sample
of water was surrounded by many fallacies and, unless performed
under the strictest conditions, might give rise to false con-
clusions, bacteriologists turned their attention to the study of
the kinds or species of bacteria, and attempts were made to
distinguish between the microbes normally present in a water
and those derived from sewage — i.e.., to distinguish between
water-organisms and sewage-organisms.

The qualitative bacteriological examination, though un-
doubtedly of the first importance and likely in the long run to
be of the greatest practical value to hygienists, has been found
to be attended with exceptional difficulties. In the early days
of the study of bacteriology, and when modern methods of

a BACTERIOLOGICAL EXAMINATION OF WATER.

distinguishing bacteria were unknown or imperfectly practised,
eager investigators of the bacterial flora of waters isolated a
large number of water-organisms. Many micro-organisms were
described as distinct species which were really only varieties, the
variation being caused by differences in food, habitat, tempera-
ture, £c. The confusion thus originated gradually became so
great that many hygienists lost faith in the bacteriological
examination of water and were inclined to believe that a reliable
opinion as to the purity of a supply could be given only from
the results of a chemical examination. But the indications given
by chemical means are merely relative, the results so obtained
cannot be correctly interpreted except by reference to local
standards. There is no general standard which can be fixed for
all waters; the source and immediate surroundings of the supply
must be known before a reliable opinion as to the quality of the
water can be given. Also it is a matter of common knowledge
how quickly some waters change after collection, and this is
especially the case when the sample is obtained from a polluted
source. It is now well known that in most waters there are two
groups of micro-organisms which may largely influence the
chemical composition of the water. In the first group are
included the micro-organisms which break up organic matter
into its chemical elements and reduce nitrates and nitrites to
ammonia ; to the second group belong the oxidising micro-
organisms which convert ammonia into nitrites and nitrates.
The exact conditions under which these two groups of micro-
organisms operate are not worked out at the present time. The
presence of oxygen is supposed to have an important influence,
but there must be other conditions, as yet unknown, which
influence the final result. It appears from the researches of
Heraeus that one species of bacteria in the absence of nitrates
oxidises a part of the ammonium salts to nitrites; but if nitrates
are present they are reduced to nitrites, and ammonia notwith-
standing an abundant supply of air. Another species also
appears to oxidise ammonia to nitric acid in the presence of air,
but in the absence of oxygen reduces nitric acid to ammonia.
Winogradsky doubts the possibility of the same species possess-
ing both oxidising and reducing characteristics. It seems
probable that Heraeus1 cultures were impure, and that the

COLLECTION OF WATER. 3

reducing and oxidising powers really belong to distinct groups
of micro-organisms. Still, the fact that a potable water may
undergo entirely opposite chemical changes according as one
group or the other gains the ascendency shows how frail a basis
there is for the belief that the determination of ammonia,
nitrites and nitrates will enable a reliable opinion to be given
as to " previous sewage contamination." At the same time, it
is certain that in surface waters ammonia and nitrites are
scarcely ever present to any extent without sewage contamina-
tion. Consequently, if a chemical examination is made imme-
diately after collection of the sample it will assist us in arriving
at a proper judgment ; this is especially the case when the
pollution is solely from urine, under which condition bacteri-
ology is often at fault. Much patient work on sewage organisms
has shown that a recent contamination of a water supply by
sewage can easily be detected even when the pollution is so
slight as one part per million. Chemistry is quite unable to
indicate such a slight contamination as this. If, however, a
considerable time has elapsed since the pollution occurred it
becomes more and more difficult to detect it by bacteriological
processes ; and this is especially the case when waters of great
original purity are under examination. It is therefore evident
that a bacteriological examination has its limits of usefulness,
and a slavish adherence to it under all conditions, combined
with neglect of the hints to be obtained by chemical means, may
lead to a perfectly erroneous judgment. Still, there is one
branch of hygienic study in which bacteriology must always
reign supreme ; it is now acknowledged on all sides that the
working of sand filters for public water supplies cannot be
properly kept under control except by appealing to bacteriological
methods of examination.

From what has been said it is obvious that a bacteriological
examination of water consists essentially of two parts — viz., (a)
the quantitative analysis or determination of the number of
bacteria in a known volume of water, and (b) the qualitative
analysis or determination of the species of bacteria present in
the sample.

Before proceeding to the consideration of these analyses it
will be advisable to mention a few points in relation to the

4 BACTERIOLOGICAL EXAMINATION OF WATER.

collection of the sample of water which is to be subjected to a
bacteriological examination.

The collection of water for bacteriological examination. — Water
from a river, well, lake or service-pipe should be collected in
glass bottles closed with glass stoppers, previously sterilised by
heating in the hot-air chamber to 150° C. for three hours ; if
sterilisation cannot be effected, the bottles may be cleaned
sufficiently for all practical purposes by washing with a little
pure sulphuric acid, all traces of acidity being finally removed
by thoroughly rinsing the bottle with some of the water to be
examined. In the case of a river or lake the bottle should be
plunged below the surface before the stopper is removed ; in this
way a sample of the main body of the water will be obtained.
In some cases, as when investigating an outbreak of cholera, it
is desirable to examine the surface water. This specimen should
be collected in a separate bottle and labelled accordingly. If
the source of supply is a " service- water,1' the tap should be
opened and the water allowed to run to waste for a few minutes
before the specimen is collected ; in this way local impurities will
be washed out of the tap and the water which has been
standing in the service-pipes will be removed. It is always
desirable in the case of a " service- water " to obtain a specimen
direct from the mains. In the case of a well not in constant
use it is generally found that the sample obtained on first
pumping contains a very large number of micro-organisms ; so
that, to gain an idea of the water which enters the well from the
surrounding strata, it is advisable to pump continuously for
several hours before collecting the example for examination.
After the specimen has been obtained the glass stopper should
be covered with oiled silk and carefully tied down. When the
source of supply is a shallow stream, such as the feeder of an
upland surface water, it is often impossible to use a bottle
without disturbing the sediment. Under these conditions a
test-tube with stout walls is drawn out just below the open end ;
the contained air is then removed by prolonged heating, or more
simply by allowing a little water to enter the tube, which is then
heated until steam escapes, when the point is rapidly sealed.
The fine point of such a tube should be placed under the
surface of the stream, and then the tube opened by breaking the

COLLECTION OF WATER. 5

point with a pair of sterile forceps. The water will run in and
fill about two-thirds of the tube ; the point is then sealed up by
heating with a spirit lamp. When examining lakes and rivers
it is often desirable to take specimens from various depths.
This may be done by using MiquePs apparatus, which consists
of a glass bulb weighted below and suspended by a wire marked
in feet or metres. The neck of the bulb is drawn out to a fine
point and bent like a swan's neck. A second wire, running
through eyes on the suspension wire, is bent at the end into a
loop which is placed round the neck of the bulb. When the
required depth is shown on the suspension wire, the second wire
is pulled sharply so as to break the neck and allow the water to
enter the bulb. The same result may be obtained by weighting
the ordinary glass bottle and supporting it by means of a
string knotted at intervals of a foot. A second string is tied
round the glass stopper, which should be partially unscrewed so
as to be easily pulled out when the required depth is reached.
The samples should be examined immediately after collection ;
if this cannot be done they must be packed in ice for trans-
mission to the laboratory. It has been found that wJien the
temperature is kept below 5° C. there is practically no increase
in the number of micro-organisms in the water. This effect
appears to be due, not to an immobilising.of the bacteria, but
rather to the fact that cold acts differently on certain species, so
that the deaths of some organisms are balanced by the births
of others. In order to keep the temperature of the sample
below 5° C- during its conveyance to the laboratory, Miquel
recommends the following procedure to be adopted. The glass
bottle containing the sample is placed in a cylindrical metal box
just large enough to hold the bottle firmly. The metal box is
placed inside a second slightly larger metal box and the space
between the two boxes is filled with sawdust. The whole is
then placed in a much larger metal box, which is filled with
from six to eight pounds of ice broken up into pieces about
the size of a walnut. The third metal box is placed in a wooden
box and the space between them is filled with sawdust. The
lid of the wooden box, which is provided with a handle, is then
clamped down. Pakes has suggested a very useful box for the
same purpose. The box is made of wood and lined with

6 BACTERIOLOGICAL EXAMINATION OF WATER.

felt ; it contains two metal cylinders placed one inside the
other ; the inner cylinder contains the sample, which is packed
round with wool ; the outer cylinder is much larger, and is
divided by partitions into sections, which are filled with
pounded ice. A metal lid is placed over, and securely clamped
to, the everted lips of both the inner and outer metal cylinders.
A thick layer of felt is then placed over the metal lid. The
wooden box is finally closed with a strong well-fitting lid, on
which a handle is placed.

CHAPTER II.

THE QUANTITATIVE BACTERIOLOGICAL
ANALYSIS.

THE determination of the number of micro-organisms in a known
volume of water is mainly effected at the present day by cultures
on solid media. Miquel, however, has suggested a dilution
method, in which a known volume of water is diluted amongst a
series of broth tubes, the object being to obtain ultimately a
series of broth tubes, each of which shall not receive more than
one microbe. This method is exceedingly laborious, and quite
unsuitable for hygienic work. Most workers now employ what
is known as Koch's method of plate cultivation. In this method
a small quantity of water is intimately mixed with fluid gelatine
or agar-agar, which is then spread out in a thin layer on a glass
plate, or preferably in a Petri dish. Sometimes the mixture of
water and fluid gelatine is simply rolled round the sides of a test-
tube and solidified, producing what is known as an Esmarch roll
culture. The frequent presence in water of micro-organisms which
rapidly liquefy gelatine renders the Esmarch tube less serviceable
than Petri dishes. Much difference of opinion exists as to the
method of preparation of the solid media. The committee of
American bacteriologists recommended that gelatine and agar-
agar should have such a reaction as would be obtained by adding
1-5 per cent. N acid to the media after they had been rendered
neutral to pheriol-phthalein. Sedgwick and Prescott stated that

the best results were obtained by adding 0'2 of ^ acid to each

cubic centimetre of nutritive gelatine. They also found that more
colonies developed on nutritive gelatine containing 15 to 20
grammes of peptone per litre than on gelatine containing
5 grammes of peptone per litre. Reinsch's experiments with

8 BACTERIOLOGICAL EXAMINATION OF WATER.

water from the Elbe, taken below Hamburg and Altona, showed
that the greatest number of bacteria was obtained when O'l per
cent, of sodium carbonate was added to ordinary slightly alka-
line gelatine-peptone. Dahmen, working with water from the
Rhine, found that the addition of 0*15 per cent, of sodium car-
bonate to ordinary gelatine caused the development of the
greatest number of micro-organisms. My own experiments with
the South Hants water supply showed that more bacteria were
obtained on alkaline than on acid media. Two batches of
ordinary gelatine-peptone were prepared, one rendered just
alkaline to phenol-phthalein, and the other to litmus. To the
first batch N acid was added to the extent of 1*5 per cent., as
recommended by the American committee, and to the second
O'l per cent, of sodium carbonate as suggested by Reinsch.
Gelatine plates were then made with 0*5 c.c. of the South Hants
water and incubated at 22° C. The following average results
were obtained :

1 c.c. South Hants water. Acidified Gelatine. Alkaline Gelatine.

(1'5 per cent. N acid.) (0*1 per cent, sodiiirn
carbonate.)

After 48 hours incubation at 22° C. ... 59 Colonies. ... 81 Colonies.

„ 72 „ „ „ ... 88 „ ... 96

On the fourth day the plates were usually destroyed by the
liquefying organisms present in the water. All the water-organ-
isms usually found in the alkaline media appeared to grow in
the acid media, only a little more slowly. Hesse and Niedner
recommend the following medium for the bacteriological exami-
nation of water :

Agar-agar . . . . '. . 1 '25 per cent.
Albumose (Hey den) . . . . . 0'75 „
Distilled water 98'00 „

This medium requires no correction for acidity or alkalinity, and
permits the development of the greatest number of water-
organisms. The authors consider that water-plates for the
simple enumeration of colonies should be kept under observation
for three weeks, as numerous experiments have shown them that
at 20° C. in the first three days 30 per cent., in the first five days
about 70 per cent., and in the first ten days about 90 per cent.

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 9

of the colonies are able to develop in the plates. Gelatine plates,
however, can rarely be kept under observation beyond the fourth
or fifth day, owing to liquefying organisms destroying the plates.
Up to the present time the quantitative analyses of most sources
of water supply have been made by means of faintly alkaline
gelatine-peptone plates, consequently, if it is desired to compare
the results with previous analyses, this medium must be used.

THE PREPARATION OF " WATER PLATES."

The water to be examined should be gently shaken for a few
minutes ; in this way all clumps of bacteria will be broken up
and each organism will produce only one colony when it
develops in the Petri dish. Forcible shaking of the sample
must be avoided as it may cause the death of some of the micro-
organisms. The next step is to determine the amount of water
which should be used for each plate, as it is necessary that each
organism shall have room to develop without touching its
neighbours. The simplest method of arriving at an idea of the
number of bacteria present in the water is to remove ^V c<c* °f
the water from the bottle and deposit it on a cover-glass ;
then by staining and counting the microbes on the cover-glass
preparation it is possible to determine very roughly the number
of micro-organisms present in a c.c. of the water. When work-
ing with Petri dishes having a diameter of four inches, it is best
to use such an amount of water as will give not more than 300
colonies in each plate. If the water contains a very large
number of bacteria it must be diluted with sterile tap-water.
Usually three plates are made from each sample, containing
respectively TV c.c., J c.c. and J c.c. of the water diluted or not
as required. In order to measure the calculated amount of
water required for each plate, accurately graduated pipettes
must be used. One c.c. pipettes graduated in y^ths are most
useful. Each pipette must be thoroughly cleaned, the upper
end plugged with cotton-wool, and the lower end placed in a
test-tube, the mouth of which is firmly plugged with cotton-
wool. The test-tubes and pipettes are then sterilised at 150° C.
for three hours and kept in metal boxes until required for use.
The Petri dishes, wrapped in thin paper, are also sterilised at
the same temperature.

10 BACTERIOLOGICAL EXAMINATION OF WATER.

The gelatine plates should be made in the following manner :
Melt three gelatine tubes in a water-bath at a temperature
of about 30° C., sterilise the cotton- wool plugs in a Bunsen
flame and twist the plugs round so as to loosen them. Draw
up the required amount of water in a sterile pipette ; then
holding the pipette almost horizontally with the right hand,
take the tube containing the melted gelatine in the left
hand between the thumb and first finger and again sterilise
the plug. Now hold the tube as nearly horizontal as possible,
without permitting the gelatine to touch the wool ; remove
the plug with the third and fourth fingers of the right
hand, introduce the water, again sterilise the neck of the test-
tube, and replace the plug. Place the pipette on a sterile metal
stand and roll the test-tube, held vertically between the palms
of the hands, so as to thoroughly mix the water and gelatine.
The test-tube must not be shaken up and down or air-bubbles
will be introduced into the gelatine. Having thoroughly mixed
the water and gelatine, hold the test-tube in the right hand,
sterilise the wool and neck of the tube in the flame, and then
withdraw the plug with the inner margin of the left hand. Next
raise the lid of the Petri dish with the thumb, first and second
fingers of the left hand, introduce the neck of the test-tube into
the aperture, pour out the mixture of water and gelatine, replace
the lid and gently diffuse the contents into a thin layer over the
bottom of the dish. When making the plates it will be noticed
that it is almost impossible to pour out all the gelatine from the
test-tubes without opening the Petri dishes to a dangerous extent,
and as the portion of gelatine still remaining at the bottom of the
test-tube may contain a few organisms, it is best, after the plug
has been replaced, to distribute the gelatine over the walls of the
tube in as thin a layer as possible — in fact, to convert the tube
into an Esmarch roll. The plates and test-tubes are usually
incubated at 20° to 22° C. and the colonies which develop are
counted from day to day. As a rule it will be found that there
is no very material increase in the number of colonies after the
fifth day, so if possible the plates should be kept for this
period. If many liquefying organisms are present the plates
will usually be destroyed by the third or fourth day.

The simplest method of counting the number of colonies is to

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 11

divide up the outer surface of the bottom of the Petri dish into
sections by straight lines running from the centre to the
periphery ; this can easily be done with a wax pencil. A small
hand magnifying glass will be found useful in differentiating
masses of granules in the gelatine from true colonies. The
result of the enumeration should be stated as so many aerobic
colonies or organisms which have developed at 20° to 22° C. after
from forty-eight hours to five days incubation, as the case may be.
Neisser recommended that the plates should be counted under the
microscope, and showed that the number obtained by this method
after twenty-four hours incubation would equal the number
obtained by the ordinary microscopic method after seventy-two
hours incubation. In tropical climates it is often impossible to use
gelatine, for during many months of the year the air temperature
is above the melting-point of this medium, consequently the plates
remain fluid and the bacteria are not kept apart so as to permit
a proper isolation of the colonies. Lender these conditions agar-
agar must be used as the nutrient medium. Agar tubes are
melted and seeded with the water after they have been cooled
down to 40° C. ; the mixture of water and agar is rapidly poured
out, as before described, into Petri dishes and allowed to
incubate at the prevailing temperature. A simpler method is
to pour out the melted agar into a Petri dish and allow it to
set ; the calculated amount of water is then introduced by means
of a pipette and the fluid evenly distributed over the surface of
the agar by means of a platinum spreader. The high tempera-
tures which obtain in tropical climates are prejudicial to the
development of water- organisms, so that the counts obtained on
the agar plates are smaller than those obtained from the same
water seeded in gelatine during the colder months.

The interpretation of the results of a quantitative bacterio-
logical analysis is a matter of considerable difficulty. In order to
give a reliable opinion on the results obtained, it is necessary to
study the causes of the multiplication of micro-organisms and the
way in which they are influenced by light, temperature, movement,
and the chemical composition of the water. The bacterial contents
of the various sources of supply must also be examined to see if
they maintain, under natural conditions, a fairly constant
numerical composition.

CHAPTER III.

QUANTITATIVE ANALYSIS— continued.

THE MULTIPLICATION OF MICRO-ORGANISMS
IN WATER.

MIQUEL'S researches showed that in surface waters, such as rivers
containing a large number of micro-organisms, the multiplication
is slow and persistent, whilst in deep well waters and springs,
containing only a few bacteria, the multiplication is rapid but
soon followed by a marked and steady decrease.

Frankland's experiments with the rivers Thames and Lea gave
the following results :

Day after collection. After 2 days in the After 4 days in the

dark at 20° C. dark at 20° C.

Organisms per c.c. Organisms per c.c. Organisms per c.c.

Kiver Thames)
.t Hampton/ 12'25° *'386 2'°18

Rj™rlf?t } ... 7,300 2.H8 1.286

Chmgford

The number of micro-organisms in the river water underwent
a marked diminution after storage for from two to four days in
stoppered bottles.

An examination of a deep well water gave results analogous to
those obtained by Miquel.

Day after collection. After 3 days in the After 16 days in the

dark at 20° C. dark at 20° C.

Bacteria per c.c. Bacteria per c c. Bacteria per c.c.

Kent Well ... 96 ... 178,379 ... 51,843

Cramer's experiments on filtered water from Lake Zurich also
strongly supported MiquePs observations as to the changes
occurring in pure waters. He filled a sterilised flask with the

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 13

filtered water from Lake Zurich and examined specimens from
time to time with the following results :

Colonies per c.c.

Immediately after filling the flask . . :'&"'. 143

After 24 hours . . . . .'..., 12,457

„ 3 days . . . . . . . 328,543

8 „ . . . . . . , * 233,452

.. 17 „ . . . . ..' . . 17,436

,, 70 „ 2,500

A low temperature appears to arrest the multiplication of
micro-organisms. Miquel showed that when a water was kept
below 5° C. the number of bacteria did not appreciably alter.
The number was kept relatively about the same, not because the
organisms were unable to increase but because they were un-
equally affected by temperature, the deaths of some species being-
balanced by the births of others.

Wolffhugel and Reidel stated that there was a difference in
the multiplication of micro-organisms in water according as the
samples were exposed to the air or not. It was generally found
that in the vessels closed with india-rubber stoppers somewhat
less multiplication took place than in those which were closed
with cotton-wool plugs. The apparently more flourishing con-
dition of the micro-organisms in the latter case seemed to be
due to the interchange of air being less restricted.

Frankland obtained the same results when working with water
from the Thames and Loch Katrine.

The diminution in the number of bacteria in a sample of water
kept for many days, appears to be due partly to some one or
other of the food materials necessary for the growth of certain
varieties being used up, and partly to the products of the
bacteria being harmful so that the water becomes both
" impoverished and poisoned.'1

Sirotinin made some experiments on the impoverishment of
media. He planted out the Bacillus typhosus on a thin layer of
gelatine and obtained a good growth ; on removing the first
crop, however, he was unable to obtain a second until the gelatine
already used had been melted and mixed with 1 per cent,
peptone and O'l per cent, meat extract. Similar results were

14 BACTERIOLOGICAL EXAMINATION OF WATER.

obtained with B. pyogenes fcetidus, B. flavescens putidus and
B. murisepticus.

The influence of the products of one species on the growth of
another is well shown by Garre's experiments. B. fluorescens
putidus was grown on gelatine slopes and then carefully scraped
off so as to injure the medium as little as possible. B. typhosus
was then planted out on one of the scraped tubes and Sp. cholerae
on the other. The tube containing the B. typhosus remained
sterile, but the Sp. cholerae grew well. B. typhosus was then
grown on gelatine and, after the growth had been carefully
removed, B. fluorescens putidus was planted out on the scraped
surface. A good growth occurred, showing that, whereas the
products of the B. fluorescens putidus were harmful to the
B. typhosus, the converse was not the case. Working with
B. fluorescens liquefaciens I found that B. typhosus would not
grow on gelatine which had already yielded a growth of the
B. fluorescens liquefaciens, but gelatine which had already
served as a medium for the B. typhosus was still capable of
nourishing the B. fluorescens liquefaciens.

Zagari also found that anthrax bacilli would not grow
in broth which had served as nutrient material for the
Sp. cholerae.

Sirotinin appeared to believe that the restraining influence
was due either to a strong acidity or strong alkalinity produced
by the development of the organisms previously planted out on
the media, as he found that when the media were neutralised
the inhibiting influence was removed. A change in the reaction
of the medium cannot be the cause of the prej udicial influence
in all cases. In my experiments with the B. fluorescens lique-
faciens and the B. typhosus I found that after scraping off the
former organism from a gelatine slope, melting the medium and
rendering it exactly neutral had no influence on the develop-
ment of the B. typhosus : the organism still refused to grow.

THE INFLUENCE OF LIGHT ON Micuo-OiiGANisMs.

In the year 1877 Downes and Blunt published a paper on the
action of light on micro-organisms. These observers showed
that, under favourable circumstances, exposure to light entirely

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 15

prevented the growth of bacteria in Pasteur's solution ; direct
sunlight had the greatest influence, but diffused daylight also
appeared to have a restraining effect. Experiments were also
made by them which showed that the blue and violet rays of the
spectrum entirely prevented the growth of bacteria, while the
red and orange-red rays merely delayed development.

Downes and Blunt also found that spores were uninjured by
exposure to light in a vacuum, and came to the conclusion that
the action of light was due to a gradual process of oxidation set
up by the sun's rays in the presence of oxygen.

Duclaux experimented with Tyrothrix scaber and Micrococci,
and demonstrated that the power of resistance to light varied
with the species, the kind of nutrient medium and the intensity
of the light.

Arloing showed that anthrax spores were killed more quickly
than anthrax bacilli. Roux confirmed Arloing's results, and
pointed out that the action of light might be due to the pro-
duction of oxygen.

Engelmann showed that all bacteria are not unfavourably
influenced by light, the Bacterium photo-metricum being
motionless in the dark but mobile when exposed to light.

Janowski studied the influence of light on the B. typhosus,
and found that diffused daylight delayed its growth both on
gelatine and in broth. The same cultures were killed in four
to ten hours by exposure to sunlight, although the temperature
of the media had not reached 40° C. Further investigations
showed that the chemical rays were especially harmful.

Buchner introduced B. typhosus, B. coli and B. pyocyaneus
into water containing a little Liebig's extract and exposed the
cultures to direct sunlight and diffused daylight. His results
are shown in the following table :

B. TYPIIOSUS. TS. COLT.

Exposed Kept in Exposed Kept in
to light, the dark. to light, the dark. Conditions.
Number of bacteria at

. 3,000 4,600 4.600 ,

the commencement . ) Direct

Number of bacteria |. 0 7 600 0 12.600 I 8unli«ht-

after two days . . J
Number at the com- ^ ooo ^m ^QQ \ jy.

mencement ... V
After three hours . J 0 5,000 0 9,800 J sunllSht-

16 BACTERIOLOGICAL EXAMINATION OF WATER.

At the commencement.
After four hours .

At the commencement

After two days

At the commencement

After three hours

At the commencement

After four hours .

B. TYPHOSUS. B. COLT.

Exposed
to light.

Kept in Exposed
the dark. to light.

Kept in
the dark. Conditions.

2.048
0

2,200 1,472
1,024 3,328

9,856 ) Diffused
11,776 / daylight.

B. PYOCYANEUS.

Exposed to light. Kept in the dark. Conditions.

50,200

9,600

22,600

0

3,328
0

38,400

| Direct

26i>,000

/ sunlight.

21,000

) Direct

21,000

J sunlight.

2,048

) Diffused

2,432

J daylight.

Buchner also arranged a series of experiments to determine
the effect of sunlight on cultures placed at different depths in
water. He prepared agar and gelatine capsules containing
B. typhosus, B. pyocyaneus and Sp. cholerae, and closed them
with india-rubber. On the upper surface of the capsules a dark
paper cross was fastened. The capsules were then placed in
water exposed to light with the following results :

Distance of the plate
from the surface
of the water.
O'l metre.
1-1 ..
1-6

CULTURE.

Sp. cholerre.
B. pyocyaneus.
B. typhosus.

DEVELOPMENT OF COLONIES.

In the covered part
of the plate.
Very good.

2-6

3-1

B. pyocyaneus.

B. typhosus.

Better growth
on the part
posed to light.

{Slightly better than^j
the part exposed j-
to light.

In the part exposed

to light.

Nil.

Good.

It appears therefore that in clear water the influence of light
on bacteria extends to a depth of two metres.

Dieudonne worked with B. prodigiosus and B. fluorescens, and
showed that it was the light and not the heat rays which had a
destroying effect. The most powerful rays were the ultra-violet,
violet and blue rays ; the green rays had less power, and the
yellow and red rays appeared to have no action. Culture media
also appeared to be affected, hydrogen peroxide being developed
from the water present.

Frankland and Ward investigated the influence of light on

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 17

anthrax spores. They found that the sun's rays in December
killed the spores in gelatine plates after a few hours exposure,
although the temperature was so low that the plates were not
liquefied. The spores also seemed to be more resistant to the
action of sunshine when diffused in water than when sown in
ordinary culture materials. Ward showed that the spores were
killed by the direct action of sunlight and not through changes
in the media.

The following conclusions seem to be justified by the results
obtained up to the present time:

(1) There is no doubt that light, and especially sunlight, has
an injurious action on micro-organisms, and that it acts directly
on bacteria as well as by producing changes in the culture
media.

(2) The action of light is greatly increased by the presence of
air and moisture, the process being largely an oxidation attended
by the development of hydrogen peroxide.

(3) The precise duration of exposure required to destroy
micro-organisms must necessarily vary with the intensity of the
light and the inherent vitality of the various species.

(4) The influence of sunlight probably extends through water
to a distance of about two metres, but undue reliance must not
be placed on its bactericidal powers even in shallow streams, as
the micro-organisms may be attached to suspended particles, and
so protected to a great extent from the action of light.

THE INFLUENCE OF MOVEMENT AND REST ON MICRO-
ORGANISMS.

Cramer investigated the action of movement by shaking a
specimen of water for four hours. He found 80 bacteria per c.c.
in the water which had been standing for the same time with-
out shaking, and 87 bacteria per c.c. in the water which had
been shaken.

Leone found no difference in the number of bacteria in a water
before and after shaking.

Miquel obtained similar results when working with the water
from the Varne and Dhuis. Tiemann and Gartner found such
slight differences that shaking was considered to have very little
influence. Schmidt could not find any difference in the growth

18 BACTERIOLOGICAL EXAMINATION OF WATER.

of B. prodigiosus, M. candicans, M. staphylococcus anreus, M.
staphylococcus albus, or B. typhosus, after shaking by hand or
in an apparatus. The growth of B. violaceus, however, seemed
to be restrained, and the liquefaction produced by Sp. Finkler
Prior appeared to be slower. Meltzer concluded from his
researches that many bacteria were killed by strong shaking in a
machine, while others remained alive and multiplied.

From these observations it is apparent that the movement
which micro-organisms in water sustain under ordinary circum-
stances has no appreciable influence on their vitality.

ON THE SEDIMENTATION OF MICRO-ORGANISMS.

Sedimentation plays a very important part in changing the
bacterial contents of water supplies. The effect of subsidence
has been investigated by many workers.

Bolton allowed a tall flask to stand at a low temperature, and
then took specimens from the surface, the centre, and the bottom
of the flask. The following results were obtained :

From the surface. From the centre. From the bottom.

No. of bacteria No. of bacteria No of bacteria

perc.c. per c.c. per c.c.

First specimen after \ 2,120 *8,460

twenty hours. / 2,240 44.980

Second specimen after \ 23.760 7,320 13,060

twenty-four hours. / 24,000 7,500 14.000

Third specimen after \ 11,740 25,940

two days. / 11,140 27.520

Fourth specimen after \ 1,720 1,920

three days. / 2,040 1,200

Fifth specimen after \ 2,280 9,580

four days. / 3,840 10,549

Three other specimens were examined after seven months, no
bacteria were found on the surface in any of the flasks, but after
shaking, 540 to 760 organisms per c.c. were counted. Hueppe
examined a service supply which contained only 16 bacteria when
drawn from the pipes. After allowing the water to stand for
two months, 11,280 bacteria per c.c. were found at the surface
and 123,750 bacteria per c.c. at the bottom. In these researches
the observers worked with a mixture of organisms, but there is
no doubt that different species behave differently as regards
sedimentation.

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 19

Tiemann and Gartner made many experiments with pure cul-
tures. Their results are given in the following tables :

Number of bacteria hi 1 c.c.
Flace from which after
ORGANISM. the specimen *

was taken. 1 day. 2 days. 3 days. 4 days.

A motile bacillus produc- rl c.m. under j
mg green pigment. the surface. J

The column of fluid was-l F ^ 173,000 232,000 r 286,000 384,000

20 c.m. high and main- 1 -I (after shaking)

tained at 18° to 20° C. [ >m' J I 288,000 368,000

1 c.m. under \ 25,700
the surface./ 28,200

From the 1 25,200". , ,,

Kafter shaking, 22,000).
bottom. ) no nnn ' v

Noil-motile coccus ob-
tained from the air.

Column of fluid 20 c.m.
high.

In order to test the effect of a greater depth of fluid, a glass
tube 60 c.m. high and 2'5 c.m. wide was filled with sterilised
service water containing 11,000 motile bacilli per c.c. and kept
at 50° C. for ten days. The following results were obtained :

C 910.000 per c.c.
1 c.m. under the surface { g^m ^

Direct from the bottom { JJJjJjj "

Next a glass tube 54 c.m. high and 2 '5 c.m. wide was filled
with sterilised water containing 13,000 white micrococci per c.c. ;
the results were as follows :

After 10 days After 20 days After 30 days

at 5° C. at 12° C. at 10° C.

1 c.m. under the surface, j Q Q

bacteria per c.c. J

From the bottom, j ^ Q Q
bacteria per c.c. J

These experiments appeared to show that neither motile nor
non-motile micro-organisms sank to the bottom.

Scheurlen placed in a tube, 17 c.m. high, suspensions of B.
typhosus, B. megaterium, Proteus vulgaris, Proteus mirabilis,
and Sp. Cholerae Asiaticse. At the end of four days he found
that the three first microbes had in great part sunk to the
bottom, whilst the last two appeared to remain equally distri-
buted through the water.

It is possible that oxygen, being necessary for the maintenance
of vitality of aerobic organisms, exercises an influence on the

20 BACTERIOLOGICAL EXAMINATION OF WATER.

sedimentation of different species. Very few observations have
been made on the sedimentation of bacteria in large masses of
water. Large lakes, away from the shore, appear to contain
very few micro-organisms ; yet during rain large numbers of
bacteria are brought down by rivers, and the mud at the bottom
of lakes is found to be very rich in micro-organisms.

Frankland's experiments with Thames water at the works of
the West Middlesex Company show the effects produced by
storage extremely well. Samples were collected (a) from the
Thames water at Hampton; (b) from the same water after
passing through one storage reservoir; and (c) after passing
through two storage reservoirs. The following results were
obtained :

No. of organisms
per n.c.

(#) Thames water at Hampton 1.437

(&) „ after passing through one reservoir . 318

(c) ,. „ „ two reservoirs . 177

It thus appears probable that a water after receiving a large
number of micro-organisms will, in a very short time, show a
great diminution in its bacterial contents. But sedimentation
is not the only factor in the change ; currents, oxygen require-
ments, inorganic and organic matters sinking to the bottom, all
exercise great influence. Moreover many of the bacteria are
not natural inhabitants of water but are derived from earth,
house-washings, &c., which are rich in nutritive material.
Consequently, with the sinking of suspended material and the
loss of food material, these foreign bacteria may quickly die out.

The self-purification of water may be due to the following
causes :

(1) To the death of those bacteria which either do not belong
to the so-called water-bacteria or are sensitive to the influence
of light.

(2) To a sedimentation of non-motile microbes and the spores
of motile bacteria.

(3) To a spontaneous sinking of bacteria with solid materials
which form nutritive centres.

(4) To a mechanical sinking of bacteria with the subsidence of
heavy materials.

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 21

THE DURATION OF LIFE OF MICRO-ORGANISMS.

The life of micro-organisms may extend over a long period ;
this is especially the case when they possess the power of form-
ing spores. Duclaux found living spores of Tyrothrix after
five years existence in a fluid medium. He also showed that 15
out of 65 vessels which Pasteur used in his experiments on
spontaneous generation were capable of giving rise to new
cultures after being closed for from twenty to twenty-five years.
The reaction of the fluid in the fertile flasks was slightly alka-
line whilst the reaction in the sterile flasks was either strongly
acid or strongly alkaline. Miquel found after nine years only
220 living organisms in a flask which had originally contained
4,800 organisms. Another flask containing Vanne water which
originally showed 66 microbes per c.c. was found quite sterile
after being closed for ten years.

Roth placed water, from a well, containing 5000 bacteria per
c.c. in sterile flasks and found

After 2 days closure 80,000 organisms per c.c.
„ 3 „ „ 120,000 „ " „

„ 5 „ „ 60,000
„ 10 „ „ 43,000
„ 4 weeks „ 2,200 „ „

The presence of carbonic acid gas in a water has a great
influence on the duration of life of micro-organisms. Liborius
in his researches on anaerobes, showed that the development of
colonies was arrested if the air present in the nutritive gelatine
was driven out by carbonic acid gas. Leone found in a water
containing carbonic acid gas that out of 786 bacteria originally
present, only 87 could be detected after five days.

Hochstetter^s researches showed that carbonic acid gas acted
differently according to the species exposed to its influence ;
some organisms were killed, others merely restrained in their
growth. In Seltzer water anthrax bacilli only lived for one hour,
but anthrax spores lived for several days. In water containing
carbonic acid gas the B. typhosus lived for five days, but the
Spirillum cholerae could not be detected after twenty-four
hours. The effect was not due to chemical changes nor to
pressure, but simply to the presence of carbonic acid gas.

22 BACTERIOLOGICAL EXAMINATION OF WATER.

THE INFLUENCE OF THE CHEMICAL COMPOSITION OF A WATER
ON THE NUMBER OF MICRO-ORGANISMS.

The relation of the chemical composition of a water supply
to the number of micro-organisms has been investigated by
many observers, especially in Germany.

Proskauer examined the unfiltered water from Stralaw and
Tegel Works, and the filtered supply delivered to the houses.
His results were as follows :

STRALAW.

TEGEL.

Time of the
investigation.

Org-anisms
per c.c.

Mgms. of
KMnO4.

Chlorine

Organisms
per c.c.

Mgms. of
KMnO4.

Chlorine.

Tarts per 100,000.

Tarts per 100,000.

April, 1888

9,400

2-27

1-775

620

1-83

1-59

May „

1,400

3-06

1-775

88

2-14

1-59

June „

110.000

2-85

2-130

48

1-46

1-77

July ,,

10,200

2-13

2-307

13

1-58

1-77

Aug. „

1,632

3-15

2-791

23

2-14

1-57

Sept. „

10,400

2-93

2-928

52

1-89

1-49

Oct. „

8,400

2-32

3-195

14

1-62

1-59

Nov. „

144,000

2-86

3-017

14

1-79

1-42

Dec. „

220

2-92

2-662

25

1-79

1-59

Jan. 1889

18,500

1-98

2-041

56

1-70

1-50

Feb. „

12,500

2-10

2-485

160

1-58

1-42

March „

190,000

2-10

2-488

68

1.24

0-88

FILTERED WATER.

Time of the
investigation.

Organisms Mgms. of
per c. c. KMnO4.

Chlorine.

April, 1888

18 1-64

1-77

May „

66

•67

1-59

June „

22 ]

•49

1-77

July „

65

•16

1-59

Aug. „

43

•77

1-59

Sept. „

131 1

•53

1-49

Oct. „

12 ]

•39

1-59

Nov. „

6

•49

1-42

Dec. „

12

•53

1-42

Jan. 1889

21

•53

1-59

Feb. „

7 1-17

1-52

March „

110 1-19

1-77

G. Frank obtained similar results when examining the Spree
water from its entrance into Berlin up to Sacrow. From these
experiments it is apparent that there is no constant direct

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 23

relation between the number of organisms and the chemical
composition of the water.

Bolton's work pointed in the same direction ; he distinctly
said that the chemical composition of a water had no relation to
the number of micro-organisms.

Tiemann and Gartner believe that, in relation to nutritive
material, bacteria may be arranged in two groups. The first
group includes the water bacteria, which have very slight
dependence on the nutritive material present, and under favour-
able conditions multiply rapidly. The second group embraces
those organisms which are greatly dependent on the nutritive
material present, and are derived from the earth and living
beings.

With regard to the organic matter the difficulties are
very great ; we are not able to determine exactly which of the
constituents form suitable food for bacteria, indeed Duclaux
has said that the quality of the organic matter is far more
important than the quantity. Using the oxygen derived from
potassium permanganate as a test, Fischer found in one case 7*4
parts per 100,000 of oxygen absorbed with 800 bacteria present,
and in another case only 0*5 part per 100,000 of oxygen
absorbed with 360,000 bacteria present. Tiemann and Gartner
investigated the effect of inorganic substances, especially lime
salts and total solids on the bacterial contents ; the results
obtained were as follows :

Number of wells.

Hardness.

Organisms per c.c.

1

0-10

536

30

10-20

1,695

14

20-30

1,910

1

30-40

720

Total soliJs. Organisms per c.c.

0-50 81

50-100 261

100-150 250

200-500 46

These results show that often a high bacterial count is
associated with high total solids, but with an excess of solids the
organisms may diminish again.

A review of all the experiments appears to justify the

24 BACTERIOLOGICAL EXAMINATION OF WATER.

statement that there is no direct and constant relation between
the chemical composition and the number of micro-organisms ;
but, speaking generally, it may be said that those waters which
contain much organic matter usually show a high bacterial
count.

THE ACTION OF ELECTRICITY ON BACTERIA.

The action of electricity on bacteria was first investigated by
Schiel, who judged the action of the electric current by an
examination of its effects on the motility of micro-organisms.
Colin and Benno Mendelsohn used a purely mineral solution
containing 5 grammes of potassium phosphate, 5 grammes of
sulphate of magnesia, 10 grammes of neutral tartrate of
ammonia, and 0'5 gramme of calcium chloride per litre. The
nutritive fluid was placed in a U-tube, then inoculated with the
micro-organism to be studied, and a current from several
elements of Marie-Davy pile passed through it. If a growth
occurred the current was deemed to have had no action. If no
growth took place this might have been due to the death of the
bacteria or to changes having taken place in the fluid which
rendered it an unfavourable medium. To judge these effects a
loopful was withdrawn from the U-tube and planted out on some
fresh medium, and a new culture was then inoculated into the
tube. Cohn and Mendelsohn found that when the action of the
current was short and feeble its effect on bacteria was nil.
When the current from two powerful elements acted for twenty-
four hours the bacteria were not killed, but the fluid was so
altered that bacteria would not grow in it. When the current
from five elements acted for twenty-four hours the bacteria
were killed at both poles, and the fluid was rendered sterile and
profoundly modified. In order to prevent the decomposition of
the fluid by the current, induction currents were tried but
without any appreciable result. Experiments were then made
by passing an electric current through a potato sown with
B. prodigiosus. Chemical effects were also produced in the
potato, just as in the nutritive fluid, so that though the bacillus
was killed, its death could not be ascribed to the physical action
of the current. Prochownick and Spaeth covered the plates of
the battery with nutritive agar, which were then infected with

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 25

the bacteria to be studied. When development had taken
place, the plates were placed in salt solution and the current
allowed to pass. It was found that the positive pole was more
bactericidal than the negative pole, and that the effect depended
on the intensity and duration of the current. A current of 50
milliamperes passing for a quarter of an hour did not kill the
Staphylococcus pyogenes aureus, but a current of 60 milliamperes
killed it at once. In order to kill a spore-bearing bacillus a
current of 200 to 230 milliamperes was required for two hours.
The bactericidal power of the positive pole appeared to be due
to the chlorine liberated from the salt solution. The effect of
the current, as in the previous experiments, was due to a chemical
and not to a physical* action. According to Fischer strong
electric currents are fatal to bacteria, their protoplasm being
killed, although some of the deleterious effect may be due to
the electric dissociation of the media and to the increase of
temperature caused by the current.

CHAPTER IV.

QUANTITATIVE ANALYSIS— continued.

THE BACTERIAL CONTENTS OF THE VARIOUS
SOURCES OF SUPPLY.

Snow. — Janowski examined snow which had been lying on the
ground for some time ; the superficial layers were removed in
order to avoid organisms which might have been derived from
the air. The sample was then taken, placed in a test-tube and
melted. One c.c. of snow-water being used for each plate the
following results were obtained :

24 hours old. 48 hours old. 72 hours old. 96 hours old.

Sample. Air temperature Air temperature Air temperature Air temperature

6° C. 6'5"C. 10° C. 11 '6° C.

I. 2 18 228 145

II. 4 20 212

Schmelck found in the melted water from snow on an iceberg
in Norway 2 bacteria and 2 moulds per c.c. Janowski also
examined freshly-fallen snow, and from one c.c. of the melted
snow-water obtained the number of micro-organisms given in the
following table :

Taken ou Feb. Taken 011 Feb. Taken on Feb. Taken on Feb.

„ 2, 1888. 20, 1888. 28, 1888. 29, 1888.

Air temperature Air temperature Air temperature Air temperature

7-20° C. 11° C. 12° C. 3-4° C.

I. 34 203 140 134

II. 38 384 168 463

Thus, even at a low temperature, snow has been found to
contain micro-organisms, and the number appears to increase
with the length of time the snow has been lying on the ground.

Ice. — Heyroth examined ice obtained from various places in
Berlin. Blocks of ice were broken and portions removed from
the centre, were then melted in sterile test-tubes and the ice-

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 27

water plated out in gelatine. The following results were
obtained :

Plotzen Lake on September 19, 1885 . . 490 organisms per c.c.

October 5, 1885 . . . 4,900

Kiver Spree on May 17, 1886 . , 171 „ ,

' .' . . 1,780

Wei'ssen Lake . . . . , , 735 „ „

Pond water ....... 14,400 „

Hail. — Buj wid and Foutin investigated the bacterial contents
of hailstones. In a very large stone 21,000 organisms per c.c.
were found, but in a smaller one, about the size of a walnut, only
729 bacteria per c.c. were counted.

Rain. — Very few bacteriological examinations of rain have
been made. Miquel, at Mountsouris, found the bacteria to
average 4'3 per c.c. during the years 1883-1886. In the middle
of Paris 19 bacteria per c.c. were found.

Rivers. — The Thames and Lea were examined by Frankland
for each month during the years 1886, 1887, and 1888. The
following table gives the number of bacteria found in Thames
water collected at Hampton.

Micro-organisms per c.c
Month.

January .

February .

March

April

May

June

July

August

September

October .

November

December

Similar results were obtained with water from the Lea. Both
rivers contained fewer bacteria during the summer months.
This is due to the fact that in the summer the rivers are chiefly
dependent on springs for supply, while in the winter they
receive surface washings from cultivated land.

Miquel examined the Seine, Marne, and Ourcq, and found the

1886.

1887.

1888.

45,000

30,800

92,000

15,800

6,700

40,000

11,415

30,900

66,000

12,250

52,100

13,000

4,800

2,100

1,900

8,300

2,200

3,500

3,000

2,500

1,070

6,100

7,200

3,000

8,400

16,700

1,740

8,600

6,700

1,130

56,000

81,000

11,700

63,000

19,000

10,600

28 BACTERIOLOGICAL EXAMINATION OF WATER.

least number of micro-organisms during the months of June,
July, and August.

Frankel and Dietrich examined the water at Marburg both
chemically and bacteriologically. The first specimen was taken
above the town, the second and third specimens from a place
where many drains existed, the fourth one kilometre below the
town, and the fifth 7'5 kilometres below the town. The results
are shown in the following table :

Specimen.

Total
solids.

Fixed
solids.

Volatile
solids.

NH3.

NO2.

01.

KMnO4.

Bacteria
per c.c.

(1)

126

107

20

nil

nil

7.1

5.3

1,200

(2)

126

105

20

trace

55

7.8

5.9

14,000

(3)

123

101

21

„

,,

8.5

5.3

4,320

(4)

127

106

21

,,

55

8.5

5.3

1,575

(5)

124:

103

21

„

,,

8.5

5.9

2,300

Rosenburg studied the Main above and below Wurzburg :

Above the town, Below the town,

bacteria per c.c. bacteria per c.c,

February 520 15,500

355 2,950

680 16,000

780 6,600

640 6,400

720 18,000

565 17,200

1,020 14.000

680 22,000

The River Ure was examined by Frankland ; a specimen taken
above Ripon yielded 1800 bacteria per c.c., but a second speci-
men taken below the town was found to contain 33,400 bacteria
per c.c. In a report to the Corporation of Aberdeen, Frank-
land stated that the River Dee above Braemar contained only
88 bacteria per c.c., but below Old Mar Castle, having received
sewage from Braemar, it contained 2829 bacteria per c.c.

Lakes. — Water from lakes usually contains fewer micro-
organisms than river water. Cramer examined fifty specimens
of water obtained from Lake Zurich during October and
December 1884, and January 1885. During this period the
average count was 184 organisms per c.c. From the 13th to
the 24th of June 1885, he examined forty-two specimens, and
obtained an average result of only 71 organisms per c.c.

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 29

Karlinski studied a lake in Herzegovinia and found near the shore
16,000 bacteria per c.c. At the centre of the lake, however,
3000 organisms per c.c. were found at the surface ; at a depth
of five metres 1000 ; at ten metres 600 ; and at fifteen metres
only about 200 a3robic organisms were detected.

Wells. — Great differences have been found in the number of
bacteria present in shallow wells. Egger examined a large
number of wells in Mainz with the following results :

No. of wells.

No. of bacteria.

No. of wells.

No. of bacteria.

1

0

3

0-5

4

5-10

10

10-20

16

20-100

9

100-200

7

200-300

2

300-400

1

400-500

4

500-1000

3

1000-2000

Maschek obtained the following results in Leitmeritz :

No. of wells.

No. of bacteria. No. of wells.

No. of bacteria.

1

'10-20

2

30-50

1

100-300

8

300-500

15

500-1000

21

1000-2000

B. Fischer's examinations in Kiel gave the following counts :

51 wells
34 ,

Up to 500
500-1,000

64 wells
22 ,

1,000-10,000
10,000-20,000

Deep wells as a rule contain few bacteria. Frankland found
in a Kent well, sunk in the chalk, between 6 and 26 organisms
per c.c. ; and in his report to the Local Government Board on
the monthly bacteriological examination of the London Water
Supply for the years 1886, 1887, 1888, he gave the results
shown in the following table :

Number of bacteria per c.c.

Wells.

Jan. Feb. Mar. Apr. May June July Aug. Sep. Oct. Nov. Dec.

1886.

New wells , —

II

12

3 —

1887.

Bath well

9

19

80

26

27

12

11

Garden well

. 48

20

4

4

—

24

IS

New well

. 12

10

5

12

20

14

8

1888.

Bath well

6

47

6

33

7

17

8

Garden well

5

19

8

4

27

71

5

New well

. 12

4

5

7

8

20

4

59 27 30

9

DO

3

5
65

4

18
19

11

6
12
67

34

30 BACTERIOLOGICAL EXAMINATION OF WATER.

Egger counted only four organisms per c.c. in the artesian
wells in Mainz, and Hueppe found only 4 bacteria per c.c. in
the deep wells in Wiesbaden. Heraeus, C. Frankel, and Maschek
made numerous studies of wells after they had been at rest for
some time and after pumping. Heraeus found 5000 organisms
in a well which had been at rest for 36 hours, but, after the well
had been emptied by continuous pumping, a sample collected
showed only 35 bacteria per c.c. Frankel and Maschek obtained
similar results.

Springs are closely allied to deep wells as regards their
bacterial contents. Cramer found in Zurich springs 9, 17, 17,
31, 36, 182, 15, 25 and 45 bacteria per c.c. Frommelt counted
in the water supply of the town of Altenburg, which was derived
from springs, only 28, 35, and 25 bacteria per c.c. Buchner
found the springs in a garden at Brunnthal to contain only
from 4 to 35 colonies per c.c.

A brief consideration of these results shows that, with the
exception of deep wells and springs, it is impossible to decide
from the quantitative examination alone whether a water supply
has been polluted or not. This is due to the fact that the
great variations in the results are attributable not only to
possible sewage contamination, but also to the influence of the
physical and chemical conditions of the supply, which operate
sometimes so as to increase the number of micro-organisms
present, and at other times lead to a rapid disappearance of
the microbes. It is also evident that the general standards
given by many observers cannot be used for all waters. Mace
certainly says that the number of micro-organisms present does
not give any accurate information as to the value of a water ;
at the same time he suggests the following classification, which
gives the mean results of a long series of examinations :

Very good waters contain from 0-50 rnicro-organisms per c c.

Good „ „ „ 50-500

Mediocre „ „ „ 500-3,000 „ „ „

Bad „ „ „ 3,000-10,000 „ „

Very bad „ „ „ 10,000-100,000

The only sound way to judge a supply is to examine each
source separately at different times of the year, especially
before and after rain, in summer and in winter ; in this way

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 31

local standards for comparison will be obtained which will have
great practical value. For general reference water-supplies
may be arranged in the following groups : (a) spring and deep
well waters ; (b) upland surface waters ; (c) shallow well waters ;
(d) river waters.

(a) The supply is obtained from main springs and deep wells.
These two sources are classed together because they both repre-
sent water which has percolated through a considerable depth
of ground ; they are really filtered supplies containing, as a rule,
very few bacteria. Tiemann and Gartner consider that water
derived from these sources should never contain more than 50
micro-organisms per c.c., and the analyses previously given
show that this is a reasonable limit ; as the water has percolated
through a considerable depth of ground it keeps a fairly
uniform temperature throughout the year, and is very little
influenced by climatic conditions. Consequently if deep wells
and springs are kept free from surface pollution, and the out-
crops of the water-bearing strata are above suspicion, there is
every reason to expect a constant bacterial composition of these
supplies.

(b) The supply is derived from an upland surface. The
water is usually collected in an impounding reservoir and may
be distributed without filtration. Such a supply must be
studied locally by determining the bacterial contents of un-
polluted feeders at different times of the year. If to this study
there be added an examination of the reservoir at its upper end
where the various feeders enter, and at the lower end where the
water is discharged after sedimentation has helped to reduce the
number of micro-organisms, a very good idea of the true bac-
terial contents of the supply will be obtained. Tables can then
be made for reference which will form an effective control of
the supply. It often happens, however, that the upland surface
grounds are obviously open to pollution from human habita-
tions, &c. In such a condition of things it will be of little use
to determine the average bacterial contents of the feeders,
because even daily examinations would not enable the supply to
be effectively safe-guarded. The water must then be subjected
to the artificial conditions of filtration.

(c) The water is derived from shallow wells. The analyses

32 BACTERIOLOGICAL EXAMINATION OF WATER.

already given show how impossible it is to fix any standard for
shallow wells. Unless properly covered, imperviously steined
to a sufficient depth and surrounded by at least an acre of
virgin ground, shallow wells must be considered dangerous
sources of supply. The number of micro-organisms in water
from such a source is also liable to extreme variation from
climatic conditions independently of pollution, so that practically
local standards are of little value.

(d) The supply is derived from a river. The bacterial con-
tents of rivers vary within very wide limits, depending upon
seasons of the year and proximity to towns, &c. As a rule,
they are dangerous sources of supply, for it appears that there
are few, if any, rivers which by natural processes can purify
themselves sufficiently to justify their consumption without
artificial treatment.

CHAPTER V.

QUANTITATIVE ANALYSIS— continued.

THE RELATION OF A QUANTITATIVE BACTERIOLOGICAL ANALYSIS
TO FILTRATION OF WATER THROUGH SAND.

THE purification of water by filtration through sand is chiefly a
biological process, the object being to remove micro-organisms
from the supply. Piefke was the first to show that sand alone
could not remove micro-organisms from water ; filters made of
sterilised sand were found to increase the number of bacteria in
the filtrate during the first few days of working, but, with the
formation of a gelatinous layer on the surface, the true filtering
action commenced. This gelatinous layer consists of zooglcea of
bacteria combined with suspended materials in the water ; it is
extremely friable and readily broken by excessive pressure on the
surface or disturbance of the body of the filter bed. The degree
of fineness and the uniformity of the sand grains are also of
importance in securing a good filtrate. By using fine sand the
current of water passing through the bed is rendered more
uniform, and the walls of the lacunar spaces are approximated,
permitting molecular action to take place, and giving greater
firmness to the gelatinous layer. Hence the work of a sand
filter is partly mechanical and partly vital. By the growth of
bacteria in the bed food material is used up, and products of
bacterial life, which have a powerful elFect in arresting the
growth of bacteria through the filter, are eliminated. It must
not be supposed that a sand filter will arrest all the micro-
organisms contained in a water applied to its surface. At one
time it was considered that a filter bed consisted of two super-
imposed systems, viz., an upper system which arrested all bac-
teria, and a lower system formed by the inferior layers of the
bed which contained a few bacteria clinging to the materials. It

34 BACTERIOLOGICAL EXAMINATION OF WATER.

was thought that the bacteria which appeared in the filtrate were
derived solely from the inferior system and had no relation to
the bacteria applied to the surface of the bed. FrankePs experi-
ments with B. violaceus and the Lawrence experiments with B.
prodigiosus showed conclusively that applied bacteria do find
their way into the filtrate. Friinkel poured water charged with
B. violaceus on to a ripe filter, and found colonies of the organ-
ism in the filtrate. In order to avoid the objection that the
B. violaceus might have gradually grown through the filter owing
to the nutriment supplied by the broth culture, Frankel made
his cultures in broth so dilute that their addition increased the
nutritive powers of the water only in a very infinitesimal degree.
The State Board of Health, Massachusetts, has carefully inves-
tigated the working of sand filters by means of experimental
beds at Lawrence. The Board employed the terms " bacterial
efficiency," " bacterial purification," and " hygienic efficiency " to
express the results of the action of sand filters. "Bacterial
efficiency " is the percentage of the number of bacteria in the
applied water which fails to appear in the effluent. The number
of bacteria in the effluent, however, includes those which have
their origin in the lower portions of the filter and its under-drains,
as well as those which have passed directly through the filter from
top to bottom. At the present time there are no well-defined
methods by which the two classes of bacteria can be readily
separated. " Bacterial purification " is the percentage removal by
filtration of the applied bacteria. The expression bears no relation
to those bacteria in the effluent which have their origin within the
filter or under-drains; consequently to gain an idea of the bacterial
purification produced by a filter it is necessary to employ a special
micro-organism. The Board employed the B. prodigiosus because
it appeared incapable of growth within the filters and their under-
drains and could be readily detected in the effluents. The following
table shows some of the results obtained at Massachusetts :

Percentage of bacteria which

Dentil of sind Form of Ratc pcr appeared in the nitrate.

/na- nitration. acre daily. , * N

B. Trodigiosus. Water bacteria.

5 feet continuous 7,660,000 0'171 1'48

„ „ 7,700,000 0-148 M9

12 inches ., 3,700,000 0'337 2'34

5 feet intermittent 3,600,000 0'463 3- 10

„ continuous 5,550,000 0'183 T04

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 35

The Board employed the term "hygienic efficiency" to
indicate the percentage removal by filtration of the applied
bacteria capable of producing disease. The hygienic efficiency
of the Lawrence filter beds, tested with the B. typhosus, was as
follows :

Material of
filter.

S.VU!1.

5-25 feet
5 feet

2 feet
1 foot

Loam.

2 inches

I inch

Size of

Filtration

Form of

sand.

per acre daily.

filtration.

0-48

980,000

Intermittent

„

1,000,000

0-26

460 000

Continuous

55

400,000

0-19

1,560,000

Intermittent

0-19

2,100,000

Continuous

0-19

1,540,000

'

0-19

1.500000

0-19

1,540,000

j

Percentage of B.

removed.

99-95

99-00

100-00

100-00

100-00

97-20

100-00

99-16

99-00

Bacterial efficiency appears to depend largely on the rate of
filtration. Low rates are safer than high rates ; but up to a
certain limit the rate does not exert much influence. At Altona
Kiimmel obtained the following; results :

Kate of filtration.

2 inches per hour (4 feet per day)
4 „ , (8 ., „ )

Bacteria per c.c. in the effluent.
• . . . 11-97
, . - 5-79

7-72

Kiimmel did not regard four inches per hour as beyond doubt
the maximum rate of safe filtration, but considered that the
danger of passing pathogenic organisms was more unlikely at the
lower than at the higher rates, and that the best velocity was
not the same for all waters. Differences in the mineral, vege-
table, and animal admixtures were of great importance in this
question.

The Massachusetts Board of Health believed that the factor
which caused the rate of filtration to become practically nil was
the age of the filter. Increasing the rate of filtration ten to
twenty per cent, might cause a temporary increase in the number of
bacteria if the filter were new and composed of coarse grains ; but
with ordinary ripe filters no increase in the bacteria was noticed.
Stopping the filtration for thirty-six hours, and then increasing
it, had no effect on old filters, but on new filters with coarse
grains it had a temporary effect. Lowering the rate and then

36 BACTERIOLOGICAL EXAMINATION OF WATER.

suddenly increasing it had no effect except on new coarse filters.
With gradually increasing age and consequent thickening of the
gelatinous layer in the filter, the rate of filtration, without a
dangerous head, became so low that cleaning of the surface
became inevitable. The depth of sand removed from the
surface, and the mode in which the filter was again put into
operation after cleaning, were found to be the principal factors
influencing bacterial efficiency.

In Germany the gelatinous layer on the surface appears to be
regarded as essential, and the main body of the filter is simply
looked upon as an under-drain and support for the surface film,
having a steadying influence on the process of filtration during
times of abnormal stress. The Lawrence experiments also
showed that the surface film removed more bacteria than any
other layer in the filter. Still, in many cases, filters were found
to remove bacteria in the absence of any surface coating. Also
0*1 to 0*3 inch could be scraped off a continuous filter without
affecting its action, and coarse filters did not give perfect results
when the coating was so thick as to almost completely clog the
filter.

From a practical point of view the influence of the removal of
clogging on bacterial efficiency depends on (a) the depth of the
material which is removed from the surface by scraping, and on
(&) the mechanical disturbance of the main body of the sand
which may be caused by the scraping and the process of refil-
ling. Until the scraping reaches about one inch the influence of
the removal is slight, but when more than one inch is removed
the change in the bacterial contents of the effluent becomes
marked. As the depth of material removed increases, not only
do the bacteria in the effluent increase in number, but also the
period of lessened efficiency of the filter becomes longer. The
most efficient method of putting filters into operation after
treatment to relieve clogging, has been carefully studied at
Lawrence. Filling from the top did not appear to affect small
experimental filters ; but on the large scale, in order to get the
upper layers of the filter dry enough to walk on, it was found
necessary to lower the water to a greater depth below the
surface, thereby furnishing an opportunity for mechanical
disturbance of the main body of the sand when its pores are

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 37

again filled with water. The best plan is to fill the bed from
below with filtered water to the top of the sand and then run on
raw unfiltered water, which is allowed to stand for ten to twelve
hours before filtration is started. Kiimmel found that when a
filter had new sand placed on it the numbers of bacteria in the
effluent were as follows :

Before cleaning . . . > . 42 bacteria per c.c.

One day after cleaning . . • . . 1,880 „

Two days after cleaning .... 752 .,

Three days after cleaning .... 208 .,

Four days after cleaning . . . .156 .,

Five days after cleaning .... 102 ,,

Six days after cleaning . . . . 84

He therefore insists on wasting the water for several days after
filling. The Massachusetts Board, however, considers that it is
best to have the fine sand in a filter of such a depth that it can
be cleaned and used for a year without fresh sand being put on
it, otherwise there is a tendency for clogging to occur at the
junction of the new and old sand. The Board found in their
experiments that it was not necessary to waste the water, pro-
vided there was no disturbance in the main body of the filter.
Piefke, however, considers that it is best to waste the water for
twenty-four hours after scraping.

As regards the influence of the depth of the filtering material
upon bacterial efficiency, it has been found that filters contain-
ing two feet, and even one foot, of sand yield satisfactory results
so long as there are no disturbing influences. But if these
occur, their effects on the efficiency of the filtration are more
marked arid of longer duration in shallow than in deep filters.
The depth of fine sand in a filter should never be reduced below
twelve inches. The size of the sand grains has a powerful
influence on bacterial efficiency ; this is shown by differences in
the length of time required by new filters to yield effluents of
normal bacterial contents. As the size of the sand grains
increases the effect of deep scraping and unequal rates of filtra-
tion become more marked even in filters which have been some
years in operation. Good results may be obtained with grains
ranging from 0'14 to 0'38 mm. ; but the size must be uniform

38 BACTERIOLOGICAL EXAMINATION OF WATER.

throughout the filter. Sand as coarse as 0*48 mm. requires a
longer period of operation to yield normal results.

The question as to whether a filter should be used continu-
ously or intermittently is considered by the Massachusetts
Board to depend on the relation of the dissolved oxygen to the
organic matter in the water. Some waters are saturated with
oxygen and contain only a small quantity of organic matter ;
these may be filtered continuously with satisfactory results.
Successful filtration by the ordinary continuous method is out of
the question when the water contains much organic matter and
very little or no free oxygen. Whether the beds are worked
continuously or intermittently it is acknowledged that an
effluent under satisfactory conditions will not contain more than
100 bacteria per c.c. At the Altona works the filtered water
usually showed counts under 30 per c.c. ; 50 to 70 per c.c. were
rarely found. The London water companies have not been so
successful, those companies which derive their water from the
Thames and Lea being hampered by insufficient storage. The
impurity of the water in the rivers is so largely increased during
periods of flood as to render it a matter of vital importance that
there shall be sufficient provision of subsidence reservoirs to
enable the necessary daily supplies being taken while keeping
the intakes closed. A certain amount of suspended organic
material is required to assist in forming the gelatinous film in
the filter bed, but an excess soon blocks up the bed and necessi-
tates frequent cleaning and consequent disturbance of the body
of the filter. It was at one time thought that double filtration
might improve river supplies ; but in the process of double
filtration the organic matter will be almost entirely arrested on
the first filter leaving only a small amount of material to form
the film in the second filter, and as a result the partially purified
water from the first bed passes through the second bed with only
a very slight improvement. The exact conditions which will
give the best results must be worked out more or less experi-
mentally for each supply. It is also of the first importance
that there shall be means of making a daily bacteriological
examination of the filtered water from each bed. Filtered water
containing more than 100 organisms per c.c. must not be passed
into the reservoir for pure water ; the beds must be so

QUANTITATIVE BACTERIOLOGICAL ANALYSIS. 39

constructed that imperfectly filtered water cannot mix with tht,-
pure water.

Reinsch and Kummel have devised an apparatus by which
samples of water for bacteriological examination can be taken
from filter-beds of old construction. The importance of being
able to examine each bed is well shown by the following results
of one of their investigations of filter-beds Nos. £ and 3 at the
Altona water-works ; No. 2 bed had been cleaned and a fresh
layer of sand 60 c.m. thick put on the bed :

-p t Bacteria per c.c. Bacteria per c.c.
ill water from lilter. in water from fllter.

No. 2.

No. 3.

June 26

. 520

10

.. 27 .

. 394

6

n 28 .

. 224

10

„ 29 .

96

10

, 30 .

56

6

July 1 .

. 34

4

When examining the filtered water at least three gelatine
plates should be made and counts taken after forty-eight hours
incubation at 20° C.

During the last two or three years the Massachusetts State
Board of Health has made experiments in order to determine
whether B. coli was present in the filtered water or not. "From
the studies of B. coli in filtered water it was thought probable
that a rate which would be entirely safe, judging only from the
determinations of the total number of bacteria present, might
really be too high to produce safe water, judging from the
B. coli determinations." The Board considers that the typhoid
bacillus cannot be present naturally in water without the
occurrence of large numbers of B. coli com munis, and regards
the absence of this bacillus from the filtered water as conclusive
evidence of the absence of the typhoid germ. But as
the Mem mack river, which receives the sewage from several
large cities, only contains fifty colon bacilli per cubic centimetre
at the intake of the Lawrence city filter, it is evident that the
typhoid bacilli contained in the river water must be very few in
number as compared with B. coli, even when the cities have
typhoid fever in an epidemic form. Consequently the volume
of water which must be declared free from B. coli in order to

40 BACTERIOLOGICAL EXAMINATION OF WATER.

establish the absence of B. typhosus becomes a very important
question. In view of the above considerations the work of the
Lawrence city filter is of interest. During the year 1899, 180
samples of the river water at the intake of the filter were
examined, and B. coli were found in all of them, the number
per cubic centimetre averaging 47 for the entire year. During
the same period 189 samples of the filtered water collected at
the pumping-station were also examined for B. coli, and it was
found in 1 c.c. of 45 of these samples. Twenty-seven of the
samples in which it was found, however, were collected during
the months of January and February. " Probably never more
than one colony per c.c. was present in any of these samples of
effluent from the city filter." It is evident from the above
results that seasonal variations of temperature affect the work of
sand filters in eliminating B. coli communis, the poorer work
being done during the colder months of the year. The experi-
ments of the State Board of Health are not yet completed ;
their additional studies gave somewhat contradictory results, so
the very important questions as to whether B. coli must be
completely absent from, or whether it may be allowed in, a cer-
tain volume of the filtered water in order to make sure of
the absence of the B. typhosus, must be considered still
unsettled.

CHAPTER VI.

THE QUALITATIVE BACTERIOLOGICAL ANALYSIS.

THE most important part of the bacteriological examination of
water supplies consists in the isolation and recognition of the
various kinds of bacteria present ; unfortunately, much confusion
exists as to the species of micro-organisms which normally
inhabit waters from various sources. Many organisms have been
described as different species which are really only varieties, the
variation being caused by differences in food, habitat, tempera-
ture, &c. Marshall Ward believes that many of the bacteria
isolated from impure waters behave as weakened forms, and
considers that the description of these enfeebled forms explains
many of the species found in the various works on water-
bacteria. Of late years, however, there has been a great increase
in the number of tests to which an organism must be submitted
in order to determine the species to which it belongs. Conse-
quently it is only quite lately that bacteria have been separated
into distinct species, characterised by Avell-defined biological and
chemical reactions. Following on the elaboration of these dif-
ferential tests, there has been a marked diminution in the
reports of the successful isolation of pathogenic organisms, such
as the B. typhosus, from water supplies supposed to have been
infected with the germs of specific disease.

Bacteriologists now realise that the isolation of such an
organism as the B. typhosus is a matter of considerable difficulty,
and that it can only be accomplished under exceptionally
favourable circumstances. Admitting this to be the case, and
considering from a hygienic point of view that it is highly
desirable, in the absence of the discovery of a specific pathogenic
organism, to be able to state that a suspected water has been
polluted with sewage, many bacteriologists have recently studied

42 BACTERIOLOGICAL EXAMINATION OF WATER.

the flora of sewage with great care. As a result of their work a
number of organisms have been described which appear to be so
characteristic of sewage as to justify the term sewage-bacteria in
contradistinction to water-bacteria. In the present state of
our knowledge, therefore, it appears most advantageous, from a
hygienic point of view, to arrange the micro-organisms found in
water in the following three classes :

Class I. — Micro-organisms which are usually found in pure

waters — viz., the so-called water-bacteria.
Class II. — Micro-organisms which are common in sewage, but

are rarely found in pure waters.
Class III. — Micro-organisms which are the cause of specific disease

in human beings, and which have been isolated from water

supplies.

CLASS I.

MICRO-ORGANISMS WHICH ARE FOUND IN
PURE WATERS.

A very large number of water-bacteria have been isolated, but
the biological characteristics of many of them are so vague that
it will serve no useful purpose to attempt to describe them.
For hygienic work it seems better to try and arrange the water-
bacteria into groups of certain well-defined types, and then to
mention the variations from the type which have been
described.

GROUP I.

In this group are included the microbes which produce a
green colour in nutrient media ; they may be again sub-divided
into those which liquefy and those which do not liquefy gelatine.

BACILLI WHICH LIQUEFY' GELATINE AND PRODUCE A GREEN
FLUORESCENCE IN NUTRIENT MEDIA.

B. Fluorescens Liquefaciens.

This microbe is exceedingly common in water from all sources,
and has the following characteristics :

Colonies on Gelatine Plates. — After twenty-four hours at 20° C.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 43

fine films appear on the surface with granular contents ; after
forty-eight hours the films become converted into shallow cups
surrounded by a green fluorescence. Under a low power the
colonies appear to have a margin set with very fine rays, and in
the centre a membrane may be seen ; later the membrane
disappears, and as the liquefaction extends the fine rays are
replaced by a hazy granular appearance. The colonies in the
depth of the gelatine are circular or oval in shape, and by
transmitted light appear to have a greenish-brown tinge.

Stab-gelatine. — The gelatine is liquefied in the form of a
funnel, and the medium acquires a green colour.

Agar-slope. — A smooth white growth, the agar assumes a
green colour.

Broth. — There is a diffuse growth, a slight pellicle forms on
the surface and a thick deposit appears at the bottom of the
tube ; the broth acquires a green colour.

Potato. — A yellowish-brown growth.

Glucose-gelatine. — No gas formation.

Nitrate-broth. — After seven days incubation at 22° C., large
quantities of ammonia are produced.

Milk. — A clot is formed which is slowly digested, the whey
assumes a green colour.

Peptone-water. — No indol reaction is obtained.

Litmus-whey. — There is no formation of acid. The blue
colour is changed to green ; after seven days incubation the
medium has a faintly alkaline reaction.

Microscopical Appearance. — A very short bacillus, generally
found in pairs. It may form chains. No spores observed. It
does not stain with Gram's method, and is very motile.

The above description gives the characteristics of the type of
this organism which is most commonly found in water. Varieties,
however, are frequently found which vary considerably from the
type; some liquefy the gelatine in forty-eight hours, others do not
produce this result until the expiration of fourteen days. The
optimum temperature is usually 20°-25° C. ; many varieties will
not grow at 37° C. ; others grow well at the latter temperature
and produce a copious growth on agar. The oxygen require-
ments vary greatly. Sometimes liquefaction of gelatine takes
place rapidly in the form of a deep funnel ; at other times it is

44 BACTERIOLOGICAL EXAMINATION OF WATER.

horizontal and limited to the surface layers. The colour of the
pigment produced in the various media may be bright green or
only greenish-yellow. The growth on potato usually shows a
yellowish-brown colour, but at times it may have a rose tint or
deep brown colour. Also some varieties which I have isolated do
not coagulate or digest milk. The B. cloacae fluorescens, the B.
fluorescens stercoralis (Laws and Andrewes), the B. fluorescens
Schuylkilliensis, B. fluorescens mutabilis (Wright), B. butyri
fluorescens (Lafar), B. fluorescens nivalis (Schmolck), and B. vis-
cosus (Frankland) are merely varieties of the B. fluorescens lique-
faciens. The B. viridans, described by Zimmermann, is a small
non-motile bacillus, which produces a steel-green colour in media,
but no fluorescence is observed. It grows well at 30° C., is aerobic,
and does not stain with Gram. It rapidly liquefies gelatine, but
in gelatine-stab the liquefaction only occurs horizontally at the
surface ; there is no growth along the line of inoculation.

BACILLI WHICH PRODUCE A GREEN FLUORESCENCE IN NUTRIENT
MEDIA, BUT DO NOT LIQUEFY GELATINE.

These bacilli are also extremely common in water, and, when
the green .colour of the media is slowly produced, there may be
difficulties in diagnosing this group from other species.

B. Fluorescens Non-Liquefaciens.

Colonies on Gelatine Plates. — After forty-eight hours at 22° C.,
the surface colonies are thin, with an irregular margin, and
when held up to the light have a bluish, almost transparent
appearance ; the surrounding gelatine shows a green fluorescence.
Under a low power the colonies show a beautiful venation
resembling that of a leaf. The colonies in the depth of the
gelatine are round, and by transmitted light appear greenish-
brown in colour.

Gelatine-stab. — There is a growth on the surface resembling
a colony, and a thin white growth along the line of inoculation ;
the gelatine is not liquefied and acquires a bright green colour.

Agar-slope. — A white growth ; the medium assumes a green
colour.

Broth. — A diffused growth ; a pellicle appears on the surface
in forty-eight hours. The broth assumes a green colour.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 45

Potato. — A greenish-yellow growth.

Glucose-gelatine. — No gas formation.

Nitrate-broth. — No reduction of nitrates.

Milk. — Unchanged.

Peptone-water. — No indol reaction.

Litmus-whey. — No formation of acid takes place. The
medium often becomes slightly alkaline.

Microscopical Appearance. — Short thin bacilli with rounded
ends ; as a rule not motile. It does not form spores. Optimum
temperature 21°-25° C. ; it does not develop at 37° C. It does
not stain by Gram's method. The type above described cor-
responds to the B. fluorescens non-liquefaciens of Eisenberg,
B. fluorescens albus of Zimmermann arid B. aquatilis fluorescens
of Lustig. The Bacillus viridis pallescens isolated by Frick
form green sputum, and found by Tils in the Freiburg pipe-
water, differs from the type only in that the bacillus is highly
motile, and the growth on potato has a nut-brown colour.

B. Fluorescens Longus.

This variety of the B. fluorescens non-liquefaciens was isolated
by Zimmermann from the Chemnitz water-supply. Its charac-
teristics are as follows :

Colonies on Gelatine Plates. — The young colonies in the depth
are white dots with a greenish tinge. On the surface at first
they are thin, almost round, and have a mother-of-pearl
shimmer. They grow rapidly, have a greenish-yellow colour,
and look as if yellowish-white threads had been drawn through
them. Under a low power, the deep colonies are yellow, sharply
outlined with a convoluted interior ; the surface colonies show
a similar appearance, but the convolutions are broader and
resemble the convolutions of the intestine in a small animal.

Gelatine-stab. — Forms a thin expansion at first, but later
becomes thicker in the centre and soon reaches the margin of
the gelatine. At first the colour is blue, but later blue-green.

Agar. — A thin growth, and the agar becomes greenish-
yellow in colour.

Potato. — Forms a yellowish moist growth which rapidly
spreads over nearly the whole surface.

Gas-production. — Not observed .

46 BACTERIOLOGICAL EXAMINATION OF WATER.

Microscopical Appearance. — Short and long bacilli observed ;
the short forms are motile but the long ones are stationary.
Unstained spots appear in the rod, but they are probably not
spores. It grows best at room-temperature, and produces a
greyish-yellow pigment. A culture of this organism from
Krai's laboratory produced colonies closely resembling the
B. fluorescens non-liquefaciens. Its cultural reactions in the
various media also were identical with those of the latter
organism. The colour of the media, however, was more
yellowish in the case of the longus. The bacilli stained with
Gram.

B. Fluorescens Tennis.

This micro-organism was isolated by Zimmermann from
the Chemnitz water supply. It has the following biological
characteristics.

Colonies on Gelatine Plates. — The surface colonies are thin,
irregularly round with a denticulated edge ; the surrounding
gelatine acquires a green colour, and is not liquefied.

Gelatine-stab. — A thin greyish-white expansion appears on
the surface, which in about four days reaches the margin of
the tube. The stab is clearly defined, but no further growth
takes place.

Gelatine-streak. — After three days a leaf-like expansion is
produced.

Agar-slope. — A greyish growth is produced; the agar assumes
a green colour.

Potato. — A thin, greyish-yellow, smooth layer is produced,
which only after some time grows out from the line of inocu-
lation. Later, the colour of the growth becomes reddish-
brown.

Gas-production. — Not observed.

Microscopical Appearance. — Short thick bacilli with rounded
ends, often arranged in groups ; in old cultures the rod consists
of five or six links. It has only rotatory movement and is de-
colorised by Gramas method.

A culture from Krai's laboratory produced very characteristic
colonies, which had the denticulated edge well marked. The
appearance of a surface colony strongly resembled a small shell.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 47

The other cultural reactions corresponded to those already
given under the B. fluorescens non-liquefaciens. The colour
produced in gelatine, broth and agar was bright green.

B. Fluorescens Aureus.

This bacillus was also isolated by Zimmermann from the
Chemnitz water supply. It has the following cultural charac-
teristics.

Colonies in Gelatine Plates. — The colonies in the depth appear
as small yellowish-white dots; the surface colonies are yellowish-
grey, smooth, not sharply outlined expansions ; they are darker
in the centre and delicately streaked at the margin. The
gelatine is not liquefied.

Gelatine-stab. — A. thin yellowish layer, which later becomes
thicker in the centre, appears on the surface, and in eight to ten
days reaches the margin of the gelatine. In the stab there is
very little growth, which later becomes brownish in colour.

Gelatine-streak. — It produces a thick ochre-yellow growth,
under which the gelatine acquires a marked brown colour. In
transmitted light the gelatine appears greenish-yellow.

Agar-slope. — A rich ochre or golden-yellow growth ; the
agar becomes darker but does not fluoresce.

Potato. — The growth is the same as on agar.

Gas-production. — Not observed.

Microscopical Appearance. — A short bacillus with rounded
ends, about twice as long as broad, found in pairs, rarely in
greater numbers. It is very motile. It produces a yellow-ochre
pigment, and stains faintly with Gram.

A culture of this organism from Krai's laboratory produced
colonies exactly like those described above. On potato and
agar the growth had a yellow-ochre colour. The growth
in gelatine was yellowish, and the gelatine acquired a green
fluorescence. The microbe did not coagulate milk, produce
indol, nor form gas in glucose-gelatine.

B. Fluorescens Crassus.

This microbe is described byFliigge in "Die Mikroorganismen."
The colonies are very granular in the depth of the gelatine ;
on the surface they are round, slightly opaque, and have a

4S BACTERIOLOGICAL EXAMINATION OF WATER.

nail-head appearance. The bacillus is not motile. It is said to be
identical with the B. iris of Frick, and to be often found in air
and water.

I have frequently isolated this organism from water. The
nail- headed colonies when fished and planted out in gelatine
produce a marked green fluorescence of the medium ; the growth
on the surface sometimes remains thick, at other times it pro-
duces a thin layer like the stab of B. fluorescens non-liquefaciens.
The organism noes not coagulate milk, produce indol, nor form
gas in glucose-gelatine. The growth on potato is greenish-
yellow. It does not reduce nitrates appreciably. In nearly all
its cultural reactions it resembles B. fluorescens non-liquefaciens,
and according to Fliigge it is only a variety of this organism.

The bacilli above described form the most important varieties
of the B. fluorescens non-liquefaciens which are found in pure
water. They are important because in the early days of bac-
teriological study they were sometimes mistaken for the B.
typhosus. When the green fluorescence is not present, which
happens if the organism is enfeebled, the colonies have a super-
ficial resemblance to those of B. typhosus, and as the bacilli do
not coagulate milk, produce indol, nor form gas in glucose
media, and some varieties are decolorised by Gram's method of
staining, it is easy to understand that they might be mistaken
for typhoid bacilli if the further tests given under B. typhosus
were not carried out.

The B. fluorescens putridus described by Fliigge is also
closely allied to the B. fluorescens non-liquefaciens, but as it is
considered by many authorities (Mace, &c.) to be characteristic
of foul waters, it has been included under the sewage bacteria.

The B. erythrosporous and the B. pyocyaneus also produce
pigment which gives rise to a greenish fluorescence in media ;
these organisms are rarely, if ever, found in pure waters^ and
have therefore been included in the sewage group.

GROUP II.

This group includes the varieties of the B. aquatilis sulcatus
described by Weichselbaum. These bacilli are frequently found
in water derived from wells, springs, and rivers. The cultural
characteristics of the most common variety are as follows :

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 49

B. Aquatilis Sulcatus.

Colonies on Gelatine Plates. — After forty-eight hours incuba-
tion at 22° C. the colonies appear on the surface as bluish-white
films with an irregular margin ; the centre is thicker than the
periphery. Under a low power numerous fine lines are seen
passing from the centre to the periphery, producing an appear-
ance closely resembling the ridges and valleys on the colonies of
B. typhosus. Later, the centre of the colony acquires a yellow
colour. The gelatine is not liquefied.

Gelatine-stab. — The surface growth exactly resembles a colony ;
there is a scanty growth along the stab.

Agar-slope. — A white smooth growth ; later, becomes yellow
in colour.

Glucose-gelatine (shake). — No gas formation.

Peptone-water. — No indol reaction obtained.

Milk. — Unchanged.

Potato. — A thin yellow film appears.

Nitrate-broth. — No reduction of nitrates.

Broth. — Diffuse growth, no pellicle appears on the surface.

Microscopical Appearance. — A small motile bacillus with
rounded ends, closely resembling B. typhosus. It does not form
spores and is decolorised by Gram. It will grow at 5° C., but
develops very feebly at 37° C.

The above description corresponds with B. Aquatilis sulcatus V.
Four other varieties were described by Weichselbaum. No. I.
differs from the type in that at 37° C. it produces a thick
white expansion on agar, and on potato it produces at first only
a moist appearance, which later becomes cream-coloured. No. II.
shows colonies which are thicker than the above, and the system
of lines is hardly visible ; the growth on potato has a yellow-blue
colour which later becomes yellowish-grey or yellowish-brown,
and is often very luxuriant. No. III. produces colonies re-
sembling the type ; but the growth on potato has a bright yellow-
colour. Bacillus No. V. shows typical colonies, but does not
produce any growth on potato.

The B. sulcatus is an important organism to recognise, owing
to its strong resemblance to the typhoid bacillus ; it is, however,
easily differentiated by the colonies acquiring a yellow colour

50 BACTERIOLOGICAL EXAMINATION OF WATER.

its quicker growth on gelatine plates, its power of developing at
5° C., and the absence of all traces of agglutination with anti-
typhoid serum.

Two other organisms have been described which have received
the name sulcatus ; they are easily differentiated from the bacilli
isolated by Weichselbaum.

B. Sulcatus Liquefaciens.

This organism is described by Fliigge, and said to be frequently
found in water. It is a medium-sized motile bacillus which does
not form spores. The colonies in the depth are small, round,
slightly granular and yellowish in colour. The surface colonies
are larger, thin and transparent, and have an irregular margin ;
the surface of each colony also shows a fine system of ridges and
furrows like the typhoid bacillus. Liquefaction slowly sets in
and the colonies sink in the gelatine. On agar there is a thin
grey transparent growth. On potato a yellowish- brown layer
appears.

B. Aquatilis Sulcatus.

This bacillus is often found in water supplies and in many
respects resembles the organisms in Group III. It is a rather
large bacillus, which forms spores, and has a slow waddling move-
ment like B. subtilis. It does not stain with Gram. The
colonies in gelatine plates closely resemble those of B. subtilis,
but the liquefaction is not so rapid. On agar it produces a dry
wrinkled growth. In broth there is a diffused growth, and later
a wrinkled film forms on the surface. Milk is coagulated and
slowly digested. On potato there is a white, dry, slightly folded
layer. In glucose media there is no gas formation. Indol is not
produced in peptone and salt solution, and the reaction of
litmus- whey remains unchanged.

GROUP III.

This group includes the three varieties of the "potato
bacillus," viz. : the B. mesentericus vulgatus, B. mesentericus
fuscus, and the B. mesentericus ruber. Fliigge includes in the
group of the " hay bacilli " the above organisms and also the
following : B. subtilis, B. liodermus, B. mycoides, B. ramosus,

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 51

B. megaterium. They have many points in common, and
varieties which seem intermediate between the various members
of the group are constantly isolated from water supplies. It
therefore appears best to place all these bacilli in one large
group.

B. Mesentericus Vulgatus.

This micro-organism has the following cultural characteristics :

Colonies on Gelatine Plates. — At the end of twenty-four hours
the colonies develop on the surface as small opaque white spots,
which at the end of forty-eight to seventy-two hours appear
sunk in small cups of liquefied gelatine. Examined at this stage
under a low power an opaque membrane is seen in the centre
surrounded by a clear circular zone ; the periphery of the colony,
outside the clear zone, is set with fine rays running out into the
gelatine. Sometimes this peripheral zone of hairs is not seen.

Stab-gelatine. — The gelatine is liquefied in the form of a
funnel.

Glucose-gelatine (shake). — No gas formation.

Agar-slope. — A thin, white, dry wrinkled growth.

Potato. — A thick white growth which becomes wrinkled and
furrowed.

Peptone-water. — No indol reaction obtained.

Broth. — There is at first a diffuse growth ; later, a wrinkled
pellicle appears on the surface and the broth clears below it.

Nitrate-broth. — Nitrates are reduced to nitrites.

Milk. — At 37° C. milk is coagulated ; the casein is digested
after several days incubation.

Litmus-whey. — After seven days incubation, the blue colour
disappears and the medium is rendered faintly alkaline.

Microscopical Appearance. — A small motile bacillus with
rounded ends. It forms short ellipsoidal spores. This bacillus
is smaller than the hay bacillus, but often has the same waggling
movement. It is stained by Grain's method.

B. Mesentericus Fuscus.

This bacillus resembles the vulgatus in most of its cultures.
The growths on agar and potato, however, are thinner and have

52 BACTERIOLOGICAL EXAMINATION OF WATER.

a brown colour. It reduces nitrates more vigorously, and is
smaller and more motile than the vulgatus.

B. Mesentericus Rufoer.

This bacillus was described by Globig. The colonies on the
surface and in the depth show quite early numerous irregular
processes running in all directions from the periphery ; later, on
the surface colony a clear zone appears between the membranous
centre and the peripheral rays, and the colony in this stage
closely resembles the above varieties. The agar growth in the
early stage shows processes running out from the margin ; later,
a dry wrinkled film is produced. On potato at 37° C. a wrinkled
growth appears and the potato acquires a red colour. The red
colour is not formed at 22° C. In other media the cultures
resemble those of the vulgatus. The bacillus is thinner and
more motile than the vulgatus, and possesses a powerful reducing
action on nitrates, large quantities of ammonia being produced
after seven days incubation.

B. Liodermus.

Fliigge described this organism, which is closely related to the
potato bacillus. On potato it produces a gummy layer, which
later is thrown into folds. The gummy material is soluble in
water and precipitated by alcohol like gum arabic. Milk is
coagulated and peptonised.

B. Subtilis.

This bacillus was first isolated by Ehrenberg from an infusion
of hay, hence it is often called the hay bacillus.

Colonies on Gelatine Plates. — The colonies at first are small
white points, which under a low power appear as irregularly
round, yellowish-brown membranes, from which fine hair-like
processes pass off into the gelatine. Later, the colonies appear
as small cups of liquefied gelatine, with little or no central
membrane ; from the margin of the colony hair-like processes
are still seen passing into the gelatine. In the depth the colonies
often appear as granular centres from which many long and
curved processes spread out into the gelatine ; sometimes very

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 58

irregular forms are seen in which the bacilli appear packed in
parallel rows.

Gelatine- stab. — The gelatine is liquefied in the form of a
funnel ; a flocculent deposit appears at the bottom and a tough
pellicle on the surface.

Agar-slope. — A thick white growth appears, which later
becomes wrinkled.

Glucose-gelatine. — No gas formation.

Nitrate-broth. — No reduction of nitrates.

Broth. — A pellicle forms on the surface, which later becomes
wrinkled and furrowed.

Potato. — A thick white growth, which is often thrown into
large transverse folds ; later, dry white granules may appear
on the surface.

Peptone-water. — No indol reaction is obtained.

Litmus-whey. — After seven days incubation the blue colour
disappears and the medium is rendered faintly alkaline.

Milk. — At 37° C. milk is peptonised and partially digested.

Microscopical Appearance. — A large bacillus with rounded
ends, often in pairs and chains. It has a slow waggling movement,
and forms oval spores. It is strongly aerobic, and stains with
Gram. Sewage varieties of B. subtilis and B. mesentericus have
been described by Houston.

B. Mycoides.

This bacillus is said to be identical with the B. ramosus in all
its cultures. It is found in sewage and also in surface waters.

Colonies on Gelatine Plates. — The colonies appear as white
patches, from which numerous irregular processes extend on all
sides into the gelatine, producing an almost mould-like growth.
The gelatine becomes liquefied after about seventy-two hours
incubation.

Gelatine-stab. — A white growth occurs along the stab, from
which processes extend horizontally into the gelatine, producing
a tree-like growth ; later, liq uefaction occurs.

Agar. — A white, rather dry growth, from which mould-like
processes extend at the margin.

Glucose-gelatine (shake). — No gas formation.

Peptone-water. — No indol reaction.

54 BACTERIOLOGICAL EXAMINATION OF WATER.

Milk. — Unchanged .

Broth. — Diffuse growth ; a deposit forms at the bottom, and
later a pellicle appears on the surface.

Nitrate-broth. — Powerfully reduced to nitrites and ammonia.

Potato. — A moist, white growth.

Litmus-whey. — The reaction of the medium remains un-
changed.

Microscopical Appearance. — A large bacillus, with rounded
ends; forms oval spores. It is motile; movement sluggish. It
is stained by Gram's method.

The B. megaterium (De Bary) is described with the sewage
organisms.

The B. filiformis described by Tils also belongs to this group.
A culture of this organism produced colonies on gelatine exactly
resembling moulds. In stab-gelatine a beautiful tree-like growth
was obtained. On agar there was a thick growth with fibres
passing out from the margin. The growths in peptone- water,
broth and glucose-gelatine exactly resembled those of B.
mycoides. It formed long threads, and divided into segments,
each containing a spore.

GROUP IV.

This group contains a large number of bacilli which are very
common in unfiltered water supplies. It includes the B. lique-
faciens of Eisenberg, the B. liquefaciens of Lustig, the B. liquidus
of the Franklands, the B. punctatus of Zimmermann, the B. aqua-
tilis communis of Flligge, B. liquefaciens communis of Stern-
berg, £c. The B. liquefaciens which occurs in unpolluted
waters has the following characteristics :

Colonies on Gelatine Plates. — The surface colonies appear in
twenty-four hours as small white points, which under a low
power have a regular margin and granular contents ; often there
is an appearance of convoluted threads passing from the centre
to the periphery. In forty-eight hours the colonies appear as
cup-shaped depressions, which rapidly extend; under a low
power the margin is now seen to be smooth, and each cup
contains flocculent masses made up of clumps of bacteria ; later
the margin of the colony often shows indistinctly marked
spinous processes.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 55

Stab-gelatine. — The gelatine is rapidly liquefied in the form of
a funnel.

Glucose-gelatine. — There is no gas formation.

Milk. — Remains unchanged.

Broth. — There is a diffuse growth, and an abundant deposit at
the bottom of the tube.

Nitrate-broth. — Nitrates are reduced to nitrites and ammonia.

Potato. — The growth on potato varies ; sometimes it has a
light yellow colour, at other times it has a flesh-coloured tint
changing to reddish-brown.

Peptone-water. — No indol reaction is obtained.

Agar -slope. — A white expansion.

Microscopical Appearance. — A short motile bacillus, often in
pairs. It does not form spores. It does not grow well at 37° C.

The B. liquefaciens of Eisenberg corresponds to the above
type, but does not grow at 37° C. The B. liquidus, described by
the Franklands, forms large cup-shaped depressions, and pro-
duces a thick, moist, flesh-coloured growth on potato.

The B. punctatus of Zimmermann forms cup-shaped liquefying
colonies ; in the bluish-grey liquid whitish dotted groups of
bacteria occur, which frequently appear joined to one another by
white streaks. In stab-gelatine the gelatine is liquefied in the
form of a stocking. On potato there is a brownish flesh-coloured
expansion. It grows better at 30° C. than at the ordinary
temperature.

The B. aquatilis communis is said by Fliigge to be one of the
commonest organisms in river- water and well-water, and probably
identical with the B. punctatus and B. liquidus. It is a plump,
medium sized very motile bacillus. It does not form spores and
it is not stained by Gram's method. It liquefies gelatine very
rapidly, and in forty-eight hours a stocking-shaped liquefaction
is produced. The colonies in the depth are round plates, and
on the surface form round cups which are filled with masses of
bacteria without any special arrangement ; the margin is finely
granular but not rayed. On agar at 37° C. it produces a trans-
parent grey expansion. On potato it produces a yellowish-
brown or reddish growth. The B. liquefaciens communis of
Sternberg is found in faeces ; it resembles the B. aquatilis com-
munis both morphologically and in cultures.

56 BACTERIOLOGICAL EXAMINATION OF WATER.

The B. aquatilis radiatus (B. radiatus aquatilis of Zimmer-
mann) is also very common in water. It liquefies a gelatine-stab
in the form of a stocking. The colonies have a rayed margin.
The growth on agar is grey and transparent and on potato there
is a yellowish-brown layer. Gas producing varieties have been
observed. The bacillus is morphologically like the above, but
somewhat longer forms are seen ; it is not motile.

Closely allied to this group is an organism which I have
described as the B. liquefaciens sewage variety. It differs from
the type in that it coagulates milk and produces intense gas
formation in glucose-gelatine shake. I have found this organism
very frequently in sewage and polluted waters ; it is apparently
identical with the sewage proteus of Houston, which is fully
described under the sewage organisms.

Between this sewage variety and the type there are many
connecting links formed by bacteria which resemble the type in
not coagulating milk, but approach the sewage organism by
giving rise to intense gas formation in glucose-gelatine. The
Bacillus gasoformans, described by Eisenberg, appears to belong
to this group and to be identical with the intermediate forms
just mentioned. It is a, small, very motile bacillus, which does
not form spores. In gelatine plates it produces cup-shaped
liquefaction, and in gelatine-stab the medium is rapidly
liquefied in the form of a stocking. In glucose-gelatine it
produces marked gas formation. It does not grow at 37° C.

The members of this group vary extremely in their reaction
to temperature ; some of the varieties will not grow at 37° C. ;
others grow better at blood-heat than room temperature ; but
even when the cultural characteristics are identical the relation
to temperature may not be the same; this statement also applies
to the sewage varieties.

The B. hyalinus, B. cloacae, and B. delicatulus, isolated by
Jordan from Lawrence sewage, probably belong to this group
and appear to be varieties of B. liquefaciens (sewage proteus,
Houston) sewage variety. The characteristics of these organ-
isms as described by Jordan will be found in the section devoted
to sewage organisms.

The B. devorans, isolated by Zimmermann from a much-used
well-water, produces colonies which are characterised by forming

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 57

conical depressions in the gelatine, containing an irregular
white mass at the bottom. LTnder a low power the margin of
the colony shows threads of varying length passing off from the
periphery. In stab-gelatine the gelatine becomes cleft in the
form of an air-bubble, without any liquefaction ; sometimes, how-
ever, true liquefaction occurs. There is no growth on potato.
On agar there is a uniform grey expansion. It is a small, very
motile bacillus, which usually occurs singly or in pairs. It does
not form spores. It is not stained by Gram's method and grows
at 37° C.

GROUP V.

This group comprises a number of organisms which produce
a red pigment. The type of the group is the B. prodigiosus
of Ehrenberg, which is often found in waters from various
sources.

B. Prodigiosus.

The organism has the following characteristics :

Colonies on Gelatine Plates. — The colonies in the depth of the
gelatine are round or oval, sharply outlined, and of a clear red
colour. The surface colonies, after twenty-four hours, are seen
as round granular films, which at the end of forty-eight hours
appear as circular depressions with a red deposit in the centre.
Under a low power the colonies show a red granular centre,
surrounded by a circular clear zone and a narrow granular edge
at the periphery.

Gelatine-stab. — The gelatine is liquefied in the form of a
stocking and with a red coloured deposit at the bottom.

A gar-slope. — At first there is a colourless growth which
gradually acquires a red colour.

Agdr-stab. — In the line of inoculation the growth is colour-
less, but on the surface it has a red colour; the absence of colour
in the stab is due to the deficiency of oxygen.

Potato. — It forms a brilliant red growth.

Milk. — Red pigment is produced, and the milk is coagulated
through the simultaneous production of acid and lab-ferment.

Broth. — There is a diffuse growth with a red deposit.

Glucose-gelatine. — No gas formation.

58 BACTERIOLOGICAL EXAMINATION OF WATER.

Peptone-water. — No indol reaction is obtained.

Litmus-whey. — After seven days at 22° C. the medium has a
markedly acid reaction.

Microscopical Appearance.— A. small non-motile bacillus, often
described as a micrococcus. In broth containing an acid or
antiseptics rods and chains are well marked ; these bacilli are
also motile. Scheurlen has described flagella at the sides of the
bacillus. It does not form spores, but resists drying for a long
time, when involution forms are seen. The pigment is best
produced at 20°-24° C. When cultured continuously at 37° C.,
forms are obtained which cannot produce the red pigment ; old
cultures at room temperature also lose the power of producing
pigment, which, however, can be restored by cultivating the
organism on potato. It is not stained by Gram's method.

B. Ruber Indicus.

Described by Koch and also found by Pasquale at Massowah.
It is a motile, small, thin bacillus, which does not form spores.
The colonies in the depth are golden-yellow with sharply-defined
edges ; the superficial colonies quickly liquefy the gelatine.
In gelatine-stab there is a stocking-shaped liquefaction ; on the
surface there is a brick-red membrane and a white deposit at
the bottom. On agar and potato at 35° C. it produces a brick-
red growth.

B. Ruber Balticus.

This organism was found by Breunig in the Kiel water. It
is a slender motile bacillus, which does not form spores. The
colonies in the depth are round and clear yellow in colour ; the
surface colonies are thin layers, which gradually liquefy and have
a rose-red colour. In stab-gelatine there is liquefaction at the
surface and gas development in the depth. It grows and pro-
duces colour at 35° C. ; at this temperature the growth on
potato is purple or carmine red ; at 20° C. it is orange red ; the
deeper layers are always reddish-violet. Milk at 20° C. is
coloured and eventually coagulated through acid production ;
at 37° C. it is quickly coagulated but there is no coloration.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 59

B. Ruber Berolinensis.

The red water bacillus of C. Friinkel is probably a modifica-
tion of the above, which it closely resembles in morphology and
cultures. The colour in gelatine is yellowish-red, on agar
yellow, and on potato rust-red or orange-yellow.

B. Ruber Aquatilis.

The red water bacillus of Lustig is a small motile bacillus,
two or three times as long as broad. It does not form spores,
but it is not killed by heating to 60° C. for twenty-four hours.
In the body of the rods fuchsia-red pigment granules are seen. It
forms colonies with a serrated margin ; the centre of each colony
has a raspberry-red colour. The gelatine is quickly liquefied and
the red colour extends in all directions. In gelatine-stab there is
a funnel-shaped liquefaction at the surface, in the centre of which
pigment is found. In the depth of the stab there is a growth
but no pigment formation. On agar, potato, and in broth
there is a growth and formation of a raspberry-red colour at
ordinary temperatures, but at 37° C. there is no pigment
formation.

B. Carneus.

This bacillus was isolated by Tils from the Freiburg water
supply. It is a slender, very motile bacillus and does not form
spores. On gelatine it forms white colonies which liquefy the
gelatine producing cup-shaped depressions. In gelatine-stab it
causes a funnel-shaped liquefaction, the lower part of which
becomes filled with pink-coloured masses. On potato it grows
slowly producing a dark flesh-coloured expansion. The same
organism is described by Zimmermann as the B. carnosus.

B. Lactis Erythrogenes.

Found by Hueppe in red-coloured milk and probably identical
with the B. rosa fluorescens, isolated by Tataroff from the
Dorpat water. It is a small bacillus which is not motile, and
does not form spores. Its colonies are round, greyish-yellow in
colour, and slowly liquefy the gelatine which round the colonies
acquires a reddish colour. In gelatine-stab it grows slowly, and

60 BACTERIOLOGICAL EXAMINATION OF WATER.

liquefies the gelatine ; the upper part in the dark has a red
colour, but in the light is colourless. On agar and potato it
produces a yellowish layer, and around this there is a slightly
yellowish-red colour. Milk is coagulated and peptonised by
lab-ferment ; it is at first reddish-brown in colour and finally
blood-red.

B. Rubefaciens.

Found by Zimmermann in water. It is a slender very motile
bacillus. The colonies in the depth are small, round, and have
a yellowish or brown colour ; on the surface they form flat
expansions which do not liquefy the gelatine, and have a sugges-
tion of a red colour. In gelatine-stab it forms a greyish-white
layer on the surface ; the gelatine itself has a bluish-white colour
which later becomes a light wine-red. On agar it forms a thick
bluish-grey expansion. On potato there is a yellowish or brown
layer, and the potato assumes a flesh colour.

B. Lactericius.

Described by Adametz as the "ziegelrother" bacillus which he
isolated from water. It is a bacillus three to five times as long as
broad. It is not motile and does not form spores. It does not
liquefy gelatine, and on the surface forms a thick brick-red layer;
there is very little growth in the stab. On potato it forms a
brick -red layer.

B. Rubescens.

Found in canal-water by Jordan. It is a slightly motile large
bacillus, which does not form spores. The colonies on the
surface are thick porcelain-white drops, which later become
brownish in colour ; in the depth the colonies are small. The
gelatine is not liquefied. In gelatine-stab there is a porcelain-
white nail-head projection on the surface, but only slight growth
along the stab. On agar there is a white growth, which later
becomes wrinkled, and in about three weeks a pinkish tinge is
seen. On potato there is a rapid growth, at first brown, but
slowly changing to pink ; in three weeks there is a luxuriant
flesh-coloured growth. Milk is not coagulated, and eventually
assumes a pinkish tinge at the surface. It does not reduce
nitrates.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 6l

B. Rubidus.

Found in water by Eisenberg. It is a medium-sized, motile
bacillus. It does not form spores and will only grow at room
temperature. The colonies in gelatine plates are round, finely
granular, and have a brownish-red colour in the centre. The
gelatine is gradually liquefied. On agar and potato it forms a
brownish-red growth.

GROUP VI.

In this group are contained the bacilli which produce a
yellowish or orange coloured pigment; some of them liquefy
gelatine, others do not.

B. Arborescens.

This organism is common both in surface and deep well-waters.
It has the following cultural characteristics.

Colonies on Gelatine Plates. — The colonies have a golden yellow
col our, under a low power they resemble a wheat-sheaf; some small
growths are also seen consisting of a rod branching at the ends.
Sometimes the wheat-sheaf is surrounded by a thin film with
fine processes projecting from the margin. Very often the
colonies consist simply of a mass of threads spreading out in
every direction through the gelatine.

Gelatine-stab. — The gelatine is liquefied in the form of a cup
with an accumulation of orange pigment at the bottom.

Agar-slope. — A thick golden-yellow growth.

Glucose-gelatine* — No gas formation.

Peptone. — No indol produced.

Potato. — A thick coarsely granular deep orange-coloured
growth.

Nitrate-broth. — Nitrates are reduced to ammonia.

Milk. — After seven days incubation at 22° C. the medium is
unchanged.

Litmus-wliey. — After seven days incubation at 22° C. the
medium is distinctly alkaline.

Microscopical Appearance. — A thin slender bacillus which does
not form spores and possesses only rotatory movement. It grows
feebly at 37° C. It is decolorised by Gram. The bacillus

62 BACTERIOLOGICAL EXAMINATION OF WATER.

described above differs from Frankland's organism by reducing
nitrates to ammonia. Some varieties of this organism send out
long wandering convoluted processes on the surface of gelatine
plates, and in this respect resemble the Proteus species.

B. Nubilus.

This organism occurs very frequently in water derived from
wells, rivers, and upland surfaces. It is a slender bacillus, which
possesses only rotatory movement and does not form spores. It
does not stain with Gram, and cannot develop at 37° C. or under
anaerobic conditions. The colonies in gelatine appear as cloudy
patches, with an ill-defined margin, which grow chiefly in the
depth ; under a low power each colony is seen to be made up of
interlacing threads of bacilli ; the gelatine is rapidly liquefied.
In gelatine stab there is a thick white growth, but under the
surface and along the line of inoculation delicate cloudy growths
appear, the gelatine is liquefied horizontally at the surface, and
the growth in the liquefied gelatine acquires a faint yellow
colour. On agar there is a white growth with a thin filmy
margin ; later, the growth acquires a faint yellow tint. On
potato there is a marked yellow growth and the potato becomes
discoloured. Milk is not affected by the growth of this organism,
and litmus-whey acquires a faint alkalinity. There is no gas
formation in glucose and lactose media. There is a diffused
growth in broth with a slight deposit, but no pellicle forms on
the surface. Nitrate broth is powerfully reduced, ammonia being
produced. In peptone and salt solution there is no formation
of indol.

The organism above described, which I have often found in
waters, appears to closely resemble the B. nubilus described by
P. and G. Frankland. The colonies are identical ; but the growths
in gelatine and on agar have a yellow colour, which is not seen
in the growths of the Franklands' bacillus ; also the latter
produces a film in broth, gives a very slight growth on potato,
and reduce nitrates feebly, only a small proportion of nitrites
being found. The B. gracilis, described by Zimmermann, appears
closely related to the B. nubilus. This organism, however,
liquefies gelatine very slowly, and on potato produces only a
slight moist appearance. It is also said to produce elliptical spores.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 63

B. Aquatilis (Percy Frankland).

This organism was found by Percy and G. C. Frankland in
water obtained from deep wells sunk in the chalk by the Kent
Company. It is very similar in appearance to the B. arborescens,
forming long wavy threads sometimes as long as 17 JJL ; spores
are not observed and the movement is only vibratory. In
gelatine plates the colonies in the depth at first appear smooth-
rimmed, but the contour gradually becomes more and more
irregular ; on reaching the surface slow liquefaction of the gelatine
commences and convoluted bands of threads extend from the
centre to the periphery. In gelatine tubes it grows extremely
slowly ; a yellow expansion forms on the surface but hardly any
growth appears in the depth ; later slight liquefaction takes
place. On agar there is a small, shining, yellow growth. Broth
is rendered turbid ; there is a white deposit, but no pellicle forms
on the surface. On potato there is hardly any growth. In
nitrate broth there is a good growth but the nitrates are not
reduced.

B. Aquatilis.

In water supplied by the Kent company, I have frequently
found a bacillus which appears to be closely allied to the
B. aquatilis of Frankland. The colonies of this organism in
gelatine plates appear at first as delicate circular films, which
under a low power show a series of concentric rings ; later lique-
faction of the gelatine slowly sets in and under a lower power
the concentric rings are not so marked, but a mass of convoluted
bands, passing outwards from the centre, are now seen, and the
margin appears set with wedge-shaped processes. In gelatine-
stab there is a cup-shaped liquefaction, the surface being covered
with a folded membrane. On agar there is yellowish, very
sticky, growth. Milk is not coagulated. In glucose and lactose
media, there is no gas formation. In peptone solution, after
incubation for seven days at 37° C., there is a slight indol
reaction on adding potassium nitrite and a few drops of pure
sulphuric acid. Litmus-whey, after twenty -four hours at 37° C ,
is rendered strongly alkaline. Nitrate-broth is not reduced.
Potato shows a yellow, limited growth. In broth there is a

64 BACTERIOLOGICAL EXAMINATION OF WATER.

diffused growth. It is a small, highly motile bacillus, which is
not stained by Gram. It does not form spores.

B. Ochraceus.

This organism was first described by Zimmermann.

In the depth of gelatine plates the colonies are small
and round and have a pale yellow colour; on the surface
they show a deep yellow-ochre coloured centre surrounded by a
thin film of a light yellow colour. Later, processes grow out
from the centre into the margin and the colony becomes sur-
rounded by a zone of liquefaction. On agar and potato it forms
a yellow-ochre coloured growth. It liquefies a gelatine-stab
horizontally with a deposit of yellow ochre pigment. It also
powerfully reduces nitrates. In glucose-gelatine there is no gas
formation. It is a thin, slowly motile bacillus ; grows best at
20° C. and does not form spores. It is not stained by Gram's
method.

B. Tremelloides.

Found by Tils in the Freiburg water supply. The colonies
in the depth of gelatine are smooth- rimmed yellow dots; on the
surface they form thin yellow expansions with an irregular
margin. In gelatine-stab there is a growth on the surface
resembling a colony, and along the stab a thin yellow growth
consisting of isolated colonies. On agar and potato it forms a
yellow growth. There is no gas formation in glucose-gelatine.
Indol is not produced in peptone, and milk is not coagu-
lated. It is a small, thin, motile bacillus. Gelatine is slowly
liquefied.

Yellow Bacillus.

Described by Lustig. The colonies in gelatine plates are
small round grey dots, which under a low power have a golden-
yellow colour and a serrated margin ; from the periphery of the
colony delicate extensions pass into the surrounding gelatine ; in
about ten days the colony rests in a cup of liquefied gelatine.
In gelatine-stab it forms a yellow slimy growth on the surface,
along the stab there is a yellow growth with short lateral

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 65

extensions into the gelatine ; about the fourth day there is a cup-
shaped liquefaction at the surface containing a golden-yellow
slimy deposit. On agar there is a straw-coloured growth. On
potato there is a coffee-coloured layer. It is a short, thin, non-
motile bacillus, occurring usually in long spiral or bent threads
containing many individual bacilli.

B. Helvolus.

Described by Zimmermann. The colonies in the depth of
gelatine plates are small, circular, bright yellow dots ; on the
surface there is a pale yellow expansion with an irregular margin
which in a short time rests in a depression of liquefied gelatine.
In gelatine-slab it forms a button-like growth of a Naples-yellow
colour ; the growth extends to the margin of the tube and a
saucer-like depression is produced ; there is very little growth
along the stab. On agar it produces a thick Naples-yellow
growth. On potato there is an abundant yellow expansion. It
is a thin bacillus with only rotatory movement. It stains with
Gram.

The following organisms do not liquefy gelatine :

B. Aureus.

Found in water by Adametz. In gelatine plates the surface
colonies have a golden yellow centre, the margin, however, is
lighter in colour and crinkled ; growth is slow and the gelatine
is not liquefied. In gelatine-stab the growth on the surface
resembles a colony ; along the stab there is a fine growth of a
golden-yellow colour. On potato a golden-yellow, raised,
coarsely granular growth appears. In broth there is a diffused
growth with a formation of the same golden-yellow pigment.
There is no gas formation in glucose gelatine. Nitrates are
reduced to ammonia. It is a fine bacillus of variable length.
Movement is feeble, and no spores have been observed.

B. Aurantiacus.

This bacillus was isolated by Frankland from deep wells in the
chalk. On gelatine plates it produces bright orange pin-heads.
Under the microscope the colonies in the depth are seen to be
smooth-rimmed. Growth is slow, and no liquefaction of the

E

66 BACTERIOLOGICAL EXAMINATION OF WATER.

gelatine takes place. In gelatine-tubes a shining orange-
coloured expansion forms on the surface, whilst hardly any
growth is visible in the depth. On agar there is a bright orange
expansion which does not extend much beyond the point of
inoculation. In broth there is a slightly orange-coloured
deposit, the fluid above is clear and a thin pellicle forms on the
surface. On potato it produces a thick and magnificent brilliant
red orange pigment, which is, however, restricted to the point of
inoculation. It reduces nitrates to nitrites very slightly. It is a
short, fat bacillus, forms pairs and short chains. The individual
bacilli are motile.

B. Chrysogloia.

This organism was first described by Zopf. It was isolated
from polluted water by Zimmermann. It is a thin, very motile
bacillus of variable length, and often found in chains. It does
not form spores. In gelatine plates the colonies, to the naked
eye, appear as small golden yellow drops. Under a low power
they are seen to be brownish-yellow, granular, round discs. In
gelatine-stab, on the surface, there is a circular yellow-ochre
growth, which later shows concentric zones and becomes folded ;
there is no coloured growth along the stab. In gelatine-streak
and on agar a yellow-ochre growth develops. On potato a
thick growth gradually appears, which at first has a yellow-ochre
colour, but later this changes to an orange-yellow tint. In broth
a pellicle forms on the surface and an orange-yellow deposit
appears at the bottom. The bacillus grows equally well at
blood and room temperatures. It is aerobic and stains slightly
with Gram.

B. Subflavus.

Found by Zimmermann in the Chemnitz water supply. It
forms yellowish-white dots in the depths of gelatine-plates ; on
the surface it appears as a yellowish-white shining drop, which
forms a flat expansion with an irregular and lobular periphery.
Under a low power the surface appears arranged in scales. The
gelatine is not liquefied. In gelatine tubes it forms a delicate,
greyish-yellow expansion on the surface. On agar it produces a
pale yellow growth, which later becomes chrome-yellow or
yellow-ochre. It does not grow well on potato, and produces a

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 67

faint, loamy yellow colour. It is a thin, slowly motile, small
bacillus, grows best at room-temperature, and is not stained by
Gram's method.

The Orange-red Water Bacillus.

Described by Adametz-Wichmann. In gelatine plates it
grows very slowly, forming orange-red centres, which under a
low power are finely granular and brown in colour. The
colonies in the depth appear to be colourless. In gelatine tubes
it grows slowly on the surface, forming a moist, slimy layer of an
orange-red colour. The gelatine is not liquefied. It is a long,
thin, non-motile bacillus. Its cultures appear to be identical
with those of the B. aurantiacus.

B. Luteum.

Described by List and Adametz. In gelatine plates it forms
irregular, slimy, orange-yellow centres which gradually grow
out into flat expansions ; under a low power they appear to
consist of club-shaped, coarsely granular, zooglcea masses, each
of which is made up of several pieces. No liquefaction occurs.
In gelatine tubes it forms an orange yellow expansion on the
surface, but grows slowly in the depth. It coagulates milk. It
forms elliptical cells which are not motile.

The B. fulvus described by Zimmermann appears to stand
between the liquefying and non-liquefying varieties. It is
a small non-motile bacillus. Media are coloured a gamboge-
yellow. Gelatine is liquefied after some weeks.

GROUP VII.

This group includes the micro-organisms which produce a
violet blue or indigo colour in the nutrient media.

B. Janthinus (Zopf) or B. Violaceus.

This bacillus was first isolated by Zopf ; it has also been
found by Zimmermann in the Chemnitz water supply and by
Proskauer in the Spree. In gelatine plates the colonies at first
are milk-white ; later, they grow out, and the margin is irregular
and thinner than the centre ; each colony acquires a violet colour

68 BACTERIOLOGICAL EXAMINATION OF WATER.

and the gelatine is slowly liquefied. In gelatine-stab there is a
white growth on the surface, which later assumes a violet colour ;
the gelatine is slowly liquefied ; there is very little growth along
the line of inoculation. On agar there is a yellowish-white
growth, which later acquires a violet colour. On potato
there is a marked violet-coloured growth. Milk is not coagu-
lated. Nitrates are rapidly reduced to nitrites. It is a motile,
medium sized bacillus, which does not form spores.

B. Violaceus Laurentms.

Isolated by Jordan from the Lawrence. In gelatine plates
the colonies in the depth are round and coarsely granular ; on
coming to the surface they grow out into thin irregular expan-
sions and the gelatine is almost immediately liquefied ; a round
violet spot appears in the centre surrounded by cloudy, liquefied
gelatine. In gelatine-stab the medium is rapidly liquefied
along the inoculation line ; the liquefied gelatine has a violet
colour, and there is a dark violet precipitate at the bottom of
the tube. On agar and potato there is an abundant growth of
a dark violet colour, which soon becomes jet black. In milk
there is a rapid growth, the milk acquires a violet colour and is
coagulated. In broth there is only a very poor growth, but in
nitrate-broth there is a luxuriant growth with formation of a
rich violet colour. It reduces nitrates to nitrites rather slowly.
It is a short, slight, very motile bacillus, which does not form
spores. It often occurs in pairs and chains.

B. Lividus.

Found by Zimmermann in Chemnitz water. It closely re-
sembles the B. Janthinus ; but the bacillus is non-motile and
forms spores. The colonies in gelatine plates have an irregular
margin and resemble a Liver acinus.

B. Violaceus Berolinensis.

Found rather commonly in Spree water by Plagge and
Proskauer and C. Frankel. The same bacillus was also found
by Frankland in the Thames and deep-well water. The colonies
enclosed in the gelatine appear as small air-bubbles, which have

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 69

an irregular margin, and from the commencement of the rather
quick liquefaction they send out thread-like branches ; later a
violet colour is developed. In gelatine-stab there is liquefac-
tion in the form of a funnel and a violet-coloured deposit. There
is a growth in broth and a violet-coloured deposit. On agar
there is a deep blue growth. It reduces nitrates. It is a
small motile bacillus, often in pairs, and forms spores.

B. Violaceus Lutetiensis.

Isolated by Mace from water. This organism is probably
identical with the above. It is a short bacillus, which liquefies
gelatine very rapidly.

B. Amethystinus (Eisenberg).

Found by Jolles in well-water (B. membranaceus amethys-
tinus). In gelatine plates the surface colonies resemble typhoid
colonies ; at first they are colourless ; later, they acquire a dark
violet colour, and the gelatine is only liquefied in ten to fourteen
days. When liquefaction has taken place the colony floats as a
large violet pellicle on the liquefied gelatine. In gelatine-stab
it forms a yellowish white expansion on the surface, which in
ten to fourteen days assumes a violet colour and the gelatine is
slowly liquefied. After about four weeks a thick, dark violet
pellicle covers the surface of the liquefied gelatine. On agar it
forms a yellowish-white growth, which in about ten days assumes
a dark violet colour with a metallic lustre. On potato it forms
a dirty yellow or olive-green coloured growth. In broth it
forms a violet-coloured pellicle on the surface, and the broth
itself is coloured brown. It is a small non-motile bacillus,
which does not grow at 37° C.

B. Amethystinus Mobilis.

Described by Germano. Its colonies resemble the above by
forming violet-coloured membranes resting on slowly liquefied
gelatine. It is, however, a large motile bacillus, which produces
a brown growth on potato.

B. Coeruleus (Smith).
Found by Smith in river water. It is a thin bacillus, often

70 BACTERIOLOGICAL EXAMINATION OF WATER.

in masses, and does not form spores. The colonies are round
with a bluish colour and slowly liquefy gelatine. In gelatine-
stab the medium is slowly liquefied in a funnel form. On agar
it forms a bluish growth. On potato there is a beautiful dark-
blue expansion.

B. Cosruleus (Voges).

Described by Voges. It is a small motile bacillus which is
decolourised by Gram. It has one polar flagellum, and does
not form spores. It will grow at 37° C., and form pigment.
The colonies in the depth are small ; on the surface they grow
out and have a typhoid-like form ; later, they become greyish-
blue and slowly liquefy the gelatine. In gelatine-stab it grows
slowly, and there is no pigment along the stab. In broth a
pellicle is formed, and a grey colour appears in the medium.
On agar there is a grey-coloured growth. Milk is not coagu-
lated ; the creamy layer has a sky-blue colour. On potato there
is a greyish-blue growth, which later becomes dark blue in colour.

B. Indigoferus (Voges).

Found in the Kiel water supply. It is a small motile bacillus,
which does not form spores. It is not coloured by Gram.
There is a slight growth at 37° C., but no pigment is produced
at this temperature. The colonies in the depth are small ; on
the surface, they grow out, are iridescent, and have a distinct
blue colour. In gelatine-stab there is a marked blue growth on
the surface, but no colour is formed in the stab ; the gelatine is
not liquefied. In broth there is a thin pellicle of a blue colour.
Milk is unchanged ; it has a bluish-grey colour on the surface.
On agar there is a beautiful dark-blue layer. On potato there
is a greenish-blue expansion, looking like a layer of caviar.

B. Indigonaceus (Schneider).

Found by Cliiessen in Spree water. It closely resembles the
B. indigoferus. The growth is, however, somewhat quicker.
In broth there is no pellicle. On potato the growth is deep
indigo blue when the medium is acid, and dark green when it is
alkaline.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 71

The B. indigoferus described by Zimmermann resembles the
above organisms, but is said to form spores.

GROUP VIII.

This group includes the B. albus of Eisenberg, B. ubiquitus
of Jordan, B. aquatilis of Lustig, and B. umbilicatus of Zimmer-
mann.

B. Albus.

Eisenberg describes this organism as a small motile bacillus
which does not grow at 37° C. In gelatine plates the colonies
are round and have a pin-head appearance ; the gelatine is not
liquefied. In gelatine-stab there is a slow growth ; on the sur-
face it forms a pin-head appearance. On agar it produces a
milk-white layer. On potato there is a dirty yellowish-white
growth limited to the line of inoculation.

I have found in water an organism corresponding to the above
description, which also failed to coagulate milk, and did not
produce gas in glucose media, indol in peptone- water, nor reduce
nitrates.

The B. aquatilis of Lustig resembles the B. albus in most of
its cultures, but it reduces nitrates to nitrites.

B. Ubiquitus.

This organism was isolated by Jordan from Lawrence sewage
and stated to be abundant everywhere. It is a small, short,
plump bacillus closely resembling a micrococcus. It is non-
motile, and grows as well at 37° C. as at 22° C. In gelatine
plates the colonies at first are small, round, and of a yellowish
tinge; in two days they form glistening white projections
resembling drops of milk. Under a low power the colonies
have a smooth edge and granular contents. In gelatine-stab
there is a good growth both on the surface and along the line
of inoculation ; the whole growth resembles a nail. At first the
colour of the growth is a lustrous porcelain white; later, it is a
dull brownish-grey. On agar there is a whitish-grey growth
having a somewhat metallic lustre. On potato there is a white
shining growth limited to the line of inoculation. Milk is
quickly coagulated with a strong acid reaction. In broth there

72 BACTERIOLOGICAL EXAMINATION OF WATER.

is a diffuse growth with a marked deposit, and in old cultures
the formation of a slight pellicle is seen. Nitrates are
vigorously reduced.

I have frequently found in water an organism which closely
resembled the above. Milk, however, was not coagulated by
my cultures.

B. Umbilicatus.

This organism was frequently isolated from Chemnitz water
by Zimmermann. It is a short bacillus with rounded ends,
occurring singly and in pairs ; at times it forms long chains. It
has a rotatory movement and does not form spores. Gram's
method causes decolorisation of the bacillus. The optimum
temperature varies between 18° C. and 30° C. In plate cultures,
under a low power, the colonies at first appear as yellow points,
which enlarge and become granular. On the surface the colonies
show a wart-like appearance, but later they grow out and have
an irregular margin. To the naked eye the colonies appear at
first as greyish-white drops, which soon acquire an umbilicated
centre. In stab-gelatine on the surface there is a thin
irregularly-lobed growth ; along the line of inoculation there is
a good growth, often accompanied by the development of gas
bubbles. On agar there is a thin greyish-white growth with an
irregular margin. On potato there is a thin, smooth, yellowish-
grey growth. Broth becomes opalescent, and a white sediment
forms at the bottom of the tube.

GROUP IX.

This group includes a number of organisms which produce a
brown colour in nutrient media.

B. Brunneus.

Isolated by Adametz from water. It is a small, thin, non-
motile bacillus which forms spores. The colonies in gelatine
appear on the surface as small, dirty white drops. The under-
part of the colonies is at first white, then grey, and finally
brown in colour ; the gelatine is not liquefied. In gelatine-stab
it forms a slimy growth on the surface, which later becomes

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 73

brown in colour ; along the line of inoculation there is also a
brown growth.

B. Fuscus.

Found by Zimmermann in the Chemnitz water supply. It
is a thin non-motile bacillus which varies considerably in length.
Sometimes the rods are bent ; no spore formation has been
observed. It is aerobic and grows best at 30° C. In gelatine
plates the deep colonies appear as small brown granular dots ;
on the surface the colonies form pin-heads, which later become
surrounded by a thinner margin, which is often thrown into
folds. The centre of the colony has a dark brown colour ; the
gelatine is not liquefied. In gelatine-stab it produces a project-
ing growth, which later expands and acquires a chrome-yellow
colour. On agar and potato the same chrome-yellow growth is
produced.

B. Fuscus Limbatus.

Isolated by Scheibenzuber from bad eggs ; also found by
Tataroff in water. It is a short motile bacillus, which often
forms threads. The colonies in gelatine are brownish lumps ;
some of them are surrounded by a lighter zone, two or three
times as wide as the original colony; the gelatine is not
liquefied.

In gelatine-stab there is very little growth on the surface ;
along the line of inoculation there is a growth from which buds
spread out horizontally; the gelatine in the neighbourhood is
coloured brown. On agar there is a superficial expansion, and
the medium becomes coloured brown. On potato it produces a
brown growth.

B. Hydrophilus Fuscus.

This organism was discovered by Sanarelli in a well-water of
the University of Siena. It is pathogenic for frogs and other
animals. It is a thin, very motile bacillus of variable length.
It does not form spores and is not stained by Gram ; growrs best
at 37° C. In gelatine plates the colonies are round and
transparent ; the gelatine is rapidly liquefied. In gelatine-stab
the gelatine is rapidly liquefied in the form of a funnel ; in a
very short time the whole of the tube is liquefied, and there is

74 BACTERIOLOGICAL EXAMINATION OF WATER.

a copious deposit. On agar it forms a thin greyish-blue growth,
which later assumes a brown colour ; gas bubbles appear in the
depth of the agar. There is a diffused growth in broth, and a
thin pellicle forms on the surface. On potato there is a light
yellow growth, which later becomes dark brown in colour.

GROUP X.

This group includes the micrococci which are commonly
found in water.

Micrococcus Agilis.

It is a motile micrococcus of rather large size, and is often
found as diplococci, streptococci and tetrads. In gelatine plates
the colonies appear as round pinkish-red films, which slowly
liquefy the gelatine. In gelatine-stab the same pinkish-red
growth appears on the surface and the gelatine is slowly lique-
fied. On agar and potato there is a pinkish-red growth. In
glucose-gelatine there is no gas formation. In broth there is a
diffused growth, and a pinkish-red deposit forms at the bottom
of the tube. It has no reducing action on nitrates.

The Micrococcus carneus and Micrococcus cinnabareus are
allied to the above, but they are not motile and do not liquefy
gelatine. The M. carneus produces a flesh-coloured pigment
and the M. cinnabareus a pigment the colour of sealing wax.

Micrococcus Flavus Liquefaciens.

This organism forms rather large non-motile cocci, which are
sometimes arranged as diplococci and in masses. In gelatine
plates, after forty-eight hours, small yellowish colonies are seen,
which gradually increase and become surrounded by a circle of
liquefaction ; under a low power the centre is granular, yellowish-
brown in colour, and sends out processes to the periphery like
the spokes in a carriage wheel. In gelatine-stab there is slow
liquefaction, and on the seventh day a funnel, containing a
deposit of yellow pigment, appears in the upper part of the
tube. On potato there is a moist growth of a yellow colour.
On agar there is a yellowish-coloured growth. In glucose-
gelatine there is no gas formation. Milk is unchanged.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 75

Micrococcus Aquatilis.

This organism consists of very small cocci, often arranged in
irregular heaps. In gelatine plates the surface colonies are
circular, with a narrow homogeneous marginal zone and a
darker centre, which strongly resembles a section of a Liver
acinus ; the colonies in the depth, under a low power, have a
yellow colour and resemble a mulberry in appearance. In
gelatine-stab and on agar there is a white growth, which is not
characteristic.

Micrococcus Luteus.

This organism forms large elliptical cocci. In gelatine plates
the surface colonies are raised, sulphur-yellow in colour, and
have an irregular margin ; the gelatine is not liquefied. In
gelatine-stab and on agar and potato the same sulphur-yellow
growth appears.

The Micrococcus aurantiacus closely resembles the M. luteus,
but the colour produced is orange-yellow instead of sulphur-
yellow.

Micrococcus Fervidosus.

Found by Adametz in water. It is a small round coccus,
which occurs as diplococci and also in small groups. In gelatine
plates the colonies in the depth appear after four or five days
as small white dots. Under a low power they are faint yellow
in colour and smooth rimmed. The surface colonies are trans-
parent and yellow, with a serrated edge and numerous lobular
projections; the gelatine is not liquefied. In gelatine-stab it
forms a thin finely-serrated surface expansion ; along the stab
there is a granular growth. In cane-grape and milk-sugar it
develops slowly. On agar it forms a milk-white slimy expansion.
On potato it forms a dirty white growth.

Micrococcus Flavus Tardigradus.

Described by Flugge. It forms large round cocci, which at
times exhibit characteristic dark poles ; usually arranged in
heaps. It grows extremely slowly ; after six days the deep
colonies are still very small, round or oval in shape and dark
chrome-yellow in colour. The surface colonies have a smooth

76 BACTERIOLOGICAL EXAMINATION OF WATER.

appearance, and project slightly above the gelatine ; under a
low power the deep colonies are circular, smooth rimmed, and
uniformly dark olive-green in colour ; the surface colonies at
the margin are lighter in colour, viz., more greyish-yellow.
The gelatine is not liquefied. In gelatine-stab there is a growth
after six or eight days, a row of small yellow colonies being
seen along the inoculation line.

Micrococcus Radiatus.

Described by Flugge and found by Adametz in water. It
occurs as medium-sized cocci, at times arranged in short chains,
more often in small heaps. In gelatine plates it forms in two
days large colonies, which have a yellowish-green iridescence.
Under a low power the colonies appear granular, sharply outlined,
with a slightly irregular margin ; at times a row of rays grows
out giving the colony a starfish-like appearance. The gelatine
at this stage liquefies, and in two more days the starfish rays
appear as a delicate and regularly arranged circlet of rays. In
two or three days a second circlet of rays may appear, and
eventually even a third circlet may be formed. In gelatine-stab
horizontal rays pass out from the stab, and above a funnel is
formed by slow liquefaction of the gelatine.

Micrococcus Flavus Desidens.

Found by Flugge in air and water. It consists of small cocci,
chiefly arranged as diplococci ; also found in short chains and
as triangles. In two days the colonies in gelatine appear as
yellowish white points, which often show on one side a finely
granular yellowish- brown projection. The surface colonies have
a lighter colour near the margin. After four days the colonies
in the depth are little changed ; the surface colonies, however,
have grown out to a diameter of 5 to 10 mm., and are circular
and lobular, of a pale yellow or brownish colour, forming slimy
expansions which slowly sink into a flat circular depression
formed by gradual liquefaction of the gelatine. In gelatine-
stab it forms a confluent, porcelain white, growth in the depth,
and a yellowish-brown slimy expansion on the surface, which
does not reach the margin of the tube ; in about eight days the

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 77

gelatine is softened into a thick fluid, in which the surface
growth sinks.

Micrococcus Versicolor.

Described by Flugge and found in water by Tils. It consists
of small cocci arranged in pairs or masses. In gelatine-plates
the colonies in the depth appear after forty-eight hours as round
centres which have a yellow colour ; on the surface they form
large expansions of irregular shape, often four-cornered with
lobular projections; they are slimy yellowish-green in colour
and the centre is often raised. The gelatine is not liquefied.
In gelatine-stab small yellowish round colonies appear in the
stab ; on the surface there is a mother-of-pearl growth with an
irregular frayed edge.

Micrococcus Viticulosus.

Found by Flugge in air and water. It forms oval cocci of
variable size arranged in thick zoogloea masses. In gelatine-
plates the colonies in the depth form hair- like extensions from
the centre of the colony, which unite to produce a delicate
network. Under a low power the network is seen to consist of
zooglcea masses of cocci arranged side by side. On the surface
the colonies form thin rapidly-growing layers, from which fine
threads penetrate into the depth of the gelatine. In gelatine
streak it forms radial extensions from the line of inoculation,
producing an appearance like a feather. In gelatine-stab the
same network appears in the stab, which is soon covered up by
the thin rapidly growing surface expansion.

Micrococcus Candicans.

Described by Flugge and also by Frankland. It forms rather
large, round cocci, which are arranged in irregular heaps. In
gelatine-plates the colonies in the depth are white centres ; on
the surface they are smooth, of a pure milk-white colour, and
measure 2 mm. in diameter. Under a low power the deep colonies
appear circular, granular, and deep brown in colour ; the surface
colonies are slightly irregular at the margin, deep brown in the
centre but lighter towards the periphery. In gelatine-stab there
is a nail-like growth ; Frankland states that the gelatine is

78 BACTERIOLOGICAL EXAMINATION OF WATER.

slowly liquefied as a long narrow funnel. Fliigge does not
mention any liquefaction. On agar it produces a thick Chinese
white growth (Frankland).

GROUP XI.

This group comprises the Sarcinae found in water supplies.
The most common variety appears to be the Sarcina aurantiaca
which produces small round yellow colonies in gelatine-plates ;
later the colonies appear granular and acquire an orange-gold
colour ; the gelatine is slowly liquefied. The same orange-gold
growth is seen on agar and potato. The Sarcina alba closely
resembles the above, but its cultures have a white colour. The
Sarcina lutea is very common in air and has been found by Tils
in water ; the cocci of which it is composed are larger than those
of Sarcina aurantiaca, and its cultures have a yellow instead of
an orange-gold colour. The cocci in all the Sarcinae are arranged
in twos or fours and in packets, which occupy three dimensions
in space.

GROUP XII.

This group includes the various Spirilla which have been found
in water. The Vibrio aquatilis of Gimther, the Vibrio Finkler
Prior, Vibrio Metchnikovi, and the Spirilla isolated from water
by Sanarelli, which strongly resemble the Spirillum Cholerae
Asiaticae, will be found fully described under the diagnosis of the
Spirillum Cholerae from allied organisms.

The following Spirilla have also been found in water from
various sources.

Vibrio Aureus.

Isolated by Weibel. The Spirillum is 1^ times as thick as
Koch's Comma spirillum ; it produces long and short forms, but
the tendency to produce spirals is not marked ; it readily under-
goes degeneration, is not motile, and does not stain with Gram.
The colonies on the surface of gelatine plates are circular, flat
expansions of a yellow colour. Under a low power they are
smooth-rimmed, circular, granular, and golden-yellow in colour ;
the gelatine is not liquefied. On gelatine-stab there is a bowl-
shaped yellow-ochre growth 'on the surface, and a yellow-ochre

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 79

granular growth along the stab. On agar and potato the same
yellow-ochre growth appears.

The Vibrio flavus described by Weibel, only differs from the
above in that the colour produced is pale yellow instead of
yellow-ochre. The Vibrio flavescens, also described by Weibel,
is a variety of the Vibrio aureus ; the colour of the colonies is
dirty greenish-yellow.

Spirillum Rubrum.

Isolated by Esmarch, and also found in water by Adametz. It
is about twice as thick as Koch's Vibrio and in most cultures
forms from one to three spirals ; in broth, however, it may form
twenty to fifty spiral turns. Spore formation is doubtful.

In gelatine plates the colonies grow extremely slowly ; in about
eight days small red centres are seen, which, under a low power,
appear granular with a smooth margin ; gradually the colonies
assume a wine-red tint ; the gelatine is not liquefied. In gelatine-
stab there is a growth along the stab consisting of wine-red
colonies ; on the surface there is no colour produced. On agar it
produces a greyish-white growth, which is red in the thickest
part of the growth. On potato it forms small, deep red colonies.
It renders broth slightly turbid, and a red, deposit appears at the
bottom of the tube.

Spirillum Rugula.

Found by Miiller in the buccal cavity; also isolated from
stagnant water and faeces. It is often found with the B. butyricus
and is probably anaerobic. It forms long rods, which are some-
times thin and sometimes very thick ; it may be simply bent or
form flat spiral turns ; at times it forms long chains. It has a
lively rotatory movement on its long axis. It forms spores at
one end of the bacillus and just before the appearance of the
spore a swelling is formed, giving the bacillus a comma-like
form. In gelatine plates, under anaerobic conditions, it forms
yellowish-white, ball-shaped discs ; the gelatine is liquefied. In
gelatine-stab it forms a pin-head growth. On agar at 37° C. it
forms a white, slightly-folded growth. On potato at 37° C. it
forms a wrinkled growth, which later becomes yellow in colour.

80 BACTERIOLOGICAL EXAMINATION OF WATER.

Spirillum Concentricum.

Isolated by Kitasato from putrid blood, and found by Lustig
in water. It forms small screw-shaped forms ; in broth cultures
there may be from five to twenty spiral turns. It is very motile,
with a screw -like movement. It does not form spores. In
gelatine plates it forms circular discs, composed of concentric
rings ; the gelatine is not liquefied. In gelatine-stab it grows
chiefly on the surface, forming a cloud-like expansion. On agar
it forms a diffused growth, which adheres strongly to the medium.
On potato there is no growth. In broth there is a diffused
growth, and a slimy deposit appears at the bottom of the tube,
above which the broth is clear.

The Spirillum volutans is found in stagnant water ; it forms
long threads with dark granular contents, each thread has 2J to
3J turns and has a long cilium at each end ; it is motile ; at
times, however, the threads appear quite motionless. The
Spirillum undula is found in stagnant water; it forms long
threads, which have 1J-3 turns; there is a bunch of cilia at each
end of the bacillus.

The Spirillum serpens is very common in stagnant waters ; it
forms long threads with 3 or 4 turns ; at times the bacilli are
joined into long chains. It is very motile.

The Spirillum tenue is found in stagnant water, and forms
long thin threads with 1J screw turns. It is very motile.

CHAPTER VII.

QUALITATIVE ANALYSIS— continued.
NITRIFYING AND DENITRIFYING MICRO-ORGANISMS.

THE idea that ammonia in water and soil is converted into
nitrates through the influence of micro-organisms was first
suggested in 1878 by Schloesing and Muntz. These observers
proved that nitrification was caused by bacteria, but they failed
to isolate pure cultures of the organisms in question. Waring-
ton, Emich, and other workers repeated and confirmed Schloesing
and Miintz's observations. In 1886 Heraeus published a very
important paper on this subject. He obtained pure cultures of
micro-organisms from water and soil and tested their power of
nitrifying salts of ammonia. The first twelve cultures investi-
gated showed no nitrifying action ; two of them, however, dis-
played distinct power of reducing nitrates. It appeared
probable that the reduction to the state of ammonia only
commenced when all the nitrates had been reduced to nitrites,
and that when ammonia was also present in conjunction with
nitrates the nitrogen from the ammonia would be assimilated
before the nitrates were attacked. By adding 50 grammes of
garden earth to 500 c.c. of water containing one gramme of
ammonium carbonate, powerful oxidation was set up ; the fluid
appeared clear, but on the surface a thin iridescent membrane
was noticed. When plated in gelatine the membrane produced
two species of bacteria of special importance. The first species
was very common, and to the naked eye the colonies appeared in
the depth as small points and on the surface as milk-white, pin-
head growths, which, under a low power, resembled smooth,
round, yellow scales. The colonies of the second species closely
resembled the first, but were more oval in form, and of a darker
brown colour. The behaviour of these organisms was tested in

F

82 BACTERIOLOGICAL EXAMINATION OF WATER.

a medium free from organic matter and containing 0'05 gramme
potassium phosphate, 0*01 gramme magnesium sulphate, 0'05
gramme calcium chloride, and 1 gramme ammonium carbonate
per litre ; the same medium was also employed with the addition
of 1 gramme of glucose per litre. Growth and oxidation were
most marked in the medium free from sugar ; in strong contrast
to the reducing micro-organisms which worked most powerfully
in the medium containing sugar.

Heraeus also examined certain well-known pathogenic and
non-pathogenic organisms, but as these did not grow well in
media containing only salts and sugar, sterilised urine, diluted 1
in 5, and broth, diluted 1 in 10, were used for the experiments.
He found that the M. prodigiosus, B. ramosus, B. anthracis,
B. typhosus, Staphylococcus citreus, and the spirilla of Finkler
and Miller produced nitrous acid in urine.

Heraeus considered that in nature there existed nitrifying and
denitrifying organisms, and whether the oxidation or the
reduction process gained the ascendency depended on the supply
of nutriment present. If there was plenty of food material,
especially organic substances, the denitrification process was
active ; but if the food supply was poor the oxidising organisms
gained the upper hand. When no other source of nitrogen
except nitrates existed in the water it seemed possible that the
oxidising bacteria might reduce the nitrates in order to obtain
food necessary for their growth ; the same reduction also
appeared to take place when the entrance of air into the solutions
was prevented.

Winogradsky considered that the amount of nitrites and
nitrates obtained by Heraeus was not sufficiently great to
establish beyond question the presence of a true nitrification
process. Nitrites and nitrates are constantly present in the air
of rooms and are readily absorbed by alkaline fluids ; conse-
quently, unless this source of error is avoided and a sufficiently
large quantity of an ammonium salt is completely oxidised by
the pure cultures under examination, it is impossible to ascribe
to them true oxidising powers. These conditions were not ful-
filled by the organisms investigated by Heraeus, so it is doubtful
whether he really succeeded in isolating true nitrifying
bacteria.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 83

P. F. and G. C. Frankland tested Heraeus' statements that
M. prodigiosns and B. ramosus possessed nitrifying powers, but
tailed to obtain any evidence of nitrifying action ; on the
contrary both organisms were found to be capable of reducing
nitric to nitrous acid. The Franklands believed that the indica-
tions of nitrous acid which Heraeus obtained on growing these
organisms in diluted urine were due to reduction of the small
quantity of nitrates almost invariably found in normal urine,
and not to any oxidation of the ammonia at all. They also
experimented with no less than thirty-three different cultures
obtained from air and water, but in every case the results were
negative. They found, however, that nitrification was readily
induced when a small quantity of ordinary garden soil was
introduced into a solution containing potassium phosphate,
crystallised magnesium sulphate, fused calcium chloride,
ammonium chloride, and carbonate of lime. By passing a few
drops of the nitrified fluid through a series of dilutions in
distilled water, a fluid was obtained which contained numerous
micro-organisms of a very short bacillary form which could not
be cultivated on gelatine-peptone. A portion of this fluid
induced strong nitrification in a sterile ammoniacal solution. It
was noticed that the solutions which had undergone nitrification
by the organisms contained in the fluid remained perfectly clear,
whilst solutions which underwent nitrification from a mixture of
the nitrifying and non-nitrifying bacteria generally exhibited a
thin surface film. The pure culture of the nitrifying organism,
obtained by the method of fractional dilution, showed only a
very short bacillus, about 0'8 /*. long, hardly longer than broad,
and exhibiting only vibratory motion. This micro-organism
transformed ammoniacal into nitrous nitrogen in amounts which
could be quantitatively estimated by the usual chemical means.
The formation of nitric nitrogen was not observed in solutions
inoculated with a pure culture of the organism in question.

Winogradsky, in 1890, made some very important contribu-
tions to our knowledge of nitrifying organisms. In his earliest
experiments he employed a medium containing 1 gramme of
ammonium sulphate, 1 gramme of potassium phosphate, to the
litre of water from Lake Zurich ; to 100 c.c. of this fluid, placed
in a flask, 0*5 to 1 gramme of basic carbonate of magnesia,

84 BACTERIOLOGICAL EXAMINATION OF WATER.

suspended in a little distilled water and sterilised by boiling,
was added. A small drop of a recently nitrified liquid placed in
a flask produced marked nitrification, and at the expiration of
fifteen days all traces of ammonia had disappeared. The oxida-
tion was found to be produced by zoogloeae of bacteria which
formed on the surface of the magnesium carbonate at the bottom
of the flask. The bacteria which appeared on the surface of the
fluid in the flask had no nitrifying action.

In later experiments Winogradsky prepared an absolutely pure
sulphate of ammonium and employed calcium carbonate instead
of magnesium carbonate. Portions of the zooglcea mass dis-
tributed over gelatine plates produced no growth, but the deposit
of crystals of calcium carbonate introduced into flasks containing
the salt solution produced marked nitrification after the lapse of
three weeks, the delay in the oxidation process being caused by
the small quantity of the nitrifying organisms placed in the
flask. In order to obtain plate cultures of nitrifying organisms
Winogradsky employed a silica which he prepared in the follow-
ing manner : Commercial sodium silicate was diluted with three
times its volume of water and 100 c.c. of the dilution added to
50 c.c. of dilute hydrochloric acid, the mixture was then dialysed.
The dialyser was kept in running water for twenty-four hours
and then in distilled water, frequently changed, for two days.
The solution so obtained was placed in a flask plugged with
cotton wool and sterilised. The following solution of salts was
required to produce gelatinisation : Sulphate of ammonia 0'4
gramme, sulphate of magnesium 0*05 gramme, phosphate of
potassium 0*1 gramme, chloride of calcium a trace, carbonate of
soda 0'6 to 0'9 gramme, distilled water 100 c.c. The two
sulphates with the chloride and the phosphate with the
carbonate were dissolved and sterilised separately, and then
mixed when cool. Small glass dishes, having a diameter of 5
centimetres, were employed for the preparation of plates. The
solution of silica was evaporated to about half its volume, or
until two or three drops mixed with one drop of the saline
solution produced gelatinisation in about five minutes. The
silica solution was then placed in the glass dishes by means of a
pipette and one-third to one-half its volume of the salt solution
added. Sometimes magnesium carbonate was used instead of

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 85

sodium carbonate in the saline solution. This salt rather
destroyed the transparency of the medium but drew attention to
the colonies which attacked the magnesium carbonate and then
seemed to be surrounded by a limpid zone. The colonies in the
depth of the jelly appeared to the naked eye as small white
points.

Duclaux was the first to suggest that the process of
nitrification occurred in two stages, the first being the oxidation
of ammonia to nitrites, and the second the conversion of nitrite
into nitrate. Warington's experiments led him to entertain
the same idea, but he failed to isolate separate nitrous and
nitric organisms. Winogradsky, however, succeeded in cultivat-
ing the organisms which produced these chemical changes. He
divided the organisms which act in the first stage into two
classes — i.e., the Nitrosococcus and Nitrosomonas. The Nitroso-
coccus was found in earth obtained from Europe, Africa, and
Japan ; it appeared as a non-motile spherical cell, about 3 fj. in
diameter. The Nitrosomonas was isolated from earth found in
Java ; it was motile and had a short ellipsoidal form. The
organism of the second stage, or Nitrobacter, was found to be a
small non-motile rod, about 0*5 m. long and 0'25 /n broad. It
grew readily in the saline fluid containing potassium nitrate
instead of an ammonium salt, and also on silica jelly plates.
Burri and Stutzer have also isolated a nitrate producing
organism from earth by means of silica jelly plates. It is larger
than Winogradsky's Nitrobacter, and is motile. It can be
cultivated in broth and gelatine, but then loses its nitrifying
action. Winogradsky found that, as a rule, the oxidation of
nitrates did not commence until all the ammonia had dis-
appeared. Occasionally oxidation of ammonia and nitrites
were found at the same time, but the oxidation of the nitrites
always commenced later than the oxidation of the ammonia.
The nitric ferment had no action on ammonia, and in the
absence of the nitrous ferment nitrification did not take place.
Nitrification in earth was then compared with the same process in
fluid media. It was found that normally only nitrates were pro-
duced in soil, nitrites being rarely detected. It was also noticed
that the oxidation of nitrites took place immediately after their
formation, even in the presence of large quantities of ammonia.

86 BACTERIOLOGICAL EXAMINATION OF WATER.

The differences observed appear to be due to the great porosity
of earth ; as compared with a fluid the surfaces exposed in soil
are enormous, and the nitrate ferment is never repressed by a
want of oxygen unless the ground is saturated with moisture or
dosed with too much ammonia.

The power of reducing nitrates to nitrites and ammonia, the
converse of the oxidation processes just considered, appears to be
common to many bacteria. The B. fluorescens liquefaciens,
B. mycoides, and the varieties of B. arborescens and B. nubilus,
which I have isolated from water, reduce nitrates very vigorously.
Special denitrifying bacteria, however, have been isolated by
Stutzer, Bum, and Jensen.

Bacillus Denitriflcans I.

This organism was isolated by Stutzer and Burri from horses'
faeces. It is a small motile bacillus, which does not form spores.
In gelatine plates there is a limited growth on the surface, and
small round colonies appear in the depth ; the gelatine is not
liquefied. Stroked over a gelatine-slope it produces a dull
bluish grey layer. On agar plates the colonies are large, round,
and very thin. On an agar slope there is a thin grey layer. On
potato, especially at blood temperature, there is a limited
reddish-brown growth. In symbiosis with B. coli or B. typhosus
it reduces nitrates very vigorously.

Bacillus Denitriflcans II.

This bacillus was isolated by Stutzer and Burri from straw.
It is a small motile bacillus, which does not form spores, and
grows best at blood temperature. The surface colonies in
gelatine plates are very characteristic ; they appear as large,
coarse growths, marked with radial ridges, which branch as they
pass outwards and then bend in and interlace at the margin ;
the gelatine is not liquefied. On agar a similar characteristic
growth is seen. On potato there is a slimy, reddish-coloured
growth. In broth a pellicle forms on the surface. This
organism acts in symbiosis with B. coli, in the same manner as
Bacillus I. According to Weissenberg, the B. coli reduces the
nitrates to nitrites, which are then acted upon by the denitrifying
bacilli.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 87

Bacillus Denitrificans (Jensen).

A culture of this organism obtained from Krai's laboratory
gave the following cultural characteristics : The surface colonies
in gelatine plates were very thin and translucent, with a foliated
margin and a dark granular centre. In broth there was a
diffused growth and a slight film on the surface. On potato a
yellowish-brown rather slimy looking growth appeared. In
glucose and lactose media there was no gas formation. Milk
remained unchanged in appearance. On an agar-slope there
was a thick growth along the line of inoculation surrounded by
a thin transparent film with a crenated margin. In stab-gelatine
there was a thick white growth on the surface and along the
stab. In peptone solution no indol reaction was obtained on
adding a nitrite and a little pure sulphuric acid. In nitrate-
broth there was a good growth and the nitrates were found to
be reduced ; the reduction process was found to be very marked
when the bacillus was grown in conjunction with B. coli.
Microscopically the bacillus resembled B. subtilis in its move-
ment, but was smaller in size. It did not stain Gram; no growth
was obtained under anaerobic conditions.

CHAPTER VIII.

QUALITATIVE ANALYSIS— continued.

CLASS II.

THIS class contains the micro-organisms which are common in
sewage, but are rarely found in pure water supplies. Many of
them are derived from the intestinal canal of human beings and

o

animals ; others are associated with putrefactive processes. The
recognition and proper interpretation of these organisms is a
matter of the first importance. It often happens during the exami-
nation of water supplies suspected to have caused zymotic disease,
that chemical analysis, enumeration of micro-organisms, and
search for pathogenic organisms, all fail to give sufficient
data upon which a reliable opinion may be based as to the
purity of the supply. Under these conditions the detection of
the organisms in this class will either enable a proper decision
to be made, or suggest a rigid examination of the supply
when possibly unsuspected sources of contamination may be
detected.

Bacillus Coli Communis.

The most important organism in the sewage class is the B. coli
communis, often called Escherich's bacillus, after the name of its
discoverer. Studies of sewage and normal faeces, however, have
shown that the B. coli usually described in the text-books is the
type of a large group containing many varieties ; the cultural
characteristics, morphology, and serum reactions of the typical
bacillus will be first considered, after which the more important
varieties will be compared with the type. The typical B. coli
communis is usually described as follows :

Gelatine Plates. — In the depth the colonies are circular or

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 89

oval in shape and dark-brown in colour. On the surface after
twenty-four to forty-eight hours they appear as thin bluish-
grey transparent films with an irregular margin. Under a low
power they sometimes resemble flakes of glass with faint surface
markings, but more commonly they appear somewhat opaque
with a faint yellowish-brown colour, and show tracings of
ridges and depressions running from the centre to the peri-
phery ; at times wavy lines parallel to the margin are seen.
The gelatine is not liquefied.

Gelatine-stab. — On the surface an expansion appears like the
surface colonies just described, only whiter and thicker ; along
the stab there is a white growth.

Gelatine-streak. — A broad white growth with a crenate margin.

Agar. — A thick white growth, not characteristic.

Potato. — As a rule a thick brownish-yellow layer is produced ;
occasionally the growth is an almost transparent film like that
of B. typhosus.

Mill:. — Generally coagulated after twenty-four to forty-eight
hours incubation at 37° C.

Litmus-wliey . — Rendered acid ; after seven days incubation at
37° C., the amount of acid produced varies extremely, it always

requires more than 6 per cent. ^ alkali and usually from 20 to

40 per cent, of ^ alkali is required, to neutralise it.

Glucose-gelatine "shake" (22° C.)l Abundant gas formation

Glucose-agar-stab (37° C.) J in 24 hours.

Witters Peptone and Salt solution. — After 7 days incubation at
37° C. indol reaction is given on the addition of 1 c.c. of
potassium nitrite solution (0'02 per cent.) and a few drops of
pure sulphuric acid ; very often the reaction is obtained after
48 hours incubation.

Broth.— Rendered turbid in 24 hours at 37° C. ; later, a slight
film usually forms on the surface.

Neutral-red Glucose-agar. — Red colour changed to yellow and
greenish fluorescence produced.

Nitrate-broth (O'l % KNO3, 5 % broth).— Nitrates reduced to
nitrites.

Gelatine (25 % at 37° C.).— A distinct thick white film appears
on the surface.

90 BACTERIOLOGICAL EXAMINATION OF WATER.

Proskauer and Capaldi, No. I. — After twenty-four hours incu-
bation at 37° C. there is a growth, and the medium is rendered
strongly acid.

Proskauer and Capaldi, No. II. — After twenty-four hours incu-
bation at 37° C. there is a growth, and the reaction of the
medium is either unchanged or rendered slightly alkaline.

Glucose- and Lactose-broth. — After twenty-four hours incuba-
tion at 37° C. there is marked growth, and bubbles of gas form
on the surface.

Microscopical Characters. — It usually appears as a very short
bacillus, often strongly resembling a coccus ; but longer forms
are frequently seen especially in old cultures. As regards
motility it varies considerably ; sometimes it is almost motion-
less, at other times it appears to be as motile as the B. typhosus.
As a rule it possesses from one to three flagella, which are very
brittle and difficult to stain ; occasionally eight to twelve long-
wavy flagella are present. It does not form spores.

Staining Reactions. — Stains readily with basic dyes; it is
decolourised by Gram's method.

Reactions to Specific Sera. — It is not agglutinated by highly
dilute anti-typhoid serum.

The above description gives the chief characteristics of the
typical B. coli ; but when studies are made of sewage and stools
derived from human beings many varieties are found which do
not conform to the type. The most common varieties are the
following :

(1) The colonies exactly resemble the type ; gas is produced
in sugar media and indol formed in peptone ; but milk is not
coagulated, though the amount of acid produced in litmus- whey
may be considerable.

(2) The colonies are typical ; gas is produced in sugar media;
milk is coagulated ; but indol is not formed in peptone solution.

(3) The colonies are typical ; gas is produced in sugar media ;
but indol is not formed and milk is not coagulated.

(4) The colonies are typical ; but there is no coagulation of
milk, production of gas, nor formation of indol.

(5) The colonies are typical ; indol is produced in peptone
solution ; but there is no formation of gas in glucose media nor
coagulation of milk.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 91

Gilbert and Lion investigated the varieties of B. coli found
in the stools of sixty healthy men ; they divided the bacilli into
the following groups :

A. Motile bacilli :

(a) Very motile — cultural reactions ; milk coagulated, milk-
sugar fermented, indol reaction sometimes present.

(b) Slightly motile — cultural reactions : (a) milk coagu-
lated, milk sugar fermented, indol reaction sometimes
present ; (/3) milk sugar not fermented, milk not coagu-
lated, indol reaction present.

B. Non-motile bacilli :

(a) Milk sugar fermented, milk coagulated.

Sub-variety : (a) colonies thick and opaque, indol

sometimes present.
Sub-variety : (/3) colonies thin and transparent, indol

sometimes present.

(b) Milk sugar not fermented, milk not coagulated, no
indol produced.

Gordon has made a careful study of the varieties of B. coli ;
he investigated over one hundred organisms, and classified
them according to (1) their reactions, and (2) their flagella
averages, and in this way differentia ted sixteen varieties. Two
other groups of organisms were also investigated, which
resembled B. coli in morphology; but the members of the
first group produced alkali in litmus-tinted media, and the
members of the second group slowly liquefied gelatine. The
following table taken from Gordon's paper in the Journal of
Pathology and Bacteriology gives the reactions of the different
varieties of B. coli :

92 BACTERIOLOGICAL EXAMINATION OF WATER.

4f

-f ,_ ,_| (M

OS C<1 CO CO rH rH i-H

i.

«MM «-.

ii

S a

CO i— 1 GO CO rH O

CO -* CO CO CO -* •*

CO

-f rH rH CO CO

§

£

a

a r r , R S

M

ri *

<Ji M ^

±

o

1 * r2i

1 '££•&.'£&• ,2

5 * r^

*H

C 0 „ „ „ ,. J- fc

" Cl O O Cl O ^"

fe C 0

1

Is* o

2 >> 3 5

3k^^^^-r - g

- 0 P v^ 0 y rS

111

m o

PQ (5 O

w

'o

-

'o _ 'o

« -2

r§ . ..H

'o ^ 3

^

^ R p§1

«

HH 0

pH O i*^

O rH O

&

<H

ft ft

0

^ r*i ? ^
T3 rS *V

CC 01

1 o |

g
5*

O2 CC 00

co

o ? S r r r ^

^

t^ <M O ° P

*^

-M O

-^ ° ^

4» 4» 49

1

o° 11 *!

5 5

^0

O

j=> ,2 ^2 ^
ooo

0

? CC CC Ct

d ci ci -*j ^> es

^ e?^ cea ^ S4

^0 T3 o 'O o ^3

OS

5

-U -i -U

o ~ o

IN

CO CO r-( 0 1C

O CO 'o »-l^- 'o ~f

^

'o "^ '~^> •" *^

? 11^ o^

^° =S 1 si ?

(M

"o

0^0

0 CO °O

O O O

O

O

Glucose
gelatine
shake.

if... ..

02 00

,?« sis

o CD C5

r^H ^

cc

^ 0

s p s p r

K«

as o:

|||

ji;.. .'.

^i ^ bfl

:;

r p p pp

0 i^32

C^l ft

Source.

1-3 lloll
|1 5|ll||

r ^ ^ £ ^ r ;r

0) r=i

dp

fl 'o

I1

I -In?

g" >, ^H C<1 CO

•* »c: co t> co o O'

.

(M CO HH 1C CC

,5

S "* x

5

•-S ^ r?

fe

8 T -p B *

- - - -

"

« 1 *

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 93

fe

fe >* >» fe >»

^

|l

-a

^-"^ ^

73

»s

I

-)H t^ ^ ^ O

IM i— I <>i <7<1

1

A

a

O'

^

O rH CO CO CO

-*

i

CO

Reaction.

'«

^ c s:

1

J&

1

|

rH

02 ^

?I11 =

o
G

PQ

o ---Jo -

bJO 3 6D

g|

I P

PQ gj.

O M

O

^

'o "o "o

(H

I ' I

o
•9

p

I|l| I

-

.

ae

>i

M jc as as TC

6

°^

0

4

^,1^,1 ^

OQ

o „ ^

(M

>O CO 7<l b- (M

O

M

0

*

O

3553 3

1

o'

^

BQ

OQ 02

^

o

^0

'o

0

O5 r-H 1? ^

rD
t--

i

^

5

00

1

flj O

|.

1*

ce

"^ c5

^3

$

T3

§ -3 -X

^^ cXJ ^

^

bC ^ ^

1_(

IfJ

i

s

1

>.

^»

^

^ w ^

c3

§

75

5 -5 ,$

1-1 cc

T3

&

^

s * rt

i I"3

i *°

J? K

d

0

So ^ §3

^

1 I -2 «

tT,

'D !^H

1

1 S 1

§

^ " "I si

0 ^ o ?

|S

1

bJD

- n

§

a;

o

'a

3 i— i

(N CO -f rH »M

rH

1

^rH C^ CO

— i

rH

«

.H

>j>

1—1

P

^

i

5

^o

94 BACTERIOLOGICAL EXAMINATION OF WATER.

With regard to the organisms which slowly liquefied gelatine,
Gordon found that sub-cultures varied in the date of com-
mencement of the liquefaction ; some gaining, others losing,
power of producing liquefaction.

During the present year I have carefully studied the cultural
reactions of 150 varieties of B. coli isolated partly from
normal and partly from typhoid stools. Between 50 and
60 per cent of these cultures produced gas in sugar media,
soured milk, formed indol after seven days incubation at
37° C. in peptone, and gave a coloured growth on potato.
About 30 per cent produced gas in sugar media, but failed to
coagulate milk and form indol. The remaining cultures failed
in only one of the typical reactions ; the coagulation of milk was
the reaction most frequently absent. Some of the varieties
of B. coli isolated from typhoid stools were very peculiar. The
colonies up to forty-eight hours incubation were typical ; but
later on wedge-shaped processes grew out from the margin of
the colony so that it slowly acquired a rosette-like appearance,
and the gelatine slowly liquefied. When stabbed into gelatine
tuft-like processes grew out from the growth along the line of
inoculation. All these colonies produced gas in sugar media,
soured milk, and gave very marked indol reaction ; in many
respects they resembled the Proteus cloacinus, described by Laws
and Andrewes as one of the commonest organisms in sewage.

A consideration of Gordon's table suggests that possibly the
varieties may be derived from the typical B. coli as a result of
unfavourable surroundings, want of food, &c. Some support
to this view is derived from observing the changes produced in
B. coli after prolonged immersion in sterile sewage. I found
that a typical B. coli, freshly isolated from a normal stool, after
forty-two days immersion in sewage, gradually lost its power
of producing indol, and became much less resistant to the
action of carbolic acid. But agglutination experiments made
with the serum of animals immunised against B. coli, point
strongly to each race being quite distinct. I have found that
cultures of B. coli, derived from the same source and quite
indistinguishable by cultural tests, produced sera which were
quite distinct, each serum reacting to its own bacillus, but
failing to influence in any way the other cultures. Jatta's

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 95

experiments on the action of coli-serum on different races of
B. coli showed that the serum always acted in the highest
dilution on the variety used to inoculate the animal. Very often
the serum had no action on other varieties ; but when it did
act the dilution was much lower than that which agglutinated
the bacillus used for the preparation of the serum. He also
showed that ,in the stools of the same person several varieties
of B. coli might appear.

Lee Smith showed that all the B. coli cultures isolated from
the stools of a child " at the breast " were agglutinated by a
serum made with any one of the cultures. The same serum,
however, failed to agglutinate cultures of B. coli, isolated
from the stools of other children fed under the same con-
ditions. The action of anti-typhoid serum on B. coli will
be fully considered when the value of the agglutination
test for the diagnosis of the B. typhosus is discussed. Here
it will be sufficient to state that though many varieties of
B. coli are quite uninfluenced by anti-typhoid serum, still races
do occur which are agglutinated by anti-typhoid serum in
low dilutions. The virulence of the B. coli does not appear
to have any relation to the action of the anti-typhoid serum
upon it. Jatta passed cultures of the B. coli through guinea-
pigs, so as to increase their virulence ; the test serum, however,
acted on the organisms in just the same manner after the
treatment as before it. It has also been found that a coli-serum
from an immunised animal sometimes causes agglutination of
the B. typhosus, but the dilution which produces the agglutina-
tion is always much lower than that which acts on the culture
of B. coli used to produce the serum. It also appears that
those varieties of B. coli, which are influenced by anti- typhoid
serum, produce, when injected into animals, sera which act on
the B. typhosus.

The Bacterium lactis aerogenes, the Bacillus cavicida, the
Bacterium tholceidum, and the Bacillus Neapolitans, isolated
from fasces, appear to be closely allied to the B. coli.

The Bacterium Lactis Aerogenes was isolated by Escherich
from the intestinal tract of animals and people fed on milk,
and especially from suckling children. It was found once in
raw milk. In gelatine plates the surface colonies are convex,

96 BACTERIOLOGICAL EXAMINATION OF WATER.

isodiametric, and porcelain white in colour ; the colonies in the
depth are yellowish, round, granular centres. In gelatine-
stab it grows well along the line of inoculation, and on the
surface forms a nail-head shaped expansion ; the gelatine is not
liquefied. On potato it produces white raised colonies,
permeated with gas bubbles, which often run together and form
a cream-like layer. It produces gas in sugar media, and
coagulates milk.

Escherich does not mention the formation of indol ; but a
pure culture from Krai's laboratory, \vhich I examined, gave a
marked indol reaction in Witte's peptone. Microscopically it
appears as a non-motile short rod with rounded ends. The
most characteristic growth of this organism is the creamy
layer containing gas bubbles which appears on potato. The
cultures which I examined did not always show the convex
isodiametric colonies ; very often the surface colonies grew out
with a thin irregular margin like the typical B. coli.

Bacillus Cavicida. — This organism was isolated by Brieger
from faeces. Its colonies on gelatine plates are said to be very
characteristic, showing a white growth, with concentric rings,
like the arrangement of the scales on a tortoise-shell. On
potato it forms a yellow slimy layer, and in sugar media
produces propionic acid. It does not liquefy gelatine. Micro-
scopically it appears as short rods, which do not form spores.
A culture of this bacillus from Krai's laboratory produced
gas in glucose-gelatine shake, and traces of indol in peptone
solution. It did not, however, coagulate milk. It appeared
to correspond to one of the varieties of B. coli.

Bacterium Tholoeideum was isolated by Gessner from the
intestinal canal of healthy men, and strongly resembles the
B. lactis aerogenes. In gelatine plates the surface colonies appear
as nail-headed, slimy, opaque growths ; later, they form large
expansions, lose their slimy character, and show a grey centre
surrounded by concentric rings. Under a lower power they are
circular, with a colourless, bright shining margin ; towards the
centre fine radiating brownish lines are seen. The deep
colonies are whetstone-shaped, of a yellowish-white colour.
Under a lo\v power they appear at first olive-green in colour ;
later, they become dark grey-green, and resemble date-stones in

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 97

shape. Gelatine-stab culture shows on the surface a moist
shining convex growth, and along the line of inoculation a
yellowish-white thick band, with closely packed, small, round
extensions having thick button-shaped ends. On potato it
forms a thick, yellowish expansion, and on one occasion bubbles
of gas were seen in a culture four days old. It does riot
liquefy gelatine, and microscopically it appears as a short rod
with rounded ends.

B. Neapolitanus. — This organism was discovered by
Emmerich in the blood and organs of cholera patients in
Naples. Weisser found it in the normal stools of human
beings and in air. On the surface it forms thin round colonies,
yellow in the centre but whiter at the margin ; in the depth
the colonies are whetstone in shape, and have a yellowish-
brown colour. It does not liquefy gelatine. In gelatine-stab
culture it grows like B. typhosus. On potato it forms a
slimy, yellowish-brown layer. Gas production has not been
observed. Microscopically it is about the same size as
B. typhosus, but it is not motile. It does not form spores,
and is decolorised by Gram.

Deeleman has described four coliform organisms, isolated from
urine and faeces, which are important from a diagnostic point
of view. The chief reactions of these organisms are as follows :

B. Coloides Virescens. — In gelatine plates the colonies,
under a low power, show a leaf-like picture, with irregular,
well-defined edges. On gelatine-streak there is a greyish-white
growth, which has a greenish iridescence ; the gelatine is not
liquefied. On agar plates there is a diffuse veil-like growth
without any markings, which later shows a feeble greenish
iridescence. On potato there is a yellow, slightly raised
growth, limited to the line of inoculation. In broth there is a
diffused growth with formation of a greenish iridescent
pellicle and a moderate deposit. It ferments grape-sugar,
cane-sugar, and milk-sugar. Milk is coagulated in twenty-four
hours. In 10 c.c. of litmus-whey, after ten days at 37° C., it
produces from 13'6 to 14*3 c.c. of N acid. It gives no indol
reaction. It is pathogenic for mice and guinea-pigs. It is not
stained by Gram's method, and microscopically appears as a
small motile bacillus.

98 BACTERIOLOGICAL EXAMINATION OF WATER.

B. Coloides Rubescens. — In gelatine plates under a low
power the colonies show a leaf-like pattern. On gelatine-stroke
culture there is a greyish- white growth with a reddish
iridescence ; the gelatine is not liquefied. On agar it forms a
diffused veil-like growth, which shows a reddish iridescence.
On potato there is a restricted yellow growth. In broth it
produces a diffused growth with a reddish pellicle. It ferments
grape-sugar and milk-sugar, but not cane-sugar ; milk is com-
pletely coagulated in twenty-four hours. In 10 c.c. litmus-whey,
after ten days at 37° C., it forms from 11'4 to TO c.c. N acid.
It gives a strong indol reaction. It is decolorised by Gram,
and microscopically appears as a small motile bacillus.

B. Vesicse. — In gelatine plates the colonies have a net-like
appearance, with a very delicate system of ridges and furrows.
On gelatine-streak there is a dull-grey opaque granular growth.
In broth there is a diffused growth, with a delicate pellicle on
the surface. It does not ferment grape-sugar, cane-sugar,
nor milk-sugar. It coagulates milk and produces indol. In
10 c.c. litmus-whey, after ten days at 37° C., it produces 7'5
to 8*75 c. c. of N acid. It is a small motile bacillus, which does
not strain with Gram.

B. Fsecalis Alkaligenes, variety. — In gelatine plates it
produces colonies like the B. vesicse. On gelatine-streak
culture there is a smooth, greyish-white growth, and the
gelatine is not liquefied. On agar it produces a greyish-white
growth. On potato it forms a circumscribed, slightly raised,
yellowish-brown growth ; on the fifth day there is gas pro-
duction, and later the potato substratum acquires a brown colour.
A diffused growth occurs in broth, and a tough pellicle is formed
on the surface. Cane-sugar and grape-sugar are fermented, but
milk-sugar remains unchanged. Milk is not coagulated, and
acquires an alkaline reaction. In 10 c.c. of litmus-whey, 8'8 to
11*0 c.c. of N alkali are produced after ten days incubation at
37° C. Indol is not formed. Microscopically it appears as a
small motile bacillus. It is decolorised by Gram.

METHODS EMPLOYED FOR THE ISOLATION OF THE B. COLI FROM
WATER SUPPLIES.

The various methods which are employed for the isolation of

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 99

B. coli are intended either to repress the growth of the ordinary
water-bacteria, so as to enable the B. coli to develop without
the chance of being crowded out by these microbes, or tc
provide a medium which will be so favourable to the growth
of B. coli that it will develop in advance of the competing
organisms.

The methods which are commonly used are as follows :

(1) Parietti's method. In this plan a series of broth tubes
are taken and inoculated with gradually increasing quantities of
the following mixture :

Carbolic acid ...... 5 grammes

Hydrochloric acid (pure) .... 4 grammes

Distilled water 100 c.c.

As a result of the admixture the broth tubes contain from
0'05 to 0*3 per cent, of carbolic acid. Many bacteriologists
omit the hydrochloric acid from the solution, and do not use
the tubes which contain more than from 0'15 to 0*2 per cent,
of carbolic acid. When the B. typhosus is supposed to be
present with B. coli it is wise to employ only the tubes which
contain 0'05 per cent, of carbolic acid; the reason for this
will be explained under the section devoted to B. typhosus.
The broth tubes are then seeded with varying quan-
tities of water, and incubated at 37° C. All the tubes which
show growth after one to three days incubation at 37° C. are
then plated out in ordinary or carbolic acid (0*05 per cent,
gelatine, and the colonies which develop are " fished " and
submitted to the various tests described under B. coli. In my
own experiments I found that B. coli, freshly isolated from
stools, would grow in 0*2 per cent, carbol-broth but if the bacillus
were allowed to remain in sterile sewage for six weeks it would
not grow in more than O'l per cent, carbol-broth.

(2) Vincent suggested cultivating the seeded broth tubes at
4&° C., at which temperature the B. coli will grow, but water
organisms are either unable to develop or do so very feebly.

(3) Fakes recommends that the water should be cultivated
under anaerobic conditions (Buchner's tube or hydrogen gas) in
glucose-formate broth (2 per cent, glucose, 0*4 per cent, sodium
formate) at 42° C.

100 BACTERIOLOGICAL EXAMINATION OF WATER.

(4) In the investigations of the State Board of Health,
Massachusetts, the method employed in testing waters for B.
coli varied slightly with the water under examination. For
waters in which B. coli was very rarely found, a qualitative test
was used ; while, for waters generally containing that organism
in some numbers, a quantitative method was used.

The qualitative test consists in inoculating a Smithes fer-
mentation tube (a bent tube with one end closed), containing
glucose-bouillon, with one cubic centimetre of the water. This
tube is incubated at 38° C., and if after twelve hours growth
gas has collected in the closed arm the tube is taken out and
species tests made. The absence of gas is considered final,
there being no B. coli in the water. In order to make the
species tests, some of the bouillon from the fermented tube is
taken out and diluted with sterile water; a portion is then
plated on lactose-litmus-agar. This plate is grown for twelve
hours at 38° C. ; if at the end of this time any red colonies are
found on the plate a few of these are fished, planted out on
agar-slopes, and grown for twelve hours as before. If the culture
resembles B. coli it is then sub-cultured in milk, nitrate solu-
tion, Dunham's solution (1 per cent, salt and 1 per cent. Witters
peptone), and Smith's solution (meat bouillon containing 2 per
cent, glucose and 1 per cent, peptone) ; a gelatine-stab is also
made. The Smith tubes, nitrate solution and milk, are grown
for twelve hours at 38° C. The Dunham's solution is grown at
the same temperature for three days, when it is tested for indol
in the usual manner. The gelatine must be grown at 20° C. for
at least seven days, as some species of bacteria resembling B.
coli liquefy gelatine very slowly.

In quantitative work, when dealing with waters in wrhich B.
coli is known to be present, the preliminary test in Smith's
solution is omitted. One cubic centimetre of the water is
plated direct on lactose-litmus-agar, and the plates grown for
twelve hours at 38° C. At the end of this time the plates are
taken out and the number of red colonies counted. The red
colonies are fished and sub-cultured as in the qualitative
method. The qualitative method is considered the more
delicate, as duplicate tests often showed B. coli present by the
gas test but absent by the plate method. This is probably due

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 101

to the influence of the slightly alkaline media, and to the fact
that better development is always obtained in fluid than in
solid media. The State Board of Health points out that " the
accuracy of these tests depends largely on attention to details,
time of growth, temperature, reaction, and composition of
media. B. coli will develop well at a temperature of 38° to
40° C. in from eight to twelve hours. The thermal death point
of most species of water-bacteria lies below that temperature,
and some species, which would grow, require more time, usually
thirty-six to forty-eight hours, for development."" According
to the Board the cultural characteristics of the Colon bacillus
are as follows :

Agar. — On the surface of glycerine-agar it produces a
luxuriant, smooth, moist, white growth, which is never stringy
to the needle.

Gelatine. — It grows well in nutrient gelatine, without lique-
fying that medium.

Lactose-litmus-agar. — In neutral agar, containing milk-
sugar and blue litmus, it grows rapidly, producing acid and
turning the litmus red around the colonies.

Milk. — Grown in milk, acid is produced and the milk is quickly
coagulated.

Smittis Solution. — In bouillon containing glucose fermentation
at once takes place; when the medium is placed in a bent tube with
one arm closed the gas is collected in the closed arm of the tube.

Nitrate Solution. — In nutrient solution, containing small
amounts of potassium nitrate, the nitrate is quickly and com-
pletely reduced to nitrite.

Indol. — In solutions containing small amounts of peptone
and salt, or in bouillon free from carbo-hydrates and containing
peptone arid salt, indol is produced. " A culture which gives
characteristic reactions with all of the above media belongs to
the Colon group undoubtedly.'''1

(5) Eisner and Holz recommended a medium consisting of
faintly acid potato gelatine for the differentiation of B. coli and
B. typhosus, and stated that water-bacteria did not grow well
on it. Further experience of this medium soon showed that
other bacteria were able to develop almost as luxuriantly as B.
coli. Kenny's modification of the potato medium has a constant

102 BACTERIOLOGICAL EXAMINATION OF WATER.

chemical composition, and B. coli forms very characteristic
colonies in it.

The methods above mentioned will be found discussed in
greater detail in the section devoted to the B. typhosus. My
own experience has shown that if a pure water supply, i.e., one
containing few water-bacteria, has been recently polluted with
fresh sewage, it is comparatively easy to isolate the B. coli by
any of the methods described. If, however, the water contains
a large number of water-bacteria and the pollution is not recent
or has been caused by old sewage, it is sometimes by no means
easy to isolate the B. coli from the supply. The B. coli
does not live indefinitely in water ; the numbers, as a rule, di-
minish rapidly unless considerable quantities of nitrates are
present. Consequently, if some time has elapsed since the
pollution, it may be necessary to examine considerable quan-
tities of the water in order to isolate the bacillus ; in other
words, the water has to be concentrated by pumping one or two
litres through a Berkefeld bougie, and the deposit on the candle,
after diffusion in 10 c.c. of sterile water, must then be examined.
But under these conditions the water-bacteria are present in
enormous numbers, and I have found that carbolised media are
quite unable to sufficiently restrain the growth of these organ-
isms unless about 0.15 to 0.2 per cent, of carbolic acid is used :
but if this amount is employed there is always a chance that
the B. coli, enfeebled by existence in sewage, may not be able
to develop. Under these conditions, I have obtained the best
results with Fakes1 method of cultivation, which combines
anaerobic conditions with incubation at 42° C. But even with
this procedure many organisms may be found besides B. coli, so
that it is necessary to make several dilutions, i.e., at least three
or four gelatine plates, from the glucose-formate broth tube
after twenty-four hours incubation at 42° C. It is also very
useful to plate the broth in litmus-lactose-agar and then sub-
culture the colonies which turn the blue litmus red.

In order to deal with larger quantities of water, Abba sug-
gested the following solution :

Lactose . ... . 200 grammes.

Peptone. . . . . 100 „

Sodium chloride ... 50 „

Water , 1.000 c.c.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 103

This is heated in the steam steriliser for half an hour, then
filtered and again sterilised. A litre of the suspected water is
taken arid 100 c.c. of this solution added to it, together with
0*5 c.c. of a 1 per cent, alcoholic solution of phenolphthalein,
and sufficient (about 2 to 3 c.c.) of a cold saturated solution of
sodium carbonate to render the mixture a permanent pink
colour. The mixture is divided into five or six Erlenmeyer
flasks and incubated at 37° C. At the same time sterile agar
plates are prepared and placed in the incubator with the flasks.
After twelve, sixteen, or twenty-four hours the fluid in one or
more of the flasks will be decolorised if the Colon bacillus be
present, and from the upper layer of the fluid a small loopful is
removed and rubbed over one of the sterile agar plates ; the
process is repeated for the other colourless flasks, the plates in-
cubated, and sub-cultures made from likely colonies.

THE VALUE OF THE B. COLI COMMUNIS AS A SIGN OF SEWAGE
CONTAMINATION OF A WATER SUPPLY.

The interpretation which is to be placed on the discovery of
B. coli in a water supply has been much debated. Kruse, in
1894, stated that the " Bacterium coli is in no way characteristic
of the faeces of men and animals. Such bacteria are found
everywhere — in the air, in earth, and in water from the most
varied sources." Freudenreich and Miquel believe that B. coli
can be found in any water if a sufficient quantity is cultivated.
Levy and Bruns, however, considered that the B. coli found in
pure waters differed from the B. coli derived from faeces in the
absence of pathogenicity . The chief arguments, therefore, which
have been advanced against the acceptance of B. coli as an
indication of sewage contamination are as follows : (1) It
is abundant everywhere; (2) it multiplies readily outside the
animal body ; (3) it occurs in the excreta of mammals and
birds as well as in the intestine of man. If B. coli be abundant
everywhere, it is strange that Houston only found it in four out
of twenty-one samples of soil derived from different sources, viz.,
from orchards, gardens, moorland, pasture, sand-pits, sea-shore,
fields, &c. It is true that nine of the samples yielded in phenol-
gelatine colonies which resembled those of B. coli ; but six of
these failed to respond to the gelatine shake-culture test, and

104- BACTERIOLOGICAL EXAMINATION OF WATER.

the remaining three failed in one or more of the reactions
considered characteristic of the typical B. coli. The statement
that B. coli exists abundantly in all waters and soils appears to
be based on a very elastic interpretation of the characteristics of
B. coli. That this is a probable explanation is rendered manifest
by the experiments just published by Weissenfeld. This observer
examined water from thirty good wells, and found B. coli in
1 c.c. of seven of the specimens but only in one litre of the
remaining samples. He, however, found B. coli in 1 c.c.
of the water from twenty-four out of twenty-six bad
wells. Weissenfeld gives the cultural characteristics of
B. coli as follows : — Superficial colonies on gelatine like
a vine-leaf; produces gas in sugar-agar-stab ; more or less
motile ; often not motile ; not stained by Gram's method. He
considers the souring of milk and the production of indol to be
of no value as distinguishing characteristics. These cultural
reactions, in my opinion, are not sufficient to differentiate the
B. coli from a large class of coliform organisms which are found
in very many waters from perfectly unpolluted sources. Whether
these coliform organisms are derived from a typical B. coli,
which by prolonged immersion in pure water has lost the power
of souring milk and producing indol, it is impossible to say ;
but the important point is that these organisms are found alone
and without any admixture of the typical forms of B. coli in
pure waters. Now when pollution of a water is produced by
sewage, probably at least half the cultures of B. coli present
will show all the typical characteristics of this organism, the
remaining cultures presenting the varying reactions of the
atypical members of this group. In my experience pollution
by sewage is never characterised by the presence solely of the
variety, which only produces gas in sugar media. The other
typical members of the group must also be present to warrant
the suspicion of sewage contamination.

The statement that B. coli multiplies outside the animal
body is only true under certain very favourable conditions.
There is every reason to believe that when the physical conditions
are unfavourable its growth is inhibited, and it soon loses its
vitality and dies out. Also it is not justifiable to assume that the
excreta of animals are harmless to man, and in any case a water

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 105

so polluted could not be considered desirable for drinking /
purposes.

Lastly, many English bacteriologists will not accept the presence
of the typical B. coli as an indication of sewage contamination
without reference to the amount of water in which it is found, n
Houston states that " it is not the mere presence of B. coli that
should tend to condemn a water, but its relative abundance
therein." This opens a difficult question, viz., What quantity
of water containing typical B. coli is to be considered indicative
of contamination with sewage ? Most ^clexLologisU would 1
agree that if the typical bacillus were found in 1 c.c. of the
water the supply ought to be condemned ; but if the bacillus
could not be .detected, except in such large quantities .as one
or two litres, the water might be passed as fit for drinking
purposes. Now, on examining a supply supposed to be polluted;
it often happens that typical B. coli cannot be found except in
50, 100, or 200 c.c. of the water. Is such a water to be con-
sidered contaminated ? Experimental investigations have shown
that when a water supply has been recently polluted by sewage,
even in a dilution of 1 in 100,000, it is quite easy to isolate the
B. coli from 1 c.c. of the water ; but if several weeks have
elapsed since the pollution occurred, the bacillus cannot be
isolated unless the water is concentrated. The purer the water
the more quickly the B. coli dies out; but, even at the end of
two months, I have always been able to find the bacillus in
200 c.c. of the water. Consequently I would say that a water
which contained B. coli so sparsely that 200 c.c. required to be
tested in order to find it, had probably been polluted with
sewage, but the contamination was not of recent date. Pakes1
work, however, led him to the following conclusions :

" Drinking-water from a deep well should contain no B. coli
in any quantity. Water from other sources which contains the
B. coli in 20 c.c. or less should be condemned; that which contains
the organism in any quantity between 20 and 50 should be
looked upon as suspicious, between 50 and 100 as slightly
suspicious, and only in greater quantities than 100 c.c. as probably
safe. If no B. coli can be obtained from the whole two litres
I should consider it absolutely safe." In the case of a recent
contamination of a drinking-water, there is no doubt that

106 BACTERIOLOGICAL EXAMINATION OF WATER.

B. coli affords a much more delicate test of pollution than any
chemical examination which can be made. Experiments have
shown me that when pure water is polluted with a highly
concentrated sewage the limit of the dilution in which the
contamination can be detected chemically is 1 in 1000. Dilutions
of 1 in 10,000 and 1 in 100,000 showed no chemical signs of
pollution ; but when examinations were made for B. coli, this
organism was detected with the greatest ease in 1 c.c. of the
1 in 100,000 dilution.

Klein and Houston examined eight samples of sewage obtained
from different sources in various dilutions. They found that,
as a rule, a dilution of 1 in 1000 could be detected chemically,
but the dilutions of 1 in 10,000 and 1 in 100,000 gave no
chemical indication of pollution ; but in all the dilutions the
presence of B. coli was usually demonstrated directly without
resorting to the " filter brushing method."

The statement of Levy and Bruns that B. coli from stools
are pathogenic, whereas those found in pure waters are not,
has been re-investigated by Weissenfeld. Blachstein, in
1893, pointed out that if a tube of broth be inoculated with
1 c.c. of good drinking water and incubated at 30° C. for
forty-eight hours, 2 c.c. of the mixture can be injected
intra-venously into a rabbit or intra-peritoneally into a guinea-
pig without inconvenience to either animal. If, however, the
water be from a contaminated source, such as the Seine, both
animals, as a rule, rapidly succumb. The lethal effects of the
impure water were attributed to the B. coli and its varieties.
Sims Woodhead and Cartwright Wood injected guinea-pigs
subcutaneously instead of intra-peritoneally, and took the
presence or absence of local reaction as the test of contamination
or purity because they found it was quite the exception for
water supplies in England to be so contaminated as to exert a
lethal action. Even when the organisms present in tap-water
were allowed to grow for weeks in broth and so heap up their
products, they produced practically no effect.

Weissenfeld's experiments were made either with pure broth
cultures of the B. coli, which he isolated from the wells
before mentioned, or else with a mixture of the waters and
broth cultures as in Blachstein^s experiments ; 1 c.c. of a

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 107

two-day s-old culture was injected intra-peritoneally into guinea-
pigs of medium weight. He found that varieties of B. coli
isolated from pure waters sometimes killed the guinea-pigs,
while some of the varieties isolated from the impure waters
had no effect on the animals, and therefore concluded that the
isolation of a virulent B. coli from a water supply did not
necessarily indicate that the water had been contaminated with
faeces.

CHAPTER IX.

QUALITATIVE ANALYSIS — continued. "

The Bacillus Enteritidis Sporogenes.

Ix October 1895, on the occasion of an epidemic outbreak of
diarrhoea amongst patients of St. Bartholomew's Hospital,
Klein isolated an anaerobic spore-bearing bacillus — B. enteritidis
sporogenes — which when injected subcutaneously into guinea-
pigs, caused acute fatal disease with sanguineous exudation.
Andrewes discovered the same bacillus in the milk supplied to
the patients and in the bowel discharges of severe cases of
diarrhoea ; he did not find it, however, in the " evacuations of
casual commonplace diarrhoea." Klein also stated in his first
report that the spores of the microbe could not be found in
the intestinal evacuations of healthy individuals. Later
investigations showed that the B. enteritidis sporogenes has a
wide distribution in nature, being found in the evacuations of
cases of diarrhoea, in sewage, water, soil, dust polluted with
sewage, and in horses1 dung. Klein now states that most
specimens of sewage contain on an average about 500 to
600 spores of this organism per cubic centimetre. The
B. enteritidis sporogenes is an anaerobic organism, and can
be best isolated by Klein's method, which is as follows :

" A small portion of any material suspected to contain the
organism is placed in a test-tube containing 15 c.c. of recently
sterilised milk ; this is heated to 80° C. for ten to fifteen minutes,
and then cooled. The milk-tube is next placed in a large
cylindrical test-tube (Buchners cylinder) containing about 120
grains of pyrogallic acid, about 10 c.c. of strong liquor
potassae are added, and the Buchner test-tube closed as
quickly as possible with a well-fitting india-rubber stopper,
which is then sealed with melted paraffin. The whole

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 109

apparatus is then incubated at 37° C. ; the next day, or at
latest after thirty to thirty-six hours, the milk will be found
changed in a characteristic manner. The cream is torn or
altogether dissociated by the development 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 formerly occupied by the milk
now contains a colourless, thin, watery whey, with a few casein
lumps adhering here and there to the sides of the tube. When
the tube is opened the whey has a smell of butyric acid, and is
acid in reaction. Under the microscope the whey is found
to contain numerous rods, some motile, others motionless."

The microbe, when obtained from a typical milk culture, con-
sists of rods, measuring from 1'6 to 4*8 ju in length and 0'8 u.
in breadth ; it may form short chains. In respect of length
and thickness the bacilli closely resembles the B. aiithracis ;
they are distinctly thicker than the bacilli of malignant oedema.
The bacilli form spores, but they are not seen in the milk
cultures; in liquefying gelatine cultures, in serum cultures,
and in the subcutaneous exudation from inoculated animals,
spores under certain conditions are readily formed. When
fully developed they are oval bodies, 1'6 fj. in length, and
as much as I/O n or even 1'2 /u in thickness. The B. enteritidis
sporogenes is pathogenic; if 1 c.c. of the whey produced
in the milk culture be injected into the groin of a guinea-pig,
weighing from 200 to 300 grammes, swelling appears in the
groin after six hours, which gradually extends on to the abdomen
and thigh, usually the animal is found dead after eighteen to
twenty-four hours. The skin then will be found separated from
the muscles over a considerable area, and the subcutaneous
tissue gangrenous, containing an offensive sanguineous
exudation. If the examination is made immediately after
death the exudation will be found to contain numerous
cylindrical bacilli ; but if the examination is deferred for
twenty-four hours, large, oval or egg-shaped spores will be
found, both free in the fluid and at the end of the bacilli. The
microbe grows well in 2 per cent, glucose-agar under anaerobic
conditions. In this medium there is very marked gas formation ;
the colonies in the depth are round, white in reflected light,

110 BACTERIOLOGICAL EXAMINATION OF WATER.

but brown and granular in transmitted light. The agar is
not liquefied, and the bacilli do not form spores. On the
surface of agar anaerobic plates, after twenty-four to forty-eight
hours incubation the colonies are circular, flat, moist, grey
and translucent. " Examined under a low power, each colony
is a little thicker and less transparent in the middle part,
thinned out and more transparent in the marginal portion.
At the same time the colony is distinctly granular, the granules
being placed closer and denser in the middle part, and less
close in the marginal part." Spores are not formed in these
colonies. In glucose-gelatine the bacillus produces gas arid
causes liquefaction of the medium. When spores are planted
in grape-sugar gelatine, which is then heated to 80° C. for ten
to fifteen minutes, and then cooled, sealed, and incubated at
22° C., fine, dot-like colonies appear in the lower part of the
tube, which after forty-eight hours form spherical liquefied
masses of gelatine. If bacilli, free from spores, are planted in
sugar-gelatine, the colonies appear non-liquefying, or liquefy
very late and slowly. When tested as to its staining
reactions, the bacillus is found to be stained by the ordinary
basic aniline dyes, and also by Gram's method.

The above description gives the chief characteristics of the
typical B. enteritidis sporogenes ; but Klein found sometimes
that the spores from subcutaneous exudation, serum and
gelatine cultures, when placed in milk and heated to 80° C. for
ten to fifteen minutes, produced no change at all in the milk for
the first two days. About the third or fourth day, however, the
milk began to clear under the unchanged layer of cream. By
the end of a week or ten days the milk culture showed on the
top a layer of unaltered cream, below which the medium
consisted in the main of a clear or slightly turbid pale yellow
watery fluid ; at the bottom of the tube there was a white mass
of caseine. The whey from this milk culture was found to be
offensive, alkaline in reaction, and to contain bacilli and spores
which had no pathogenic action. Such an atypical milk culture
was at first considered to be produced by B. enteriditis
sporogenes which had lost its virulence. But further experi-
ments showed that the appearances produced in the atypical
milk culture were caused by a new bacillus, called by Klein the

QUALITATIVE BACTERIOLOGICAL ANALYSIS. Ill

B. cadaveris, which resembles the B. enteritidis sporogenes, in
that it is an obligatory anaerobe. It produces gas in glucose-
agar without causing liquefaction, but liquefies glucose-gelatine
and solidified blood serum ; it forms drum-sticks and terminal
spores, and is conspicuously motile. Striking differences
between the two organisms are seen when anaerobic cultures
are made on the slanting surface of ordinary agar or on a
surface anaerobic plate culture of agar. It is then found that
the B. cadaveris forms angular dark granular colonies, from
which extend numerous filamentous processes, which later on,
as incubation proceeds, form a more or less densely reticulated
mass of considerable extent ; at the same time this dark
granular material is embedded on a lobulated, finely granular,
translucent, filmy basis. In some cultures the filmy basis is
absent altogether. On the other hand the colonies of
B. enteritidis are circular discs, and there is never the reticulated
expansion from the centre, which is seen in the colonies of the
B. cadaveris. The B. cadaveris also forms spores on agar and
in milk, which is never the case with B. enteritidis. The milk
culture of B. cadaveris also has a putrid odour, and its reaction
is amphoteric , or perhaps a little alkaline. " The milk cultures,
liquefied gelatine cultures, agar cultures and liquefied serum
cultures of the B. cadaveris are without pathogenic effect on
guinea-pigs ; as much as 2 and 3 c.c. of a liquefied
sugar-gelatine, or of a milk culture, can be injected without
physiological effect." According to Klein, the B. enteritidis can
be distinguished from the bacillus of malignant oedema, &c., by
the following points :

B. Enteritidis Sporogenes.

(a) It is thicker ; (b) spores at the end of the bacillus ; (c)
flagella chiefly in a bundle at the end of the bacillus;
(d) gangrenous condition of the tissue at site of injection in
a guinea-pig ; (e) butyric acid smell and acid reaction of milk
culture, which does not contain spores ; (f ) stains with Gram ;
(g) gelatine-stab has not lateral off-shoots.

Bacillus of Malignant (Edema,
(a) Thinner ; (b) spores in the middle of the bacillus ;

112 BACTERIOLOGICAL EXAMINATION OF WATER.

(c) flagella disposed at the sides, not specially at the end of the
bacillus; (d) subcutaneous tissue haemorrhagic ; (e) gas not
formed so abundantly in milk culture ; slow separation of the
fluid and coagulated part, which commences near the cream ;
the whey is not acid, has no butryic acid smell, and contains
spores ; (f) it does not stain with Gram ; (g) gelatine-stab
possessed of many short lateral off-shoots.

Bacillus of Symptomatic Anthrax.

(a) It does not stain with Gram ; (b) gelatine-stab has many
short lateral off-shoots ; (c) it does not produce subcutaneous
gangrene, but a haemorrhagic swelling containing gas bubbles.

Bacillus Butyricus of Botkin.

This bacillus resembles the B. enteritidis sporogeues morpho-
logically and culturally in sugar-gelatine, in agar, and particu-
larly in milk ; but it is without pathogenic action when injected
in large doses subcutaneously into guinea-pigs or rabbits.

The characteristics and means of diagnosis of the B.
enteritidis sporogenes have been considered in detail, as many
bacteriologists claim that the discovery of this bacillus has given
a fresh impetus to the bacteriological examination of water.
Houston states : u It is safe to anticipate that this discovery is
destined to largely enhance the value of the bacteriological test
of potable waters. For the spores of this anaerobe cannot
be demonstrated in pure water (by this is meant that the
bacillus is absent, not only in 1 c.c. of a pure water, but that
it may be absent in 500 c.c. or more of such water), whereas
in impure water it can readily be shown to be present, and in
raw sewage it is specially abundant. My own records, which
extend over a year, and deal with a very large number of
samples of sewage, show that the spores of this microbe are
present in numbers varying usually from 100 to 1000 or more
per cubic centimetre. It cannot fairly be said of this anaerobe
that it is likely to multiply outside the animal body, and the
only possible objection to accepting its presence as an index of
dangerous contamination of water is the fact that its being a
sporing anaerobe weakens somewhat its usefulness as evidence
of recent and, therefore, presumably specially dangerous,

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 113

pollution. Nevertheless, and as far as can be seen at present, it
not only equals, but far exceeds any known test, chemical or
bacteriological, for inferring the wholesomeness or otherwise of
a drinking water." Klein and Houston examined eight
specimens of sewage as to the total number of microbes, the
number of B. coli, and the number of spores of virulent
B. enteritidis contained in 1 c.c. of the fluid. Some of their
results are given in the following table :

Sample of sewage.

No. of microbes
per 1 c.c. of crude

Xo. of B. coli
per 1 c.c. of crude

No. of spores of
virulent B. enteri-
tidis per 1 c.c. of

sewage.

sewage.

crude sewage.

Chiefly domestic .

14,240,000

260,000

2000

Mixed ....

7,800.000

180,000

200

Chiefly domestic .

4,800^000

500,000

2000

Mixed with a large

amount of trade

liquid ....

36,000,000

1,100,000

400

Hospital

2,800,000

200,000

30

Domestic," with a

large amount of

trade liquid

4,100,000

500,000

56

Wholly domestic

28,000.000

2,000,000

50

Mixed ....

21,000,000

1,000,000

35

From this table it appears that there is no definite relation
between the total number of bacteria and the number either of
B. coli or of spores of B. enteritidis ; also there is no parallelism
between the number of B. coli and the number of spores of
B. enteritidis. It is also evident that to detect this organism
in a polluted water it may be necessary to resort to " the filter
brushing method," and to concentrate the water to small bulk.
Unless the pollution is very great, i.e., a dilution of sewage
less than 1-1000, it may be impossible to detect B. sporogenes
in 1 c.c. of a water polluted with sewage containing even
2000 spores per c.c. But as some specimens of sewage
contain only from 30-35 spores per c.c., a considerable
quantity of the water must be examined in order to detect the
B. enteritidis. As a matter of routine, it would seem advisable
to concentrate always at least 2000 c.c. of a polluted water,
by pumping through a Pasteur's filter, and diffusing the deposit
in 5 c.c. of sterile water. One c.c. of this should be added
to each of five milk tubes, which are then heated to 80° C. for

H

114 BACTERIOLOGICAL EXAMINATION OF WATER.

ten to fifteen minutes, cooled, and cultivated anserobically in
Buchner's tubes. If the typical changes are seen in any one
of the milk tubes, the presence of B. enteritidis is probable ;
but to obtain a decisive result the pathogenic action of the
organism must be tested by injecting 1 c.c. of the whey into
a guinea-pig. The fact that Botkin's bacillus butyricus exactly
resembles the B. enteritidis, culturally and morphologically,
renders the animal experiment absolutely imperative ; unfortu-
nately, the necessity of resorting to an animal experiment
militates against the extensive use of this test in ordinary
bacteriological examinations for hygienic purposes. Still, if
the typical appearances in milk are found in conjunction with
the presence of typical B. coli in a small quantity of the water,
the fact of sewage contamination may be considered sufficiently
proved for all practical purposes.

CHAPTER X.

QUALITATIVE ANALYSIS— continued.

SPECIES OF STREPTOCOCCI AND STAPHYLOCOCCI.

THE value of these organisms as indicative of sewage con-
tamination was prominently brought forward by Houston in
the supplement to the report of the Local Government Board
for 1898-99. Laws and Andrewes, however, in 1894, pointed
out that a small streptococcus was the commonest organism
present in fresh sewage from St. Bartholomew's Hospital. The
colonies of this organism were minute, grew slowly, and possessed
the power of coagulating milk in twenty-four to thirty-six
hours, when incubated at 37° C. In a specimen of sewage from
Snow Hill the same streptococcus was very abundant. An old
specimen of sewage from Barking showed numerous colonies of
the streptococcus and a yellow staphylococcus, but did not
appear to contain any B. coli. This last observation is of the
greatest importance in relation to the contamination of potable
waters by old sewage. In my own work I have repeatedly
examined waters which, from their local surroundings, must
have been polluted with sewage, and yet I have been unable
to find the slightest traces of B. coli in them ; streptococci
were, however, nearly always present. On considering the
question, I thought that possibly B. coli died out in sewage
kept under the conditions of an old and little-used cesspool ;
and on testing the point experimentally I found that B. coli
gradually disappeared from many specimens of sewage kept
in the dark at the temperature of an outside verandah. But
the commonest organisms which persisted in these old specimens
of sewage were varieties of streptococci and staphylococci, and I
proceeded to examine the cultural reactions of the commonest
types of these microbes, believing that their isolation might be

116 BACTERIOLOGICAL EXAMINATION OF WATER.

of great service in solving the difficult problem of sewage con-
tamination. The following are the chief types observed up to
the present time :

Streptococcus A.

Gelatine Plates. — The surface colonies were extremely small,
with an irregular margin, from which here and there "streamers'"
ran out over the surface of the gelatine. An "impression"
preparation showed that the colony consisted of small cocci
tightly packed together at the centre, but forming short and
long chains at the periphery. The gelatine was not liquefied.

Agar-slope. — The growth consisted of numerous small isolated
colonies with a circular outline and granular centre.

Gelatine-stab. — On the surface there was a very small circum-
scribed film, and a fine thin growth along the stab consisting of
small colonies massed together. The gelatine was not liquefied.

Mill:. — Clotted after six days incubation at 37° C.

Litmus-whey. — Faintly acid after twenty-four hours incubation

at 37° C. The acidity required I'l c.c. per cent, of ^ alkali
to neutralise it. After seven days incubation the acidity equalled
2 '4 per cent. ^ alkali.

Broth. — After twenty -four hours incubation at 37° C. a diffuse
growth appeared with a slight deposit. After ten days the
broth was not clear.

Peptone and Salt Solution. — After seven days incubation at
37° C. no indol was produced.

Glucose-gelatine. — No gas formation.

Gelatine (ordinary stab) at 37° C. — A diffused growth and
slight deposit.

Potato. — Moist glistening surface ; growth hardly apparent.
A portion examined under the microscope showed small cocci
in chains and masses.

Microscopical Characters. — A twenty-four-hour broth culture
showed medium-sized cocci arranged in short chains and small
masses.

Staining Reactions. — Stains with basic aniline dyes and by
Gram's method.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 117

Streptococcus B.

Gelatine Plates. — Small circular colonies, with clear sharp
edges, thicker in the centre than at the periphery ; under a low
power granular, but no streamers observed. Gelatine was not
liquefied.

Agar -slope. — The growth consisted of numerous small isolated
circular colonies.

Gelatine-stab. — On the surface a very small circumscribed
film, and along the stab a fine growth consisting of small
colonies massed together.

Gelatine-stab at 37° C. — After twenty-four hours incubation
at 37° C, the tube was nearly clear with a thick deposit at the
bottom.

Glucose-gelatine. — No gas formation.

Milk. — Coagulated after forty-eight hours incubation at
37° C.

Litmus-whey. — After twenty-four hours incubation at 37° C.

the acidity equalled 3'3 per cent. -^ alkali ; after seven days

incubation the acidity equalled 1 3 per cent. -^ alkali.

Peptone and Salt Solution. — No formation of indol.

Broth. — A diffused growth with a marked deposit at the
bottom.

Potato. — Moist, glistening surface, growth hardly apparent.

Staining1 Reactions. — Stains with Gram's method.

Microscopical Characters. — Twenty-four hours broth culture
showed chains consisting of seven to eleven cocci.

Streptococcus C.

Gelatine Plates. — Small circular colonies with clear, sharp
edges, thicker at the centre than the periphery ; no streamers
observed ; gelatine not liquefied.

Agar-slope. — Very small circular colonies, uniting below to
form a thin greyish-white growth.

Gelatine-stab. — On the surface a very small circumscribed
film, and along the stab a very fine growth, consisting of small
colonies massed together.

118 BACTERIOLOGICAL EXAMINATION OF WATER.

Gelatine-stab at 37° C. — Diffused growth through the tube.

Glucose-gelatine. — No gas formation.

Milk. — Coagulated in forty-eight hours.

Litmus-whey. — After twenty-four hours incubation at 37° C.

the acidity equalled 5 per cent. ~ alkali. After seven days

incubation at 37° C. the acidity equalled 14 per cent. ^ alkali.

Peptone and Salt Solution. — No formation of indol.

Broth. — A diffused growth with deposit at the bottom of the
cube ; after ten days growth the tube was still turbid.

Potato. — Glistening transparent growth.

Staining Reactions. — Stains with Gram's method.

Microscopical Characters. — Twenty-four hours broth culture
showed very long chains, consisting of at least thirty cocci.
NOTE. — Streptococcus C. was only distinguished from B. by the
small size of its colonies on agar and the extreme length
of the chains in the broth culture.

Streptococcus D.

This culture very closely resembled streptococcus C. in the
size of its colonies on agar, its re-actions in milk, litmus-whey,
and glucose-gelatine, and on potato. It, however, only pro-
duced very short chains in a twenty-four hours broth culture,
and gave an indol reaction in peptone.

Streptococcus E.

Gelatine Plates. — Colonies circular, larger and more opaque
the above varieties, consisting of very small cocci.

Agar-slope. — A thick, white, opaque growth ; at the margin,
large, white, and opaque, discrete colonies like those on the
gelatine plates.

Gelatine-stab. — Thick, white, opaque growth on the surface
and along the stab. After fourteen days at 22° C. gelatine
slowly liquefied.

Gelatine-stab at 37° C. — Tube clear, with a deposit at the
bottom after twenty-four hours incubation.

Broth.— At the end of twenty-four hours at 37° C. faintly

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 119

turbid ; after forty-eight hours the broth cleared and a deposit
formed at the bottom ; on shaking the tube, long thin spiral
threads rose up from the bottom.

Milk. — Unchanged after ten days incubation.
Litmus-whey. — No acid produced. After ten days incuba-
tion at 37° C. the reaction was slightly alkaline.
Glucose-gelatine. — No gas formation.

Peptone and Salt Solution. — After seven days incubation at
37° C., a marked indol reaction.

Potato. — At first only a thin glistening film, but after ten
days incubation at 37° C., a distinct yellow growth was
apparent.

Staining' Reactions. — Stained by Gram's method.
Microscopical Appearances. — A twenty-four hours broth
culture showed short chains and masses of very small cocci.
NOTE. — This organism seemed to be intermediate between the
streptococci and the staphylococci. In its growth on
potato and liquefaction of gelatine it resembled sta-
phylo-cocci. But the growth in broth, milk, and lit-
mus-whey, and the microscopical appearances showed that
the culture was also closely related to the Streptococcus
pyogenes.

None of the cultures when injected into guinea-pigs produced
any effects, either constitutional or local.

Staphylococcus P.

Gelatine Plates. — Large, white, opaque, circular colonies, which
later acquired a lemon-yellow colour, and slowly liquefied the
gelatine.

Agar-slope. — A thick white growth, which later acquired a
lemon-yellow colour.

Broth. — Diffuse growth, with a slight deposit at the bottom.

Milk. — Unchanged.

Potato. — A rather dry, light yellow growth.

Litmus-whey. — No acid produced.

Peptone and Salt Solution. — No indol produced.

Glucose-gelatine. — No gas formation.

Gelatine-stab at 37° C. — A diffused growth with a small
deposit.

120 BACTERIOLOGICAL EXAMINATION OF WATER.

Gelatine-stab at 22° C. — White thick growth on the surface
and along the stab ; later the growth acquired a light yellow
colour, and the gelatine slowly liquefied.

Staining Reactions. — Stained by Gram's method.

Microscopical Characters. — Large cocci in masses and in pairs ;
no chains observed.

This organism appeared to be closely allied to the Staphylo-
coccus citreus.

As a result of my experiments, I arrived at the conclusion
that streptococci presenting the above characteristics might
be extremely useful to the water-bacteriologist ; for if they
were found in the absence of B. coli, they would probably
indicate pollution of the water-supply by old sewage.

Houston, however, is inclined to regard streptococci as
indicating recent and objectionable pollution. He states that
streptococci " as a class may be thought of as germs especially
liable to discouragement by unfavourable physical conditions ;
and, indeed, as surviving only when the conditions are almost
ideally propitious. In the present state of our knowledge,
therefore, the presence of streptococci in a substance, be it soil
or sewage or water, suggests recent association of certain
ingredients of that substance with an animal host.1' The
following descriptions of streptococci, which were isolated
from water, are given to show the characteristics of the
organisms described by Houston.

Streptococcus I9. Surface Colonies on Gelatine Plates. —
Yellowish-grey, transparent, showing wavy granulation, edge
sinuous, and made up of loops of streptococci. No lique-
faction. Surface-agar Plates. — Yellowish-brown, semi-trans-
parent colonies of irregular shape and small size. The chains
of cocci give the colonies a wavy, granulated appearance. The
loops of cocci are readily seen at the periphery. Broth.—
Clear ; at foot of tube there are woolly masses, which tend to
cohere, and are somewhat viscous. Litmus-mill:. — Practically
no change even after six days. Gelatine-slope. — Trans-
parent coarsely granular colonies, which tend to remain
separate and not form a continuous growth. No liquefaction.
Morphology. — Stains with Gram. Long chains of cocci ; also
masses made up of chains.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 121

Streptococcus, J10. Surface- gelatine Plates. — Colonies
minute, circular, transparent, yellowish -grey in colour, showing
very faint granulation. The edge is clean, there are no loops
visible, no liquefaction. Surface-agar Plates. — The colonies
are darker, less transparent, and the granulation, which is wavy
in character, is much more marked. The colonies are usually
more or less circular in shape ; the edge is usually clean. Broth.
— Diffuse but not very abundant cloudiness in twenty-four
hours. At foot of tube a white and somewhat viscous deposit.
Litmus-mill;. — In three days distinct acidity but no clot; no
clot sixth day. Gelatine-slope. — Minute, circular, transparent
colonies. Morphology. — Stains with Gram. Cocci chiefly in
adherent masses, but also separate chains.

Streptococcus Kn. Surf ace- gelatine Plates. — Minute,
transparent, yellowish-grey, nucleated, faintly granular colonies
with clean edge ; no liquefaction. Surface-agar Plates. The
colonies are darker and less transparent and show a wavy and
coarse granulation ; some of the colonies are quite circular with
clean or only very slightly frayed edge. Other colonies are
irregular in shape with a broken-up edge. Broth. — Remains
quite clear, on the sides of the tube and at the foot fluffy white
masses are to be seen, which are of a somewhat viscous nature.
Litmus-milk. — Practically no change visible after eight days at
37° C. Gelatine-slope. — Minute circular transparent colonies.
Morphology. — Stains with Gram. Cocci chiefly in masses, many
of them of large size and of irregular shape ; ? also chains.

Streptococcus L12. — This appeared to be practically the
same as Kn, but it grew more slowly.

Streptococcus C3. Surface-gelatine Plates. — Circular, trans-
parent, minute, faintly granular, yellowish- white colonies. No
loops of cocci visible at the edge; of the colonies; no liquefac-
tion. Surface-cigar Plates. — More or less circular minute
colonies of yellowish-white colour, faintly granular, with un-
broken edge. Broth. — Only slight diffuse cloudiness, not much
deposit at the foot of the tube. Litmus-milk. — Very feeble
acidity, no clot after fifteen days at 37° C. Gelatine-slope.—
Minute greyish-white circular colonies, nearly transparent.
Morphology. — Stains with Gram. Chains of cocci of medium
length, not matted in masses.

122 BACTERIOLOGICAL EXAMINATION OF WATER.

Houston also isolated seven other varieties of streptococci
from " soil-washings " which were characterised by colonies with
clean edges. The broth cultures were clear, with a deposit at
the bottom of the tube. Four of the organisms produced
strong acidity but no clotting in milk. None of the cultures
showed pyogenic or pathogenic properties.

In the third Report to the London County Council on the
Bacterial Treatment of Crude Sewage, Houston stated that the
type of streptococcus most abundant in sewage presented the
following characteristics : —

Sewage Streptococcus (Houston).

Source. — Barking and Crossness crude sewage and effluents
from bacterial coke-beds.

Abundance. — Usually more than 1000 per c.c. of crude
sewage and effluents.

Temperature. — Grows well at blood heat.

Morphology. — Stains well by Gram's method. The chains of
cocci are usually short.

Agar and Gelatine-plate Cultures. — The colonies are small and
transparent. They are nearly circular in shape with a clean
edge. The granulation is faint. The colonies in agar are
usually more granular and darker-looking than in gelatine
cultures. The gelatine is not liquefied.

Streak Cultures. (Agar at 37° C. and gelatine at 20° C.) —
The growth usually shows itself as a delicate white film, which
on close observation is seen to be made up of numerous separate
minute transparent-looking colonies. The gelatine is not lique-
fied.

Broth Cultures (37° C.). — Abundant diffuse cloudiness. On
gently shaking the tube a viscous white deposit rises from the
foot of the tube in a spiral form.

Litmus-mill: Cultures (37° C.). — Acidity, but usually no clot.

Roscoe and Lunt isolated from sewage a streptococcus, which
appears closely allied to the varieties I have already described.
The cultural characteristics of this organism are as follows :

Streptococcus Mirabilis (Roscoe and Lunt).

Gelatine Plates. — It grows badly. The colonies in the depth
of the gelatine after four days incubation are mere dots or

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 123

gnarled and convoluted thread-like masses. Surface colonies
exhibit a faint and transparent expansion. Under a low power
the colony is seen to consist of a mass of long threads, sometimes
throwing out processes into the surrounding gelatine, which is
not liquefied.

Gelatine Tubes. — Produces a faint transparent film almost
invisible to the naked eye.

Agar. — Growth resembles that on gelatine.

Potatoes. — Growth inappreciable.

Broth. — In forty-eight hours a fine mass resembling delicate
cotton-wool collects at the bottom of the tube and the broth
above remains perfectly clear.

Microscopical Characters. — Streptococci forming very long
chains. The individual cocci are 0'4 /* thick. It is not motile.

Three other varieties of streptococci have been described,
which differ from the above in that they liquefy gelatine. They
were not very fully investigated by their discoverers, so that it
is difficult to follow the descriptions. They are as follows :

The Streptococcus Albus. — This organism was isolated by
Maschek from water. On gelatine plates it forms flat expansions
with a white periphery ; under a low power a small dark yellow
cloud is visible in the centre. Liquefaction proceeds rapidly.
In gelatine tubes it forms a flat expansion, which rapidly liquefies
the gelatine and produces a white deposit. On potatoes it grows
rapidly, producing a slimy expansion.

The Streptococcus Vermiformis. — This organism was also
isolated from water by Maschek. On gelatine plates it forms
yellowish- white centres, which sink into the gelatine. The
centre is light, whilst the periphery is composed of a dark ring.
Under a low power the contents of the colony are granular,
whilst the rim exhibits a radiated structure. It liquefies the
gelatine very rapidly. On potatoes it forms a dirty yellow
expansion, which grows very rapidly. The individual cocci
are almost always arranged in filaments which show a slow
vermiform movement.

The Streptococcus Coli Gracilis. — This streptococcus was
found by Escherich in the intestinal canal and faeces of flesh-
eaters. It consists of cocci with an average diameter of
0*2-0'4 fj. ; in recent gelatine cultures it shows S-shaped forms

124 BACTERIOLOGICAL EXAMINATION OF WATER.

consisting of six to twenty cocci. On gelatine plates it forms,
at first, round, sharply-outlined colonies; later they are dark
and surrounded by a broad clear funnel of liquefaction. In
gelatine stab there is a funnel-shaped liquefaction. After
six to eight days, at the bottom of the completely liquefied
gelatine there appears a white, finely granular deposit. On
agar there is a scanty surface growth. On potato a scanty
growth appears in the form of small white raised points. It
grows best at blood-heat, and gas production has not been
observed.

The varieties of streptococci which I have found most
commonly in polluted water-supplies correspond to the strepto-
coccus B., streptococcus E., and the "sewage streptococcus11
of Houston. In many of the waters which contained these
organisms the B. coli was not detected, and an opinion was
formed that the waters in question had been polluted with old
sewage ; the opinion was usually found to be justified by a local
examination of the supply. My experience does not support
the contention that streptococci probably indicate a dangerous
contamination. As I have already mentioned, the sewage
streptococci appear to maintain their vitality in sewage for a
much longer time than B. coli. Specimens of barrack sewage
preserved in a laboratory cupboard for three months and then
diluted 1-100 or 1-1000 with tap-water, show, when examined
by the usual methods, large numbers of streptococci but few
or no B. coli. Old sewage from which B. coli has disap-
peared will be unlikely to present conditions favourable to
the prolonged vitality of the B. typhosus or Sp. cholera?.
Consequently it appears that streptococci alone cannot be
considered as necessarily indicating a dangerous contamina-
tion. It is true that when the dilutions of old sewage are
kept for a few days the streptococci rapidly disappear ; so
their presence in a water-supply undoubtedly indicates a
recent contamination, but the contamination is not necessarily
dangerous unless the streptococci are accompanied by B. coli.

CHAPTER XL

QUALITATIVE ANALYSIS— continued.

THE PROTEUS GROUP.

MEMBERS of the proteus group are often found in sewage, but
are very rare in pure waters. They do not necessarily indicate
sewage contamination, but if they are found in conjunction
with B. coli there is very little doubt that such pollution has
occurred. The three chief members of the group were first
isolated from putrid meat by Hauser.

Proteus Vulgaris.

Gelatine Plates. — The young colonies are very variable in
outline ; they are generally yellowish-brown in colour and
rapidly liquefy the gelatine. Under a low power the edge of
the colony may be seen to be set with fine bristles, which appear
to be continued into the centre of the colony ; from the margin
wandering processes grow out over the gelatine, and in these
the bacilli are seen to be packed side by side. Sometimes
instead of the colony with a bristly edge, there appears an oval
or spindle-shaped form from which processes or " swarmers,"
often bent on themselves, run out all over the gelatine. In the
depth of the gelatine oval colonies are seen in which the bacilli
appear packed in concentric circles. Zooglcea forms are also
common. After twenty-four hours incubation the plates to the
naked eye often show nothing but circular or irregular pits of
liquefaction with slimy contents.

Gelatine-stab. — The gelatine is rapidly liquefied in a more or
less funnel-shaped manner, and at the bottom of the liquefac-
tion an opaque flocculent mass is seen.

Agar. — A moist greyish-white growth.

Potato. — A dirty yellowish-white moist growth.

126 BACTERIOLOGICAL EXAMINATION OF WATER.

Glucose-gelatine. — No gas formation.

Broth. — Uniformly turbid.

Peptone and Salt Solution. — No indol reaction produced.

Milk. — Unchanged.

Nitrate-broth. — Nitrites and ammonia produced.

Litmus-ivhey. — Reaction becomes slightly alkaline.

Microscopical Characters. — Thin motile rod varying greatly
in length. It does not form spores and grows best at from
20° to 24° C. It is decolorised by Gram's method of
staining. Involution forms are often seen.

Proteus Zenkeri.

Gelatine Plates. — The surface colonies are yellowish- white, and
from the margin processes extend in every direction, so that the
whole colony somewhat resembles a mould. Under a low power
many of the processes may be seen to be beaded ; others are
spiral, and join with processes from neighbouring colonies.
The gelatine is not liquefied for a long time, and then it is only
slightly marked. Zoogloea forms are said not to be seen in
the depth.

Gelatine-streak. — Along the streak a white growth appears,
from which fine processes run out on both sides to the margin
of the gelatine ; often the branches are curved with the con-
cavity upwards.

Gelatine-stab. — A white surface growth, with processes
running out to the margin ; along the stab a white growth
appears, from which numerous hair-like branches penetrate
horizontally into the gelatine.

Agar. — A thin white growth.

Potato. — Yellowish-white growth.

Milk. — Unchanged.

Peptone and Salt Solution. — No indol reaction produced.

Glucose-gelatine. — No gas formation.

Microscopical Characters. — A thin motile bacillus, varying
greatly in length ; some forms are almost like cocci. It does
not form spores. It grows best at 20° to 24° C.

Proteus Mirabilis.
Gelatine Plates. — The surface colonies, after twenty-four

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 127

hours incubation, appear as thin white expansions, more or less
circular in outline, with granular contents ; later, the margin
becomes irregular, and wandering processes pass out in all
directions, slowly spreading over the gelatine. Zooglcea forms
are commonly seen. Gelatine is slowly liquefied.

Gelatine-stab. — On the surface there is a thin white growth,
from which processes run out to the margin ; along the stab
there is a white growth from which processes penetrate almost
horizontally into the surrounding gelatine. The gelatine is
very slowrly liquefied.

Gelatine-streak. — There is a white growth along the line of
inoculation, from which processes run out to the margin, pro-
ducing a growth somewhat like a feather. The growth is very
like that of Proteus Zenkeri, but the processes are shorter and
thicker. The gelatine is slowly liquefied.

A gar. — A thin white growth.

Potato. — Greenish-yellow, moist, slightly raised growth.

Glucose-gelatine. — No gas formation.

Mill:. — Unchanged.

Nitrate-broth. — Nitrites and ammonia produced.

Peptone and Salt Solution. — Noindol reaction produced.

Microscopical Characters. — A motile bacillus ; often resembles
a coccus, and at times grows out to a length of 3 to 4 /LL.
Involution forms are common.

Proteus Zopfl.

A culture of this organism, which I obtained from Krai's
laboratory, gave reactions in the various media, closely
resembling the Proteus Zenkeri. The colonies in gelatine
plates showed the same beaded wandering processes. The
s warmers were, however, not quite so marked, and on gelatine
streaks the lateral processes were shorter and thicker, more
resembling Proteus mirabilis.

All the varieties of Proteus when " plated " in 5 per cent,
gelatine show, after a few hours incubation and before lique-
faction has commenced, a peculiar motility of the bacilli.

MICRO-ORGANISMS SUGGESTIVE OF SEWAGE CONTAMINATION.
The following organisms appear to be frequently present in
waters contaminated with sewage, but are rarely, if ever, found

128 BACTERIOLOGICAL EXAMINATION OF WATER.

in pure supplies. Waters containing these bacteria should be
regarded with suspicion, and the source of the supply carefully
investigated.

Bacillus Megaterium.

This organism was first isolated by De Bary from boiled
cabbage leaves. It was found by Tils in the Freiburg water-
supply. I have also found it in a water derived from a
polluted upland surface.

Gelatine Plates. — The surface colonies appear as circular
masses with a " woolly" margin ; later they appear depressed in
the gelatine owing to liquefaction. Under a low power they
show a dark granular centre surrounded by circular lines, from
which fine branching processes pass off at the margin.

Gelatine-stab. — A delicate growth at first appears along the
needle track, from which fine hair-like processes spread out into
the gelatine. Liquefaction rapidly takes place, and a funnel-
like excavation is formed, containing an opaque yellowish-white
layer at the bottom.

Agar. — A thick white growth appears with a feathery
margin.

Potato. — A thick yellowish-white growth.

Glucose- gelatine. — No gas formation.

Milk. — Coagulated, and then partially digested.

Peptone and Salt Solution. — No indol reaction obtained.

Nitrate-broth. — Powerful reducing action, large quantities of
ammonia being produced.

Microscopical Characters. — A fat bacillus, often very long, but
by the use of dehydrating agents, separate divisions can be seen,
so that the bacillus then appears to be made up of a series of
short rods, each containing a spore. It possesses a slow waggling
movement, and is stained by Gram's method.

Bacillus Lactis Cyanogenus.

This bacillus was first isolated from milk by Hueppe.
Jordan found it frequently in Lawrence sewage.

Gelatine Plates. — The surface colonies are circular and dirty
white in colour ; later the colour changes to an iron-grey, and
the surrounding gelatine acquires a deep blue colour. Under a

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 129

low power the colonies have a yellowish colour and a dark,
almost opaque, granular centre.

Gelatine-stab. — There is a white growth on the surface and
along the line of inoculation. The gelatine is not liquefied,
and acquires a steel-blue colour, which later becomes almost
black.

Agar. — A greyish-white growth appears, and the medium
becomes almost brown in colour.

Glucose-gelatine. — No gas formation. The gelatine acquires
a deep Prussian-blue colour.

Milk. — Not coagulated after several days incubation, and
becomes slightly alkaline, with a faint bluish tinge. In unsteril-
ised milk it produces a sky-blue colour.

Broth. — A diffused growth. Later, a pellicle forms on the
surface, and the broth acquires a blue colour.

Potato. — It forms a yellow restricted growth, and the potato
assumes a greyish-blue colour.

Blood-serum. — It grows, but does not produce any pig-
ment.

Microscopical Characters. — A thin motile bacillus, which forms
spores at the ends of the rods. According to Fliigge, the spores
seen by Hueppe are only involution forms. It grows at room
temperature and blood-heat. It does not stain with Gram.
NOTE. — Jordan stated that the specimens which he isolated
from sewage did not grow as well at 37° C. as at 21° C.,
and were not " very motile." He also failed to observe any
true spores.

Bacillus Acidi Lactici.

This bacillus was isolated by Hueppe from sour milk, and
was found in the Freiburg water by Tils.

Gelatine Plates. — The surface colonies appear as circular white
masses with an irregular outline and resemble those of B. coli ;
later, leaf-like expansions grow out from the margin. The
gelatine is not liquefied.

Gelatine-stab. — There is a white growth on the surface, which
later shows leaf-like expansions. Along the stab there is a
white growth consisting of small white centres.

Agar-slope. — A thick, wrhite, adhesive growth.

130 BACTERIOLOGICAL EXAMINATION OF WATER.

Potato. — Thick yellowish-brown growth.

Milk. — Coagulated in twenty-four hours at 37° C.

Peptone and Salt Solution. — A marked indol reaction obtained.

Broth. — A diffuse growth. A slight surface pellicle is formed.

Nitrate-broth. — Nitrates are largely reduced to nitrites.

Glucose-gelatine. — Marked gas formation.

Microscopical Characters. — Small non-motile bacillus, usually
in pairs. Spore formation present.

It will not grow below 10° C. ; requires from 10° to 12° C. ;
at 15° C. acid is produced, which ceases at 45'5° C.

Bacillus Urese.

This micro-organism was isolated from ammoniacal urine by
Jaksch. It was found by Tils in Freiburg water, and by Lustig
in river water.

Gelatine Plates. — The surface colonies form small, transparent,
finely granular films, with an irregular margin, somewhat re-
sembling the colonies of B. typhosus.

Gelatine-stab. — A surface growth like the colonies ; also a fine
white growth along the stab.

Glucose-gelatine. — No gas formation.

Milk. — LTnchanged.

Potato. — A thin white growth.

Agar. — A white expansion, not characteristic.

Broth. — A diffused growth and a deposit forms at the bottom
of the tube.

Peptone and Salt Solution. — No indol reaction obtained.

Microscopical Characters. — A small motile bacillus. It converts
urea into ammonium carbonate.

Bacillus Fluorescens Putridus.

This microbe was isolated from water by Flugge. It is
considered by Mace to be characteristic of foul waters. A
culture from Krai's laboratory gave the following cultural
reactions :

Gelatine Plates. — The colonies both in the depth and on
the surface are small ; but, later on the surface they grow
out as greyish-white expansions, with a circular or slightly
irregular margin. They are thicker at the centre than at the

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 131

margin, and often there is a slight knob in the centre of the
colony. Under a low power they appear round, finely granular,
with a dark centre ; the colour is yellow in the centre, and
clear grey at the margin. The gelatine around the colonies
acquires a green colour.

Gelatine-stab. — Surface growth is white and opaque ; along
the stab there is also a delicate white growth. The gelatine
assumes a bright green colour.

Agar. — A thick white growth; the agar acquires a green
colour.

Glucose-gelatine (shake). — No gas formation.
Peptone and Salt Solution. — No indol reaction obtained.
Broth. — Diffuse growth ; after several days a pellicle appears
on the surface. The broth acquires a green colour.
Nitrate-broth. — No reduction of nitrates.

Potato. — At first there is a greenish-yellow growth, which
later forms a brownish, slimy expansion.
Milk. — Unchanged.

Microscopical Appearance. — A small motile bacillus with
rounded ends, often in pairs. It grows best at room temperature,
and does not form spores.

In my experience this organism is often found in polluted
waters. Some of the varieties isolated, when sub-cultured in
broth, produced pathogenic effects when 1 c.c. was injected
sub-cutaneously into guinea-pigs. Lepierre isolated a similar
organism from a cistern at Coimbra, but the bacillus was not
motile.

Bacillus Erythrosporus.

This micro-organism was isolated from putrefying liquids
by Eidam ; it has also been found in water by Bolton and
Migula.

Gelatine Plates. — The surface colonies form white centres,
with a slightly irregular margin. Under a low power the centre
is dark brown in colour ; the margin shows radial streation,
and has a yellowish-green colour. The gelatine is not liquefied,
and around each colony acquires a green fluorescence.

Gelatine-stab. — There is a white growth along the stab and on
the surface ; the gelatine acquires a green colour.

132 BACTERIOLOGICAL EXAMINATION OF WATER.

Glucose-gelatine (shake). — No gas formation.

Milk. — Unchanged .

Peptone and Salt Solution. — No indol reaction obtained.

Nitrate-broth. — The nitrates are not reduced.

Broth. — A diffuse growth, the broth acquires a green
colour.

Potato. — A yellowish-brown restricted growth, later, nut-
brown in colour.

Microscopical Appearance. — Thin motile bacillus, with
rounded ends. From two to eight spores, which have a red
colour, appear in each rod. The red spores are visible when
the bacillus has been stained with methylene blue.

Bacillus Pyocyaneus.

The colour of green pus is due to the products of this
bacillus ; it was described by Gessard and Charrin. The same
bacillus was found by Tils in the Freiburg water-supply.

Gelatine Plates. — The colonies develop rapidly. They are not
sharply circumscribed, and usually present a fringe of delicate
filaments, surrounding a dark granular centre ; as growth
advances the gelatine is liquefied, and acquires a bright green
colour. Lender a low power the centre of the colony appears
filled with dark rounded masses, and surrounded by a wide,
thin, filmy growth, which has an irregular margin.

Gelatine-stab. — There is rapid liquefaction of the gelatine,
which in the earliest stage takes the form of a long narrow
funnel. The gelatine assumes a bright green colour.

Agar. — There is a dry greenish- white growth ; the agar is
bright green.

Potato. — There is a dry brownish growth which, when touched
with a platinum needle, sometimes becomes green ; the colour
lasts about ten minutes. This is known as the chameleon phe-
nomenon of Ernst.

Broth. — There is a floccular growth ; a pellicle forms on the
surface, and the broth assumes a green colour.

Milk. — An acid reaction is produced and the milk coagulated.

Peptone and Salt Solution. — Indol is produced.

Microscopical Appearance. — It is a delicate rod with rounded
ends ; sometimes it is very short and resembles a coccus ; in old

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 133

cultures longer forms are seen. It is very motile and possesses
a flagellum at one end. According to Flugge it forms spores,
though this is denied by other observers.

Micrococcus Urese.

This name was given by Pasteur to the organism which
converted urea into ammonium carbonate. Leube described
the following biological characteristics : — It is a medium-sized
coccus, which usually occurs as diplococci, sometimes as tetrads
and long chains. In gelatine plates the colonies appear as
smooth, mother-of-pearl like growths, which in ten days grow
out to the size of a sixpence and look like drops of wax ; under
a low power the margin is finely granular and the centre
opaque ; the gelatine is not liquefied. In gelatine stab it
forms a thin, tough, thread-like growth along the inoculation
line.

Micrococcus Urese Liquefaciens.

This micro-organism was isolated in Fliigge's Institute ; it
also rapidly converts urea into ammonium carbonate. It forms
large cocci, which are arranged in chains or masses. After two
days the deep colonies appear as small white points, which, under
a low power, are seen to be granular ; on reaching the surface
the colonies form yellowish-brown discs, often with a cavity in
the centre ; the margin is irregular, and liquefaction slowly
takes place. In gelatine stab there is a white growth along the
stab, the gelatine is soon liquefied, and the tube becomes filled
with fluid containing a yellowish-white deposit.

Cladothrix Dichotoma.

This organism was first described by Cohn. It is found in
fresh and stagnant water, and especially in water rich in organic
matter ; it is also common in sewage.

Gelatine Plates. — Small yellow dots appear in four or five days ;
each growth is surrounded by a brown colour, which gradually
extends over the gelatine. Under a low power, each colony is
seen to consist of a mass of branching threads closely felted
together and strongly resembling a mould. Later a white
mouldy growth forms on the surface of the colonies, and the
gelatine is slowly liquefied.

134- BACTERIOLOGICAL EXAMINATION OF WATER.

Gelatine-stab. — Growth occurs at the surface and along the
line of inoculation ; the gelatine acquires a brown colour around
the growth, and is slowly liquefied at the surface in the form of a
cup. A white incrustation appears in the liquefied mass after a
few days.

Agar. — There is a thick adherent growth, which later
becomes covered with a white incrustation. The medium
acquires a brown colour.

Broth. — A white growth appears on the surface and a deposit
forms at the bottom ; the intermediate portions of the medium
are quite clear.

Glucose-gelatine (shake). — There is no gas formation.

Milk. — A white growth appears on the surface, but the milk
is not coagulated.

Potato. — There is a localised greenish-yellow growth, which
becomes covered with a white incrustation.

Microscopical Appearance. — In a hanging-drop long branching
filaments, with hyaline contents, are seen. The long filaments
break up into smaller segments, which appear to be motile and
have a tendency to form zooglo2a masses.

CHAPTER XII.

QUALITATIVE ANALYSIS— continued.

" SEWAGE-BACTERIA " DESCRIBED BY E. O. JORDAX.

IN 1890 Jordan made a special study of bacteria occurring in
sewage. He believed that sewage was probably inhabited by
species peculiarly adapted to the conditions of life, and thought
there was good reason for speaking of "sewage bacteria," i.e., those
bacteria normally found in sewage, and composing its special
flora. The following micro-organisms were found in the Law-
rence sewage, and appeared to be peculiar to this medium.
The descriptions are taken from Jordan's paper in the Report
of the Massachusetts State Board of Health for 1890.

Bacillus Cloacae.

This bacillus appeared to be one of the commonest bacteria
in sewage.

Plate-cultures. — In twenty-four to forty-eight hours a round,
yellowish colony becomes visible in the gelatine. On coming to
the surface this forms a slight bluish expansion, with irregularly
notched edges, and almost immediately begins to liquefy the
gelatine. Under a low power of the microscope, the colony is
seen to have a dark centre, an outer translucent zone and a
darker edge ; the interior is finely granular. The whole plate
is soon liquefied (in from three to four days).

Gelatine-tubes. — A rapid growth, liquefying the gelatine. A
good growth along the inoculation line. An iridescent scum
forms on the surface, and there is a heavy, whitish, flocculent
precipitate. Grows as well in slightly acid gelatine as in the
slightly alkaline medium.

Agar-tubes. — A moist, slimy, porcelain-white surface growth.
Excellent growth along the inoculation line.

136 BACTERIOLOGICAL EXAMINATION OF WATER.

Potato-cultures. — In two days a good yellowish-white growth,
which projects from the surface.

Milk. — In about four days the milk is coagulated, and gives
a strong acid reaction.

Broth. — The broth becomes very cloudy in two days. In ten
to fourteen days considerable whitish precipitate is produced.
A slight skin forms on the surface, but falls to the bottom on
shaking the tube. After some time — two to three weeks — the
broth is still cloudy.

Nitrate-broth. — Reduces nitrates in broth vigorously.

Microscopical Characters. — Short, plump, oval bacilli, with well-
rounded ends. No spore formation observed. Quite variable in
size ; slightly longer and thicker on potato than on agar. They
are frequently grouped in pairs. About 0'8 u to 1*9 /u long, 0'7 a
to 1 IUL broad. Very motile. Grow better at 37° C. than at 21° to
23° C. Require oxygen ; grow very scantily under a mica plate.

Bacillus Superficialis.

Found frequently in Lawrence sewage.

Plate-cultures. — The colonies become plainly visible to the
naked eye in about two days. Under a low power the colony
itself is seen to be approximately round, but is divided by irre-
gular lines into angular lumps, giving a somewhat cracked
appearance to the whole colony. On coming to the surface a
regular, round, homogeneous, finely granular expansion is
formed. To the naked eye the whole colony then appears as a
projecting translucent drop. The colony grows slowly, and
gradually liquefies the gelatine. When the liquefaction has pro-
ceeded for several days the colony has a yellowish-brown opaque
centre and a translucent edge.

Gelatine-tubes. — A very slow growth, taking about ten days
to liquefy the gelatine to the walls of the tube. The growth is
almost wholly on the surface, and there is only the scantiest
growth along the inoculation line. Even after standing for
several weeks only the gelatine in the upper part of the test
tube is affected by the growth.

Agar -tubes. — On agar there is a moist, lustrous, grey, trans-
lucent growth. After several weeks the growth is still smooth
and shiny and has assumed a light brown tint.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 13?

Potato-culture. — Potato does not seem to be a favourable
medium for the development of this species. Repeated trials
at different temperatures have failed to induce a growth.

Milk. — There is no visible change in twenty days. The
reaction, however, is slightly acid.

Broth. — Becomes turbid very slowly. A scanty white pre-
cipitate is formed after some time. No skin.

Nitrate-broth. — There is no reduction to be observed in forty
days.

Microscopical Characters. — Fair-sized, plump bacilli, with
rounded ends. Generally occur singly or in pairs. No spore
formation observed. About 2'£ JJL long, 1 JJL broad ; motile.
Grow better at 37° C. than at 21° C., and develop feebly under
a mica plate.

Bacillus Rubescens.

Isolated from Lawrence sewage.

Plate-cultures. — A slow growth. The young colonies beneath
the gelatine are usually round, sometimes oval. On coming to
the surface they rise into a projecting porcelain- white drop.
The colonies increase in size slowly, and eventually take on a
slight brownish tint.

Gelatine-tubes. — The growth is slow and mainly on the surface,
where a porcelain- white, nail-head projection is found. There
is only a slight growth along the line of inoculation. There
is good growth in slightly acid gelatine.

Agar-tubes. — There is a rapid surface growth, white and
lustrous. At first the growth is smooth, but later becomes
crinkly, and the whole skin is much wrinkled. In cultures
about three weeks old a slight pinkish tinge can be seen.

Potato-cultures. — There is a rapid luxuriant growth on potato.
At first the colour of the growth is light brown, but this slowly
changes to pink. In three weeks there is a luxuriant, projecting
growth, tinted a delicate flesh-pink. The potato itself is not
coloured.

Milk. — The milk is not coagulated, and gives a good alkaline
reaction. In long-standing cultures a slight pinkish tinge is
observed at the surface of the milk.

138 BACTERIOLOGICAL EXAMINATION OF WATER.

Broth. — The broth becomes slightly turbid and a heavy
white precipitate is formed. In several weeks a thick tenacious
skin forms on the surface ; the main body of the broth is then
clear.

Nitrate-broth. — No change in the nitrates after fifty days.

Microscopical Characters. — Large, long bacilli with well-rounded
ends. Often occur in pairs and short strings. Many of the
individual bacilli are slightly curved. No spore formation
observed. About 4 JJL long and 0*9 u broad. A slow, sluggish
movement. Grow better at 21° C. than at 37° C., and develop
very poorly under the mica plate.

Bacillus Hyalinus.

Isolated from the sand of Tank 13 at a time when the tank
was nitrifying well. Found in large numbers in the sand.

Plate-cultures. — Fast-growing. In twenty-four hours the
colonies can be plainly seen with the naked eye. In all, even
in the smaller colonies, there is a dark centre surrounded by
a broad translucent zone, which gives a hazy appearance to the
colonies. With a low power the interior is seen to be coarsely
fibril lar, with short fibrils radiating from the edge. In two
days the colonies are large — about one and a half centimetres
in diameter ; the contour is evenly round, the interior slightly
translucent ; the rim is distinct, opaque, yellowish ; the edge
still shows radiating fibrils.

Gelatine-tubes. — In two days a long, narrow, funnel-shaped
growth. The gelatine is rapidly liquefied, and is at first cloudy,
with a precipitate at the bottom of the funnel. In about eight
days the tube has assumed a highly characteristic appearance :
a lustrous, tenacious, white scum, a slight, flocculent precipitate,
and perfectly transparent liquefied gelatine between. Grows
very well in acid gelatine.

Agar-tubes. — A rapid, dry, spreading, greyish growth. Dull,
tough, and rather thin. When about four days old warty pro-
jections appear.

Potato-cultures. — In two days the growth is just visible; in
four days good, spreading, whitish-grey growth. Later, small
protuberances appear on the surface.

Milk. — In seven days strongly coagulated, reaction acid.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 139

Broth. — Soon becomes cloudy, a stringy precipitate, and a
thick skin on the surface.

Nitrate-broth. — Reduces nitrates vigorously and rapidly.

Microscopical Characters. — Large, long, stout bacilli, with
rounded ends. Usually gathered in short strings. No spore
formation observed. About 3'6 to 4 JJL long and T5 /* broad.
Highly motile; grow better at 37° C. than at 21° to 23°:
develop well under a mica plate.

Bacillus Delicatulus.

Isolated from the effluents of nearly all the tanks.

Plate-cultures. — When young, whitish, homogeneous, with a
regular, radiating edge. In two days the gelatine becomes
liquefied; later, the centre becomes darker than the surrounding
zone.

Gelatine-tubes. — In two days the gelatine is liquefied well down
into the inoculation line. The gelatine is completely liquefied
in about seven days. There is a thick whitish skin on the
surface, and a heavy, flocculent, brownish precipitate at the
bottom.

Agar-tiibes. — At first a crinkly greyish growth which, when
older, becomes porcelain- white and glistening. Grows well both
on the surface and below.

Potato-cultures. — A grey spreading growth ; not projecting.

Milk. — The milk is coagulated, and gives a strong acid
reaction.

Broth. — Soon becomes cloudy. A white precipitate and a
white scum.

Nitrate-broth. — Reduces nitrates to nitrites very rapidly and
completely.

Microscopical Characters. — Medium-sized, plump bacilli; often
joined in pairs and in short strings, 2 /LL long and 1 /LL broad.
Spore formation was not observed. V7ery motile. Grows better
at 37° C. than 21° C. This species is very sensitive to low
temperatures ; it refuses to grow at about 15° C. The cultures
in tubes appear to live only a short time.

SEWAGE-BACTERIA DESCRIBED BY LAWS AND ANDREWKS.
In 1894 Laws and Andrewes presented to the London County

UO BACTERIOLOGICAL EXAMINATION OF WATER.

Council a report on the micro-organisms of sewage. They in-
vestigated specimens of sewage from various sources, and de-
scribed the following new species. The descriptions are taken
from the above-mentioned report :—

Proteus Cloacinus.

Source. — Found in every sample of sewage.

Form and Arrangement. — A short bacillus of variable length ;
0'4 to 0*5 u in breadth ; stained specimens from a two-days old
agar culture vary from 0'5 to 1 u in length ; the shorter forms
almost resembling cocci. On gelatine some longer forms are
seen up to 1'6 u in length, and these may be as much as 0'6 /i in
breadth. In a one-day-old broth culture the average length is
0*8 /LI. It is actively motile. It does not form spores, and
grows well at 37° C. and 20° C.

Gelatine Plates. — The colonies are visible to the naked eye in
nineteen hours, though still minute. Under the microscope the
surface colonies are thin, transparent, and irregular in outline ;
the deeper colonies globular and brownish by transmitted light.
In two days the surface colonies are more circular and regular
in outline. The largest are about 1 mm. in diameter ; a few
show a thin spreading edge, but no " swarming islands " are
given off. No further change occurs, except that the colonies
increase in size up to & mm. in diameter, and become raised at
the edge ; they are of a greyish-white colour. No liquefaction
occurs even at the end of a fortnight. The plates have a sour
smell.

Gelatine-streak. — Rapid growth with irregular edges, but
shows no " swarms " as a rule. The growth strongly resembles
that of B. coli communis. Liquefaction occurs in about three
weeks. Occasionally the gelatine is liquefied in three to five
days, and in these cases swarming islands are thrown out like
the common Proteus.

Gelatine-stab. — Growth is rapid and well-marked along the
needle track. Scanty gas bubbles are formed. Liquefaction
first commences in from two to three weeks. The growth is
white and nodular in the depth.

Glucose-gelatine (shake). — There is profuse formation of gas
bubbles in twenty-four hours.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 141

Agar-plates. — Incubated at 37° C. show exceedingly rapid
and characteristic growth. A single colony spreads in thirty-
six hours as a thin irregular dendritic film, covering almost the
whole surface of the plate. The film does not thicken on
further incubation.

Agar-fitreak culture. — Growth is very rapid both at 37° C.
and 20° C. Even at the latter temperature the whole surface
of the tube is covered in twenty-four hours by a thin homo-
geneous semi-transparent greyish film, advancing at the edge by
irregular processes. On the second day the growth is somewhat
thicker ahd whiter, and has reached its full development. There
is a scanty development of gas bubbles in the depth of the
agar-agar, but no odour of putrefaction is ever perceptible.

Potato. — Grows rapidly, forming a slimy layer of a yellowish-
grey colour.

Broth. — In twenty-four hours the broth is turbid, and in a
few days a slight film is present on the surface and a yellowish-
white deposit at the bottom of the tube. In a week there is a
marked putrefactive odour. Tested in eleven days no trace of
indol reaction was obtained.

Mill:. — Grows without producing any change in this medium.

Micrococcus Aurora.

Source. — Found in fresh sewage from St. Bartholomew's
Hospital.

Form and Arrangement. — Slightly oval cocci, 0*7 to 0'8 ju in
diameter, often as diplococci separated by a cleft, rarely in
short chains of three or four elements. In broth cultures the
cocci are often arranged in irregular packets.

Motility. — Not motile.

Gelatine Plates. — Colonies first appear on the third day,
minute and circular, not yet pink in colour. Under the micro-
scope they appear granular and slightly irregular in outline.
On the fifth day the surface colonies are less than 0*5 mm. in
diameter, circular, prominent, and pinkish. After a week the
largest colonies are 0*7 mm. in diameter, those in the depth
0'25 mm., and of a pale pink tint. Later the colonies attain a
diameter of 1 to 1'5 mm., and liquefaction slowly occurs after
two to three weeks.

142 BACTERIOLOGICAL EXAMINATION OF WATER.

Gelatine-streak. — After two days a narrow, prominent line of
growth of a pale pink tint, which eventually attains a width of
1*5 to 2 mm., and a bright pink colour. Liquefaction occurs
after fourteen to sixteen days.

Gelatine-stab. — Definite growth in three days, almost entirely
confined to the surface ; in the depth of the gelatine little or
no development occurs. A flat layer, 4 mm. in diameter and
of a light pink colour, is present on the surface at the end of a
week. Liquefaction is extremely tardy, but some depression of
the surface occurs after a fortnight, and distinct liquefaction
begins in sixteen days.

Agar-ugar. — Streak -cultures at 20° C. show distinct growth
in two days, semi-transparent and barely pink. At the end of
a week the growth is opaque and of a bright pink colour. At
37° C. no growth occurs.

Potato. — A thick layer of a brilliant deep pink colour, which
eventually covers the whole surface.

Broth. — The fluid gradually becomes turbid, and after a week
there is a light pink sediment at the bottom of the tube.

Temperature.— Grows well at 20° to 22° C. No growth on
any medium at 37° C.

Spore-formation. — Not o bserved .

Gas-production. — Not observed.

Need of oxygen. — Growth very scanty away from the
surface.

Pigment-production. — Produces a bright pink pigment, readily
soluble in alcohol, only slightly so in water.

Relations. — Appears most closely related to Micrococcus
agilis and Micrococcus roseus, but differs from the former in the
absence of motility and the lighter pink colour of the cultures,
and from the latter in the absence of any growth at 37° C.

Bacillus Cloacae Fluorescens.

Source. — Found in Homerton sewage and sewage from St.
Bartholomew's Hospital.

Form and Arrangement. — Motile rods, on an average twice as
long as broad. Breadth, 0'6 to 0'75 ju ; length, 1 to 2 u. The
rods are sometimes attached in chains. It is actively motile.
It is unable to grow at 37° C. No spore formation.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 148

Gelatine Plates. — Minute colonies become visible in twenty-
four hours. In two days the surface colonies begin to liquefy,
and are circular, 2 to 2-|- mm. in diameter, and under the micro-
scope are seen to have a very delicate, transparent, irregular,
spreading margin not yet liquefied . The circular liquefied portion
is turbid, with a granular deposit in the centre, and shows slight
but distinct greenish fluorescence. On the third day the colonies
are 5 to 6 mm. in diameter, and partly fused with one
another; the whole plate being largely liquefied with bright
yellowish-green fluorescence, and having a sour, unpleasant smell.

Gelatine-stab. — Growth chiefly on the surface ; hardly any to
be seen along the depth of the track. Liquefaction begins at
the surface on the second day in a cone, extending some way
along the needle tract. It gradually spreads, chiefly at the
surface, till it reaches the wall of the tube in seven days. In
fourteen days the gelatine is liquefied from 10 to 12 mm. from
the surface, and spreads downwards in a horizontal plane. The
liquefied portion has a yellowish-green fluorescence.

Agar. — No growth at 37° C. At 20° C. whitish, semi-opaque
growth, with an irregular edge. The agar has a brilliant
yellowish-green fluorescence.

Potato. — A dirty brown layer, spreading from the line of
inoculation.

Broth. — Rapid growth causing dense turbidity. A thin
film forms on the surface, and the top layer of the broth
acquires a marked greenish fluorescence.

Milk. — No coagulation, but after four days the upper layer
becomes slightly yellowish and less opaque, owing to gradual
digestion.

Gas-production. — Not observed. Growth scanty away from
the surface. Oxygen seems necessary for the production of the
fluorescent pigment.

Bacillus Pluorescens Stercoralis.

Source. — Homerton and Barking sewage.

Form and Arrangement. — Motile rods, almost twice as long
as broad, but of variable length. The breadth varies from
0*5 fj. to 0*7 jU, and the length from 0*8 to 1*6 fj. ; longer forms
rarely occur. Actively motile.

144 BACTERIOLOGICAL EXAMINATION OF WATER.

Gelatine Plates. — Tiny circular colonies are visible under the
microscope in nineteen hours. In twenty-four hours they are
visible to the naked eye. In two days the surface colonies are
liquefied and appear as uniformly turbid circles, with a whitish
flocculent deposit at bottom. The deeper colonies are white,
spherical, and still very minute. In three days the colonies
have increased up to 10 mm., and are fused in places, the
plate being mostly liquefied with distinct greenish-yellow
fluorescence, and having a sour, faecal smell. On the fourth
day the plate is completely liquefied, and the fluorescence
brighter.

Gelatine-stab cultures. — In nineteen hours there is already
liquefaction at the surface, as a small flat hemisphere, 2 mm.
in diameter, which in two days attains a diameter of 6 or
7 mm., and in three days reaches the walls of the tube,
though not to a greater depth than 3 mm. Little growth
occurs along the needle track. The liquefaction always
advances in a horizontal plane. Greenish fluorescence is
marked in the liquefied portion by the fifth day; a thin
pellicle forms on the surface, and there is a thick whitish
deposit at the bottom of the liquefied portion.

Agar-agar. — No growth occurs at 37° C. At 20° C.
growth is rapid. In streak cultures there is a greyish-white
growth, with a thick, slightly irregular edge. The agar
acquires a brilliant greenish-yellow fluorescence.

Broth. — The growth is similar to B. fluorescens liquefaciens ;
marked green fluorescence appears in the surface layers after
the third day.

Milk. — The milk does not coagulate, but is gradually
digested, the upper layer becoming yellowish and transparent
after two days ; this change gradually spreads downwards, and
the upper layers show a yellowish-green tint.

Gas production has not been observed.

This organism is closely allied to the B. fluorescens liquefaciens,
but differs from it in the mode and rate of liquefaction of
gelatine, and in the greater brilliancy of the fluorescence it
produces on agar cultures, in which it resembles B. pyocyaneus.
From the last mentioned species it differs in its refusal to grow
at 37° C.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 145

SEWAGE BACTERIA, ISOLATED BY HOUSTON, FROM LONDON
CRUDE SEWAGE AND THE EFFLUENTS FROM COKE-BEDS.

The latest contribution to the study of sewage flora has been
made by Houston, who, during his studies of the bacterial
treatment of sewage, isolated a number of micro-organisms
which appeared to be new species and peculiar to sewage. The
following descriptions are taken from the second report on the
bacterial treatment of sewage, presented to the London County
Council.

Sewage Proteus.

Source. — Very abundant in London crude sewage ; frequently
as many as 100,000 per c.c. of crude sewage.

Gelatine-plate Cultures. — In less than twenty-four hours at
20° C. the surface colonies appear as delicate granular films
of an irregular shape. In two days the colonies look like
" punched out " circles, containing liquefied gelatine and greyish-
white bacterial deposit. The masses of bacteria lying in the
liquefied gelatine give the colonies a mottled appearance.
Under a low power the individual bacilli can be made out, and
their active movement watched. The colonies are usually
exactly circular in shape, with well -defined borders, and no
" swarming islands " appear to be given off as in the Proteus
vulgaris. By the third or fourth day the plate is completely
liquefied.

Gelatine-stab Culture. — The growth is very characteristic, in
twenty-four hours at 20° C. liquefaction has occurred all the
way down the path of the needle, and minute bubbles of gas
may be seen rising to the surface through the turbid liquefied
gelatine. In forty-eight hours the liquefaction is more pro-
nounced, and numerous bubbles may be seen at the surface, and
also bubbles in the solid gelatine. The bacteria collect at the
foot of the liquefied portion, as a greyish-white deposit. In
four or five days the whole of the gelatine is converted into a
greyish- white liquid.

Gelatine-shake Cultures. — In twenty -four hours at 20° C.
numerous gas bubbles are formed, and the gelatine is liquefied
near the surface.

146 BACTERIOLOGICAL EXAMINATION OF WATER.

Agar-plate Cultures. — The colonies are more or less circular
in shape, greyish-white in colour, and have a moist glistening
appearance.

Agar-stab Culture. — There is growth all down the stab ; gas
bubbles may form in the medium ; on the surface the growth is
like a surface colony.

Agar-streak Culture. — A greyish-white moist glistening growth.

Potato Culture. — A slimy, thin, yellowish-white growth.

Broth. — Abundant, diffuse, cloudy growth.

Litmus-milk. — The opaque bluish-purple coloured fluid is
changed into semi-transparent reddish-coloured fluid. If clotting
occurs, it is imperfect in character as possibly casein is dissolved
as soon as it is 'formed.

Indol Reaction. — Usually no indol is formed in broth cultures,
but in one such culture kept for twelve days a feeble reaction
was obtained.

Reduction of' Nitrates. — Rapid reduction of nitrates to nitrites
in twenty-four hours at 20° C.

Microscopical Appearances. — Small bacillus with rounded ends,
solitary,1 in pairs, or sometimes in short chains; involution
forms may be seen. It is actively motile, possesses a single
flagellum, and does not form spores. The original culture
grew better at 20° C. than at 37° C.

Remarks. — This organism differs from Proteus vulgaris; it
has been called sewage Proteus, owing to its prevalence in
sewage and its superficial resemblance to the members of the
Proteus group. Some of the cultures isolated from sewage
were pathogenic to guinea-pigs ; others grew well at 37° C.

Bacillus Frondosus.

Isolated from London crude sewage.

Gelatine-plate Cultures. — The surface colonies appear as white
coarsely-granular films of irregular shape, which send out curious
processes of varied form, resembling bits of sea- weed flattened
out. Liquefaction is slow and is shown by slight pitting near the
centre of each colony. The deep colonies are not characteristic.
Under a low power the superficial colonies have a laminated
appearance, and show at the spreading edge processes of most
varied shapes, which form patterns of great delicacy and beauty.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 147

Gelatine-stab Culture. — The growth at the surface is like a
superficial colony ; the growth down the stab is not character-
istic. ' Liquefaction sets in slowly.

Gelatine-streak Culture. — A coarsely-granular film appears,
which peripherally shoots out processes of very irregular shape.

Gelatine-shake Culture. — No bubbles of gas are formed.

Agar-plate Cultures. — The colonies are white in colour and
do not grow out in the same characteristic manner as on
gelatine.

Agar-streak Culture. — A greyish- white film, which at the
margin resembles the corresponding growth on gelatine.

Potato Cultures. — A transparent colourless film.

Broth Cultures. — Grows slowly ; a white deposit forms at the
bottom of the tube, leaving the liquid above fairly clear.

Litmus-milk Cultures. — In milk at 20° C. no change is visible
in three days ; later, the milk turns slightly acid ; no clotting is
visible even after one month.

Indol Reaction. — No indol is formed in broth after six days
incubation at 20° C.

Reduction of Nitrates. — No reduction takes place in four
days at 20° C.

Microscopical Appearances. — Large motile bacillus with rounded
ends ; solitary, in pairs, and in chains of varying length. It
forms spores. There is no growth at 37° C.

Bacillus Fusiformis.

Isolated from London crude sewage.

Gelatine-plate Cultures. — The growth is very slow. The
surface colonies are circular, of an opaque porcelain-white
glistening appearance. The periphery is slightly transparent.
By transmitted light the colonies are yellowish in colour.
Under a low power, no delicate details of growth can be made
out. No liquefaction of the gelatine takes place.

Gelatine-stab Cultures. — A white growth appears along the
line of the stab, and on the surface a scanty yellowish-white
layer slowlv develops.

Gelatine-shake Cultures. — No gas bubbles are formed.

Agar-streak Cultures. — A thin whitish layer is slowly formed,
which is not characteristic.

148 BACTERIOLOGICAL EXAMINATION OF WATER.

Agar-plate Cultures. — The colonies are white and more or less
circular in shape ; the growth is not characteristic.

Potato Cultures. — A porcelain-white growth slowly develops,
afterwards the colour becomes dirty yellowish -white, and the
bacterial layer becomes unequally thickened.

Broth Cultures. — The growth is scanty and not characteristic.

Litmus-milk Cultures. — No clotting occurs ; after sixteen days
very feeble acidity is produced.

Indol Reaction. — No indol reaction is obtained in broth
cultures.

Reduction of Nitrates. — No reduction after twelve days at 20°C.

Microscopical Characters. — Large motile bacillus, with rounded
ends ; solitary, in pairs, and in chains. It forms large spindle-
shaped spores.

Remarks. — This micro-organism has a somewhat negative
character in all the media ordinarily in use. It has been called
B. fusiformis, owing to the spores being spindle shaped.

Bacillus Subtilissimus.

Isolated from crude sewage.

Gelatine-plate Cultures. — The surface colonies are very cha-
racteristic, and grow so quickly that a single colony may cover
nearly a whole plate in two days. The growth is film-like,
exceedingly thin and transparent, dull grey in colour, and very
faintly granular. Under a low power the colony resembles
B. coli, but the growth is so delicate that it is difficult to see
the delicate veining of the bacterial film. The surface colonies
are more or less circular in shape, but the spreading margin is
nearly always irregular.

Gelatine-streak. — The growth is like an elongated surface
colony. In less than twenty-four hours a delicate film has
spread nearly to the walls of the tube. The spreading edge is
very irregular, and in older cultures may present a terraced
appearance.

Gelatine-shake Culture. — No gas bubbles are formed in the
gelatine.

Agar-streak. — A white film is formed on the surface, having
a markedly irregular edge. The lateral expansion is less than
in the case of gelatine cultures.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 149

Broth Cultures. — Uniform turbidity occurs in twenty-four
hours at 20° C., and a very faint scum forms on the surface.

Litmus-milk Cultures. — No clotting or production of acidity.

Indol. — No indol is formed even after twenty days at 20° C. !

Microscopical Appearances. — In most cultures it appears as a
large micrococcus. In the spreading edge of the colony the
elements are distinctly longer than broad ; nearer the centre
they are oval, and frequently united in pairs ; at or about the
centre they are perfectly spherical.

Remarks. — This microbe has been called B. subtilissimus on
account of the thin, almost gauze-like, character of the surface
colonies in gelatine-plate cultures.

Bacillus Membraneus Patulus.

Isolated from crude sewage and effluent from coke-beds.

Gelatine Plates. — The surface colonies appear as coarsely
granular, greyish-white films of somewhat irregular shape.
From the spreading edge of the colony processes extend in a
tortuous fashion over the surface of the medium, often forming
patterns of great delicacy and beauty. Beneath the surface
film-like growth slow liquefaction of the gelatine occurs.
Under a low power the colonies present a characteristic,
granular, and striated appearance.

Gelatine-streak. — In less than two days at 20° C. a coarsely
granular film forms on the surface of the gelatine, which spreads
rather rapidly ; from the margin processes are given off' which
wind over the gelatine in a characteristic way. Soon a longi-
tudinal pitting of the bacterial film along the line of inocu-
lation is observed, and the growth sinks down to the foot of the
tube.

Agar-streak. — In one night at 37° C. the growth appears as
a coarse, granular, semi-transparent, greyish-white film. In old
cultures the surface assumes a tuberculated appearance.

Gelatine-stab. — The growth varies ; sometimes there is lique-
faction down the line of the stab with tuft-like processes
extending into the solid medium, and at other times there is no
liquefaction, and the processes extend to the wall of the tube.
The growth on the surface is like a surface colony in a gelatine
plate.

150 BACTERIOLOGICAL EXAMINATION OF WATER.

Gelatine-shake Culture. — There is no gas formation.

Broth. — It grows rapidly at 37° C., producing a flocculent
growth. On the surface a skin is formed, which sinks on
shaking the tube.

Indol — There is no formation of indol in broth cultures.

Litmus-milk. — A gelatinous clot is formed and the medium
becomes faintly acid.

Potato. — A dirty, faint, yellowish-grey coloured growth.

Nitrate-broth. — Great reduction of nitrates in one night
at 37° C.

Microscopical Characters, $c. — A non-motile, very large
bacillus, which forms long chains. The cultures resist heating
to 80° C., but a satisfactory double-stained spore preparation
has not been obtained. It grows well at 37° C. and at the
room temperature.

Bacillus Capillareus.

Isolated from crude sewage and effluents from coke-beds.

Gelatine-plate Cultures. — The colonies in the depth have a
characteristic fluffy appearance, and when they reach the
surface quickly liquefy the gelatine. The growth is
filamentous in the depth, and on the surface from the spreading
edge of the colonies delicate film-like processes are given off,
which extend over the surface of the medium to form irregular
patterns. Later, the final details are lost owing to the rapid
liquefaction of the gelatine, the colonies appearing as large,
more or less circular, areas of liquefaction with greyish- white
contents.

Agar-streak. — In twenty-four hours an opaque-white growth
of limited extent appears, which along the spreading edge is
slightly transparent and granular-looking.

Gelatine-streak. — A longitudinal furrow appears due to the
liquefaction of the gelatine; from the edge of the furrow
delicate processes may be seen, extending in irregular fashion
over the solid surface of the gelatine.

Gelatine-stab. — Liquefaction takes place in funnel form, and
extends down the stab to an extent varying in different cultures.
Along the line of inoculation feathery processes are given off,
which extend into the solid gelatine for a short distance. The

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 151

masses of bacteria gradually sink to the foot of the liquefied
medium.

Gelatine-shake. — No gas is formed. Sometimes the colonies
have a star-shaped appearance, but at other times they are
globular.

Broth. — At 37° C. there is a diffuse cloudiness ; later a skin
forms on the surface, and an abundant white deposit forms at
the foot of the tube, leaving the liquid above fairly clear.

Indol. — No indol is formed in broth cultures.

Litmus -mill:. — No decided change after seventy-two hours at
37° C. In five days weak gelatinous clot, but no acidity.

Potato. — By the second day at 37° C. a fairly abundant
creamy coating has formed on the surface, and later becomes
of a dirty-brown colour.

Nitrate-broth. — Reduces nitrates to nitrites.

Microscopical Appearances, fyc. — A large bacillus forming
long chains. It is motile. Old cultures resist heating to 80° C.;
but a satisfactory preparation of spores has not been obtained.
It grows luxuriantly at 37° C. and at room temperature.

Bacillus Mesentericus (Sewage variety E.).

Isolated from London crude sewage. Gelatine plates heated
to 80° C. for ten minutes.

Gelatine-plate Cultures. — The growth is not very rapid. The
deep colonies have a somewhat star-like appearance ; the super-
ficial colonies appear as bluish-white, delicate granular films,
which are almost colon-like in character. They are somewhat
irregular in shape, and from the wavy transparent edge
irregular processes are given off'. Under a low power the deep
colonies show a central darkish yellow spot from which root-like
processes are given off'. The surface colonies are transparent,
granular, and striated ; and from the edge delicate processes are
given off', which spread over the gelatine, forming curious and
intricate patterns. The colonies do not attain a large size,
and liquefaction only slowly sets in.

Gelatine-stab. — The growth at the surface is slow and resembles
a superficial colony. Growth takes place all the Avay down the
stab accompanied by slow liquefaction; delicate tuft-like fila-
ments are given off' all down the line of inoculation.

1,52 BACTERIOLOGICAL EXAMINATION OF WATER.

Gelatine-streak. — -A delicate granular film of limited extent
forms on the surface of the gelatine. After some days the
film changes its appearance, and shows numberless fine processes
radiating outwards and upwards from the central line. Tuft-
like processes also extend into the solid gelatine. Liquefaction
slowly sets in and shows itself as a longitudinal furrow.

Gelatine-shake Cultures. — No gas formation. The gelatine
is slowly liquefied.

Agar-streak. — In twenty-four hours at 37° C. a dirty yellowish-
white film has covered nearly the whole oblique surface. The
spreading edge may extend as irregular processes, forming leaf-
like patterns. Later, the film darkens in colour and becomes
wrinkled.

Potato Cultures. — The growth is extremely characteristic. At
37° C. in twenty-four hours the whole surface of the potato is
covered with a thin, yellowish-white film, which has a character-
istic, folded, wrinkled appearance. The substance of the potato
takes on a bright pink colpur. Later, the film becomes thicker,
more deeply pitted and wrinkled, and the colour becomes brown.

Broth. — In twenty-four hours at 37° C. a greyish-white
wrinkled film is formed at the surface, and the liquid below is
nearly quite clear.

Litmus-milk. — In seventy-two hours a weak clot has formed
and the liquid near the surface is semi-transparent. On shaking
the tube a pinkish tinge develops.

Indol Reaction. — No indol is formed.

Nitrate-broth. — Great reduction of nitrates to nitrites in
twenty-four hours.

Microscopical Appearances. — Long bacillus with rounded ends,
solitary, and in pairs and long chains. It is actively motile,
possesses numerous flagella, and forms spores. It grows very
rapidly at 37° C.

Remarks. — This bacillus appears to be identical with B. mesen-
tericus ruber.

Bacillus Mesentericus (Sewage variety I.).

Isolated from London crude sewage. Gelatine-plate cultures
heated to 80° C. for ten minutes. It is constantly present in
sewage in the form of spores.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 153

Gelatine-plate Cultures. — The growth is characteristically
rapid. The deep colonies quickly reach the surface, to form
saucer-shaped areas of liquefaction ; from the edge of these
processes may be given off which are almost of the nature
of " swarming islands." In the areas of liquefied gelatine the
bacteria may be gathered in clumps so as to give rise to a
mottled appearance. Under a low power the surface colonies
may be almost completely translucent and of an irregular star-
shaped form ; but soon circular areas of liquefied gelatine are
produced, in which the extraordinarily rapid movement of the
individual bacilli can be clearly watched.

Gelatine-stab. — Rapid liquefaction takes place in the form of
a funnel. The liquefied gelatine has a greyish-white flocculent
appearance, due to aggregation of little masses of bacteria.

Gelatine-streak. — The growth is not characteristic owing to
the rapid liquefaction.

Gelatine-shake Cultures. — No gas bubbles are formed.

Agar-streak. — In twenty-four hours at 37° C. a dirty,
yellowish-white film has covered nearly the whole surface of the
agar. The spreading edge may be lobed or fissured. The film
darkens in colour and becomes wrinkled.

Potato Cultures. — The growth is extremely characteristic. In
twenty-four hours at 37° C. the whole surface of the potato is
covered with a thick, greyish-white moist skin, which is thrown
into multiple folds, creasings and wrinkles. The colour
rapidly changes from greyish -white to yellow and then to
brown.

Broth Cultures. — The growth at 37° C. is characteristic even
in twenty-four hours. A greyish-white wrinkled film is formed
at the surface, and the liquid below is nearly quite clear. Later
the film thickens and acquires a reddish-brown colour; the
liquid below, which remains nearly quite transparent, also takes
on a reddish-brown colour.

Litmus-milk. — In forty-eight hours there is no clot, but the
milk is rapidly becoming transparent ; in seventy-two hours
the whole of the contents of the tube are semi-transparent and
of a dirty yellowish colour. On gently shaking the tube, the
liquid assumes a crushed-strawberry colour.

Indol. — No indol is formed.

154 BACTERIOLOGICAL EXAMINATION OF WATER.

X it rate-broth. — No reduction of nitrates to nitrites in twenty-
four hours at 37° C.

Microscopical Characters. — Medium-sized bacillus with rounded
ends, solitary, in pairs, and in chains. Very motile, possesses
numerous flagella, and forms spores. Grows with great rapidity
at 37° C.

Remarks. — This micro-organism is probably identical with
Bacillus mesenterictis vulgatus.

Bacillus Subtilis (Sewage variety A.).

Isolated from crude sewage and effluents from coke-beds.

Gelatine-plate Cultures. — In two days at 20° C. it forms circular
greyish- white areas of liquefaction ; masses of bacteria in the lique-
fied gelatine may give the colony a mottled appearance. Under a
low power the bacilli at the margin of the colony are seen to bore
into the gelatine, side by side, in a highly characteristic manner.

Gelatine-stab. — In two days at 20° C. the gelatine is liquefied
down to the foot of the stab. A scum forms on the surface,
but no distinct skin is formed.

Agar-streak. — In two days at 37° C. there is an abundant
creamy-white layer covering nearly the whole surface, which
shows numerous minute circular areas where the growth, instead
of being opaque, is semi-transparent.

Gelatine-shake. — There is no gas formation.

Broth Cultures. — Diffuse cloudiness in twenty-four hours at
37° C. A scum forms on the surface, which readily falls to the
foot of the tube on shaking.

Indol. — There is no formation of indol.

Litmus-milk. — In two days at 37° C. there is complete dis-
coloration and a clot has formed ; no redness is visible. In five
days a white clot occupies about one-third of the bulk of the
medium, which appears to be slowly peptonised. The liquid
surrounding the clot is pale yellow in colour, semi-transparent,
and without pink coloration.

Potato Cultures. — In two days at 37° C. a dirty-white layer
is formed with a yellowish tint.

Microscopical Appearances. — Large, long bacillus with rounded
ends, frequently in long chains. It forms spores and has
a waddling, sluggish movement. It grows well at 37° C.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 155

Bacillus Subtilis (Sewage variety B.).

Isolated from crude sewage and effluents from coke-beds.

Gelatine-plate Cultures. — Rapidly forms greyish- white circles
of liquefied gelatine. The masses of bacteria lying in the
liquefied gelatine may present a rosette or star-shaped appear-
ance. The parallel arrangement of the bacilli at the periphery
of the colony is absent, or not so well marked as in the above
variety. A skin forms on the surface of the liquefied gelatine.

Agar-streak. — In twenty hours at 20° C. there is a greyish-
white layer. In forty-eight hours the growth is somewhat dry
and granular looking, and in four days ridges stand oat from
the surface of the medium about one-sixteenth of an inch.
They are usually arranged in more or less transverse folds.

Gelatine-stab Cultures. — Rapid liquefaction all the way down
the stab, but as the growth proceeds the liquefaction spreads
in a cylindrical fashion from above downwards. A distinct
skin forms on the surface, which eventually sinks in the
liquefied gelatine.

Gelatine-silage. — There is no gas formation.

Broth. — Diffuse cloudiness ; on the surface a skin forms,
which is brittle, and sinks on shaking the tube.

Indol. — There is no formation of indol.

Litmus-milk. — The pale-blue colour gradually fades, but no
clotting or redness occurs. In eight days the milk is almost
transparent, and of a pale, dirty yellow colour.

Potato Cultures. — A white rather dry-looking coat is formed.
Later, portions of the growth become upraised, and some-
times present a worm-like appearance.

Microscopical Characters. — Large, long bacillus with rounded
ends, often in long chains. It forms spores and has a
waddling, sluggish movement. It shows little or no growth
at 37° C., but grows luxuriantly at 20° C,

Thermophylic Bacteria.

The micro-organisms which are capable of growing luxuriantly
at a temperature of 60° to 70° C. are of considerable interest to
the water bacteriologist. They are found in the alimentary
canal of human beings and animals, in sewage, and in polluted

156 BACTERIOLOGICAL EXAMINATION OF WATER.

waters, but do not appear to be present in pure waters. In
1888 Globig showed that a number of bacteria were capable of
growing at a temperature of 50° to 70° C. He isolated these
organisms by inoculating potatoes with soil, and then incubating
them at varying temperatures. In the same year Miquel
described the B. thermophylus, which he found to be very
prevalent in sewage-polluted water, as many as 1000 per c.c.
being detected in water from the Seine collected at the Bridge
of Austerlitz. According to Miquel the organism was not
present in spring-water ; it was occasionally found in air, but its
normal habitat appeared to be sewage. When drops of drain-
water were inoculated into broth, and then maintained at 69° C.,
the medium was found to become turbid in twenty-four hours,
and contained the B. thermophylus. On agar at 452° to 45° C.
the bacillus gave rise to a white raised meniscus-shaped growth,
which, on microscopic examination, showed the presence of a
short, plump bacillus, with a highly refracting spore at one
end. In broth it grew best at a temperature of 65° to 70° C.,
and produced an abundant white deposit at the bottom of the
tube, the liquid above becoming clear. The microscopic appear-
ance of the bacillus appeared to vary with the temperature at
which it was cultivated. At 50° C. the bacillus was short, with
an oval spore at one end ; at 60° C. it formed filaments, and
only a few spores appeared; at 71° to 72° C. no spores were
found, and the bacillus appeared swollen. It was not motile,
and produced no pathogenic effects in animals. In 1894
Macfadyen and Blaxall isolated thermophylic bacteria from
specimens of earth, derived partly from the surface and partly
from depths up to five feet. They also found the organisms in
Thames mud and in faeces from human beings and mice. In
1895 Rabinowitsch repeated and extended Macfadyen and
BlaxalPs work. This observer used potato as the nutrient
material for isolating the bacteria. After twenty hours incu-
bation at 62° C. four to eight white colonies appeared on the
inoculated potatoes ; after forty hours incubation yellowish-
grey, brown, and reddish-brown colonies also developed. Pure
cultures were obtained by inoculating the growths on plates
containing three per cent. agar. The thermophylic bacteria
were found in earth, snow, the River Spree, and in the excre-

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 157

merit of horses, cows, dogs, guinea-pigs, pigeons, mice, frogs, &c.
They were detected in gradually increasing numbers in the
mouth, stomach, small and large intestines. Rabinowitsch
considered that the bacteria were facultative anaerobes. At
high temperatures they grew well as aerobes, but very slowly as
anaerobes ; at lower temperatures, on the other hand, they grew
better under anaerobic than aerobic conditions, especially when
broth was used as the nutrient medium. The following table
gives the characteristics of the eight varieties of thermophylic
bacteria which developed at 62° C. :

Number.

Growth oil
potato.

Colonies on
plates.

Microscopical
characters.

Spores.

Reaction in
broth.

1

White
colonies

Coarsely
granular
colonies with
dentate

Non -motile
rods, often
growing into
threads

Oval spores
at one end of
the bacilli

Produces
acid

margin

2

Yellowish-
grey colonies
with irregu-
lar margins

Greenish
colonies, not
so coarsely
granular as 1

Non-motile,
slightly
curved rods

Spores in
the centre of
the bacillus

Produces
alkali

3

Brown
colonies

Small, round,
greyish,
sharply out-
lined colonies

Thick,
non-motile
rods

Oval spores
at one end of
the bacillus

Produces
acid

4

Reddish
colonies

Colourless
colonies, with

Non-motile,
often

Round spores
in the centre

Produces a
trace of

many thin
out-runners

producing
threads

of the bacilli

alkali .

5

Scanty
growth of a
brownish-
grey colour

Colourless
colonies,
granular in
the centre

Non-motile
rods

Oval spores
at one end of
the bacilli

Produces a
little acid

6

Moist, grey
colonies

Greenish-
grey colonies,
granular in
the centre

11

11

Produces a
considerable
quantity of
alkali

7

Whitish -grey
colonies

Coarsely
granular
colonies, with

»

»

11

a dentate

margin

8

Greyish-
brown moist
colonies

Round,
granular,
sharply out-
lined colonies

11

Spores in the
centre of
the bacilli

Produces
a trace of
acid

!58 BACTERIOLOGICAL EXAMINATION OF WATER.

None of these varieties were pathogenic to mice or pigeons,
The spores were not destroyed by exposure to current steam for
five or six hours. Also specimens of earth, after drying in an
oven for four or five months, produced distinct growths when
planted out on potato. The optimum temperature of the
thermophylic bacteria appeared to be between 60° and 70° C.
Macfadyen and Blaxall, in their later experiments with ther-
mophylic bacteria, employed salt potato-agar as a means of
isolating the different varieties of these organisms. They found
that the colonies of the bacteria remained discrete on this
medium, so it was possible to obtain readily pure cultures of
the organisms. "The potato-agar is prepared as follows:
Potatoes are first steamed, peeled, and pounded. To 100
grammes of potato is added one litre of tap-water, and the
mass is steamed for half an hour and filtered. To the filtrate
1*5 to 2 per cent, of agar is added, and the whole autoclaved
for fifteen minutes. It was found an advantage to add 1 per
cent, of salt. After neutralisation with soda, and further
steaming, the potato-agar is filtered into test-tubes and
sterilised once more." From the primary colonies on the salt
potato-agar sub-cultures were made on the ordinary culture
media. The various organisms were found to stain with
ordinary dyes, but, on the whole, carbol-methylene blue gave
the best results. It was usually found advisable before staining
to treat the cover-slip preparation with dilute acetic acid ; this
cleared away a zooglcea-like membrane, which often interfered
with clear staining. Macfadyen and Blaxall described fourteen
thermophylic bacteria.

Bacillus I. — Was isolated from animal dejecta, and produced
colonies on agar closely resembling those of the B. anthracis.
It liquefied gelatine on the fourteenth day. There was no
growth on potato, and no action was produced in milk. In
broth there was a viscous deposit, and the supernatant broth
was clear. It gave no indol reaction. It was a large, non-
motile bacillus, which did not stain with Gram, and contained
a spore towards the centre of the slightly swollen rod.

Bacillus II. — Also isolated from animal dejecta, produced
dull, white, round, smooth colonies. In broth there was a
deposit and a slight surface film; indol reaction was obtained.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 159

There was no growth on potato, and milk was rendered viscous.
It was a large, non-motile bacillus, with an oval spore towards
one end. It stained with Gram.

Bacillus III. — Isolated from soil ; was characterised by small
pin-head colonies. There was a viscous deposit and a stringy
pellicle in broth, and a positive indol reaction was obtained.
On potato there was a cream-coloured growth, and milk was
converted into a solid curd. In was a large, non-motile bacillus,
with a terminal spore, and stained with Gram.

Bacillus IV. — From ensilage; produced dull, white spherical
colonies. There was an indol reaction in broth, and milk was
converted into a solid curd. There was no growth on potato.
It was a large bacillus, with oscillatory movements. It stained
with Gram, and contained a small oval spore towards the
slightly swollen end.

Bacillus V. — From Thames mud ; gave rise to spherical,
yellowish-white colonies, with regular margin. Under a low
power the centre appeared dark-brown and finely granular ; the
outer zone was pale yellow, coarsely granular, with ridged
markings. Its reactions in broth, milk, and on potato were
the same as No. IV. The bacillus in form was more slender
than No. IV., and did riot stain with Gram.

Bacillus VL — Was obtained from Plymouth sea-water, and
produced dull, white, feathery colonies. It was a slender, non-
motile bacillus with a drum-stick spore.

Bacillus VII. — From soil ; produced round, dry, feathery
colonies. There was no perceptible growth in broth. On
potato there was a vigorous, moist, brick-dust-coloured growth.
Milk was converted into a solid curd. The bacillus was non-
motile ; about the size of the anthrax bacillus, and appeared
in pairs and short chains. It stained with Gram, and had a
large terminal spore.

Bacillus VIII. — From soil ; gave rise to yellowish-white
spherical colonies. On agar there was a pale lemon-coloured
growth. On potato a moist, cream-coloured growth. There
was an indol reaction in broth, and milk was coagulated with
separation of whey. It was a large bacillus with oscillatory
movement, and contained a large oval central spore.

Bacillus IX. — From soil ; produced spherical colonies, with a

160 BACTERIOLOGICAL EXAMINATION OF WATER.

dull ground-glass appearance. On agar and potato there was
a white, crusty growth. Milk was coagulated, with separation
of whey. An indol reaction was obtained in broth. It was a
large, non-motile bacillus with a tendency to long thread
formation. It possessed a very small terminal spore, and stained
with Gram.

Bacillus X. — From soil; was characterised by dull, white, circular
colonies. On potato there was a dry, felt-like, reddish-brown
growth. Milk was converted into a solid curd. There was no
indol reaction in broth. It was a very large motile bacillus
with a large, terminal oval spore.

Bacillus XI. — From soil ; formed round, dull, white colonies
with a moist, glistening appearance. On potato there was
a moist, brownish-yellow growth. Milk was coagulated with
separation of whey. There was an indol reaction in broth. It
was a large bacillus, which formed long chains. It stained with
Gram and had .a terminal oval spore.

Bacillus XII. — From soil ; produced stellate, yellowish-white
colonies. On potato there was a dry, pipe-clay coloured growth.
Milk was converted into a solid curd. There was an indol
reaction in broth. It was a large, non-motile bacillus with a
drum-stick spore. It did not stain with Gram.

Bacillus XIII. — From soil ; produced round, yellowish-white
colonies. On potato there was a moist, yellowish-brown growth.
Milk was coagulated with separation of whey. In broth an
indol reaction was obtained without the addition of a nitrite !
It was a large, non-motile bacillus with square ends. It stained
with Gram, and possessed a spore towards the swollen end of the
bacillus.

Bacillus XIV. — From soil ; gave rise to a thin surface film over
agar, with discrete and circular colonies at the margin. On
potato there was a moist, slate-coloured growth. Milk was
coagulated, with separation of whey. There was an indol
reaction in broth. The organism was motile, formed short
slender rods, surrounded by a zoogloea. It stained with Gram
and contained a spore in the centre of the bacillus. All the
above organisms grew well at 55° to 65° C. on ordinary media.

Macfadyen and BlaxalFs experiments did not support Rabino-
witsch's statement that at low temperature thermophylic bacteria

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 161

grow more quickly under anaerobic than aerobic conditions.
The exact conditions under which these organisms find, in
nature, temperatures favourable to their growth, are not yet
known ; they may possibly be of a chemical nature. Houston
found that thermophylic bacteria were present in London crude
sewage in abundance. These organisms are, however, found in
soils which have not been polluted recently, so their presence in
potable waters cannot be said to point to recent contamination.
But as all the workers on this subject have failed to detect
thermophylic bacteria in water from wells and springs, the
presence of these organisms in supplies derived from such sources
may safely be assumed to indicate gross contamination. Their
absence, however, cannot testify to purity or safety.

CHAPTER XIII.

QUALITATIVE ANALYSIS— continued.

CLASS III.

THE most important of all the micro-organisms which appear
in water are those which produce specific disease in human
beings. It is not absolutely proved that the causes of most
infectious diseases may not appear in water ; but at the present
time it is exceedingly difficult to show the presence of any of
them except the spirillum of cholera, and perhaps the
B. typhosus.

Bacillus Typhosus.

This micro - organism was first observed by E berth, and
more fully investigated by Gaffky. Its characteristics are as
follows :

Gelatine Plates. — The surface colonies appear as thin bluish
films with an irregular margin. Under a low power markings
are seen, which look like ridges and valleys, running irregu-
larly from the centre to the periphery ; sometimes the mark-
ings are very faint, and the colony resembles a very thin film of
glass. The colonies in the depth of the gelatine are round or
oval in shape, more or less translucent, and appear filled with
highly refracting granules. As a rule, the colonies in gelatine,
when incubated at 20° to 22° C., grow very slowly, requiring
seventy-two to ninety-six hours incubation before they show
their characteristic appearances. The gelatine is not liquefied.
Agtir Plates. - - The surface colonies are white, semi-
opaque, with a circular or slightly irregular margin. Under
a low power surface markings of ridges and valleys are seen,
but they are not usually so distinct as in the gelatine plates.
Sometimes the centre appears thicker than the periphery,
giving the colony a somewhat ringed appearance. When

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 163

examined with the naked eye, the growth on agar at 37° C. is
almost as rapid as that of B. coli, and it is often impossible
to distinguish between colonies of B. coli and B. typhosus on
agar plates. Still, as a rule, the colonies of the latter organism
are much smaller and more transparent than those of the former.

Gelatine-stab. — On the surface there is a thin circumscribed
growth like a surface colony, and along the line of inoculation
a thin greyish growth appears, ending below in discrete white
masses.

Gelatine-slope. — A thin, narrow growth developes along the
line of inoculation ; the margin is slightly irregular, but not
so markedly crenate as the corresponding growth of B. coli.

Agar-slope. — A thin, greyish-white growth, usually more
transparent than that of B. coli.

Potato. — A smooth glistening film develops, which is often
difficult to recognise with the naked eye. If the potato has an
alkaline reaction, a faintly brownish coloured growth may
appear.

Sugar Gelatine (Shake Culture). — There is no formation of
gas in lactose media, but glucose-gelatine is said to be some-
times fermented.

Witters Peptone and Salt Solution. — After seven days in-
cubation at 37° C. there is, as a rule, no nitroso-indol reaction
when the medium is tested with 1 c.c. of potassium nitrite
(0*02^), and a few drops of pure sulphuric acid. Occasionally,
after standing some few hours, a faint rope tint is shown by
undoubtedly pure cultures of B. typhosus.

Milk. — After one month's incubation there is no change in
the appearance of this medium.

Neutral-red Glucose-agar . — In this medium, as a stab or
shake culture at 37° C., the B. typhosus produces no fluor-
escence or change in colour. One loop of a broth culture
of the suspected organism is added to the melted agar, cooled
to 40° C., when a shake is to be made. Scheffler recommended
0'3 gramme of glucose and 1 c.c. of a concentrated aqueous
solution of neutral-red, to be added to 100 c.c. of melted agar.
B. coli, when inoculated into this medium, produces a marked
greenish fluorescence after twenty-four, or at latest forty-eight.
hours incubation at 37° C.

164 BACTERIOLOGICAL EXAMINATION OF WATER.

Litmus-whey. — After seven days incubation at 37° C. there is
a slight development of acid, which, however, never requires

more than 6 per cent, of ^ alkali to neutralise it.

Broth. — A diffused growth takes place; an imperfect film
appears on the surface after several days incubation.

Gelatine (25 per cent.). — After incubation at 37° C. no film
appears on the surface, but a diffused growth takes place
throughout the medium.

Proskauer and Capaldi's Medium, No. I. (Asparagin and
Salts). — After twenty-four hours incubation at 37° C. there is
no growth.

Proskauer and Capaldi's Medium., No. II. (Witters Peptone
and Mannite). — After twenty-four hours incubation at 37° C.
there is a distinct growth, and the medium acquires a strongly
acid reaction.

Microscopical Characters. — A bacillus about 3 /u long and
1 ni broad, highly motile, with a quick, serpentine move-
ment.

Staining Reactions. — Stains readily with basic dyes. It is
decolorised by Gram's method.

Flagella. — Eight to twelve long wavy flagella, disposed all
round the bacillus.

Spore-formation. — It does not form spores, and is destroyed
when the cultures are heated to a temperature of 65° C. for ten
minutes.

Reactions with Anti-typhoid Serum. — The B. typhosus is quickly
agglutinated by a highly dilute anti-typhoid serum. A small
dose of anti-typhoid serum injected into the peritoneum of a
guinea-pig, along with a lethal dose of B. typhosus, protects the
animal from the effects of this organism.

The most important of the above tests are the following :

(1) The Slow Growth on Gelatine Plates. — This test is of great
value ; none of the varieties of B. coli grow so slowly on gela-
tine plates as the B. typhosus. The B. enteritidis of Gartner,
and the organisms of the so-called Gartner group, are the only
bacilli which really grow like B. typhosus on gelatine plates,
though even these appear to grow more quickly than the true
typhoid organisms.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 165

(2) The Absence of Gas in Lactose Media. — By means of
this test the typical members of the Colon group are at once
excluded from the typhoid group. Lactose media are recom-
mended for this test, as some bacteriologists declare that the
B. typhosus may ferment glucose media. Lip to the present
time I have never found the B. typhosus produce gas bubbles
in a glucose-gelatine " shake " culture, or in a glucose-agar-stab.

(3) The Absence of Indol In Witters Peptone and Salt Solution.
— This test is not quite so exclusive as the tests just mentioned.
Two typical cultures of B. typhosus (one from Krai's laboratory)
in my possession give traces of indol after seven days growth at
37° C. in this medium. Also there are varieties of B. coli which
fail to produce indol when cultivated in this manner. Witters
peptone is a much better medium than broth for this test. I
have often found indol produced in the peptone medium when
not a trace could be discovered in broth. Muscle sugar is some-
times found in broth, and bacteria first convert this into acid
and then neutralise it by producing alkali. The indol reaction
is not obtained when the bacteria are unable to neutralise the
acid produced. Theobald Smith recommends the use of
dextrose-free broth, which is obtained by inoculating bouillon
with B. coli. After twenty-four hours the broth is sterilised
and filtered, and peptone and salt added in the ordinary
manner.

(4) The Unchanged Appearance of Milk. — I have never found a
true culture of B. typhosus which coagulated milk even after
incubation at 37° C. for one month. At the same time there
are many varieties of B. coli which do not change milk ; the
same is true of varieties of Gartner's bacillus.

(5) The Amount of Acid Produced in Litmus-whey. — After
incubation for one week at 37° C. the amount of acid is very
small, as compared with the typical members of the Colon group.
At the same time there are atypical coliform organisms which
produce less acid than the B. typhosus, so the test is not
conclusive.

(6) The Growth on Potato. — The colourless growth on potato
is a time-honoured test of great value, but the character of
the growth depends on the reaction of the potato. If the
potato has an acid reaction, the growth is colourless and trans-

166 BACTERIOLOGICAL EXAMINATION OF WATER.

parent; if the reaction is alkaline, the growth acquires a
brownish colour. This coloured growth is never so marked as
that obtained when B. coli is grown on the same potato.
Some varieties of B. coli, however, produce a colourless growth
exactly like that of B. typhosus. The best way to employ
the test is to inoculate the same potato on one side with B.
tvphosus and on the other side with the microbe under inves-
tigation.

(7) Proskauer and Capaldi's Media. — These media are of the
greatest value as a means of diagnosis. No. I. medium has the
following composition :

Asparagin ....... O20 per cent.

Mannite . * . ... . . 0-20 „

Sodium chloride .' . . .'..." . 0-02 „

Magrnesium sulphate . ... . . 0*01 „

Calcium chloride . . . 0'02 „

Potassium mono-phosphate .... 0'20 „

The reagents are dissolved in distilled water and sterilised
for an hour and a half in the steam steriliser. The medium,
which has a slightly acid reaction, is then carefully neutralised
with caustic soda. No. II. medium contains :

Witte's peptone . . . . . 2-0 per cent.

Mannite .... . . . .01.,

The reagents, as in No. I. medium, are dissolved in distilled
water and sterilised in the steam steriliser for an hour and a half.
The medium will then be found to have an alkaline reaction ;
it must be carefully neutralised with citric acid. A solution of
litmus is then added until the media acquire a deep-red colour.
They are then sterilised again for half an hour, filtered and
placed in sterile test-tubes, which are plugged with cotton wool
and covered with a cap of parchment paper, which is fastened to
the tube by an india-rubber ring.

If these media are inoculated with the B. typhosus and B.
coli respectively and incubated at 37° C. for twenty hours, the
following changes will be noticed :

No. I. medium. No. II. medium.

/No growth or change hO
J3. lyphosus-5 reaction j Growth, reaction strongly acid.

/"Growth, reaction neutral or
B. Coli . Growth, reaction acid x fahit] a]k;iljm._

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 167

After two or three days incubation the acid reaction of
B. coli in No. I. medium changes to an alkaline reaction.

Most of the gas-producing varieties of B. coli conform to the
reaction described by Proskauer and Capaldi.

The B. enteritidis of Gartner also behaves like a B. coli.
Varieties of B. coli which I isolated from water and which
strongly resembled the B. typhosus in nearly every test failed to
give the characteristic reaction in these media ; they all grew
in No. I. medium and rendered No. II. medium alkaline in
twenty hours. The B. typhosus simulans varieties (a, b, c,
and d) described by Houston were also, according to Horton-
Smith, at once differentiated from the typhoid bacillus by means
of these media.

Radzievsky has lately tested Proskauer and Capaldi's reactions
with sixty-six varieties of the B. coli. All the microbes grew
in No. I. medium, and in No. II. medium the reaction was
neutral or slightly acid.

Proskauer and Capaldi's tests are thus seen to be extremely
valuable, as by them the B. typhosus can at once be distinguished
from a large number of bacteria which strongly resemble it.
At the same time it cannot be said that a bacillus which does
not grow in No. I. medium and produces an acid reaction in
No. II. medium is necessarily the B. typhosus. I have quite
lately isolated from polluted water three varieties of B. coli
which produced gas in glucose media, and yet reacted in Pros-
kauer and Capaldi's media exactly like the true B. typhosus.

(8) The Reactions with Anti-typhoid Serum. — (a) The aggluti-
nation test. When serum from an animal immunised by injections
of B. typhosus is mixed with a broth culture of B. typhosus, or
an emulsion in broth of a twenty-four hours growth of the
typhoid bacillus on agar, it is found that the bacilli become
immobilised and heaped together in motionless masses, or in
other words agglutinated. The reaction &o produced was at
first considered specific, but further study soon showed that
the sera of healthy human beings and animals also possessed
the power of agglutinating the typhoid bacillus. WidaFs
experiments showed (1) that the serum from a horse, monkey,
or rabbit, diluted from 1-30 to 1-50, might agglutinate
the B. typhosus ; (2) that the serum of guinea-pigs had usually

168 BACTERIOLOGICAL EXAMINATION OF WATER.

no effect, but exceptionally it showed an agglutinative action in
a dilution of 1-5. Biberstein and Stern examined the sera of
persons who were not suffering from typhoid fever. The following
results were obtained :

BIBERSTEIN. STERN-SKL.OWER.

Dilution B. Typhosus culture. Dilution B. Typhosus culture.

of serum. Agglutinated in of serum. Agglutinated in

1-10 24 per cent, of the cases 1-10 25 per cent, of the cases

1-20 6 „ „ ,. 1-20 8 „ „ „'

1-30 2 „ „ „ 1-30 2 „ „ „

1-40 0 „ „ „ 1-40 1

In my own experiments with normal serum from a horse I
found that the B. typhosus was always completely agglutinated
by the serum diluted 1-20, and a marked reaction was obtained
when the serum was diluted 1-80.

From these studies it is evident that, when an unknown serum
is being tested for a specific reaction with a known bacillus
(Widal's reaction), the errors introduced by the action of the
agglutinins present in normal blood may be avoided by working
with the unknown serum in a dilution of 1-50 ; but when
attempts are made to diagnose an unknown bacillus by the
action of a known specific serum upon it, the problem is more
difficult. Much recent work on this subject has shown that at
least three variable quantities enter into the reaction, viz. :
(1) the nature of the bacillus ; (2) the strength of the specific
serum ; and (3) the time during which the serum is allowed to
act on the bacillus. It will be necessary to consider these three
points in detail :

(1) The Nature or Species of Bacillus operated on by the
Specific Serum. — In the early days of serum diagnosis it was
thought that a specific serum only agglutinated the bacillus
which had caused the serum to assume its specific characters.
But a very little study soon showed that a specific serum, such
as an anti-typhoid serum, would agglutinate other bacilli besides
the B. typhosus ; and of late years an enormous amount of
work has been done in order to find out the essential points
required to establish a really specific reaction. The B. coli and
its varieties and B. enteritidis of Gartner being closely allied to
the B. typhosus in their cultural characteristics, have necessarily
received much attention, and many experiments have been made

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 169

to contrast the action of anti-typhoid serum on these organisms
and the B. typhosus.

Typhoid bacilli in collection
obtained from

Dilution of the serum.

^TT oT7

rrro"

100000

^CTTTJTTO"

irwW

Lou vain ....

+

+ +

.

+

Paris .....

+

+

+ +

+

+

Lille

+

+

+ +

+

+

Berlin . . - .

0

0

0 0

0

0

Gand

+

+

+ +

+

±

Vienna I. ....

4.

T

+ +

4-

+

Vienna II. . . ...

-1-

_|_

_l_ _|_

+

~

±

Breslau I. ....

_l_

_|_

+ +

+

Breslau II

+

+

+ +

+

+

Breslau III

+

+

+

+

+

±

l!£ Case / No-X

0

0

0

0

0

0

lf£ . t 2 to 16

+

+

+

+

+

±

§JJJ1 Case • 1 to 25

+

+

+

+

+

±

»!? 1 Case . 1 to 22

+

+

•+

+

+

±

3 o ( C No. 34
fif|gj Case I. No. 35

+

+

J

+

+

±
±

'| * $ " cultures I The
* 5 I ^ remainder

}oor±

0 or ±

0

0

0

0

«•« ( Case II. ( v0'*

+

+

+

+

+

±

-JL- 143 J ^:1()

"*"

+

+

~^

+

—

J" S a cultures
-^ g J ^remainder

| 0 or ±

0 or +

0

0

0

0

||l Caselll. ( None of
s| 36 \ themag-
l cultures ( glutinated

Uor±

0 or ±

0

0

0

0

-IS f

!£|jl 300 J^oneof

11 P| CUltm>eS [glutted

j-Oor ±

0 or +

0

0

0

0

« o 8 * | J 145 cultures .

0 or +

0 or +

0

0

0

0

NOTE. — The sign + indicates agglutination with deposit and clearing of the fluid.
The sign + indicates agglutination without deposit.
The sign 0 indicates the absence of all reaction.

Van de Velde's experiments on the value of agglutination in

170 BACTERIOLOGICAL EXAMINATION OF WATER.

the identification of Eberth's bacilli are often quoted. The
agglutination tests were made in a series of tubes of small
calibre, so that each tube held a column of fluid from J to 1 c.m.
in height. The serum was added to the culture contained in
the tubes, which were then incubated at 37° to 40° C. After
thirty to forty minutes the tubes were examined with the naked
eye. The anti-typhoid serum employed was extremely powerful,
having been obtained by injecting a horse for two years with
cultures obtained from the same specimen of B. typhosus. The
results obtained are shown in the table on page 169.

The cultural characteristics of these various bacilli were care-
fully studied, and Van de Velde found that the bacilli which
corresponded in their cultures to the B. typhosus were those
which were agglutinated by very small doses of the anti-typhoid
serum. The culture " Berlin " was the only exception, and this on
further study was found to give an indol reaction in Witte's pep-
tone solution, so it was excluded from the list of true typhoid
organisms. On the other hand, all the bacilli which corresponded
to the cultures of B. coli were not agglutinated by the anti-
typhoid serum.

Beco"s researches on the value of agglutination by anti-typhoid
serum as a means of diagnosis between the B. typhosus and coli-
form races form a strong contrast to Van de Velde's experiments.
Beco employed young broth cultures of B. typhosus and diluted
the anti-typhoid serum with distilled water or physiological
fluid. A known quantity of the culture was distributed amongst
a series of glasses, and the serum was added with a sterilised
pipette so as to give the dilutions mentioned in the table. The
glasses were examined by the naked eye, and from time to time
a loopful was taken and examined in a hanging drop ; at the
end of two hours the test was considered finished. When
agglutination occurred the homogeneous aspect of the culture
disappeared and small granular masses were seen in the fluid ;
the appearances in the hanging-drop agreed with the changes
observed by the naked eye. The specimens were also submitted
to the action of formalin according to Malvoz\s technique. Beco's
results are shown in the following table :

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 1 71

Micro-organisms.

Origin.

Action of
formalin.

Action of anti-
typhoid serum and
dilution employed.

B. typhosus I. ,

Typhoid stools

+

+ , 1-100,000

II.

55

+

+ 5 »

III.

0

+ 5

IV.

Collection

4-

+ 5 55

V.

55

0

+ 5

VI.

55

+

+ 5

VII.

T)rphoid stool

+

+ 5

VIII.

55

+

+ 5

IX.

55

0

+ 5

',, x. . • ,.

55

+

+ 5

XL . .

+

+ 5

XII. . .

55

+

+ 5

„ XIII. . .

55

+

+ , 1-10000

XIV.

55

+

+ 5 5

!, xv.

55

+

+ 5

XVI.

?5

+

+ 5

„ XVII.

55

+

+ 5

,. XVIII.

55

+

+ 5

B. coli I.

55

+

+, 1-10

Paracoli II.

55

+

+ , 1-100

B. coli III.

55

+

+ , 1-10

IV.

55

+

+ 5 55

V. .

55

+

+ 5 55

VI.

55

+

+ 5 55

VII.

55

0

+ 5 55

VIII.

55

+

+ 5 55

IX.

55

0

0

XIII.

0

+ , 1-100

XIV.

55

0

+ , 1-1,000

XV.

55

0

+ , 1-10,000

XVI.

+

+ , 1-1,000

XVII.

55

+

+ 5

XVIII.

0

+ . 1-100

XIX.

Normal stools

0

0

XX.

55

0

0

XXI.

0

0

XXII.

Urine, cystitis

0

+ , 1-1,000

XXIX.

Normal stools

0

+ , 1-10

XXX.

55

0

+ , 1-1,000

XXXI.

55

0

+ , 1-10,000

XXXII.

55

0

+ 5

, XXXIII.

55

+

0

XXXIV.

Meuse water

0

+ , 1-100

XXV.

55

0

+ 5 „

XXVI.

0

+ 5

B. fluorescens liquefaciens

Typhoid stool

+

+ , 1-10,000

„ non-liquefaciens
Proteus vulgaris ,

Collection

+
+

+ , 1-1.000

+, 1-100

„ mirabilis

55

+

+ 5 55

„ figurans •;

+

+ 5 '5

Proteus . • •

Normal stool

0

+ , 1-1,000

As a result of his work, Beco concluded that all the speci-
mens of B. typhosus, whatever their origin, were agglutinated

172 BACTERIOLOGICAL EXAMINATION OF WATER.

by the anti-typhoid serum in the high dilution of 1-100,000
(the table, however, shows several specimens of B. typhosus
which were not agglutinated by dilutions beyond 1-10,000).
The typical B. coli of Escherich varied extremely in its reaction
to the serum ; some were only agglutinated by a dilution of
1-10, others reacted to a dilution of 1-10,000. Taking Beco's
results as published by him in the above table, the agglutination
test would not appear to be an infallible means of diagnosis, as
varieties of B. coli and B. typhosus were agglutinated by the
same dilution of the anti-typhoid serum.

Jatta has quite recently re-investigated this subject and
arrived at some important results. He found that the serum
of animals immunised by injections of B. typhosus agglutinated
this organism more strongly than twenty-eight different speci-
mens of B. coli which he investigated. The serum behaved
very differently to the various specimens of B. coli ; the results
Jatta obtained are shown in the following table :

Serum of an immunised dog :

When diluted 1-10,000, agglutinated the B. typhosus.

When undiluted, had no action on fifteen out of the twenty-four

cultures of B. coli.

When diluted 1-10, agglutinated one specimen of B. coli.
When diluted 1-100, agglutinated five specimens of B. coli.
When diluted 1-300, agglutinated four specimens of B. coli.

Serum of an immunised sheep :

When diluted 1-1,000, agglutinated the B. typhosus.
When undiluted, had no action on ten specimens of B. coli.
When diluted 1-10, agglutinated five specimens of B. coli.
When diluted 1-30, agglutinated six specimens of B. coli.
When diluted 1-100, agglutinated seven specimens of B. coli.

The serum of a typhoid patient when tested with the speci-
mens of B. coli. gave the following results :

Serum of patient :

Diluted more than 1-300, agglutinated the B. typhosus.
Diluted 1-300, agglutinated seven specimens of B. coli.
Diluted 1-100. agglutinated four specimens of B. coli.
Diluted 1-10, agglutinated four specimens of B. coli.

Jatta's observations appeared to show that as the aggluti-
nating action of the serum on the B. typhosus increased or
diminished, the action on B. coli also increased or diminished.
The serum of the immunised dog, mentioned above, tested

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 173

on July 25, and again on October 2, gave the following
results :

JUL.Y '

OCTOBER 2.

Dilution of serum, which

Dilution of serum, which

Serum.

Culture.

completely agglutinated
the culture.

completely agglutinated
the culture.

Dog

B. typhosus
B. coli

1-10.000
1-300

1-1,000
1-100

1-300

1-100

1-300

1-100

1-300

1-100

1-100

1-30

1-100

1-30

"

'

1-100

1-30

These experiments are of great interest, as the dog was
inoculated with B. typhosus, and the results could not be in-
fluenced by a secondary Coli infection. When the serum of a
typhoid patient is tested with B. typhosus and B. coli, it is
sometimes found that the serum reacts in a higher dilution with
the B. coli than with the B. typhosus. Biberstein found that
in five cases of enteric fever which he investigated, the serum
reacted in higher dilutions with the B. coli than the B. typhosus.
He considered that the patients were suffering from a secondary
Coli infection. Stern and Widal have recorded similar results.
Consequently when the agglutination test is used to distinguish
B. coli from B. typhosus, it appears absolutely necessary that
the serum of an immunised animal should be used. If the
serum of a patient is employed, there is always the possibility
that he may be suffering from a mixed infection, and conse-
quently the serum may react to both B. coli and B. typhosus.

Working with sera derived from immunised horses, I have
lately tested the reaction to the specific agglutinins of 150
cultures of B. coli derived partly from healthy stools, and partly
from the stools of typhoid patients. The results obtained with
the cultures derived from healthy stools are shown in the
following table :

174 BACTERIOLOGICAL EXAMINATION OF WATER

ACTION OF ANTI-TYPHOID SEKUM ON CULTUEES OF
B. COLI FKOM HEALTHY DEJECTA.

Xo. of the culture.

Dilutions of the serum.

1-50

1-100

1-200

1-500

1-1000

Culture 1 .

0

0

0

0

0

2 . .

0

0

0

0

0

3 .

0

0

0

0

0

± - ••

0

0

0

0

0

„ 5 .

0

0

0

0

0

6 . .

o

0

0

0

0

7 . .

0

0

0

0

0

8 .

+

+

—

0

0

9 .

0

0

0

0

0

10 .

0

0

0

0

0

Cultures 11 to 22 .

0

0

0

0

0

Culture 23 .

+

+

0

0

Cultures 24 to 29 .

0

0

0

0

0

Culture 30 .

+

+

0

0

31 to 41 .

0

0

0

0

0

Culture 42 .

+

±

0

0

0

43 .

+

±

+

—

0

Cultures 44 and 45

0

0

0

0

0

Culture 46 .

4-

+

±

—

0

47 .

+

±

+

—

0

48 .

+

+

±

—

0

Cultures 49 to 51 .

0

0

0

0

0

Culture 52 .

+

0

0

0

Cultures 53, 54, 55

0

0

0

0

0

Culture 56 .

+

+

±

—

0

57 .

0

0

0

0

0

Cultures 58 and 59

±

±

±

—

0

Culture 60 .

0

0

0

0

0

Cultures 61 to 63 .

+

+

0

0

0

Culture 64 .

+

0

0

0

0

„ 65 .

+

±

±

—

0

66 ..

—

0

0

0

0

„ 67 .

0

0

0

0

0

68 .

+

—

0

0

0

Cultures 69 and 70

0

0

0

0

0

The sign + indicates complete agglutination.

„ + ,, nearly complete agglutination.

„ „ partial agglutination (a few clumps but many motile

bacilli).
„ 0 „ no traces of agglutination.

The results given in the above table closely agree with
Jatta's experiments, and show that the limit of the dilution of
anti-typhoid serum which causes agglutination of the B. coli
from healthy stools is about 1—200. The results, however, with
B. coli from typhoid stools were somewhat different, and suggest

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 17.5

that some of these bacilli may become temporarily susceptible
to agglutination by a highly dilute anti-typhoid serum.

Experiments with B. fluorescens liquefaciens and B. fhiorescens
putridus have shown that the limit of the dilution of anti-typhoid
serum which agglutinates these bacilli is about 1—100.

Coliform organisms which reacted strongly to dilute anti-
typhoid serum, have been recently isolated from water-supplies
by C. Sternberg. One organism was completely agglutinated
by serum diluted 1—1400, and another by serum diluted 1—600.
A stronger specimen of anti-typhoid serum which completely
agglutinated the B. typhosus, when diluted 1-10,000, did not
agglutinate the coliform organisms in a dilution above 1-5000.

It thus becomes evident that to bring out the specific re-
action it is necessary to dilute the serum from the immunised
animal. This opens up the difficult question as to what dilu-
tion is to be considered necessary.

Many bacteriologists believe that the B. typhosus, no matter
from what source it is obtained, will always react in the same
way to the same serum. Biberstein working with ten different
cultures, Bensaude and Achard with twenty different cultures,
and Widal with twenty-six cultures, found no difference in
the agglutination of the different cultures with anti-typhoid
serum.

Jatta also failed to detect any difference in his cultures which
had been grown on favourable media for periods varying from
some months to many years.

Kolle, however, stated that the agglutination produced by a
serum was inversely proportional to the virulence of the bacilli.
Gruber advised that a culture of little virulence should be taken
for the diagnosis of a serum, because such a bacillus was sensitive
to its action. He also thought that by means of a serum, the
agglutinating action of which had been exactly estimated, it
was possible to decide as to the virulence of a culture. Mills
also found that the susceptibility of a bacillus to agglutination
was inversely proportional to its virulence.

My own experiments showed distinct differences in cultures
from different sources. As all the races were tested in the same
manner with the same serum the results are perfectly com-
parable. At the same time the virulence of these cultures, when

176 BACTERIOLOGICAL EXAMINATION OF WATER.

tested in Wright's laboratary at Netley, did not bear out Rollers
and Gruber's contention.

Culture G. of B. typhosus, which was not agglutinated by
the anti-typhoid serum, when diluted above 1-500, possessed
little or no virulence as compared with culture 13 p, which was
markedly agglutinated by the serum diluted 1-10,000. Culture
Gx which was isolated from the spleen of a typhoid patient,
at first proved highly resistant to the action of the anti-typhoid
serum, but after being preserved in milk for six months it re-
acted to the serum diluted 1-1000, and yet with this diminished
resistance to the anti-typhoid serum there appeared to be no
diminution in its virulence. Consequently it appears that there
is no constant relation between the virulence of the B. typhosus
and its reaction to anti-typhoid serum.

All my cultures, with the exception of Gx, were completely
agglutinated by anti-typhoid serum diluted 1-500, and this I
should propose as a working limit, as it would exclude all forms
of Coli isolated from normal stools. At the same time the
existence of such cultures as Gx shows that it is impossible to
lay down a hard and fast rule. This culture was carefully
worked out and compared with other undoubted cultures of
B. typhosus before it was finally accepted as a true B. typhosus.
The results of the various tests are shown in the table on
page 232, from which it appears that Gx responds to all the
cultural reactions considered typical of B. typhosus. Cultures
of this nature are probably rare, as I have only met with this
one during several years' work at the subject.

Sacquepee has described " Eberthiform " bacilli, which, when
isolated from water, typhoid stools, and also from the spleens of
fatal cases of typhoid fever, did not appear susceptible to agglu-
tination by anti-typhoid sera. The Eberthiform bacilli also
differed from the true B. typhosus by producing a brownish
growth on potato. When these bacilli were preserved in closed
tubes, without access of light or air, they were gradually trans-
formed into typical typhoid bacilli, producing a colourless
growth on potato and reacting readily to anti-typhoid sera.
The time required for this transformation varied between
six and ten months, and the susceptibility to agglutination
appeared to be gradually acquired during this period. The

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 177

Eberthiform bacilli when injected into animals, before the
transformation had taken place, were found capable of producing a
serum which in high dilutions readily agglutinated typical
typhoid bacilli. Sacquepee believes that Eberthiform bacilli
found in water have been derived from typhoid stools, the
transformation from the true typhoid bacillus to the Eber-
thiform variety having taken place within the human body.
He could not convert B. coli into the Eberthiform variety,
but found that true typhoid bacilli introduced into collodion
bags and then placed in the peritoneal cavity of a white rat,
which had been strongly immunised by injections of B. typhosus,
were converted into the Eberthiform variety. The typhoid
bacilli so treated lost their susceptibility to specific agglutinins,
and produced a coloured growth on potato. SacquepeVs results
are very interesting, but require confirmation by other bacterio-
logists. In the present state of our knowledge I think a dilution
of 1-500 of anti-typhoid serum may be accepted as a fairly safe
working guide for the diagnosis of the true B. typhosus.

(2) The Strength of the Specific Serum. — Having considered
the first variable quantity of the agglutination test, viz., the
bacillus operated upon, it is necessary to mention a few points
in relation to the strength of the specific serum, because its
potency is intimately related to the dilution of the anti-typhoid
serum, which is to be employed in the routine test. It is well
known that sera prepared by immunising animals with cultures of
B. typhosus vary greatly in their power of causing agglutination
of this bacillus. At one time it was supposed that the
results were largely dependent on the virulence of the culture
used for the process of immunisation. Later experiments have
thrown some doubt on this theory, so the question must still be
considered sub judice. Whatever the explanation may be, the
fact remains that sera as prepared vary extremely in their
agglutinating action. The action of an anti-typhoid serum
appears to be more marked on the variety of the bacillus, by
means of which it has been prepared, than upon other varieties
of the organism ; that is to say, a serum has a " special "
action upon its own organism, and a non-special action varying
in degree on other varieties of the organism. Recognising
these facts, attempts have been made to prepare polyvalent

M

178 BACTERIOLOGICAL EXAMINATION OF WATER.

sera, and to study how far the value of an anti-typhoid
serum is affected by the employment of several varieties (races)
of B. typhosus, either together or in succession. Ainley
Walker found that there was no cumulative action or
addition of agglutinating powers of the constituent monovalent
sera in the production of a polyvalent serum. That is to say,
the valency of polyvalent sera against any one of the varieties of
typhoid used in its production is neither greater nor less than
that of the monovalent serum obtained by the use of that
variety alone. Further, it appears that " the polyvalent serum
has no more powerful action on an independent variety of the
bacillus than the most powerful towards that bacillus of its
monovalent constituents." The " special "' agglutinative power
already obtained towards one variety of typhoid is found to
diminish when the injections are discontinued, in spite of the im-
munisation with another variety of the bacillus being continued.
For purposes of diagnosis either monovalent or polyvalent
sera may be used, though perhaps the latter would be more
generally useful. Ainley Walker found that polyvalent rabbits'
sera could be prepared which, when diluted in 1-3500, com-
pletely agglutinated the most pathogenic variety of the typhoid
bacillus against which it was tested. Horses'* sera, prepared
by immunisation extending over one year, when diluted
1-500,000 completely agglutinated the same variety of the
typhoid bacillus. Anti-typhoid monovalent. sera, prepared by
immunising rabbits and guinea-pigs for short periods, do not
usually bear dilution beyond 1-1000. It is quite easy to prepare
sera, which, when diluted 1-1000 will completely agglutinate the
B. typhosus ; but to obtain sera of greater strength requires
considerable time for the process of immunisation. Also, I have
found that " special "" sera, which act in a high dilution, slowly
lose their agglutinative action when kept at room tem-
perature. One serum in particular, which, when diluted
1-2,000,000 completely agglutinated different races of the
B. typhosus, slowly lost its agglutinative action, and at the
expiration of three months it ceased to influence the typhoid
bacillus when employed in a dilution above 1-10,000.

Recognising these facts, it is important that every sample of
anti-typhoid serum shall be standardised with the stock cultures

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 179

of different races of the B. typhosus, before it is used to
diagnose an unknown bacillus. A serum which fails to agglutinate
the stock cultures when used in a dilution of 1—1000 should be
rejected as being too weak to secure satisfactory results.

(3) The Time during which the Serum is allowed to Act on the
Bacillus. — The third factor of the agglutination reaction becomes
of great importance when weak anti-typhoid sera are employed.
If a quickly acting powerful serum is used for the test the time
limit of two hours is quite sufficient for all practical purposes.
But sometimes a powerful serum cannot be obtained, and

POWEEFUL ANTI-TYPHOID SEBUM.

B. TYPHOSUS.

B. COLT.

Dilution of tnc
serum.

Hanging drop

Capillary tube

Hanging- drop

Capillary tube

Ci hours).

(24 hours).

(2 hours).

(24 hours).

1-10

+

+

+

+

1-20

+

+

+

+

1-40

+

+

_j_

+

1-80

+

+

±

+

1-160

+

+

+

+

1-320

+

+

0

+

1-640

+ +

0

0

1-1280

+

+

0

0

1-2800

+

+

0

0

1-5000

+

+

0

0

1-10.000

+

+

0

0

1-100,000

+

+

0

0

1-2.000,000

+

+

0

0

WEAK ANTI-TYPHOID SEBUM.

B. TYPHOSUS.

B. COLI.

serum.

Hanging drop

Capillary tube

Hanging- drop

Capillary tube

(2 hours).

(24 hours).

(2 hours).

(24 hours).

1-10

+

+

+

+

1-20

+

+

+

+

1-40

+

+

+

+

1-80

0

+

0

±

1-160

0

+

0

+

1-320

0

+

0

±

1-640

0

+

0

0

1-1280

0

+

0

0

The sign + indicates complete agglutination.

„ ± „ incomplete agglutination (large clumps, but a few

motile bacilli).
., 0 „ no trace of agglutination.

180 BACTERIOLOGICAL EXAMINATION OF WATER.

under these conditions a longer exposure is necessary to
bring out the specific reaction. This is well shown by the
table on page 179.

GENERAL CONCLUSIONS ON THE REACTION OF AGGLUTINATION.

Summarising our present knowledge on the subject of agglu-
tination the following practical rules should be observed when
testing the reaction of an unknown organism to anti-typhoid

•e

serum.

(1) The serum from a patient suffering from enteric fever
must not be used, as it is possible that he may be suffering from
a mixed coli and typhoid infection.

(2) The serum of an animal immunised by injections of
B. typhosus should be employed. It should be standardised
against a stock culture of B. typhosus, and if, when diluted
1-1000, it fails to agglutinate this bacillus, it should be rejected.

(3) The agglutination reaction cannot be considered specific
unless it is produced by the anti-typhoid serum, diluted 1-500
and upwards.

(4) The specific reaction must be considered as only one link
in the chain of evidence, and the result of the test must be
taken in conjunction with the cultural reactions of the micro-
•organism under investigation.

Several methods have been suggested for the performance of
the agglutination test. Widal recommended the following
procedures for testing an unknown serum with the B. typhosus :

(a) Mix the serum with sterile broth in the required propor-
tions; then inoculate the mixture with the B. typhosus and
incubate at 37° C. After four to seven hours granular masses
appear, and at the end of twenty-four hours the bacilli are
usually found in little white heaps at the bottom of the tube
and the fluid appears nearly clear.

(b) Add the serum in the required proportion to a twenty-
four hours broth culture of B. typhosus and incubate at 37° C.
After several hours the culture loses its uniform opacity,
granular masses appear, and finally the broth clears by precipi-
tation of the masses to the bottom of the tube.

(c) Add the serum in the required proportion to a twenty-
four hours broth culture, or an agar culture made into an

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 181

emulsion with broth, and examine in a hanging-drop. This is
called the rapid method as compared with the two former,
which are slow methods.

(d) Wright has introduced a method of making the dilutions
by means of glass capillary pipettes. Normal salt solution is
used to dilute the serum, and a twenty-four hours agar growth
of B. typhosus is made into an emulsion with normal salt solu-
tion. The dilutions of serum, thoroughly mixed with an equal
volume of emulsion (which, of course, doubles the dilution of
serum already prepared), are placed in capillary pipettes, which
are then sealed at one end and kept at the room temperature
for twenty-four hours. When agglutination is " complete" the
bacilli are found as a firm sharply- outlined mass at the bottom
of the pipette ; if, however, the agglutination is " incomplete "
granular masses are seen throughout the tube. The appearances
must always be compared with a control tube containing only
the emulsion diluted with an equal volume of salt solution, which
has been left at the room temperature for the same time.

Wright's method of making the dilutions is admirable, but I
have often found that when normal salt solution is used to make
the emulsion, and also to dilute the serum, the sharpness or
sensibility of the reaction is much diminished. The following
method of procedure is therefore recommended :

Place a little of the anti-typhoid serum in a watch-glass,
insert the point of a capillary tube in the fluid and allow this to
run up the tube for about an inch and a half, mark on the
glass with a wax pencil the upper limit of the fluid, and then
blow out the serum into a watch-glass. Next dilute the serum
1-5. This is easily done by inserting the point of the pipette
in a little sterile broth (contained in a watch-glass), which is
allowed to run up to the mark on the tube. The broth is then
blown out into the watch-glass containing the serum, three
other equal quantities of broth, obtained in the same manner,
are next added to the serum, and the broth and serum are then
thoroughly mixed. Some of the dilution of 1-5 is now converted
into 1-25, by taking a portion of the 1-5 dilution and adding to
it four equal portions of broth. The 1-25 dilution is made into
1-50 by taking a portion and mixing with it an equal portion
of broth. The 1-50 dilution is converted into a dilution of

182 BACTERIOLOGICAL EXAMINATION OF WATER.

1-250 by taking a known volume of it in the pipette as before
and adding four equal volumes of broth. The 1-250 dilution
is converted into 1-500 by taking a portion of the dilution and
mixing with it an equal portion of broth. In this manner the
serum is diluted with broth in the following proportions: 1-5,
1-25, 1-50, 1-250, 1-500.

A twenty-four hours agar culture of the suspected organism
is now rubbed up with broth in a watch-glass until a perfect
emulsion is made. The end of a pipette is now placed in the
emulsion, which is allowed to run up the tube until a column of
any desired length, usually one-third of the tube, is obtained.
The upper limit of the column is marked as before, and the
culture is then blown out into a clean watch-glass ; a
column of fluid of equal length obtained by the same pipette
from the serum diluted 1-5 is now added, and the whole
thoroughly mixed. In this way a mixture of serum and culture
is obtained, the dilution of the serum being 1—10. The mix-
ture of serum and culture is finally drawn up into the pipette
until it is about half filled ; the point is then rapidly sealed in
the Bunsen flame. The same procedure is followed with the
other dilutions of the serum, and in this wav tubes containing
serum, mixed with an equal volume of culture, are obtained in
the following dilutions, viz., 1-10, 1-50, 1-100, 1-500, 1-1000.
For ordinary diagnostic work these dilutions are most useful,
but any other dilutions desired can be readily obtained by
applying the same principles. The tubes thus obtained are
preserved at the temperature of the laboratory and examined
at the end of two hours, and again at the end of twenty-four
hours. The results obtained at the end of two hours by the
naked eye can be confirmed in the hanging-drop by snipping off
the end of the capillary tube and blowing out a drop of the
mixed culture and serum on to a cover-glass and examining it
in the ordinary way. The capillary tube should be sealed up
again and preserved for twenty-four hours. If it is preferred,
a drop of the mixed serum and culture from the watch-glass
can be at once put up in a hanging-drop and kept under obser-
vation for two hours. In my own work I have found that in
the hanging-drop so prepared the clumps of bacteria often settle
to the bottom of the drop by the end of two hours, and it is

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 183

then difficult, if not impossible, to focus the clumps accurately,
consequently I prefer to open the capillary tubes and examine
the contents in the manner described ; the clumps are then seen
perfectly, as there is not time for them to settle down to the
bottom of the drop. Sometimes when the capillary tubes are
examined with the naked eye a deposit is seen at the bottom
of the tube, and without practice mere sedimentation of the
bacilli might be confused with agglutination, To avoid this
source of error, the tubes may be opened and the contents
examined by the hanging-drop method ; or more simply, the
question of agglutination may be decided by placing the tubes
horizontally on a table for five or ten minutes. If the clumps
are the result of true agglutination, the masses at the bottom of
the tubes will not have changed their shape, and the outline will
be sharp and convex ; if, however, the clumps have been formed
merely by sedimentation, the masses will have changed their
shape and the upper margin will be oblique, gradually tapering
off* to the lower side of the tube.

The results obtained by the twenty -four hours observation
are more delicate than those obtained at the end of two hours ;
but as many bacteriologists perform the test in the hanging-
drop and limit the observation to two hours, it is always well to
examine the tubes at the end of this time, so that the results
may be compared with those obtained by other observers.
Temperature has an effect on the agglutinative action of anti-
typhoid serum. Biberstein found that a serum diluted 1-14,000
clumped the B. typhosus when the preparation was kept at 37°
C., but at the room temperature the agglutinative action was
limited to a dilution of 1-12,000. I obtained very similar
results with my own experiments. The serum always acted in
a slightly higher dilution at 37° C. than at the room tempera-
ture. Consequently when results are stated it is always
important to give the temperature at which the experiment
was made.

Again, the results obtained with an emulsion made from a
twenty-four hours agar growth of B. typhosus are not compar-
able with those obtained with a twenty-four hours broth culture
of B. typhosus. I have found very marked differences in the
results obtained by the two methods when applied to the same

184 BACTERIOLOGICAL EXAMINATION OF WATER.

bacillus. Broth cultures also are not suitable for use in capillary
tubes ; at the end of twenty-four hours the growth is rarely
sufficient to give a marked reaction in the tubes, and if the
culture be incubated for a longer time false clumping is likely
to occur in the culture. P'or these reasons it is always best to use
a twenty-four hours growth on agar, and make an emulsion from
it bv rubbing up one or two loopfuls of growth in sterile broth.

Pfeiffer's Test. — This consists in injecting into the peritoneum
of a guinea-pig varying quantities of a twenty-four hours agar
growth of the microbe under investigation so as to determine
the fatal dose. Ten times the fatal dose, mixed with a little
serum (0*01 c.c. to O'OOl c.c. according to the strength of the
serum) from a highly immunised animal, is then injected into
the peritoneal cavity of a guinea-pig. If the bacillus under
examination be the B. typhosus the animal will remain well
whilst a control animal injected only with the bacillus will die.
In the peritoneal cavity of the animal which recovers the bacilli
are first converted into granular masses, and then disappear
under the lysogenic action of the specific serum. If the bacillus
be a variety of the B. coli, the animal will infallibly die ju^t
like the control animal. This test is of the highest importance ;
but, unfortunately, owing to the difficulties of the experiment it
is not applicable to general hygienic work. The test also has
this limitation, the typhoid bacillus may have lost its virulence,
so that a fatal result may not be obtained when the control
animal is injected with the agar culture.

Results obtained by Immunising an Animal with the Micro-
organism under Examination. — True races of the B. typhosus
when injected into an animal gradually produce a serum which
will agglutinate other races of B. typhosus. The extent to
which the serum so produced can be diluted, and still agglutinate
other races of B. typhosus, depends on the race of B. typhosus
used for, and the length of time occupied by, the process of
immunisation. Experiments at Netley showed that the least
pathogenic race of B. typhosus produced a serum which, when
diluted 1-100, completely agglutinated all the other races of
the typhoid bacillus in the laboratory collection. I have never
been able to isolate from water a micro-organism belonging to
the pseudo- typhoid group which would produce a serum capable

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 185

of agglutinating the B. typhosus. In doubtful cases I regard
this test of immunisation as a very important means of diagnosis
between the B. typhosus and varieties of B. coli. Fodor and
Rigler have shown that O'l c.c. and 0'5 c.c. of a typhoid broth
culture injected into a guinea-pig weighing 300 grammes
produce a serum with less agglutinating power than that
obtained when 1 c.c. is injected. They also believed from the
results of their experiments that increasing the amount of broth
culture to 5 c.c. per 300 gramme guinea-pig did not increase
the agglutinating power of the serum obtained, and that the
best results were derived from a serum taken from the animal
ten days after the subcutaneous injection of the broth culture.
These statements, however, did not hold good for the sera
prepared for me in the pathological laboratory at Netley. It
was always found that 3 c.c. of typhoid broth culture per 300
gramme guinea-pig produced a more powerful serum than 1 c.c.
of broth culture per 300 gramme guinea-pig. The whole question
of immunisation of experimental animals is very intricate ; much
depends on the health of the individual animal, it is very easy
to give too big an initial dose, and time must be given for the
animal to recover from the effects of the injection before pro-
ceeding to give another. By gradually increasing the dose, and
extending the injections over a period of three weeks, a more
powerful serum is usually obtained than when a single big dose
is given. Fodor and Rigler also believe that it is quite possible
for typhoid bacilli to become so altered that they will neither
become agglutinated by typhoid serum nor produce a serum
with agglutinating action ^hen injected into animals. In my
own experiments I have found that true typhoid bacilli may be
left in water and sewage for several weeks and still retain their
susceptibility to typhoid serum and the power of producing a
serum when injected into animals.

THE RELATION OF TYPHOID BACILLI TO OTHER PATHOGENIC
AND NON-PATHOGENIC BACTERIA.

Freudenreich has made some interesting experiments on this
subject. He inoculated flasks, containing 200 to 300 grammes
of broth, with different bacteria. After these had developed
the contents of the flasks were filtered through a Pasteur-

18fi BACTERIOLOGICAL EXAMINATION OF WATER.

Chamberland bougie, and the filtrate divided amongst a number
of small flasks. As a rule, the flasks were kept for from four to
.six weeks before being filtered. The flasks containing the
filtrate were then incubated to ascertain whether the broth had
been completely freed from bacilli. If there was no turbidity
after incubation of the filtrate, the flasks were inoculated with
different bacilli. It was found that when the B. typhosus was
inoculated into flasks containing the toxins of the following
bacilli, the results recorded below were obtained :

In the toxins of :

Staphylococcus pyogenes aureus, the B. typhosus grew very feebly.
Staphylococcus pyogenes albus, the B. typhosus did not grow.
Staphylococcus pyogenes foetidus, the B. typhosus did not grow.
B. pyocyaneus, the B. typhosus did not grow.
B. typhosus, the B. typhosus grew very feebly.

B. of symptomatic anthrax, the growth of the B. typhosus was delayed.
Chicken cholera, the growth of the B. typhosus was very feeble.
Pneumonococcus of Friedlander, the growth of the B. typhosus was very

feeble.

B. phosphorescens, the B. typhosus did not grow.
•Spirillum of Asiatic cholera, the B. typhosus grew feebly.
Spirillum of Finkler-Prior, the B. typhosus grew well.
Spirillum of Miller, the B. typhosus grew very feebly.
Spirillum of Deneke, the B. typhosus grew very feebly.

In flasks containing the toxins of the B. typhosus it was
found that the following bacilli grew well, viz., Staphylococcus
pyogenes aureus, albus, citreus, and foetidus ; Micrococcus
tetragenus ; Micrococcus prodigiosus ; Micrococcus rose us ;
Bacillus of blue milk ; Bacillus pyocyaneus ; Bacillus of sympto-
matic anthrax ; Bacillus megaterium ; Bacterium phosphorescens ;
Pneumonococcus of Friedlander ; Spirillum of Asiatic cholera ;
Spirillum of Finkler-Prior; Spirillum of Miller; Spirillum of
Deneke. The micro-organism of chicken cholera grew feebly, and
the B. typhosus very feebly in typhoid toxins. The influence of
the products of B. fluorescens putridus and B. fltiorescens lique-
faciens on the B. typhosus have already been described. The
influence of the toxins of B. coli on B. typhosus will be fully
considered under the section devoted to the experimental
researches on the duration of life of the B. typhosus in water
and sewage.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 187

THE RELATION OF THE TYPHOID BACILLUS TO ACIDS AND
ALKALIES IN NUTRIENT MEDIA.

The effects of the presence of acids and alkalies in nutrient
media on the development of the B. typhosus have been care-
fully investigated by Kitasato. Broth and gelatine media were
carefully neutralised, and varying quantities of the substances
under examination added. The medium was then inoculated
with a culture of the typhoid bacillus. When gelatine was
employed " roll cultures " were made, and the development of
the colonies of B. typhosus examined in the usual manner.

The sign + indicates growth, + growth restrained, — no growth.

Acids.

+ ±

per cent.

per cent.

per cent.

Sulphuric acid

0-049

0-065

0-08

Hydrochloric acid .

o-i

0-158

0-2

Nitric acid ....

0-1

0-157

0-2

Sulphurous acid

0-09

0-2

0-28

Phosphoric acid

0-15

0-224

0-3

Acetic acid ....

0-2

0-255

0-3

Carbolic acid ....

0-2

0-258

0-34

Formic acid .

0-22

0-278

0-35(5

Oxalic acid ....

0-23

0-285

0-366

Lactic acid ....

0.27

0-36

0-4

Tartaric acid . • . . . j

Citric acid . . . . j-

0-338

0-384

0-47(5

Malic acid . . . . J

Tannic acid ....

1-3

1 -6(5

Boric acid ....

1-5

2-0

2-7

Alkalies.

+

+

—

per cent.

per cent.

per cent.

Caustic lime ....

0-0725

0-0805

0-0966

Caustic potash . . . \
Caustic soda . . . . /

o-i

0-14

0-18

Ammonia . ...

0-148

0-2

0-3

Lithium carbonate .

0-514

0-6

0-666

Potassium carbonate

0-566

0-74

0-81

Barium hydrate

0-65

0-83

1-00

Ammonium carbonate .

0.72

0-845

1-0

Sodium carbonate .

2-00

2-2

2-47

Salts.

+

+

—

per cent.

per cent.

per cent.

Potassium iodide .

6-66

8-0

9-23

Potassium bromide

8-0

9-23

10-37

Potassium chloride

9-1

10-6

12-0

188 BACTERIOLOGICAL EX A MI NATION OF WATER.

When broth was used, the medium was inoculated with the
typhoid bacillus and kept at 36° C. until a good growth was
obtained. The chemical substance was then added, and the
culture kept at the room temperature for four or five hours. At
the end of this time, and ten to fifteen hours later, half a cubic
centimetre was plated out in a " roll culture." The results
obtained by Kitasato are given in the table on page 187.

Kitasato's results with carbolic acid do not agree with my
observations. The percentage of caustic lime required to kill the
typhoid bacillus is also much greater than that given by other
observers. Liborius found that 0'0074 per cent, of caustic lime
destroyed the typhoid bacillus when growing in broth. Kitasato
explains the discrepancy by the fact that he used concentrated
broth, whereas Liborius employed broth diluted fifteen times,
with distilled wrater. The phosphates present in the undiluted
broth required to be neutralised by the lime before this substance
could act on the bacilli. By using a diluted broth, Kitasato.
found that 0'0074 per cent, of lime destroyed the typhoid bacilli.

MICRO-ORGANISMS WHICH RESEMBLE THE B. TYPHOSUS.

There are many organisms which may be mistaken for the
B. typhosus unless great care is taken to apply the tests already
mentioned. In polluted waters B. coli and its varieties occur,,
and, before the modern tests were elaborated, there is no doubt
that this great group of micro-organisms was often confused
with the B. typhosus. There is, however, at the present day
not much difficulty in separating the great bulk of these-
organisms from the typhoid bacillus. The following table gives,
the reactions of the typical B. typhosus and B. coli :

B. COLI COMMUNIS. B. TYPHOSUS.

(1) Surface Colonies, Gelatine Plates. (1) Much thinner than those of
— Thicker, and grow more rapidly B. coli, and grow more slowly. After-
than those of B. typhosus. After forty-eight hours incubation at 22° C,.
forty-eight hours incubation at 22° C. they are hardly visible to the naked
they are usually large and character- eye.

istic.

(2) Gelati tie-stab. — Quick growth (2) Slow growth on the surface like-
on the surface and along the line of the colonies ; along the line of inocu-
inoculation. lation the growth is much thinner, and

often ends below in a few white points,
consisting of discrete colonies.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 189

B. COLT COMMUNIS.

(3) Gelatine-slope. — Thick, broad
greyish-white growth with a cremated
margin.

(4) Witters Peptone and Salt Solu-
tion.— Indol produced,

(5) Mint.— Coagulated.

(6) Litmus-whey, one week at 37°
C. — Acid produced, usually requiring

B. TYPHOSUS.

(3) Thin narrow greyish- white
growth, crenated margin not marked
as in B coli.

(4) No formation of indol.

(5) Unchanged after a month.

(6) Very small amount of acid
produced, requiring not more than <5

N

from 20 to 40 per cent, of , ~ alkali to per cent, of ;— alkali to neutralise it.

neutralise it.

(7) Neutral -red Glucose -agar. —
Marked green fluorescence.

(8) Glucose - gelatine and Lactose-
gelatine Shake Cultures, and Glucose-
agar-stab. — Marked gas formation.

(9) Gelatine, 25 per cent, incubated
at 37° C. — Thick film appears on the
surface.

(10) Potato. — As a rule, a thick
yellowish-brown growth.

(11) Proskauer and Capald'C s Media ,
No. I., after twenty hours. — Growth,
medium acid.

No. II. , after twenty hours. — Growth ,
medium neutral or faintly alkaline.

(12) titrate - broth. — Nitrate re-
duced to nitrite.

(13) Microscopical Appearances. —
A small bacillus often like a coccus,
not motile as a rule.

(14) Flagella.— Usually 1 to 3, short
and brittle ; sometimes 8 to 12, long
and wavy.

(15) Agglutination. — As a rule, no
agglutination with a dilute anti-ty-
phoid serum.

(7) No change.

(8) No gas formation.

(9) No film appears on the surface,
but a general growth takes place
throughout the tube.

(10) Thin transparent growth hardly
visible to the naked eye.

(11) No. I., no growth or change in
the reaction of the medium.

No. II.. Growth, medium acid.

(12) Keduction of nitrate not so
marked.

(13) Usually longer than B. coli :
highly motile, with a quick serpentine
movement.

(14) Usually eight to twelve, long
and wavy.

(15) Marked agglutination with di-
lute anti-typhoid serum.

The " atypical " members of the Coli group which fail to coagu-
late milk and do not produce indol sometimes give rise to difficul-
ties in diagnosis, but most of them, as a rule, can be distinguished
by the production of gas in glucose media. In polluted waters,
however, coliform organisms occur which are not so readily dis-
tinguished. The four varieties of B. typhosus simtilans, de-
scribed by Houston, are instances of this class of micro-organisms.
When investigated they showed the following characteristics :

190 BACTERIOLOGICAL EXAMINATION OF WATER.

B. Typhosus Simulans (Houston).

A.

B

C.

D.

Morphology

Long thin,
highly motile
bacilli

Long thin,
highly motile
bacilli

Long thin,
highly motile
bacilli

Long thin,
highly motile
bacilli

Surface colonies .

Greyish-
white thin
films with an
irregular
margin,
slow growth

Greyish-
white thin
films with an
irregular
margin,
slow growth

Greyish-
white thin
films with an
irregular
margin,
slow growth

Greyish-
white thin
films with an
irregular
margin,
slow growth

Shake- gelatine .

No gas

No gas

No gas

No gas

Lit in us- milk

formation of indol
after jive days

No acid, no
clot

Trace of indol

No acid, no
clot

No indol

No acid, no
clot

Trace of indol

No acid, no
clot

No indol

Growth on potato

Transparent
colourless
growth

Transparent
colourless
growth

Transparent
colourless
growth

Faint
yellowish-
coloured
growth

Flayella

3 to 6

3 to 6

3 to 9

3 to 9

A (/(/ lii tlnatlo n ic Ith
anti-typhoid serum .

Nil

Nil

Nil

Nil

The following organisms which I have recently isolated from
a polluted water are even more difficult, as some of these react
to anti-typhoid serum :

B. Typhosus Simulans (Culture A.).

Gelatine Plates. — The surface colonies appeared as thin bluish
films with an irregular margin. Under a low power the centre
of the colonies had a yellowish tinge, and surface markings of
ridges and valleys were seen running out from the centre to the
periphery.

Gelatine stab. — Surface growth resembled the colonies ; a thin
white growth along the stab.

Milk. — Unchanged after a month.

Litmus-whey. — Faint acidity, requiring less than 6 per cent.

of -^ alkali to neutralise it.

Glucose-gelatine Shake. — No gas formation.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 191

Witters Peptone and Salt Solution. — No indoi produced after
seven days incubation at 37° C.

Potato. — Growth at first transparent, later acquired a faint
yellowish tinge.

Gelatine (25 per cent.) at 37° C. — No film on the surface ; a
diffused growth through the tube.

Proskauer and Capaldi, No. I. — Growth, but no change in the
reaction of the medium.

Proskauer and Capaldi^ No. II. — Growth, medium rendered
slightly alkaline.

Microscopical Characters. — A small highly motile bacillus^
about the same size as B. typhosus.

Staining Reactions. — Stained by basic aniline dyes ; decolor-
ised by Gram's method.

Reaction to Anti-typhoid Serum. — Anti-typhoid serum, even,
in a dilution of 1-10, produced no agglutination.

This micro-organism only differed from the B. typhosus in that
it failed to grow typically in Proskauer and Capaldi's media, and
did not react to anti-typhoid serum.

B. Typhosus Sinmlans (Cultures B3, B4, B5, and B6).

These cultures all gave the following characteristics, and
appeared to be the same micro-organism :

Gelatine Plates. — Identical with those of culture A.

Gelatine-stab. — Surface growth the same as the colonies, which
appeared to grow more slowly than B. coli ; a thin growth
along the stab.

Gelatine-streak. — A thin growth, whichdid not show thecrenated
margin of B. coli, but was distinctly more copious than that of
B. typhosus planted out at the same time.

Witters Peptone and Salt Solution. — No indol produced after
seven days incubation at 37° C.

Glucose-gelatine Shake. — No gas formation.

Milk. — Unchanged after one month.

Potato. — Growth at first transparent, later acquired a yellowish
tinge.

Gelatine (25 per cent.) at 37° C. — No film appeared on the
surface, but a diffused growth appeared throughout the tube.

192 BACTERIOLOGICAL EXAMINATION OF WATER.

ProsTcauer and Capaldi, No. I. — Growth, but no change in
the reaction.

Proskauer and Capaldi, No. II. — Growth, but medium rendered
slightly alkaline.

Microscopical Characters. — A small, highly motile bacillus,
closely resembling the B. typhosus.

Staining Reactions. — Stains with basic aniline dyes ; decolor-
ised by Gram's method.

Reaction with Anti-typhoid Serum. — Practically complete
agglutination with anti-typhoid serum, diluted 1-80.

This micro-organism closely resembled the B. typhosus in
many of its reactions, but it did not give the typical reactions
in Proskauer and Capaldi's media and the anti-typhoid serum
which, when diluted 1—2,000,000, caused complete agglutination
of the true B. typhosus, had no complete agglutinating action
on the B. typhosus simulans when diluted above 1-80.

The B. Fsecalis Alkaligenes, isolated by Petruschky, is said
to resemble the B. typhosus in the following points: (1)
Active motility ; (£) number of flagella ; (3) decolorisation bv
Gram ; (4) appearance of the colonies on gelatine plates ; (5)
growth in milk without fermentation ; (6) growth in sugar
media without the production of gas ; (7) agglutination with
anti-typhoid serum ; (8) failure to produce indol. It is, how-
ever, supposed to be distinguished by producing a brown growth
on potato, and rendering litmus-whey alkaline after forty-eight
hours incubation.

A pure culture of this organism was obtained from Krai's
laboratory and carefully studied. It was a small motile bacillus,
closely resembling the B. typhosus. It did not stain with
Gram. In broth a surface pellicle was produced, which fell to
the foot on shaking the tube. In glucose-agar at 37° C., and in
lactose- and glucose-gelatine shake cultures at 22° C., there was
no formation of gas. Indol was not formed in Witte's peptone
and salt-solution after seven days incubation at 37° C. Litmus-
whey showed no appreciable change in reaction after seven days
incubation at 37° C. On potato a yellowish-brown growth was
produced. Milk was unchanged after seven days incubation.
In gelatine-stab the growth was slow, and strongly resembled

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 193

that of the B. typhosus ; the medium, however, acquired a
greenish tinge at the surface. In an agar-stab the same greenish
tint was seen in the medium ; on an agar-slope the growth was
thin and greyish in tint, the discrete colonies were small. In
gelatine plates the colonies grew slowly : those in the depth
were circular in shape, of yellowish-brown colour, and coarsely
granular in the centre; on the surface, the colonies showed a
wavv irregular margin and a dark centre, from which ridges and
valleys ran out to the periphery. The colonies in the depth
resembled B. coli ; those on the surface, however, were more like
the B. typhosus, both in slowness of growth and the small size
and appearance of the colonies. A growth took place after
twenty-four hours in both of Proskauer and Capaldi's media ; in
No. I. the reaction was alkaline, and in No. II. neutral. A
twenty-four hours agar growth was tested as to agglutination with
anti-typhoid serum ; the reaction was incomplete when the
serum was diluted 1-100. The same serum diluted 1—1000
completely agglutinated a stock culture of B. typhosus.

It will be noticed that this culture failed to give the cha-
racteristic alkaline reaction in litmus- whey. <jrermano and
Maurea also found that the formation of alkali by Petruschky's
bacillus was inconstant. The B. fyecalis alkaligenes appears to
be fairly common in faeces, so it is important to distinguish the
organism from the B. typhosus. The diagnosis can be readily
made, in the absence of the formation of alkali, by the brown
growth on potato and the absence of reaction to a highly
dilute anti-typhoid serum.

The B. Enteritidis, isolated by Gartner from cases of meat
poisoning, seems to be intermediate between the B. coli com-
munis and the B. typhosus. In gelatine plates the surface
colonies sometimes exactly resemble those of B. typhosus, at
other times the wavy margin and ridge and valley appearance
are not marked ; the colonies then appear coarsely granular
with a nearly circular outline. The colonies grow more quickly
than those of the B. typhosus, but more slowly than those of
B. coli. In gelatine-stab and streak the growths resemble those
of B. typhosus, but are more copious. In glticose-agar-stab at
37° C. gas is always produced, but in giucose-gelatine-shake at
22° G. it is sometimes absent. Gas does not appear

194 BACTERIOLOGICAL EXAMINATION OF WATER.

formed in lactose-gelatine-shake cultures. Milk is not coagu-
lated. There is no formation of indol in Witte's peptone and
salt solution after seven days incubation at 37° C. The growth
on potato varies ; sometimes it is colourless and transparent,
like that of B. typhosus, at other times there is a yellowish
slimy growth, resembling that produced by B. coli. A growth
takes place after twenty hours incubation in both of Proskauer
and Capsldi's media. In No. I. the reaction is sometimes acid
and sometimes alkaline ; in No. II. the reaction is either neutral
or alkaline. The behaviour of the bacillus in litmus-whey is
peculiar and very characteristic. It rapidly produces a large
quantity of acid ; this is followed later by the formation of
alkali, which neutralises the acid, so that the litmus-whey is
acid for the first two or three days and then becomes alkaline.
In microscopical characters the bacillus is almost identical with
B. typhosus. It is not stained by Gramas method. The flagella,
as a rule, are fewer than those of B. typhosus, but more numerous
than those of B. coli. In its reaction to anti-typhoid serum the
B. enteritidis shows less susceptibility to the specific agglutinins.
than the B. typhosus. Durham states that B. enteritidis is not
agglutinated by anti-typhoid serum diluted 1—50,000. As a
rule, this is true, but on one occasion I obtained a culture which
was completely agglutinated by anti-typhoid serum in this
dilution.

The B. enteritidis was found by Karlinski four times in normal
faeces, and Germane and Maurea have found bacilli closely
resembling it in water.

The Bacillus Breslaviensis belongs to the Gartner group, and
was found by Van Ermenghem and Kansche in outbreaks of
meat poisoning in Morseele and Breslau. It is very motile^
possesses from four to twelve flagella, and does not stain with
Gram. The colonies are coliform, broth becomes turbid, indol
is not produced, and milk is not coagulated. On potato there
is a thick yellowish growth. In agar and broth containing
grape-sugar, milk-sugar, or cane-sugar, gas is produced, but
the quantity is small, except in the media containing grape-
sugar.

The Bacillus Friedebergensis described by Gaffky and Paak,
;,s also closely allied to Gartner's bacillus. It is about one-third

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 195

the size of the B. typhosus and only slightly motile. Indol is
not produced and milk is not coagulated.

Schottmiiller has described a number of cases of " paratyphus.1"
The symptoms very closely resembled those seen in true cases
of enteric fever. The temperature charts were so characteristic
that the diagnosis of enteric fever seemed almost beyond cavil.
Rose-spots, enlargement of the spleen, and diarrhrea were also
observed in some of the cases, but in others the abdominal
symptoms were not so marked. From twenty to twenty -five
cubic centimetres of blood were removed from each patient and
made into plates on agar. The colonies which resulted were
carefully studied ; in every case motile bacilli which did not
stain with Gram were isolated. On gelatine plates and Pior-
kowski's alkaline urine medium the colonies resembled those of
B. coli. On potato a greyish-brown, thick growth, was observed.
In peptone and salt solution there was good growth but no
indol reaction was observed. Gas formation was seen in glucose-
broth and glucose-agar. Neutral-red agar was changed to a
yellow colour, and a greenish fluorescence was observed. Litmus-
whey was rendered acid for the first two or three days, but a
markedly alkaline action appeared later. When tested with
the serum from an undoubted case of enteric fever the bacilli
showed no traces of agglutination, but when tested with serum
collected from each of the cases all the bacilli became agglu-
tinated. The reaction was observed not only between the
bacilli and serum from the same case, but also between the
bacilli and sera from all the cases. The effective dilutions
of the sera varied from 1-100 to 1-10,000. All the cases
recovered, so an examination of the spleen and internal organs
could not be made. At the same time the bacteriological results
obtained seem to prove conclusively that the cases of "para-
typhus" were caused by a variety of the bacillus of
Gartner.

The B. aquatilis sulcatus varieties described by Weichselbaum
prod uce surface colonies which at first resemble those of B. typhosus
but on later development acquire a yellow colour. These bacilli
do not develop indol, coagulate milk, nor produce gas in glucose
media. Some of the varieties produce a yellow growth on
potato. They are readily distinguished by failing to agglutinate

196 BACTERIOLOGICAL EXAMINATION OF WATER.

with anti-typhoid serum and also by growing at 0° C., at which
temperature the B. typhosus does not develop.

The B. lactis serogenes resembles the B. typhosus in the
appearance of its colonies, but it is not agglutinated by anti-
typhoid serum. It also develops indol, coagulates milk, and
produces gas in glucose media.

The B. cavicida described by Brieger also resembles the
B. typhosus in the appearance of its colonies, but it develops
indol and produces gas in glucose media.

The B. fluorescens non-liquefaciens group of organisms
occasionally gives rise to difficulties, as the surface colonies have
a superficial resemblance to those of the B. typhosus. As a
rule, the marked venation on the surface colonies and the green
colour of the gelatine around them enables a diagnosis to be
made without difficulty. Sometimes, however, the venation is
not marked and the green colour is not developed for many
days. Under these conditions a mistake may be made, as these
bacilli do not produce gas, coagulate milk, nor give rise to gas
in glucose media. They are, however, not agglutinated by
anti-typhoid serum in a high dilution, though some members
of the group are agglutinated by the serum when diluted
1-80.

The B. Acidi Lactici produces surface colonies somewhat
like those of B. typhosus, but it is a spore-bearing organism,
coagulates milk, produces indol, and forms gas in glucose
media.

The B. Urese also gives rise to surface colonies which closely
resemble those of the B. typhosus. It is distinguished by not
reacting to the agglutination test and by its power of convert-
ing urea into ammonium carbonate.

The B. Psittacosis, which has been isolated from Parrots,
belongs to the typhoid group of organisms. It is an actively
motile bacillus, which does not stain by Gram's method. The
colonies on gelatine plates closely resemble those of the B.
typhosus. It does not ferment lactose, nor curdle milk, nor
produce indol in Witte's peptone and salt solution. It, how-
ever, produces a coloured growth on potato like the B. coli,
which serves to distinguish it from the B. typhosus.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 197

Coliform Organisms Isolated by C. Sternberg from Water.

No. I. Gelatine Plates. — Typical growth.

Broth. — Diffused growth without any surface pellicle.

Witters Peptone and Salt Solution. — No indol produced.

Milk. — Unchanged after seventeen days.

Litmus-whey. — 5*5 per cent. acid.

Sugar-media. — Gas formation.

Potato. — Thick greyish- white growth.

Microscopical Characters. — Short and long motile

rods, not stained by Gram.
Anti-typhoid serum. — Completely agglutinated by

the serum diluted 1-1400.

No. II. Resembled No. I. in most of the tests, but the
acidity was 9 per cent., and the limit of the dilution which
caused agglutination was 1-600.

METHODS WHICH HAVE BEEN PROPOSED FOR THE ISOLATION
OF THE B. TYPHOSUS FROM WATER.

In order to study gelatine plates and make sub-cultures of
suspicious organisms, numerous efforts have been made to repress
the growth of the saprophytic microbes without hindering the
development of the B. typhosus. Chantemesse and Widal pro-
posed to add 0'25 per cent, of carbolic acid to the nutrient
gelatine. Holz, however, showed that 0*20 per cent, of carbolic
acid prevented the development of the B. typhosus in Esmarch
roll tubes and on Koch's plates; the same effect was not ob-
tained when O'llT per cent, of carbolic acid was employed,
the B. typhosus always grew well, and the liquefying bacteria
were destroyed. Tiemann and Gartner found 0'25 per cent, of
carbolic acid very useful, and showed that the growth of the
typhoid bacillus was only slightly hindered when this amount
of carbolic acid was added to the water for three hours. Plates
made at the end of this time were found to be almost completely
free from foreign organisms. Holz's experiments showed that
it was not safe to use more than 0'15 per cent, carbolic acid
even for the three hours exposure. Parietti suggested a
solution containing 5 grammes of phenol and 4 grammes of
pure hydrochloric acid in 100 c.c. of distilled water : varying

1.98 BACTERIOLOGICAL EXAMINATION OF WATER.

quantities of this solution, usually 0*1, 0*2, and 0'3 c.c. were
added to tubes containing 10 c.c. of nutrient broth. The tubes
were then incubated for twenty-four hours at 37° C., and if they
remained free from growth, gradually increasing quantities of
the water — from two drops to 1 c.c. — were added to the tubes
which were again incubated at 37° C. for from twenty-four to
seventy-two hours. All the tubes which showed any growth
were then plated out in gelatine ; the colonies which developed
were examined under a low power, and those resembling
B. typhosus were fished and sub-cultured in various media.
Tiemann and Gartner very often found that the typhoid bacillus
did not grow in broth containing O'lO and 0'15 percent, of car-
bolic acid, and Losener obtained similar results. Vincent recom-
mended the addition of h've drops of a 5 per cent, solution of
carbolic acid to the broth tubes, which were then incubated at
42° C. Ravitsch-Stcherba added O'l per 1000 of a-naphthol to
the nutrient medium. Losener did not obtain satisfactory results
with this method ; the surface colonies of B. typhosus were
green in colour, granular, raised in the centre, and not easily
recognised. Holz prepared a potato-gelatine medium, which
had an acid reaction, and ten grammes of the medium required
1*6 c.c. of deci-normal alkali to effect neutralisation. Eisner
added 1 per cent, of potassium iodide to the potato-gelatine ;
the advantage of this addition has been denied by many workers.
On the iodised potato-gelatine medium the B. typhosus grows
slowly, and at the end of forty-eight hours the colonies appear
as small, clear dots, like minute drops of water ; the B. coli, on
the other hand, shows much larger colonies of a dark-brown
colour. Unfortunately these appearances cannot be relied upon ;
the colonies of B. coli sometimes appear small and transparent
like those of B. typhosus. Also many liquefying organisms
develop on the medium, so that the plates cannot be kept very
long under observation when the water is seriously polluted.
The great advantage of the potato-gelatine is that it diminishes
the number of colonies which must be sub-cultured for further
study. Only the small transparent colonies need be " fished •" ;
the large brown colonies can be at once excluded from the in-
vestigation. Uffelmann suggested that ordinary gelatine should
be exactly neutralised, and then O'Ol per cent, of pure carbolic

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 199

acid and 0*002 per cent, of methyl-violet added. The colonies
of B. typhosus grow in a characteristic manner and remain
distinctly bluer than the surrounding medium. Stoddart
recommended a medium containing 0'5 per cent, agar and 5 per
cent, gelatine. A sufficient quantity of this medium is poured
into a dish or flat flask, so as to fill it to a depth of 5 mm. ;
the dish or flask is then sterilised and allowed to cool gradually
(preferably in the steriliser over night). Drops of moisture
which may have condensed on the cover are drained off, and the
centre of the medium touched with a platinum needle charged
with the material to be examined. In order to prevent con-
densation the dish is enclosed in a larger one and incubated at
35° C. for twenty-four hours. If B. typhosus is present, about
two-thirds of the dish will be found occupied by a circular
opalescence, which later extends throughout the whole dish.
Under similar conditions the B. coli produces the usual colony
of limited extent. If the inoculation is made from a mixture
of both organisms, there results a central disc of coli bacilli,
surrounded by an opalescent halo of typhoid growth, which
yields perfectly pure sub-cultures. Unfortunately many of the
varieties of B. coli, which are motile, produce exactly the same
halo appearance as the B. typhosus ; so the test cannot be relied
upon to distinguish between these two microbes.

Hankin has published a method by which he states he has
been able to isolate the B. typhosus from water supplies in
India. The method is as follows :

^ Five tubes are taken, each containing 10 c.c. of neutral
bouillon. To the first tube no addition of Parietti's solution
is made. It merely serves as a control of the capacity of the
bouillon used to permit the growth of microbes. To the
remaining tubes are added one, two, three and four drops of
Parietti's solution respectively. Each tube is then infected with
a few drops of the water or other liquid to be tested. The
tubes are covered with india-rubber caps and placed in the
incubator at 37° C. for twenty-four hours. On the following-
day a variable number of the above tubes will be found to be
turbid. One of the series has now to be chosen for further use.
The tube containing the highest number of drops of Pariettfs
solution that is yet turbid should be discarded. Usually the

200 BACTERIOLOGICAL EXAMINATION OF WATER.

tube next below this in the series should be chosen. But the
character of the growth in the different tubes must be taken
into account. As a rule, a tube that has a thick scum on the
surface, or in which growth is only visible in the deeper layers
of the bouillon, or in which bubbles are seen in the liquid,
should be discarded. A tube should be preferred in which there
is a uniform turbidity. Usually the tube containing two
or three drops of Parietti's solution is the one chosen for
further use, although with the bouillon I employ, if infected
with dirty water, growths would occur in a tube containing ten
or twelve drops of the Parietti's fluid. Tubes containing the
larger quantity of Parietti should be employed if it is required
to isolate B. coli communis. The tube of bouillon chosen as
above is then used to inoculate a second series of bouillon tubes
to which successively increasing quantities of Parietti's solution
are added. The first tube of the second series has the same
number of drops of Parietti added to it as were present in the
tube taken for further inoculation from the first series< For
instance, supposing the tube taken from the first series was the
one containing three drops of Parietti, then the first tube of the
new series will also contain three drops, the next will contain
four drops, the next five drops, and so on. Two or three drops
of the bouillon in the tube from the first series are used to
inoculate each tube of the second series. These tubes are
covered with indiarubber caps and placed in the incubator as
before. On the following day choice of a tube has to be made
from the second series . . . the tube containing the highest
number of drops of Parietti's solution that is turbid is discarded,
and one of the lower tubes is taken for further use. The tube
of bouillon chosen must now be inoculated on to agar, having a
fairly dry surface in such a way as to produce isolated surface
colonies. A small quantity of the bouillon is taken up on the
end of an inoculating needle from the surface of the liquid.
The needle is introduced into the agar tube and rubbed on the
bottom of the slanting surface of the agar : then it is moved in
a zig-zag track over the remaining portion of the surface of the
agar. At least three agar tubes should be inoculated in this
way. After inoculation the agar is kept in the incubator. On
the following day the colonies that have developed must be

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 201

carefully examined. Each colony whose appearance is suspicious
must be inoculated on to a tube of litmus-agar . . . It is
not enough to inoculate two or three tubes from suspicious
colonies . . . five to ten tubes may be sufficient if the
water is comparatively pure. But if the water has been exposed
to very obvious contamination, the resulting colonies on the
agar will be very varied in aspect, and it will be necessary to
inoculate from them ten to twenty or an even larger number of
litmus-agar tubes." The litmus-agar contains twenty-five
grammes of litmus and thirty grammes of milk sugar to the
litre of nutrient agar. " On the day after their inoculation a
number of the litmus-agar tubes will be found to have turned
red, and may be at once discarded. After a further twenty-
four hours others may turn red, and may also be discarded.
. . . The remaining tubes that are still blue, and that have
the naked eye appearance of the growth of enteric, must now
be subjected to microscopical examination." The cultures
resembling B. typhosus are finally sub-cultured in milk, potato,
&c., and agglutination tested with anti -typhoid serum.

Using this method, Hankin states that he has frequently
isolated the B. typhosus from suspected waters in India ; and,
as most modern bacteriologists working with other methods have
failed to detect the B. typhosus in suspected waters, the pro-
cedure appears deserving of critical study. The tube which
Hankin selects from his first series is usually that containing
two to three drops of Parietti's fluid, which corresponds to
0*05 per cent, carbolic acid (1 c.c. = 30 drops), the amount
generally used by bacteriologists at the present day. The
second series, commencing with three drops, would run up to
six drops of Parietti, or in other words up to O'l per cent, of
carbolic acid, in which B. typhosus will usually grow, if it has
not been debilitated by prolonged immersion in sewage. So that
if the B. typhosus were not accompanied by B. coli it should
be detected by this method, providing of course it was originally
present in considerable quantity, so that typhoid bacilli might
be present in the small quantity of water used for the experi-
ment. If, however, B. coli is present with the B. typhosus (as
it usually is in most polluted waters) the result will be different.
B. coli will grow in 0'2 per cent, carbolic ( = 12 drops of

202 BACTERIOLOGICAL EXAMINATION OF WATER.

Parietti's fluid), and following Hankin's technique if the next
tube or next but one to this be taken it will contain ten to
eleven drops of Parietti, or 0'15 per cent, of carbolic, in which
B. typhosus often does not grow in twenty-four hours. Con-
sequently by Hankin's method B. typhosus may not be detected
when it is accompanied by B. coli. Hilbert carefully investi-
gated Hankin's method by adding a twenty-four hours broth
culture of B. typhosus to water from different sources. He
first estimated the number of bacilli which would be found in
varying dilutions of the twenty- four hours broth culture, and
came to the conclusion that with a dilution of 1-100,000,000,
three drops of the fluid would contain a typhoid bacillus. The
following table gives the results of his experiments with three
different samples of water :

Xo. and date of
the experiment.

Origin of
the Water.

Dilution.

Quantity
used.

Result
+ = B.
typhosus
detected.

No. of
drops of
Parietti's
solution
in the
tube used.

lie-
murks.

f Pipe-

}

1. 1-12-1899

water

1-10,000

3 drops

+

3

—

[(filtered)

1

2. 1-12-1899

1-1,000.000

»»

+

3

—

3. 4-12-1899

1-1,000,000

_j_

5

4. 3-1-1900

1-10,000.000

+

2

—

o. 7-12-1899

1-110.000,000

M

—

—

- —

(>. 3-1-1900

1-100,000,000

+

1

—

7. 7-12-1899

1,1,000,000,000

M

—

—

—

.f Pump in

)

8. 13-12-1899

-! Drumm-

I 1-100.000

n

+

3 and 4

—

( strasse.

J

9. 13-12-1899

55

1-1,000,000

v

+

2 and 3

—

10. 4-1-1900

1-10,000,000

J5

+

5 and 6

—

11. 4-1-1900

1-100,000,000

,,

—

—

—

12. 19-12-1899

/ Pregel
\ water

1-100,000

M

—

B.coli
found

13. 19-12-1899

??

1-1,000,000

^

—

14. 19-12-1899

1-10,000,000

"

—

In pure filtered water the B. typhosus was detected when only
three drops of a dilution of 1-100,000,000 was used for the
experiment. In water from a pump in the Drummstrasse the
same result was obtained with a dilution of 1-10,000,000. In
Pregel water, which contained the B. coli, the B. typhosus
could not be detected. Hilbert then endeavoured to tind out

QUALITATIVE BACTERIOLOGICAL ANALYSIS.

if the failure to find the B. typhosus in the Pregel water was
due to the presence of the B. coli. With this end in view he
mixed equal quantities of B. coli and B. typhosus in " pipe-
water," and also added typhoid stools, which were known to
contain B. typhosus, to the " pipe-water.11 The waters thus
polluted were investigated by Mankinds method, with the results
shown in the following table :

Xo. and date of
experiment.

Pipe-water, mixe:!
with

Dilution.

Quantity
use. I.

Result.

'Remarks.

1. 11-1-1900

Equal quantities

1-10,000

3 drops

_

B. coli

of twenty

found

hours cultures

of B. typhosus

~and B. coli

2. 11-1-1900

/ 55

1-100,000

—

3. 11-1-1900

5>

1-1,000,000

—

4. 30-1-1900

Typhoid stool

1-100

—

B. coli

found

5. 21-1-1900

M

1-1,000

—

M

6. 21-1-1900

"

1-10,000

—

It is evident from these experiments that Hankin''s method
will fail to detect the B. typhosus in water supplies polluted
with typhoid excreta. If, however, a water supply were
polluted only with urine, in which case the B. coli is usually
absent, the method would probably succeed in demonstrating
the B. typhosus, but under these conditions most of the other
methods would also be successful, and the case of a water-supply
being only polluted with urine so seldom occurs in actual
practice that Mankinds method must be considered of limited
value.

Piorkowski has lately recommended a method for isolating
the B. typhosus, which has been successfully used to determine
the presence of this organism in typhoid stools. A special
medium is employed, which is made by allowing ordinary urine
(specific gravity 1020) to turn faintly alkaline; 3*3 per cent,
gelatine and 0'5 per cent, peptone are then added, and the
medium sterilised in the "steamer" on three successive days.
After fifteen hours at 22° C. in this medium B. typhosus grows
into characteristic flagellated or Ramosus-like colonies, while the
B. coli appears only as round colonies, from which at times
short, plump processes may run out into the gelatine. Working

204 BACTERIOLOGICAL EXAMINATION OF WATER.

with Piorkowski's medium I have found that the appearances
presented by B. coli and B. typhosus are very variable. It is
very difficult to get the medium of a uniform composition, as
the time taken by different urines to become alkaline is not
constant. Also the characteristic forms of B. typhosus often
appeared in ordinary faintly alkaline gelatine, containing 33
per cent, gelatine. In 1896 Klie investigated the growth of
B. typhosus in media containing different percentages of
gelatine, and at different temperatures. He figured a very
characteristic colony of B. typhosus, growing at 19°— 21° C. in
3'3 per cent, gelatine, consisting of long threads growing out
from a central point. The appearances, however, changed
after twenty-four hours incubation, round centres appearing as
in the colonies of B. coli. Klie came to the conclusion
that the appearances of the colonies in dilute gelatine
media did not afford a sure means of diagnosis. The colony
figured by Klie, however, seems to me characteristic of
B. typhosus. None of the specimens of B. coli in my possession
ha,ve ever shown in 3'3 per cent, gelatine, after twenty-four
hours incubation at 19° to 21° C., a colony which at the centre
consisted of a point or fine line, from which long threads ran
out into the surrounding gelatine. The most motile forms of
B. coli under these conditions always showed a more or less,
thick circular dark-brown centre from which threads passed out
into the gelatine.

Mayer has lately investigated the growth of B. typhosus in
Piorkowski's medium, and also in 3'3 per cent, neutral gelatine.
In order to overcome the difficulty introduced by the difference
in the time required to render urine alkaline, Mayer recommends
the addition of a twenty -four hours culture of Proteus vulgans
to the morning urine. If the urine be kept at 22° C. it will
become faintly alkaline within fifteen to twenty hours ; gelatine
(3'3 per cent.) and peptone are then added, and the medium
sterilised in current steam at 100° C. In this medium (which
has a neutral reaction) five forms of colonies are obtained >
according to Mayer ; the fifth form or group is characteristic of
B. typhosus, and consists of colonies which have no appre-
ciable centre, or simply a thick point or streak from which
ffagella pass off' either on one side only or from all sides. This

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 205

fifth form exactly corresponds to the colonies described by Klie,
and Mayer also found that the same colonies were obtained in
a medium simply consisting of 3'3 per cent, gelatine and 0'5 per
cent, peptone, carefully neutralised with crystallised soda, or bi-
potassium phosphate. It therefore appears that the characteristic
growth of B. typhosus in Piorkowski's medium is chiefly
dependent on the small percentage of gelatine in the medium,
the temperature and length of the incubation, and the motility
of the bacillus. Mayer believes that if the characteristic form
appears, and the colony when planted out in glucose broth does
not; give rise to the production of gas, the diagnosis of B.
typhosus may be made.

Nearly all the methods described aim at giving the B. typhosus
time to 'develop, but they all fail to prevent the growth of
the B. coli and its varieties. I have lately employed a modifi-
cation of Parietti's method combined with the use of glucose-
litmus-agar plates. Two litres of the water to be examined are
concentrated by pumping through a Berkefeld or Pasteur-
ChamberJand candle, and the deposit on the candle is then
diffused by means of a sterile brush in 10 c.c. of sterile water.
One c.c. of the concentrated water is then added to each of ten
broth tubes containing O05 per cent, carbolic acid. It has been
shown that the B. typhosus which has existed for several weeks
in sewage, may not develop in broth tubes containing more than
0'05 per cent, carbolic acid ; consequently it is important not to
exceed this amount of acid. It is true that many liquefying
organisms can develop in such a broth, but this disadvantage is
more than counterbalanced by the fact that the B. typhosus, no
matter how debilitated it may be, will always be able to grow in
the medium. The broth tubes after inoculation are incubated
at 37° C., and kept under observation for seventy-two hours.
All the broth tubes which show a diffused turbidity are then
plated out on glucose-litmus-agar. This medium contains two
per cent, glucose added to ordinary agar containing sufficient
aqueous extract of litmus to give the medium a light blue
colour. For use the medium is melted, cooled to 42° C., and

then ^ NaHO added, so that each 10 c.c. of the medium has

vr

an alkalinity equal to 1'8 c.c. ^ alkali. The alkaline agar

206 BACTERIOLOGICAL EXAMINATION OF WATER.

mixture is then poured into Petri dishes and allowed to solidify.
Three plates are prepared for each turbid broth tube, one loop-
ful from which is stroked in a ring form, without recharging,
over the surfaces of all three plates. The colonies of B. typhosus
appear on the plates as small round transparent drops of -a bluish
tint, the colonies of B. coli are larger, more opaque, and red in
colour, also the medium around them shows the same red colour.
The bacilli which belong to the fluorescent group (B. fluorescens
liquefaciens and non-liquefaciens), also some of the Proteus
species, which are able to develop at 37° C., give rise to colonies
which have a deep blue colour, and are readily distinguished
from the colonies of B. typhosus and B. coli. The sewage
streptococcus produced colonies which by the naked eye cannot
be distinguished from the colonies of the typhoid bacillus. The
transparent blue colonies are then fished and examined in a
hanging-drop ; if motile bacilli like the B. typhosus are seen,
I am in the habit of preparing 1—50 and 1—500 dilutions of
anti-typhoid serum, and then adding a loopful of the 1-50
dilution to the hanging-drop already prepared, and a loopful of
the 1-500 dilution to another hanging-drop made from the
same colony. If agglutination of the bacilli is observed a
portion of the colony is then planted out on an agar-slope, and
the growth which results is then tested on all the various media
described under the cultural reactions of B. typhosus.

The special alkalinity of the glucose-lit mus-agar recommended
above was arrived at by testing the amount of acid or alkali
produced by B. typhosus, B. coli and its varieties, and the
various water bacteria. I thought at first that about 6 per

cent, of ^ alkali (0*6 c.c. per 10 c.c. tube) would suffice to
neutralise the acid produced by B. typhosus, and still leave B.
coli markedly acid. In litmus-whey tubes 6 per cent. ^ alkali

was sufficient to neutralise the acid, but on glucose-litmus-agar
poured into Petri dishes the typhoid colonies showed a marked
red colour in the presence of this amount of alkali. By gradually

increasing the amount of alkali it was found that 1'8 c.c. y^
NaHO. per 10 c.c. tube of lit mus-agar sharply differentiated
B. typhosus from most of the varieties of B. coli. There are a few
varieties of B. coli which are also neutralised by this amount of

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 207

alkali, but they are comparatively rare and, when sub-cultured,
are easily distinguished from the typhoid bacillus. By adopting
the alkaline glucose-litmus-agar the labour of investigating
plates is much diminished, with a little practice a large number
of plates can be examined in a very short time ; as a rule, only
about half a dozen colonies in each plate require to be fished.
Remy has lately suggested a modification of the potato-
gelatine medium. The new gelatine has a chemical composition
almost identical with the chemical analysis of potato ; it consists
of:

Distilled water , . . . . . 1,000 grammes

Asparagine . . . , . . . 6-00 „

Oxalic acid . . ' . 0'50

Lactic acid . . •. . . . .- O'lo „

Citric acid ... . 0-15

Di-sodic phosphate 5 '00 „

Magnesium sulphate . . . . . 2'50 „

Potassium sulphate . . . . .1-25 „

Sodium chloride 2'00 „

After sterilisation the gelatine is acidified with semi-normal
H9SO4, so that 10 c.c. of the medium shall have an acidity
which is exactly neutralised with 0'2 c.c. of a semi-normal solu-
tion of soda. Just before use, 1 c.c. of a 35 per cent, solution of
lactose and O'l of a 2'5 per cent, solution of carbolic acid are
added to each tube. In this gelatine the deep colonies of B. coli
are round, oval, or sometimes fusiform, and of a brownish-yellow
colour, fine bubbles of gas arising from the decomposition of
the lactose sometimes accompany the colonies. The superficial
colonies of B. coli are of two kinds : the first are globular, of a
brownish-yellow colour, and sometimes show vertical prolonga-
tions which are raised above the surface of the gelatine ; the
second are opaque films (spread out) with an irregular margin.
At firsh they are sometimes transparent and of a bluish colour,,
but later they rapidly become opaque. The colonies of B.
typhosus which appear at the end of two days are deep and
superficial. The deep colonies are bluish-white in colour and
much smaller than those of B. coli ; they are, however, very
distinct to the naked eye. The deep typhoid colonies are
never surrounded by bubbles of gas. The superficial colonies
are not generally visible until the third day. At first thev

208 BACTERIOLOGICAL EXAMINATION OF WATER.

have the appearance of moulds ; later they spread out, become
more blue in colour, and may attain the size of a 50-centime
piece. When the typhoid bacillus has great vitality, as for
example, if it comes from the spleen, its superficial colonies
more resemble those of B. coli. The deep colonies, however,
preserve their typhoid aspect. When the B. coli is feeble, its
deep colonies are less distinct ; they may lose their brownish
colour and become blue ; in this case they have a more
marked blue colour than the colonies of B. typhosus. When
plates are made in the medium from a fresh mixture of
B. coli and B. typhosus, there is a very marked difference
between the colonies of these organisms both in the depth
and on the surface; the development of gas bubbles,
however, is too irregular to be of much value. The great
points are the brownish colour and the opacity of the coli
colonies as compared with the transparent bluish colonies of
B. typhosus. At times the varieties of B. coli, especially when
they have been isolated from water, show colonies in the depth
which are quite indistinguishable from those of B. typhosus.

Remy^s medium is not elective ; other bacteria grow on it,
but not so well as B. coli and B. typhosus. It is, however,
fairly easy to prepare, and, being of constant composition,
might replace the Eisner potato-gelatine medium with
advantage.

Remy points out that it is better to directly plate out a
suspected water on his gelatine, containing 0*25 and 0*5 per
1000 of carbolic acid, than to pass the specimen through car-
bolic acid broth before making the plates. After ten days
association with B. coli, the B. typhosus is said to become
enfeebled, and if the weakened organism is passed through
carbolic broth, containing only 1 per 1000 of carbolic acid, it
is apt to die out. Remy has found that B. typhocus, after
being associated for thirty-eight days with B. coli, can be
easily isolated by directly plating out the water in his special
medium. But when the mixture of the two organisms is
passed through carbolic acid broth, the typhoid bacillus cannot
be found in the gelatine plates made from the carbolised broth
if more than ten days have elapsed since the admixture of the
typhoid with the colon bacilli. He also suggested that typhoid

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 209

bacilli in water have become adapted to a comparatively low
temperature, and incubation at 37° C. may be inimical to them.
Remy stated that typhoid bacilli isolated from a stool and kept
at the room temperature of 25° to 30° C. did not develop well
in broth at 37° C., but grew vigorously in broth kept at 25° to
30° C. The pernicious influence of carbolic acid on weakened
forms of the typhoid bacillus is undoubted, and many bacteri-
ologists at the present day have abandoned the use of carbolised
media and prefer to make a large number of plates with
ordinary gelatine. By employing a small quantity of water,
diluted if necessary, for each plate, they hope to obtain com-
paratively few colonies, and so give the typhoid bacillus time to
develop without the risk of being crowded out by liquefying

EXPERIMENTAL RESEARCHES ON THE DURATION OF LIFE OF
THE B. TYPHOSUS IN WATER, SEWAGE, ETC.

Kraus introduced a large number of typhoid bacilli into three
samples of unsterilised water, one being a pure pipe supply and
the other two well-waters ; the bacilli disappeared between the
fifth and seventh day in all three samples. Hueppe repeated
Kraus's experiments and obtained the same results. Karlinski
introduced 30,000 typhoid bacilli into the Innsbruck water, and
found that after seven days they had completely disappeared.
Karlinski also placed typhoid dejecta, containing typhoid bacilli,
in a cistern filled with river arid rain water. Three experiments
were made, and in all of them the B. typhosus could not be
found after the third day. Bobrow studied the duration of life
of the B. typhosus in a well-water containing 15,000 water-
organisms per c.c. ; the typhoid bacillus could not be isolated
after eight days. Hueppe introduced typhoid bacilli into un-
sterilised well-water kept at 16° to 20° C. ; out often experiments
he found the typhoid bacilli had disappeared after five days on
four occasions, after ten days on five occasions, and in one case
they appeared to live up to the thirtieth day. Frankland
made simultaneous experiments on water derived from the
River Thames, Loch Katrine, and a deep well in the chalk.
The same amount of culture was introduced into the three
specimens. In the Thames water the B. typhosus could not

210 BACTERIOLOGICAL EXAMINATION OF WATER.

be found after nine days, but in Loch Katrine water it lived
for nineteen days, and in the deep well-water for thirty -three days.
In the deep well-water the ordinary water bacteria increased far
more rapidly than in the Thames and Loch Katrine waters, so
Frankland believed that the disappearance of the B. typhosus
could not be due to the action of the water-organisms. Klein
introduced a recent culture of B. typhosus into unsterilised
tap-water derived from the West Middlesex, Kent, and New
River companies, and kept the samples at room temperature.
Eighteen days after the inoculation living typhoid bacilli were
found in O'l c.c. from each of the waters. After thirty-six days
typhoid bacilli were not detected in 1 c.c. of the New River and
Kent waters, but they were found in the West Middlesex water.
After forty-two days the typhoid bacilli appeared to have
disappeared from all three waters. In sterilised waters the
B. typhosus has usually been found to have a more prolonged
existence. Straus and Dubarry found it lived for sixty-nine
days in distilled water, for eighty-one days in sterile L'Ourcq
water, and for forty-three days in sterile Vanne water. Hoch-
stetter, however, stated that the B. typhosus only lived for five
days in distilled water and for seven days in sterile Berlin water.
Wolff huge! and Riedel, working with sterilised Panke water,
found that the B. typhosus multiplied when the temperature
was favourable, viz., above 16° C., but when the temperature
remained below 8° C. the bacillus remained alive but did not
multiply. Klein found the B. typhosus lived for eighty-five
days in sterilised water derived from the New River, West
Middlesex and Kent companies.

Numerous experiments have been made on the vitality of the
B. typhosus in sewage. Klein found that it could be detected
in ordinary sterile sewage three weeks after inoculation. Laws
and Andrewes experimented with sewage from Barking and
Crossness. At first the sewage was sterilised by boiling, but
in the later experiments the sewage was twice filtered through
a Berkefeld bougie and then heated to 60° C. in a sterile flask
for twenty minutes on two successive days. Planted out in this
sewage and incubated at 37° C. the B. typhosus appeared to
die within a few days : it could not be recovered after seventy-
four hours. Further experiments were made on the same lines,

QUALITATIVE: BACTERIOLOGICAL ANALYSIS. 211

but the sewage was now kept at 22° C. instead of 37° C. The
results obtained were as follows :

Immediately after 1

f. Y 22o colonies developed from one loopful of sewage.

After 24: hours . 250

„ 68 „ . . 140

„ 5 days . 48

„ .7 „ . 13

„ 13 „ . . 0

Experiments made with B. coli showed that this organism
was able to grow and multiply abundantly in sewage sterilised
by heat when incubated at 37° C., and that it could be cultivated
for several generations. After growing for four days in sewage
the B. coli still retained its powers of coagulating milk, of
forming gas-bubbles in gelatine and producing indol in broth.
As a result of these experiments, Laws and Andrewes concluded
that sewage, even in the absence of the normal micro-organisms
which it contains, is an unfavourable medium for the growth of
the typhoid germ, whereas the colon bacillus can grow and
multiply freely in it. Further experiments were then made
in order to gain some idea of the influence which various non-
pathogenic bacteria normally present in sewage would exert on
the life and growth of the typhoid bacillus when growing side
by side with it. Organisms were selected which did not grow
well at 37° C., viz., B. fluorescens liquefaciens, B. albus putidus,
B. fluorescens stercoralis, and B. cloacae fluorescens. The follow-
ing conclusions were arrived at :

The presence of certain non-pathogenic organisms commonly
present in sewage appeared to influence the extinction of the
B. typhosus. Of the four organisms tested B. fluorescens
stercoralis alone seemed to have any marked effect upon the
vitality of B. typhosus, and this effect was practically absent
when other organisms were present at the same time. The
mixture of the four non -pathogenic bacteria had no effect in
hastening the extinction of B. typhosus ; indeed, the reverse
appeared to be the case. My own experiments at Netley gave
somewhat different results from those obtained by Laws and
Andrewes. The experiments were performed as follows :

A specimen of sewage obtained from Yeovil was (a) placed

2 BACTERIOLOGICAL EXAMINATION OF WATER.

in sterile test tubes and sterilised by heating to 65° C. for ten
minutes on four successive days ; (b) placed in test tubes and
sterilised in the autoclave at 120° C. for fifteen minutes ; (c)
filtered through a Berkefeld bougie into sterile test tubes. All
the test tubes were inoculated with one loopful of a forty-eight
hours growth of the B. typhosus, and then kept in the laboratory
cupboard at a temperature which varied between 16° and 22° C-
The day following one loopful was removed from each of the
tubes and plated out in gelatine; after seventy-two hours
incubation at 22° C. all the plates were crowded with typhoid
colonies. Four days later a loopful was again removed and
rubbed over an agar slope ; after twenty-four hours at 37° C. a
growth appeared which, when made into an emulsion, was found
to be completely agglutinated by anti-typhoid serum. At the
end of three weeks similar results were obtained. After two
months a loopful was removed and plated out in gelatine ; very
few colonies appeared, and those which did grow were black in
colour with granular contents, and the margin was only slightly
wavy. In fact they appeared so unlike the original colonies
that it was feared some pollution had occurred. However, one
of the colonies was fished and inoculated on an agar slope ; after
twenty-four hours a growth appeared which, when tested with
anti-typhoid serum, was completely agglutinated by the serum
diluted 1-1000. Besides the change in the appearance of the
colonies the B. typhosus appeared to have less resistance to
carbolic acid, for while the original culture gave a copious
growth in broth containing 0'05 per cent, carbolic acid after
twenty-four hours incubation at 37° C., the agar growth prepared
from the discoloured colonies showed no growth after twenty-
four hours incubation at 37° C. in 0'05 per cent, carbolic acid
broth, but after forty-eight hours there was a slight loss of
transparency, and after seventy -two hours there was a marked
growth. Experiments on the same lines were made with the
same sewage diluted 1 in 10 ; identical results were obtained.
These experiments appeared to show that the B. typhosus will be
found alive after sixty days immersion in both strong and diluted
sewage containing its usual toxines and salts, but freed from
other living organisms. The power of reacting to anti-typhoid
serum will still be present, but the colonies may present a dark

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 213

crumpled appearance, and the bacillus will show diminished
resistance to carbolic acid. This last fact is of importance in
relation to Parietti's test for the isolation of the B. typhosus
from water. The behaviour of the B. coli in sterile sewage was
then studied. After immersion for forty-two days in sewage
filtered through a Berkefeld bougie it was found that the
production of indol was delayed and the bacillus seemed less
resistant to carbolic acid, for whereas the original B. coli grew
well in 0'2 per cent, carbolic acid broth after twenty-four hours
incubation at 37° C., the bacillus from the sewage showed no
growth in 0'15 per cent, carbolic acid broth after twenty-four
hours at 37° C., but grew well in 0*1 per cent, carbolic acid
broth. The action of B. coli on B. typhosus when growing side
by side in sterile sewage was next tested. A fresh twenty-four
hours culture of B. coli was added to test tubes containing sterile
sewage in which B. typhosus had been planted out twenty-four
hours earlier. After fourteen days at the laboratory temperature,
loopfuls were removed and plated in Holz's potato-gelatine,
and rubbed over the surface of ordinary gelatine solidified in
Petri dishes. No signs of the B. typhosus could be detected in
these plates. The experiment was repeated again, but in a
slightly different manner. A large loopful of a twenty-four
hours agar growth of B. typhosus was added to 10 c.c. of sterile
sewage and incubated at 37° C. for forty-eight hours ; the sewage
was then found quite opaque, and a loopful plated out in gelatine
produced innumerable colonies of the typhoid bacillus. A very
small quantity of an agar growth of B. coli was now introduced
into the mixture of sewage and typhoid bacilli, and the tube
preserved at the laboratory temperature. The B. typhosus was
isolated from the tube up to the end of five days, but after this
it could not be detected. Working on the same lines with B.
fluorescens Jiquefaciens I was not able to isolate the typhoid
bacillus after the seventh day. These experiments appeared to
show that the presence of B. coli and B. fluorescens liquefaciens
had a marked influence on the existence of the typhoid bacillus
in sewage. At the same time I was conscious that the results
were by no means conclusive, as some typhoid colonies mio-ht
have been present on the gelatine plates and yet escaped atten-
tion owing to the large number of colonies of B. coli which

214 BACTERIOLOGICAL EXAMINATION OF WATER.

developed. In later experiments by employing surface alkaline
glucose-litmus-agar plates instead of surface gelatine plates, I
have been able to sharply differentiate colonies of B. coli from
those of B. typhosus, and so obviate the defects of the earlier
experiments. By this means I found that when one loopful of
an agar growth of B. typhosus was mixed with one loopful of an
agar growth of B. coli in ten cubic-centimetres of sterile tap
water, the B. typhosus could be easily isolated after twenty-
eight days. The bacillus so obtained had lost none of its
cultural characteristics, and as regards its reaction to the specific
agglutinins contained in Berne serum it appeared, if anything,
more susceptible to agglutination after the four weeks existence
with B. coli than when it was first introduced into the tube.
The original culture was completely agglutinated by Berne
serum diluted 1-1000, and only partially agglutinated by the
same serum diluted 1-10,000 ; but the culture of B. typhosus
after association with B. coli was found to be completely
agglutinated by the Berne serum when diluted 1-10,000.
Attempts were then made to study the duration of life of the
B. typhosus in unsterilised tap-water to which B. coli was also
added. A loopful of B. typhosus was added to 10 c.c. of
unsterilised tap-water and twenty-four hours later a loopful of
B. coli was introduced. The tube was maintained at the room
temperature, and loopfuls withdrawn from time to time were
plated out on the alkaline glucose-litmus-agar medium. The
typhoid bacillus was easily isolated after twenty-one days. In
all these experiments the B. typhosus was present in large
quantity, and it seemed probable that the B. coli did not
multiply to any extent. When B. typhosus and B. coli were
added to peptone water, it was found that the B. coli under-
went rapid multiplication and no signs of the B. typhosus could
be discovered after seven days. Also when a loopful of a typical
typhoid stool was diluted in 10 c.c. of sterile tap- water and
examined at once, B. typhosus was found to be present in
considerable numbers ; but after seven days had elapsed the
typhoid bacillus could not be found.

Sidney Martin has investigated the growth of the typhoid
bacillus in the presence of particular soil organisms. Most of
the experiments were made with bacteria isolated from

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 215

Chichester soil. The organisms were obtained in pure culture,
but no attempt at naming them was made ; they were simply
indicated as Chich. 1, Chich. 2, &c. " The culture media of
experiment were a flask containing 200 c.c. of sterile distilled
water and about 10 c.c. of ordinary peptone broth ; and
sterilised soil, the soil used in each instance being that from
which the particular soil bacillus of experiment had been
isolated. The temperature at which the experiments were
performed was either that of the 37° C. incubator, the tempera-
ture of an outside shed (3°— 14° C.) or, in a few cases, the
temperature of the laboratory (13°-19° C.). The general method
of investigation consisted in inoculating the different media
with a definite quantity of a pure broth culture of one of the
soil organisms, and at the same time with an equal quantity of
pure broth culture of the typhoid bacillus, both cultures being
of the same age." The relative proportion of the two sorts of
bacteria present was determined at stated intervals by brushing
two or more agar plates with a sterile brush dipped in the
nutrient fluid. Surface colonies, generally discrete, were
obtained, and from these plates the relative proportion of each
organism present could be ascertained. Chichester 1 was a
fluorescent organism which did not liquefy gelatine. When
grown in a liquid medium at 37 C. with B. typhosus, the colonies
of both organisms could be easily detected twenty-four hours
after the inoculation; but, after six days, no colonies of
B. typhosus could be detected. The same result was obtained
when the medium was kept at 4° to 12° C. In sterile soil at
37° C. Chichester I. caused the disappearance of the typhoid
bacillus in six days, but at 8° to 12° C. the typhoid bacillus
could be detected after six days, though it was not found at the
end of fourteen days. Experiments on the same lines were
made with Chichester 2, which was a large, stout bacillus?, ami
formed spores freely. It was easily distinguished from thr
typhoid bacillus. In all the experiments conducted ai;
37° C. the typhoid bacillus at first grew more rapidly than
Chichester 2. As growth proceeded, however, Chichester
2 gradually increased more rapidly than the typhoid
bacillus, and eventually gained the upper hand. At shed
temperature, Chichester 2 maintained the ascendency at the

216* BACTERIOLOGICAL EXAMINATION OF WATER

first, and in two weeks was the only organism that could be
found ; but later the typhoid bacillus, present apparently in too
few numbers to be seen 011 the plate when growing at the shed
temperature, gradually increased, and in the final experiment
at 37° C. assumed the upper hand. The experiments made
with Chichester 3 were ot special interest. Chichester 3 was a
stout bacillus with rounded ends, forming long chains. On
potato it formed an abundant yellow growth, but it did not pro-
duce indol nor form gas in sugar media. It liquefied gelatine
slowly, and formed dense opaque colonies on agar plates. No
confusion could arise between this organism and the typhoid
bacillus. The experiments showed that Chichester 3 had
difficulty in growing in the presence of the typhoid bacillus.
It grew readily enough in a mixture of water and broth when
it was the only organism present ; but when the typhoid
bacillus was also present Chichester 3 could not establish itself,
it did not grow or was killed out altogether. This took place
at low temperature and at 37° C. When, however, Chichester
3 was grown in the sterilised soil from which it was obtained,
its chances of growth and multiplication were greatly increased,
and in one case it so far gained the mastery that no typhoid
bacilli could be found after fifteen days.

The latest contribution to the study of the B. typhosus in
relation to other micro-organisms is furnished by the researches
of Remy on the antagonism of the B. coli and the B. typhosus.
This observer believes that the failure to isolate the B. typhosus
when growing side by side with other bacteria is not due to the
disappearance of the typhoid organism but to the imperfection
of the methods employed to isolate it. He added to a
flask containing a litre of peptonised water (peptone 3 per
cent., sodium chloride 0'5 per cent.), exactly neutralised, 5 c.c.
of a twenty-four hours culture of B. typhosus and 5 c.c. of a
twenty-four hours culture of B. coli. The same procedure
was followed with a flask containing peptonised water to which,
after neutralisation, 0'5 per 1000 of H2SO4 was added. The
flasks were kept at the room temperature, and examined at
intervals by plating out the contents in Remy's special medium.
After five days association the B. typhosus was found not to
be agglutinated by anti-typhoid serum, diluted 1-10, though

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 217

originally it was agglutinated by the serum, diluted 1-70,000 ;
the B. coli maintained its characters unchanged. After fifty-
one days association the B. typhosus colonies were found to
develop more slowly than before, and the B. coli colonies were
smaller and resembled those of B. typhosus. After eighty-two
days association the B. typhosus remained as before, but the
B. coli isolated from the neutral peptone water, though it still
fermented sugar media, had lost the power of producing indol.
In the acidified flask, however, the B. coli maintained its
characters unchanged. After 100 days it was difficult to
distinguish the colonies of B. coli and B. typhosus from the
neutral flask ; all the colonies failed to produce gas and indol
and to agglutinate with anti-typhoid serum. In the flask con-
taining the acidified water, however, the B. coli still gave the
indol reaction, and fermented lactose, and the B. typhosus did
not react with anti-typhoid serum. A week later the colonies
from the neutral water were exceedingly feeble in their growth,
and, as before, failed to produce indol or gas or to agglutinate
with anti-typhoid serum. In the acidified water the B. typhosus
had now disappeared, and only B. coli could be isolated, which,
however, still produced indol and gas. In these experiments
there was a very marked difference between the results obtained
in the neutral and acid waters ; in the former medium the
B. coli and B. typhosus speedily lost their special characteristics ;
but in the latter the B. coli maintained its power of forming
gas and indol right up to the end of the experiment, and the
B. typhosus only lost its power of reacting to anti-typhoid
serum after 100 days association with B. coli. Control
experiments were then made to see if the B. coli and
B. typhosus lost any of their special characteristics when
growing separately in the neutral peptone water. After four
months the two organisms were still found alive, and they both
maintained their specific reactions. Other races of B. coli and
and B. typhosus when grown together in neutral peptone water
were not found to be affected in the same way as the B. coli
and B. typhosus which were used for the first experiments.
Remy then endeavoured to decide whether the colonies which
had perfectly negative reactions when isolated after 100
days from the neutral peptone water were B. coli or

218 BACTERIOLOGICAL EXAMINATION OF WATER.

B. typhosus. To decide this question he immunised guinea-pigs
with the cultures ; of the sera so prepared, some agglutinated
the B. typhosus, while others agglutinated the B. coli, used
originally to infect the flasks ; consequently these atypical
bacilli appeared to be true descendants of the original cultures.
As a result of these experiments Remy arrived at the following
general conclusions :

(1) The idea of the effacement of the B. typhosus by the
B. coli is not confirmed by researches on the antagonism of
these two organisms.

(2) Association can profoundly modify the properties of the
two organisms, the B. typhosus losing its sensibility towards
agglutinins and the B. coli being deprived of its specific
characteristics.

(3) The colonies of B. coli in the depth of the gelatine, after
the third or fourth week, insensibly approach in dimensions and
appearance the colonies of B. typhosus.

(4) The diminution of the vital energy of the organisms
shows itself not only in a diminution in the size of the colonies,
but also by a delay in their appearance.

(5) If the agglutination of a bacillus, presenting the characters
of the B. typhosus, by an experimental anti-typhoid serum in a
high dilution, suffices to authorise us to consider the organism
as typhoid in nature, the absence of sensibility towards the
specific agglutinins does not permit us to reject the organism
from the group of typhoid bacilli.

(6) Bacilli which possess the attributes of the B. typhosus,
but which are no longer agglutinated by experimental anti-
typhoid serum, ought to be considered as typhoid bacilli if a
guinea-pig, which has received every second day J2 c.c. of a
forty-eight hours broth culture of the organism, produces after
fifteen days a serum which agglutinates a tvpical B. typhosus in
a minimum dilution of 1-40.

(7) There may exist true typhoid bacilli which cannot be
agglutinated by anti-typhoid serum, and the typhoid nature of
which it is impossible to determine, either by the procedure given
above or by the means of diagnosis actually at our disposal.

Remy's results are very interesting, but it is strange that the
effects of the association of B. coli and B. typhosus were only

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 219

obtained with two particular cultures of these organisms, and in
neutral peptone water. My own experiments showed that the
B. typhosus, when growing with B. coli maintained its power
of reacting to anti-typhoid serum so long as it could be isolated
from the mixture. Also experiments made at Netley on the
inter-action of B. coli and B. typhosus when planted out
in large quantities in tap- water, showed that even when em-
ploying Remy's medium B. typhosus could riot be isolated after
the thirteenth day. The colonies of B. coli acquired much the
same appearances as those of B. typhosus, so that it was necessary
to sub-culture all the colonies before a diagnosis could be made
between the two organisms.

The whole question of the duration of life of the B. typhosus
in water and sewage is very complex ; temperature, the amount
of nutriment, the presence of other organisms and their toxines,
the number and vitality of the typhoid bacilli introduced, are
powerful factors which influence the result in various ways. In
most experiments the typhoid bacilli have been introduced in
considerable numbers, but examination of the stools of typhoid
patients has shown that typhoid bacilli may not be present in
any quantity when water is fouled by typhoid dejecta. Wathelet
only found ten colonies of B. typhosus among six hundred
colonies having characters common to B. coli and B. typhosus,
which he isolated from typhoid stools. Consequently it seems
extremely probable that under natural conditions comparatively
few typhoid bacilli will gain access to a water supply, and to
detect them it will be necessary to examine a considerable
quantity of the water. The bulk of the experimental evidence
seems to show that, unless the specific pollution is continuous,
it is extremely unlikely that typhoid bacilli will be detected in
a water if more than a week has elapsed since the actual pollu-
tion occurred. The outbreak of typhoid fever at Maidstone
affords considerable support to this view. Washbourn made a
bacteriological examination of the infected water about ten days
after the specific infection had taken place. He failed to find
the B. typhosus, though the presence of B. coli showed that
sewage contamination of the water had occurred. When, how-
ever, sewage containing the specific excreta passes into a cess-
pool, and percolates at once into the surrounding ground, the

220 BACTERIOLOGICAL EXAMINATION OF WATER.

conditions of life of the B. typhosus are different. Robertson's
experiments appeared to show that typhoid bacilli may exist in
earth under natural conditions for several months. Martin's
latest studies on the behaviour of B. typhosus in soil, however,
throw some doubt on these results ; everything seems to depend
on the nature of the soil and the other organisms present in it.
Still, the endemic prevalence of enteric fever year after year in
certain localities, characterised by polluted soils, renders it
extremely probable that the life of the B. typhosus in soil is
longer than in water-supplies. If this be so the bacilli may be
carried from the soil into drinking water by rising ground water
or rain, and in this way outbreaks of enteric fever may occur
long after the initial infection.

CASKS IN WHICH THE B. TYPHOSUS HAS BEEN ISOLATED FROM
WATER-SUPPLIES.

The discovery of the B. typhosus in water-supplies has often
been announced. Moers, in 1886, was the first to isolate the
bacillus from a contaminated well supplying water to a number
of people amongst whom cases of enteric fever had occurred.
Michael, in the same year, claimed to have found the bacillus
in a well-water suspected to have caused enteric fever in
Dresden. Chantemesse and Widal, Thoinot, Loir, and Vincent
have all found the typhoid bacillus in the water of the river
Seine. Fodor stated that he found the bacillus five times in the
water-supply of Budapest. Most of the early announcements
of the discovery of the B. typhosus must be accepted with great
reserve, as at that time the appearance of the colonies on gela-
tine plates was considered sufficient to prove the presence of the
bacillus, and most of the tests now required were not applied.
Cassedebat studied with great care the water-supply of Mar-
seilles in 1890, employing special methods, but he failed to
discover the typical B. typhosus of Eberth and Gaffky. He,
however, isolated three other bacilli which he called pseudo-
typhoid ; these organisms bore a close resemblance to the true
typhoid, and could only be distinguished from it with great
difficulty. Cassedebat made about two hundred and fifty cul-
tures, employing ten drops of water for each culture, and was
careful to state that his failure to discover the typhoid bacillus

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 221

did not prove anything for or against the existence of this
bacillus in water, owing to the small amount of water with
which he worked. At the same time it is evident that his so-
called pseudo-typhoid bacilli would have been described bv
earlier observers as true typhoid organisms. Klein examined the
water-supply of Worthing, where an outbreak of enteric fever
had occurred. A sample was collected drop by drop during a
period of twelve hours from the rising main in the yard of the
Worthing waterworks ; 1500 c.c. of this water were passed
through a Berkefeld filter, and the residue on the bougie was
distributed among phenolated broth tubes and phenolated
gelatine plates. " In the tubes and plates B. coli developed
abundantly, and in addition a few other colonies were detected
— two in one plate and one in another plate — which on sub-
culture presented, morphologically as well as culturally, all the
characters of enteric fever bacilli." The following features
were given as characteristic of B. coli and B. typhosus :

B. COLI. B. TYPHOSUS.

Shorter and less motile than the Longer and more motile than K.

enteric fever bacilli. coli.

Forms gas bubbles in gelatine-shake Does not form gas bubbles in gela-

culture. tine-shake culture.

Curdles milk in one to two days at Does not curdle milk.
37° C.

Forms indol in broth after several Gives no indol reaction,
days.

The tests given under B. typhosus would be considered at the
present day as quite insufficient to establish the diagnosis of the
typhoid bacillus. Consequently the Worthing epidemic must
be relegated to the large class of cases in which there is still a
doubt as to the presence of B. typhosus in the water-supply.

In 1897 Remlinger and Schneider examined thirty-seven
specimens of water obtained from wells and rivers both during
outbreaks of enteric fever and in the absence of all manifesta-
tions of this disease. Nine of the specimens contained a bacillus
presenting all the characters which they considered typical of
B. typhosus, viz., (1) aspect of cultures on gelatine ; (2) active
motility of the bacillus; (3) large number of flagella ; (4)
failure to stain by Gram's method ; (5) absence of gas forma-

222 BACTERIOLOGICAL EXAMINATION OF WATER.

tion in sugar-media; (6) milk unchanged; (7) indol not pro-
duced in peptone water ; (8) acid produced in litmus-whey not

exceeding 3 per cent, of ^ alkaline solution ; (9) typical growth

on potato ; (10) pathogenic action ; (11) agglutination with
anti-typhoid serum. Two of the specimens came from towns
(Meaux, Saint-Omer) where typhoid fever existed at the time.
The B. typhosus was only transitory in these two cases, for new
specimens of the water taken a month later did not contain the
bacillus. Six other specimens were obtained from towns where
enteric fever had existed, but at the time was not in epidemic
form. The examinations of the water from Chateaudun and
Dijon were of special interest. An epidemic of typhoid fever
appeared in Chateaudun during the winter of 1895-6 amongst
the civil and military population. The water-supply was first
examined on January 21 by means of phenolated media ; the
B. typhosus was not found. A second examination was made
on March 15 by Eisner's method ; on this occasion the
typhoid bacillus was isolated. On May 10 and again on
June 15 the B. typhosus was detected, though from the
beginning of March the outbreak lost its epidemic form and
only appeared as rare and isolated cases. Typhoid fever also
appeared in an epidemic form in Dijon during the winter of
1895-6, but examinations of the water-supply, both in 1895
and early in 1896, by phenolated media failed to demonstrate
the presence of the B. typhosus. In April 1896, however,
when enteric fever no longer existed in the town, Eisner's
method showed that a bacillus, identical with the B. typhosus,
was present in the water-supply. In May and June the same
bacillus was isolated, though no enteric fever existed in either
the civil or military population. In this outbreak water
apparently containing the B. typhosus was consumed for three
months without any epidemic of enteric fever appearing.

In 1899 Kubler and Neufeld examined specimens of water
from a farm where enteric fever existed. By means of Eisner's
method B. typhosus was isolated from the water in a well sup-
plying drinking-water to the farm. The tests on which the
authors relied for the differentiation of the typhoid bacillus
were as follows : Characteristic growth on potato, gelatine, and in
litmus-whey, absence of indol in broth, absence of gas formation

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 223

in sugar media, large number of flagella, and agglutination with
a very dilute anti-typhoid serum. Also a three hundred gramme
guinea-pig was killed by one loopful of a twenty-four houre
growth on agar ; but with the same amount of growth and
O'Ol c.c. of anti-typhoid serum, a guinea-pig of the same weight
remained quite well. About four weeks later two cultures were
obtained from the same well, which in appearance, motility,
chemical reactions, and agglutination resembled the typhoid
bacillus ; but these cultures were not pathogenic, one loopful of
a twenty-four hours agar culture had no effect when injected
into a guinea-pig. The authors regard Pfeiflfer's reaction as the
most important diagnostic sign of the B. typhosus, because it
indicates that the bacillus is in a virulent condition. In this
outbreak B. coli was not found in the well-water, and the
typhoid bacilli were present in large numbers. Kubler and
Neufeld, therefore, believed that the pollution of the water was
caused by urine from a typhoid patient, for when infection of
the water-supply was produced by faecal contamination B. coli
was usually present in large numbers. The absence of B. coli
from the water in this case rendered the isolation of the typhoid
bacillus a comparatively easy matter, and probably accounted
for the persistence of the bacillus in the water supply.

In 1899 Genersich also investigated an outbreak of typhoid
fever in Pecs, supposed to have been caused by infected water.
In the epidemic there were 209 cases and 28 deaths, which
occurred in a portion of the town supplied with water from the
Also-Rokus wells. Suspicion fell on cisterns in the Kis Rokus-
Gasse and in the " Kreuzer-kaserne." Twenty-eight gelatine
plates were made with water taken from the cisterns and the
wells. Suspicious colonies which developed in the plates were
fished and planted out in gelatine-stabs. The growths which
resulted were compared with a true typhoid culture; and the
motilily and staining reactions were also examined. The
cultures which resembled B. typhosus were then planted out on
potato, control preparations of B. coli and B. typhosus being
made at the same time. Characteristic cultures were then
inoculated into sugar-agar broth and milk. The growths in
broth were tested for the indol reaction. Agglutination experi-
ments were made with a twenty-four hours broth culture of the

*24 BACTERIOLOGICAL EXAMINATION OF WATER.

vsuspected organisms and the serum from typhoid patients, and
also with the serum from a guinea-pig (immunised by injecting
1 c.c. of a forty-eight hours broth culture per 0'3 kilogramme
of body-weight) diluted 1-50. Lastly, according to Fodor and
Rigler's recommendation, forty-eight hours broth cultures of
the suspected organisms were injected into guinea-pigs and
dogs. Specimens of blood from these animals were taken after
eight to ten days, centrifugalised, and then tested with true
cultures of B. typhosus, B. coli, and the organisms which had
been used for the immunisation. From the 28 plates 157
suspicious cultures were obtained, but of these only 61 corre-
sponded to B. typhosus in motility, staining reactions, and
growth in gelatine-stab. Out of 61 cultures, however, only 31
grew typically in sugar-agar, milk, and broth, and only 11 of
these were agglutinated by the serum from typhoid patients
diluted 1-50. Out of 9 cultures tested 7 were agglu-
tinated by the serum of the guinea-pig immunised with B.
typhosus diluted 1-50. Eight guinea-pigs and three dogs
were then inoculated with the 11 cultures which had shown
a reaction with the serum of the typhoid patients. The serum
diluted 1-50 from each of the guinea-pigs and dogs agglu-
tinated the special organisms which had been used for their
immunisation; and also, for the most part, the serum from all
the animals agglutinated all the organisms which had been used
for the injections. But, with one exception, the sera failed to
agglutinate a broth culture of a true typhoid bacillus.
Genersich could not explain the cause of this failure, but
suggested that the supposed typhoid bacilli, isolated from water,
had much less virulence than the typhoid bacilli isolated from
the spleens of typhoid patients, and so were unable to produce
a serum which would agglutinate the latter bacilli.

B. Fischer and G. Flatau in January 1901 reported the
discovery of the typhoid bacillus in water from a well in the
town of Rellingen. Plates were made with varying quantities
of the water added to ordinary and carbolic acid (0'05 per cent.)
gelatine ; srrface plates with the same media were also prepared.
On the surface plates made with carbolic acid five suspicious
colonies appeared, but only one of these failed to produce gas in
glucose-agar. This organism was then carefully studied. It

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 225

was a motile bacillus, about the same size as the typhoid bacillus,
and did not stain with Gram. On potato it produced a moist
transparent growth. Milk was not coagulated after eight days
incubation at 37° C. Indol was not produced. In Petruschky's
litmus-whey the amount of acid formed was very small. The
agglutination test was then tried with serum from a goat im-
munised by repeated injections of the B. typhosus. This serum,
diluted 1-200, caused agglutination of the bacillus which had
been used to prepare it, the clumps being visible, macro-
scopically, in a quarter of an hour. It also caused agglutination
of the bacillus isolated from water when employed in dilutions
of 1-50, 1-100, and 1-200, but the clumps were not visible,
macroscopically nor microscopically, for from one to two hours.
The same delayed agglutination was also seen when two true
typhoid cultures from different sources, i.e., a typhoid spleen
and a typhoid abscess, were tested with the serum. The bacillus
isolated from water was also tested with a serum which, in a
dilution of 1—500, caused agglutination of its own bacillus in
one to two hours. The " water bacillus" was agglutinated by this
serum diluted 1—500, but the reaction was not apparent micro-
scopically for four hours. PfeinWs test was also applied, but
with some difficulty, as the typhoid serum had not a powerful
bacteriolytic action, and the minimal fatal dose of the water
bacillus varied from 0'046 mgm. to 0'092 mgm. per 100
grammes of body-weight of the experimental animal. Experi-
ments showed, however, that 0'05 c.c. of the typhoid serum pro-
tected against ten times the lethal dose of the bacillus, whereas
the same amount of normal serum from men and goats had no
protecting action against the same dose of the bacillus. The chain
of evidence seems fairly complete, and the bacillus isolated by
Fischer and Flatau must be regarded as the true B. typhosus.
Four weeks later the authors again examined the water from
Rellingen, but failed to find any traces of the B. typhosus.

APART FROM THE ISOLATION OF THE B. TYPHOSUS, ARE THERE ANY

MEANS OF DIAGNOSING POLLUTION OF A WATER SUPPLY WITH

THE SPECIFIC DEJECTA OF CASES OF ENTERIC FEVER ?

The cases just recorded show that, with one or two exceptions,
it has been impossible to isolate the B. typhosus from suspected

226 BACTERIOLOGICAL EXAMINATION OF WATER.

water supplies during epidemics of enteric fever. The cause of
this failure has been variously interpreted. Some bacteriologists
believe it is due to the feeble vitality of the B. typhosus when
associated with other micro-oganisms, especially the B. coli, so
that the specific organism dies out of the water supply before
the outbreak of disease has directed attention to the necessity of
a bacteriological examination of the water. Other bacteriologists
would, however, attribute the failure to the imperfection of modern
bacteriological methods of investigation. A French school , headed
by Rodet and Roux, believes in the existence of transitional forms
between the typical B. typhosus and the typical B. coli. Rodet has
quite lately described such transitional forms which are of great
interest to students of epidemiology. In May 1897, Rodet isolated
from the spleen of a fatal case of enteric fever a bacillus which
resembled the typhoid bacillus in its morphology, mode of growth
on gelatine and agar, absence of gas production in sugar media, and
failure to produce indol in peptone solution. Immediately after
isolation it was not agglutinated by anti-typhoid serum, even
in a dilution of 1-10, and produced a brownish-yellow growth on
potato ; consequently the organism appeared to be an atypical
coli. The microbe was preserved for a year and its power of
reacting to anti-typhoid serum tested from time to time. It
was found that the reaction to specific agglutinins gradually
increased until, finally, the bacillus was agglutinated by a specific
serum diluted 1-100,000. In October 1897, a similar bacillus,
which showed a steady increase in its reaction to anti-typhoid
serum, was also isolated from the spleen of a fatal case of typhoid
fever. A third bacillus, which more strongly resembled B. coli
and produced gas in glucose-agar, was recovered from the spleen
of a fatal case of enteric fever in December 1897. Six months
after its isolation this microbe, which at first was not affected
by anti-typhoid serum, was found to be agglutinated by two
different sera in exactly the same manner as a typical B. typhosus.
On the other hand, Remlinger and Schneider state that they
have isolated the B. typhosus from the stools of patients wrho
were not suffering from enteric fever. Four of the bacilli
isolated were pathogenic to guinea-pigs, and preventive injections
of anti-typhoid serum preserved animals from infection. Inde-
pendently of the above bacteria the same observers stated that

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 227

they often found in water, soil, and human fasces bacilli which
strongly resembled the B. typhosus in all points except reaction
to specific agglutinins and pathogenicity to animals. They
believe that there are varieties of the true B. typhosus, and
consider that this organism exists naturally in air, water, and
earth, but how it obtains its pathogenic action is not yet
known ; possibly privation and fatigue are important factors.

Whether it be true that B. coli can be converted into
B. typhosus, as Rodet believes, or that the B. typhosus naturally
exists in water, &c., in an abortive form, as Remlinger and
Schneider consider probable, the fact remains that up to the
present time most bacteriologists have failed to isolate the
typhoid bacillus from suspected water supplies. Consequently
it would be a very important matter for hygienists if they
could, in the absence of the discovery of the B. typhosus, say
that a water supply had been infected by the specific dejecta
of enteric fever. With this end in view I have lately made a
special study of the varieties of B. coli which are associated
with the typhoid bacillus in the dejecta of patients suffering
from enteric fever. I hoped that the varieties in question
might show cultural characteristics or reactions to specific sera,
which would enable them to be distinguished from the varieties
of B. coli present in the stools of healthy people. One hundred
and fifty different cultures were examined, eighty being derived
from the stools of patients suffering from typhoid fever and seventy
from the stools of healthy men. The stools of the enteric fever
patients were obtained during the third and fourth weeks of the
disease and also from the intestines after death had occurred.
A loopful of the stool was stroked over the surface of a series of
plates containing solidified gelatine. The first two plates generally
liquefied, but the remaining plates showed discrete colonies which
developed satisfactorily. Typical colonies were fished and planted
out on a gar slopes, which were then incubated at 37° C. for
twenty-four hours. The growths which resulted were used for
agglutination experiments and planted out in the various media.

Agglutination Experiments. — The twenty-four hours growth
was made into an emulsion with broth and then mixed with an
equal amount of horse's anti-typhoid serum, so as to make
final dilutions of the mixed emulsion and serum in the

228 BACTERIOLOGICAL EXAMINATION OF WATEtt.

proportion of 1-50, 1-100, 1-200, 1-500, 1-1000. The experi-
mental investigation of the 150 cultures of B. coli extended
over a period of four months, consequently it was necessary to
employ different batches of anti-typhoid serum ; and to make the
results strictly comparable, the various specimens of serum were
standardised week by week with two stock cultures of B. typhosus.
All through the experiments the sera used gave a complete
agglutination in a dilution of 1-1000 and a marked reaction in
a dilution of 1-10,000 with the stock cultures of B. typhosus.
The results obtained with the 80 cultures of B. coli isolated
from enteric stools are shown in the following table :

TABLE A.

VARIETIES OF B. COLI FEOM ENTERIC STOOLS TESTED
WITH ANTI-TYPHOID SERUM.

No. of
culture.

Dilution of the scrum.

1-50

1-100

1-200

1-500

1-1UOO

1

+

+

+

0

0

2

+

()

0

0

3

+

0

0

0

4

+

+

+

0

0

5

0

0

0

0

0

6

+

+

+

0

0

7

0

0

0

o

0

8

o

0

0

0

0

9

0

0

0

0

0

10

0

0

0

0

0

11

0

0

0

0

0

12

+

+

±

0

0

13

0

0

o

0

0

14

0

0

0

0

0

15

+

+

+

()

0

16

+

+

+

0

0

17

0

0

0

0

0

18

0

0

o

0

0

19

0

0

0

0

0

20

0

0

0

0

0

21

0

0

0

0

0

22

0

0

0

0

0

23

()

0

0

0

0

24

0

0

0

0

0

25

0

0

o

0

0

26

0

0

o

0

0

27

+

+

±

o

28

+

+

±

0

29

+

+

±

0

30

+

+

±

—

0

QUALITATIVE BACTERIOLOGICAL ANALYSIS. -229

TABLE A. — continued.

No. of
culture.

Dilution of the serum.

1 -50

1-100

1-200

1-500

1-1000

81

+

+

+

0

32

+

+

±

—

0

33

—

0

0

0

0

34

—

0

0

0

0

35

—

0

0

0

0

36

+

+

±

0

0

37

+

+

+

0

0

38

+

+

+

0

0

39

+

+

—

0

0

40

+

+

+

0

0

41

+

+

±

0

0

42 .

+

+

+

0

0

43

+

+

±

0

0

44

+

+

+

0

0

45

0

0

0

0

0

46

0

0

0

0

0

47

0

0

0

0

0

48

0

0

0

0

0

49

0

0

0

0

0

50

+

0

0

0

51

+

+

0

0

0

52

+

+

±

()

0

53

+

+

+

0

0

54

+

+

±

0

0

55

+

+

+

0

0

56

+

+

0

0

57

+

+

+

0

0

58

+

+

+

—

0

59

+

+

0

0

0

60

+

+

+

0

0

61

+

+

+

0

0

62

+

+

+

+

+

63

+

-h

+

•f

—

64

+

+

+

+

—

65

+

+

+

+

—

66

+

+

+

+

+

67

+

+

+

+

+

68

+

+

+

—

0

69

+

+

+

+

—

70

0

0

0

0

0

71

0

o

0

0

0

72

0

0

0

0

0

73

0

0

0

0

0

74

0

0

0

0

0

75

0

0

0

0

0

76

0

0

0

0

0

77

0

0

0

0

0

78

0

0

0

0

0

79

0

0

0

0

0

80

0

0

0

0

0

NOTE. — The sign + indicates complete agglutination.

„ + „ nearly complete agglutination.

„ partial agglutination (many free bacilli).
0 „ no traces of agglutination.

230 BACTERIOLOGICAL EXAMINATION OF WATER.

This table shows that seven cultures of B. coli were completely
agglutinated by the anti-typhoid serum, diluted 1-500 ; two
cultures were markedly agglutinated by the serum, diluted
1-1000, and one culture was completely agglutinated by the
serum in this dilution. This last culture was further tested at
once with the serum in still higher dilutions, and was found
to be completely agglutinated by the serum when diluted
1-2500. The serum was then tested with four cultures of B.
typhosus isolated from the spleens of the fatal cases which
furnished the stools under investigation. The results, together
with the effects of the serum on the two stock cultures, are
shown in the following table :

TABLE B.

Cultures of
B. typhosus.

Dilution of anti-typhoid serum.

1-50

1-100

1-200

1-500

1-1000

1-10,000

G
G*

1

0

0

0

0

0
0

M

A

+

+

J

+

±

0
0

l\18p.

+

*

+

+

+

*

It will be observed that all the specimens with the exception
of Gx were completely agglutinated by the anti-typhoid serum
diluted 1—500 ; with the higher dilutions, however, the results
varied, culture B. typhosus G and culture B. coli 62 were ob-
tained from the same case, and it will be noticed that the B. coli
was completely agglutinated by the anti-typhoid serum in a higher
dilution than was effective with the B. typhosus. If the serum
employed had been obtained from the patient it would have been
easy to explain the result by supposing that the patient suffered
from a mixed " typhoid" and "coli" infection. In the present case,
however, there could be no question of a secondary coli infection,
as the anti-typhoid serum was obtained by vaccinating a horse
with pure cultures of B. typhosus. It therefore appears that
coliform organisms in typhoid stools, as a result of their sur-
roundings, may become agglutinated by a highly dilute anti-
typhoid serum. That the sensibility to agglutination is only

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 231

acquired by environment, and is not truly specific as in the case
of the B. typhosus, is shown by the fact that when the cultures
were removed from their associations and preserved for several
months on agar slopes, the sensibility to the agglutinins in the
anti-typhoid serum gradually declined, and at the end of six
months the results shown in the following table were
obtained :

TABLE C.

CULTURES OF B. COLI, FEOM TYPHOID STOOLS, RE-TESTED
WITH ANTI-TYPHOID SERUM AFTER BEING PRE-
SERVED FOR SIX MONTHS ON AGAR,

Xo. of coli

Dilution of anti-typhoid serum.

culture.

1-50

1-10U

1-200

1-500

1-1000

62

+

±

0

0

0

63

+

0

0

0

0

64

+

+

0

0

0

65

+

+

0

0

0

66

+

±

0

0

o

67

+

+

0

0

0

68

0

0

0

0

69

+

+

+

—

0

The maximum dilution which caused complete agglutination
was now only 1—100. The races of B. typiiosus, however, showed
no change, the results previously given were again obtained at
the end of six months. The culture B. typhosus Gx was
peculiar; when first isolated from the spleen it only showed
traces of agglutination with the anti-typhoid serum diluted
1-50 ; but, after being preserved in milk for six months, a sub-
culture was completely agglutinated by the an ti- typhoid serum
diluted 1-1000.

The typhoid culture Gx was carefully studied, and it was
found to possess cultural characteristics absolutely identical
with undoubted races of B. typhosus. The cultural character-
istics of Gx and the races of B. typhosus shown in Table B. are
given in Table D on pago 232.

The cultural characteristics of the coliform cultures Nos. 62
to 69 were then examined ; it was thought that as they came
within the " typhoid range" as regards agglutination they
might show an approximation to the cultural reactions of

232 BACTERIOLOGICAL EXAMINATION OF WATER.

c

.g

c g

-«

^ ^

tj

. »^M

^

1 1* — o

2

1

•S c

r-

''S o

-2

£,

.H2 2 * J

be

O)

"3

,0 trt

"S

9

c
^

PS

tc.^ •=' 7-

sf ^

2 = ^

32
1

O

'o

c

0

3

bo
1

to 2

1C

0?

Ill

|

ll

j>>

o

EH H

O

P

&

c

CO

o

£

5

(2

^j-

^ i

^

_ 0

^

M

c t-i

s .

i SS *-* 0>

-§

2

*3

^ c

tJ

il

r2

1 | © 8

c

fcJD

tj

^

"fe '^

*|

S x

-M

o

M

II ^

= -£ ?J=!

QQ

i

i

as

Q

aQ

J

o>
be
a

^

be ^.o

r^ "O

'be

02 |_r-
OJ ">

1 2

H H

£

1
H

1

G

w

»b

o* «

CO

!&

Q

be

n

o

3

p!-

t- •

r-J

C

"1

£

o g

§

*

fj il

g i c-i

'o
'd-

2

be

'C!
1

"3

Jz; 2

4H S3

12

'o

CO

be

IS o

ll

>,

e<

S3 o

2 ^ I53

be

0

0

^0

0

c

^

^=5 o
o o ss

p

l&

Te

0

H H

o

P

ib

to

CO

0

£

«

C
0

5

o 2

-g

rS r-

^

r^H

c5

<

tlfl

Ilil

1

2

be
^

Ji<
^

*d

&

i

|

TL

^
o

1

Tc

^
||

1

IIP

EH H

be

0

p

0

o
§

^s

0

c

P

/o

ci 1T3

o-§§

^

o

CO

j!

£

Q

C

^

O '

rrj

J3 2

"fe

^

S3 J;

g

a

Thin greyish-
white growt

Turbid, no filn
the surface

a

be

0

Traces of indo

Colourless gro'

Unchanged

i— i
ib

ils
111

o

2

1

2

Colonies small
growth slow

o
2

be

£

ss

r-

CO

•_, i

r,-J

o

O

93
O)

a

o g

8b

•*•* 5

— < fe

o^

^ ^ «5 o

S3

o

a

^ .2

12

^^

•* 2 n ^

?

•a

13

>

i

5 ^

"Q

o

I'll!

£
|

1

^

3 |

**

l&a

be g .2

Hi

c

ll

|

•£ ^ ^ •£

OJD

0

o

*0

11

fc-

oil

o

•3 be

JSP

0

H EH

s

p

«b

{z;

CO

0

£

1

T; ^

^

c .i,

'J2 i _ •

1

&H

.00

c -2 ^

)

a; S

g

> ce

^ ^ o

"rH

0

3

ll

"

!
i

VD
1

6

g :-

155

|s

^M

E

c

J

3 o c

i

pe

0?°

O . si

r"cS

SD

D O

i- "an

~p,'j ?

i

^ ^ 4£

cV tr '/-1

5 5°

~^j

•f ^

>-i

O *-"

&-|P|

r

3

g

^^3

^ ff£

o 1

"E

^ c

^ ^

1 2

C

it

~Z,~r. "-,

• — 00 9

i|

-^ c

; ill

5-S^

III

^^

II

|

15

li

5 S3

•< 33

o

^

i"1"

£

Q

O

s

CO

ri

HI

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 233

B. typhosus. All the cultures, however, were found to present
the chief typical reactions of B. coli ; cultures 62 and 66 only
varied from the type by failing to produce indol. Culture
No. 70, however, which did not react at all with anti-typhoid
serum, showed considerable variations from the type; it was
highly motile, did not produce acid in litmus-whey, and failed
to give the characteristic reactions of B. coli in Proskauer and
Capaidi's media.

Forty-five of the other cultures shown in Table A were also
examined as to the production of indol in peptone solution, acid
in litmus-whey, souring of milk and gas formation in sugar
media. Twelve of the cultures showed characteristic colonies
and gave gas in glucose media, but failed to produce indol, or
change milk, and the acid formed in litmus-whey was compara-
tively small, requiring only from 8 to 16 per cent, of ~ alkali

to neutralise it. As regards the three tests usually considered
typical of B. coli, viz , the production of indol, the formation
of gas in sugar media, and the souring of milk, 64 per cent,
of the colonies examined gave all three reactions, 4'5 per
cent, failed to produce indol, 4'5 per cent, also failed
to sour milk, and 27 per cent, gave only one reaction, the
production of gas in sugar media. The results obtained by
the study of the eighty varieties of B. coli derived from
typhoid stools were then compared with those obtained by an
investigation of seventy cultures of B. coli obtained from the
stools of healthy men. The same procedure was followed as
before. As regards agglutination with anti-typhoid serum not
one of the cultures from healthy men was agglutinated by the
serum diluted 1-500 ; a complete reaction was only obtained
twice \\ith the serum diluted 1—100, and a marked reaction ten
times with a dilution of 1-200. As regards cultural character-
istics 48 per cent, gave the three typical coli reactions, 52 per
cent, gave only two reactions, usually the souring of milk was
absent. The acid produced in litmus-whey required from 27 to

47 per cent, of ^ alkali to neutralise it. There were no constant
cultural characteristics by which the cultures of B. coli which
reacted to anti-typhoid serum could be distinguished from those
which were completely uninfluenced by the serum.

BACTERIOLOGICAL EXAMINATION OF WATER.

Conclusions arrived at from this research may be summarised
as follows :

(1) As regards the appearances of the cultures on the various
media: the varieties of B. coli in typhoid stools cannot be
distinguished bv cultural characteristics from the varieties of
B. coli found in healthy stools.

(2) As regards reaction to anti-typhoid serum : the varieties
of B. coli isolated from typhoid dejecta show much greater
sensibility to agglutination than the varieties of B. coli found in
normal stools. Consequently if cultures of B. coli isolated from
suspected water supplies are found to come within the typhoid
range of agglutination, there appear to be fair grounds for
assuming that the water supply in question has been polluted
with typhoid dejecta

Having ascertained that the B. coli present in the intestines
of patients suffering from enteric fever may acquire an increased
sensibility to the agglutinins existing in anti-typhoid serum, it
appeared desirable to test experimentally the possibility of
producing the same sensibility in vitro. With this object in
view a pure culture of the colon bacillus, isolated from a
typhoid stool, was grown in Berne anti-typhoid serum for six
months at blood temperature. It was then planted out on agar,
and the resulting growth tested as to its sensibility to the
agglutinins contained in Berne serum. No change in the
sensibility was observed ; the bacillus reacted to the serum
in exactly the same manner as before the experiment was
commenced. The specific serum having failed to influence the
B. coli, an experiment was then devised by which the culture
was kept under the influence of the toxins produced by the
typhoid bacillus. A sterile Berkefeld candle, placed in a glass
mantle, was connected by means of a short piece of india-rubber
tubing, previously fitted with a pinch-cock, to a sterile glass
tube passing through an india-rubber bung fitted into a
Kitasato flask. The arm of the flask arid the mouth of the
glass mantle were plugged with sterile wool. Sterile broth
was then introduced into the mantle and the flask, the former
being inoculated with a culture of B. typhosus and the latter
with a culture of B. coli, which, when isolated from a typhoid
stool, showed no sensibility to agglutination. By releasing the

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 235

pinch-cock every third day, about 10 c.c. of the toxins,
excreted by the typhoid bacillus, were allowed to filter
through into the Kitasato flask. The amount of fluid filtered
from the mantle was replaced by sterile broth. The pro-
cedure was continued for two months until the flask was
filled nearly to the level of the arm. A loopful was then
withdrawn and planted out on agar. The growth which resulted
was tested for agglutination with Berne anti-typhoid serum,
and sub-cultured in various media. No change was observed in
its sensibility to agglutination or in its cultural reactions. The
flask containing the mixture of colon bacilli and typhoid toxins
was kept at a temperature of 25° C. for a further two months.
A loopful was then withdrawn as before, and the growth tested
with Berne serum. No increased sensibility to agglutination
was noticed.

It is impossible to imitate by experiment the complex condition
of things which exists in the human body suffering from a
typhoid infection, so it is not surprising that the agglutinins
and toxins in vitro failed to influence the sensibility of the
colon bacillus to agglutination.

CHAPTER XIV.

QUALITATIVE ANALYSIS— continued.

SPIRILLUM CHOLEILE ASIATICS.

THIS micro-organism was first isolated by Koch, in 1883, from
the stools of cases of cholera which he investigated in Egypt.
Later, more extended researches were made in India, and Koch
came to the conclusion that the association of this microbe
with Asiatic cholera was constant.

On cultivation the following appearances are seen on the
various media :

Gelatine Plates. — In twenty-four to forty-eight hours the
colonies appear as minute white points, wrhich under a low power
show an irregular or furrowed margin. Later, liquefaction
of the gelatine is produced, and the colonies sink into small
cups formed by the liquefaction ; the plate then shows small
sharply-marked rings around the colonies. After about seventy-
two hours, owing to the increasing liquefaction, the colonies
appear as granular masses, with torn or broken up margins, lying
in the bottom of circular cups of liquefied gelatine. The
granular masses gradually became fainter, and the whole surface
of the plate is completely liquefied.

Agar Plates. — Under a low power the surface colonies appear
as circular discs of brownish-yellow colour, and are more trans-
parent than the colonies of most other organisms.

Gelatine-stab. — During the first twenty-four hours the growth
is hardly perceptible, but in forty-eight hours slight liquefaction
with excavation of the gelatine is seen at the surface of
the stab, and below this a white growth appears along the line
of inoculation. About the fourth day a bubble-shaped
depression is seen at the surface, and below this a funnel of
liquefaction, covered at the surface with a white growth.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 237

Agar-streak. — A greyish-white growth, which is not char-
acteristic.

Broth. — A diffuse growth takes place and a thin pellicle
forms on the surface.

Milk. — The microbe grows well at 37° C., and produces coagu-
lation of the casein accompanied by an acid reaction of the medium.

Potato. — It does not grow on slightly acid potato at 22° C. ;
but at 37° C. it produces either a yellowish-brown or a non-
glistening white growth, which is hardly perceptible except
when the potato is held up to the light in a particular position.
On potato which has been rendered alkaline by soaking in a 2
per cent, solution of sodium carbonate it grows both at 22° C.
and at 37° C., producing a yellowish-brown growth.

Peptone ( Witters) and Salt Solution. — After twelve to twenty-
four hours incubation at 37° C., both indol and a nitrite are
formed, so that on the addition of a few drops of pure sulphuric
acid a red colour is produced. This is usually known as the
" cholera-red reaction."

Microscopical Characters. — The cholera spirillum is a small
curved organism resembling a comma ; usually only one curve
is seen, but sometimes two spirilla are attached together and
an S-shape is produced. It is actively motile and possesses a
single flagellum placed at one end. It does not form spores,
and involution forms are common, especially when the organism
has been kept in water ; it may then appear short and thick and
resemble a coccus. It is not stained by Gram's method. The
optimum temperature for its growth is between 35° and 38° C. ;
it grows, but more slowly, at a temperature as low as 17° C.,
but under 16° C. no growth is visible. It is strongly aerobic.

Reaction with Anti-cholera Serum. — Pfeiffer suggested the
following test as a means of diagnosis : The blood serum of an
animal, which has been immunised to the highest degree possible
by several months treatment with cholera vibrios, is diluted
1-100 with ordinary nutrient bouillon. A loopful (2 mgm.)
of a twenty hours agar culture of the vibrio to be tested is
then added to 1 c.c. of the mixed serum and bouillon, and the
mixture injected into the peritoneal cavity of a guinea-pig
weighing 200 grammes ; every five minutes up to the end of
twenty minutes, when the experiment is ended, a portion of the

238 BACTERIOLOGICAL EXAMINATION OF WATER.

peritoneal contents is withdrawn with a glass capillary tube
and examined in hanging-drop and " cover-glass " preparation.
If the vibrios injected have been true cholera spirilla, they will
become swollen up into round motionless masses which eventu-
ally become broken down into granular masses and disappear.
If at the end of twenty minutes the peritoneal contents show
unaltered and motile vibrios, they cannot be true cholera spirilla,
but belong to some other species. If, however, the vibrios are
dead but otherwise unchanged, two conditions of things are
possible, viz., either the bacilli are true cholera vibrios which
have been acted on by the specific bactericidal substances, or they
are saprophytic vibrios which, without any question of a specific
action, have been acted on by the bactericidal substances
present in the normal guinea-pig. In order to diagnose between
these two possibilities, one loopful of the suspected culture
must be added to 1 c.c. of bouillon containing O'Ol c.c. of normal
serum, and the mixture then injected into the peritoneal cavity
of a guinea-pig. If the vibrios are found alive and motile at
the end of the time when, in the cholera serum experiment, they
were completely dissolved, then it is probable that they belong
to the cholera bacilli species. Unfortunately it is possible that
vibrios which appear to be killed by normal serum may still be
true cholera spirilla, for it has been shown that Koch^s vibrios
which are enfeebled and have lost their virulence, may be
destroyed by normal peritoneal fluid. If, therefore, the sus-
pected vibrios disappear in the control experiment, Pfeiffer
recommends that the suspected culture be used to immunise a
guinea-pig. The serum from this animal must then be tested
as to its bactericidal action on true cholera vibrios. Do'mtx,
acting on Pfeiffer's suggestion, showed that an old culture of
cholera from Calcutta produced a serum which had a specific
bactericidal effect on cholera vibrios.

The agglutination test (or Durham-Gruber reaction) may.be
performed in the same manner as described under B. typhosus.
The serum is generally used in dilutions varying from 1-10 to
1-120. This test is more suitable for general hygienic work
than Pfeiffer's reaction. It requires to be performed with care,
and, as in the case of the B. typhosus, the dilution of the serum
employed is a matter of great importance. Durham prepared

QUALITATIVE BACTERIOLOGICAL ANALYSIS.

cholera serum with different races of cholera vibrios which had
been tested as to their virulence by peritoneal injection into
guinea-pigs. He found that a serum prepared from the vibrio
with the greatest virulence reacted powerfully with all the races,
but a serum prepared from the vibrio with least virulence had
a very slight action on the most virulent vibrios. For purposes
of diagnosis it is therefore important to work with a sufficiently
potent sample of cholera serum. Durham also found that
vibrios allied to the cholera vibrio reacted with cholera serum.
The Vibrio Ivanoff reacted to the different varieties of cholera
serum exactly like a cholera vibrio of medium virulence. Also,
the influence of the various kinds of cholera serum upon the
race of Vibrio Berolinensis agreed with the results obtained
upon the different races of cholera vibrios themselves. The
Vibrio Sanarelli also reacted markedly with potent cholera
serum. Sera prepared by other vibrios, such as the Massowah,
Danubicus, and Elwers, had a slight action on cholera vibrios,
but weak dilutions of the serum had very little effect. On the
other hand, cholera serum had absolutely no effect on the Vibrio
Finkler-Prior. With regard to the relation of the agglutina-
tion test to Pfeiffer's reaction, Durham concludes that whatever
"specific," or preferably "specialised," action a serum has upon
a given vibrio, such action has already taken place before the
injection is made. The reaction of Pfeiffer's is to be regarded as
consisting of two parts : the one being a weakening of the
microbes before injection by direct action of the active serum ;
the other being the normal bactericidal influence of the guinea-
pig^s peritoneal fluid upon these weakened microbes.

It is thus evident that the serum reactions cannot be considered
absolutely diagnostic. Positive reactions, however, especially when
obtained with a highly dilute serum, are undoubtedly very strong
evidence in favour of the microbe being the true cholera spirillum.

The cultural reactions are also open to serious criticism.
The appearance of the colonies in gelatine-plate cultures has
been considered highly characteristic of the cholera spirillum,,
and special stress has been laid upon the irregular margin
of the young colonies, the margin of the young colonies of
most foreign vibrios being quite regular and forming an
unbroken circle. Metchnikoff has, however, isolated vibrios

240 BACTERIOLOGICAL EXAMINATION OF WATER.

from cheese which form young colonies in gelatine with an
irregular margin, resembling in every way those of the true
cholera microbe. He regards this characteristic as of no
importance as a means of diagnosis. The rate of liquefaction
of the gelatine, once considered an important sign, has been
abandoned since Koch himself described a case of cholera in
which the vibrios liquefied the gelatine so slightly that the
colonies appeared in the form of shields. Also, according to
Metchnikoff, true cholera spirilla are met with, which liquefy
gelatine more rapidly than the typical forms and produce
funnels filled with masses of uniformly turbid culture. The
form of the spirilla is also very variable ; side by side with
thick curved forms others are found which are thin and
hardly curved at all. Friedrich described a Shanghai variety
which closely resembled a true bacillus, and a Malta variety
which appeared to be a cocco-bacillus. Koch, now, attaches
only a secondary diagnostic value to these characteristics,
and insists above all on the indol reaction and virulence for
animals. With regard to the cholera red reaction, he states
that none of the spirilla, which can possibly be confounded
with the cholera vibrios, produce indol and nitrous acid at
the same time in their cultures. Consequently he attaches
great value to this test, but points out that the test must
be performed according to certain well-defined rules. Above
all, a proper variety of peptone must be employed, as every
peptone does not give the reaction. Bleisch has shown that
the peptones which do not give a proper reaction contain
either too much or too little nitrates. The peptone to be
employed must first be tested with a known cholera vibrio
to see if it gives the characteristic reaction ; if the peptone
contains too little or no nitrates, these may be added in
proper amount. Bleisch recommends the following solution :
Witte's peptone 2'00 gramme, pure sodium chloride 0'5 gramme,
distilled water 100 c.c., pure potassium nitrate solution (0'08
per cent.) 30 to 50 drops. He also insists on the necessity
for using pure sulphuric acid, quite free from nitrous acid.
Broth, even though it contains peptone, does not give the
reaction so well as peptone and salt solution.

With regard to the virulence test, Koch recommends that one

QUALITATIVE BACTERIOLOGICAL ANALYSIS. Ml

loopful (1*5 mgm.) of a young agar culture be added to 1 c.c. of
broth, and the mixture then injected into the peritoneal cavity
of a guinea-pig, weighing 300—350 grammes. The weight of the
animal must always be taken into consideration ; if a guinea-pig
of greater weight is employed for the experiment a larger dose
of the agar culture must be given. One or two hours after the
injection the animal is languid, loses its appetite, and there is a
rapid fall of the body temperature, sometimes preceded by
pyrexia, and death ensues from collapse. Koch states that none
of the spirilla which can possibly come into question during the
investigation of cholera, when employed in the above dose, give
rise to the same symptoms as the true cholera spirilla.

According to Metchnikoff both the indoi reaction and the
virulence test are open to question. Sanarelli and Kohlbrugge
have both isolated from water supplies vibrios which give the
cholera-red reaction; indeed Kohlbrugge's investigations appeared
to show that temperature and season of the year had an influence
on this reaction.

It is thus evident that the diagnosis of the cholera spirillum,
ordinarily fairly easy when the microbe is isolated from a case of
cholera, may be exceedingly difficult, owing to the variability of
its reactions, when it has gained access to water. Klein has
shown how modifications of the cholera vibrio can be artificially
induced. He inoculated non-sterile sea- water with an agar
culture of a St. Petersburg cholera vibrio, and then isolated it
again eleven days later. The vibrio was found to have under-
gone the following modifications :

(a) It had lost altogether its power of growing at 37° C. either
on agar or in peptone.

(b) It grew at 20° C. in peptone and salt solution, but when
tested with pure sulphuric acid, after incubation for four or
more days, it gave no trace of the cholera-red reaction.

(c) In stained specimens it appeared thicker than the St.
Petersburg vibrio.

(d) When injected in considerable doses into the peritoneal
cavity of a guinea-pig it produced no diseased condition ; the
animal remained well.

(e) When tested with cholera serum, which quickly agglu-
tinated the original St. Petersburg vibrio, it gave a negative

Q

BACTERIOLOGICAL EXAMINATION OF WATER.

result. The vibrio was then transmitted from culture to culture
for more than a year, but it retained unimpaired the cultural
characteristics just given. It failed to grow at 37° C., gave no
cholera-red reaction, and did not react with cholera serum in
vitro. These results are directly opposed to the statement of
Pfeiffer that a vibrio which does not react to cholera serum is
not a descendant of a true cholera vibrio.

THE DIAGNOSIS OF THE CHOLERA Vnmio FROM
ALLIED FORMS.

Many spirilla have been isolated from various sources which in
some of their cultural reactions closely resemble the true cholera
spirillum. The most important of these are the following :

Spirillum of Finkler-Prior.

This micro-organism was isolated at Bonn by Finkler and
Prior during an epidemic of cholera nostras. Koch asserted
that these observers had only found the microbe on one occasion ;
but, in a second paper on the same subject, Finkler and Prior
stated that they had isolated the vibrio from other cases of
cholera nostras in perfectly pure culture. The cultural charac-
teristics of the vibrio are as follows :

Gelatine Plates. — During the first stages of growth the colonies
are larger than those of the cholera spirillum, and show a smooth
circular margin free from crenations. In twenty-four hours the
gelatine commences to liquefy, and in forty-eight hours the
liquefaction is very marked, the colony which is now ragged and
torn lying in a circle of liquefaction. The double contour effect
seen in the case of the cholera colonies is little marked, as a large
clear area results from the rapid liquefaction of the gelatine.

Gelatine-stab. — In twenty-four hours there is distinct growth
along the line of inoculation, and at the upper part liquefaction
is commencing with the formation of an air bubble. In forty-
eight hours liquefaction has advanced considerably and produced
a funnel-shaped excavation with a deposit of bacteria at the
apex of the funnel.

Agar-slope. — The growth forms a moist, greyish-white layer,,
indistinguishable from that of the cholera vibrio.

Potato. — There is a growth at ordinary temperature con*

QUALITATIVE BACTERIOLOGICAL ANALYSIS.

sisting of a slimy, greyish -yellow layer, which rapidly spreads
over the potato.

Peptone and Salt Solution. — Asa rule, no indol reaction can
be obtained on the addition of sulphuric acid, as, though it
produces indol, nitrites are not manufactured in sufficient
quantity to give the cholera-red reaction. In some cases more
nitrites are produced, and the reaction is then developed slowly
with a faint reddish tint.

Microscopical Appearances. — The organism is a comma-shaped
bacillus, slightly thicker in the middle than at the ends ; it
is larger than Koch's organism, highly motile, and possesses a
single flagellum.

When injected into the peritoneum of a guinea-pig it causes
peritonitis ; and in man, after neutralisation of the acid in the
stomach, it causes intestinal disturbance (Metchnikoff*).

The vibrio. of Tinkler-Prior is easily distinguished from the
typical cholera spirillum by its form, rapid liquefaction of
gelatine, and failure to give the cholera-red reaction.

Vibrio Metchnikovi.

This microbe was discovered by Gamale'ia at Odessa, where
it produced epidemic disease amongst fowls. It is of great
interest owing to its very close resemblance to Koch's spirillum.
An organism closely resembling the vibrio Metchnikovi was
isolated by Pfuhl from water.

Gelatine Plates. — The young colonies may have a perfectly
smooth outline, though sometimes the margin is distinctly
crenated like that of the cholera colonies. The colonies
generally grow more quickly, and liquefy more rapidly, than
those of Koch's vibrio ; in forty-eight hours the margin is often
very irregular and torn, and resembles Finkler-Prior's colonies.

Gelatine-stab. — The appearances closely resemble those pro-
duced by Koch's vibrio ; liquefaction, however, occurs more
rapidly ; the rate seems to be intermediate between Finkler-
Prior's and Koch's organisms.

Agar-slope. — A greyish white, moist growth.

Potato. — A raised, dirty-brownish growth.

Broth. — Diffuse growth in twenty-four hours ; a surface
pellicle is eventually formed.

244 BACTERIOLOGICAL EXAMINATION OF WATER.

Peptone and Salt Solution. — Indol and nitrites are produced,
so that a red coloui is obtained on adding a few drops of pure
sulphuric acid.

Milk. — Becomes acid, and the casein is coagulated about the
eighth day.

Microscopical Appearances. — As a rule, it appears nearly iden-
tical with Koch's spirillum ; sometimes, however, larger and
thicker forms are seen. It is very motile, and has one flagellum.

Pigeons are readily killed by sub-cutaneous injection of very
small quantities of this organism ; inoculation of the same
quantity of Koch's organism produces practically no effect.
This test at once enables a diagnosis to be made between the
two organisms.

Spirillum of Deneke.

This spirillum was isolated by Deneke from old cheese ; it is
also known as the Spirillum tyrogenum. In morphology it is
closely allied to Koch's spirillum.

Gelatine Plates. — The colonies, to the naked eye, have a
yellowish colour, and they liquefy gelatine more rapidly than
Koch's microbe. Under a low power the margins of the young-
colonies are seen to be smooth and regular. Later, there is
often a very close resemblance to the colonies of the cholera
spirillum ; but, as a rule, the yellow colour and the more
marked liquefaction enable a diagnosis to be made.

Gelatine-stab. — Closely resembles the appearances produced by
Koch's vibrio. Liquefaction however is more rapid.

Potato. — It produces a thin yellowish layer.

Peptone and Salt. — It produces indol, but only traces of
nitrites, so that the " cholera-red " reaction is produced very
feebly and irregularly on the addition of pure sulphuric acid.

Vibrio Aquatilis.

This vibrio was first isolated by Giinther from the water of
the river Spree at Stralaw. It was also found by Kiessling in
the washings of the sand used in the water-works of Altona.

Gelatine Plates. — The colonies at first are circular, with a
perfectly smooth rim. They are brown in colour, and have
granular contents. Liquefaction takes place very slowly, and
the colony eventually sinks in a basin of liquefied gelatine.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 24,5

Gelatine-stab. — There is growth on the surface, and lique-
faction slowly takes place, forming a basin- shaped depression.
There is very little growth along the stab.

Agar. — Forms a greyish-white growth, not characteristic.

Broth. — It is said not to grow in broth at 37° C. ; and at
21° to 22° C. the growth is extremely slow.

Peptone and Salt Solution. — The cholera-red reaction is not
obtained on adding a few drops of pure sulphuric acid.

Potato.— No growth takes place at 22° C. or at 37° C.

Microscopical Appearances. — It is a highly motile vibrio
exactly resembling the cholera spirillum. It possesses one

flagellum.

Vibrio Berolinensis.

This microbe was isolated by Neisser and Glinther from
Berlin pipe-water. The specimen of water had been designedly
infected with true cholera bacilli some time before, so there was
a possibility that the Vibrio berolinensis was a variety of the
cholera spirillum, modified by existence in water.

Gelatine Plates. — It grows extremely slowly, and in twenty-
four hours the colonies are only visible with the aid of the
microscope, and are round and finely granular. Even in forty-
eight hours the colonies are not visible to the naked eye.

Gelatine-stab. — Growth extremely slow, resembles that of the
cholera vibrio.

Agar. — White growth, not characteristic.

Potato. — Growth like the cholera vibrio.

Milk. — Unchanged .

Broth. — Diffuse growth.

Peptone and Salt Solution. — Cholera-red reaction obtained.

Microscopical Characters. — A motile vibrio about the same
size as Koch's spirillum ; it posseses one flagellum. It is
decolorised by Gram's method, and is pathogenic to guinea-
pigs. It does not give Pfeiffer's reaction.

Vibrio Massowah.

This vibrio was isolated from the stools of a sick person
during a small epidemic of cholera. It was at first accepted
as a true cholera spirillum, but later researches showed that
there were differences between the two organisms. The Vibrio

246 BACTERIOLOGICAL EXAMINATION OF WATER.

Massowah possessed four flagella, liquefied gelatine very slowly,
and its colonies were quite round, sharply outlined, and had a
yellowish colour. It gave the nitroso-indol reaction after
twenty-four hours. Its pathogenic action was very like that of
the Vibrio Metchnikovi.

Vibrio Gindha.

This vibrio was isolated by Pasquale from the water of a well
in Gindha. There had been a small epidemic of cholera in the
neighbourhood of the well some months before Pasquale's in-
vestigations took place. In cultural reactions the vibrio closely
resembled Koch's vibrio, but it failed to give the nitroso-indol
reaction after twenty-four hours incubation.

Vibrio Danubicus.

This vibrio was isolated by Heider from canal water. In
many respects it resembles Koch's vibrio. But in the appearance
of the colonies, its pathogenic action, and failure to give Pfeiffer's
reaction, it differs from the true cholera microbe.

Vibrio Ivanoff.

This microbe was discovered by Ivanoff in the stools of a
typhoid patient. Pfeiffer believes that the vibrio was the result of
an accidental contamination, as Ivanoff at the time was working
on the action of disinfectants on stools containing pathogenic
organisms. The Vibrio Ivanoff gave Pfeiffer's reaction, and
corresponded in every way with Koch's organism, excepting
that it showed a tendency to grow out into fine, long, slightly
curved, comma forms.

Vibrio Phosphorescens.

Dunbar and Rumpel found phosphorescent vibrios frequently
in water from the Elbe, Spree, and Rhine. The same vibrios,
according to Rumpel, were also found in the filtered pipe-water
of Hamburg. These spirilla are very closely allied to the true
cholera spirilla, both morphologically and culturally, and also
in pathogenic action. Kutscher observed that the vibrios
isolated by Dunbar produced in the dark a bluish-green light,
which was seen on fresh agar and gelatine cultures, and also in

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 247

the peritoneal cavity of guinea-pigs which had been infected
intra-peritoneally by the microbes. The phosphorescence is not
an absolutely constant phenomenon ; it may disappear when the
microbes are cultivated for a long time on nutritive media.
Dunbar found that the phosphorescent vibrios failed to give
Pfeiffer's reaction. He tested a large number of these organisms
and found that a pathogenic action was exerted both on the
control guinea-pigs and on the guinea-pigs which had received
the vibrios plus cholera serum ; the vibrios when withdrawn
from the peritoneal cavity were found to be still highly motile.
RumpePs statement that the true cholera vibrio can be changed
into the phosphorescent variety requires confirmation.

The characteristics of Dunbar^s vibrios will be seen from the
following short description of some of these organisms:

Vibrio No. 17a was a rather plump, curved bacillus, with
rounded ends. It was highly motile and stained unevenly. On
agar there was a greyish-white, moist layer. In peptone-water
at 37° C. a marked growth appeared after fourteen hours ; a
thin membrane formed on the surface, which later became very
thick and villous ; a typical " cholera-red " reaction was ob-
tained. In gelatine-stab after four days there appeared a clear
air bubble, below which the gelatine was liquefied in a capillary
form along the stab. In gelatine plates there was no lique-
faction apparent to the naked eye, the colonies appeared as
white points. Under a low power a thin, colourless, finely
granular surface growth was seen ; the deep colonies were also
colourless and finely granular, and rather sharply outlined.
Phosphorescence was always observed.

Vibrio No. 29 in appearance exactly resembled No. I?A. On
agar a greyish-white slimy growth was seen. In peptone- water
there was a marked growth, with formation of a thick pellicle cm
the surface ; a typical " cholera-red " reaction was obtained, in
gelatine-stab after six days the gelatine was completely liquefied
to the depth of 1 c.m. ; below this the gelatine was liquefied in
a capillary form. The appearance of the colonies in gelatine
plates exactly resembled No. 17A. Phosphorescence was always
seen.

Vibrio No. 30 was plump, highly motile, and possessed a long
flagellum at one end. On agar there was a greyish slimy

i>48 BACTERIOLOGICAL EXAMINATION OF WATER.

growth. In peptone and gelatine-stab the appearances were the
same as No. 29. The colonies were also very similar to those of
17 A ; those in the depth, however, had an irregular margin and
were coarsely granular.

Vibrio No. 34 was rather larger than the above, and possessed
one long flagellum. On agar, in gelatine-stab and peptone, the
growths were the same as No. 29. In gelatine plates small dry
hollows appeared, containing at the bottom dry white colonies,
which, under a low power, exactly resembled No. I?A.

Vibrio No. 4& was rather small, curved, with rounded ends
and highly motile. On agar, in peptone, and in gelatine the
growths resembled No. 29. In gelatine plates dry hollows
which contained thin grey colonies were seen. The colonies in the
depth were sharply outlined greenish-yellow, and finely granular.

Vibrio No. 51 closely resembled No. 45 both in form and in
the appearance of the colonies in gelatine plates. In peptone
and gelatine-stab the growths were the same as I?A.

Bacillus Choleroides a

This vibrio Avas isolated by Bujwid from river water. A little
peptone was added to some of the water placed in a flask ; after
incubation for three days a loopful from the surface plated out
in gelatine showed, after three to six days at room-temperature
(10°-12° R.), very suspicious colonies resembling the cholera
spirillum. At a higher temperature, however, the colonies
grew out in a more superficial manner, and were broader and
did not sink so much into the gelatine. A smell of methyl-
merkaptan was observed. Under a low power the colonies had
a sharp, regular outline, and a smooth or very finely granular
appearance. In gelatine-stab it grew chiefly on the surface, and
liquefied only the upper layers. At a temperature of 10C-12° R.
the gelatine was more slowly liquefied, and the air-bubble,
so characteristic of the cholera vibrio, was produced. In the
depth of the stab there was very little growth. On agar-slope
there was a marked growth, with a smell resembling methyl-
merkaptan. Broth showed very little growth, and no surface
pellicle was formed. There was no indol reaction. Microscopi-
cally the vibrio was not so motile as the cholera vibrio, and
longer and shorter forms were seen.

QUALITATIVE BACTERIOLOGICAL ANALYSIS.

B. Choleroides j3.

This organism was isolated by Orlowski, Bujwid's assistant,
from a well, in the neighbourhood of which there had been
several cases of cholera. It resembled the cholera vibrio more
closely than the above, but grew more anaerobically and pro-
duced a more marked funnel-shaped liquefaction.

VIBRIOS ISOLATED FROM WATER BY SANARELLI.

Sanarelli isolated thirty-two vibrios from water. Eight were
found in the Seine below Paris; of these one gave a very
marked nitroso-indol reaction, and its biological characters were
as follows :

Morphology. — A motile, thin, curved vibrio, often presenting
a spiral form.

Culture on Gelatine. — Forms the characteristic bubble of air
after twenty-four hours, and liquefies in a typical funnel form.

Culture in Solution of Peptone-gelatine. — Rapid development ;
a pellicle forms at the surface in twenty-four hours.

Culture on Agar. — Marked development at 37° C. on agar
made with and without broth.

Culture in Broth. — Rapid growth, with formation of a super-
ficial pellicle in twenty-four hours.

Culture on Potato. — Brownish circumscribed growth.

Nitroso-indol Reaction. — Very marked in twenty-four hours.

Indol Reaction. — Very marked in twenty-four hours.

Nine other vibrios were isolated from the upper part of the
Seine where it flows through Paris. None of these gave a
marked cholera-red reaction ; the colour was either absent or
just perceptible after eight days incubation.

In the effluent from the sewage of Paris after irrigation,
vibrios were also found which gave a marked nitroso-indol
reaction. The biological characters of these organisms were as
follows:

No. XXV.— -Gennevilliers.

Morphology. — A motile vibrio, but longer and thicker than
the true cholera microbe.

Culture on Gelatine. — Development along the line of inocula-
tion, with funnel-shaped liquefaction at the surface.

250 BACTERIOLOGICAL EXAMINATION OF WATER.

Culture in Solution of Peptone- gelatine. — Diffuse growth, with
a pellicle at the surface.

Culture on Agar. — Good growth on the agar made without
broth ; no growth on the agar containing broth.

Culture in Broth. — Very feeble after twenty-four hours ; later
very abundant with a surface pellicle.

Culture on Potato. — Insignificant.

Nitroso-indol Reaction. — Negative after twenty-four hours ;
very marked after eight days.

No. XX VI.— G-ennevilliers.

Morphology. — Very thin motile vibrio.

Culture on Gelatine. — Development along the line of inocula-
tion, with a small cup-shaped liquefaction at the surface.

Culture in a Solution of Gelatine-peptone. — Abundant growth
with formation of a pellicle.

Culture on Agar. — Abundant growth on agar free from
broth ; none on the agar made with broth.

Culture in Broth. — Very small after twenty-four hours ; later,
more abundant, with formation of a surface pellicle

Culture on Potato. — Insignificant.

Nitroso-indol Reaction. — Negative after twenty-four hours ;
very marked after eight days.

No. XXVII.— Gennevilliers.

Morphology. — A motile vibrio, less curved, longer, and a little
thicker than the above.

Culture on Gelatine. — Abundant growth, with liquefaction
along the line of inoculation and characteristic bubble of air ;
after four days typical liquefaction in the form of a funnel.

Culture on Agar. — Very rapid on both forms of agar.

Culture in Broth. — Abundant, with a superficial pellicle after
twenty-four hours.

Culture on Potato. — A light grey layer.

Nitroso-indol Reaction. — Slight after twenty-four hours ; verv
marked with a scarlet colour after eight days.

In the water of the Seine at Versailles another vibrio was
found closely resembling No. XXVII. It, however, gave a very-
marked nitroso-indol reaction after twenty-four hours incubation.

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 251

The remaining cultures out of the total thirty- two isolated
showed extreme variability. Side by side with vibrios which
liquefied gelatine in a manner absolutely characteristic of the
true cholera spirillum, others were found which liquefied gelatine
very slightly, or not at all, and which multiplied with great
difficulty. Many also refused to grow on ordinary gelatine, and
also in broth, except at the room temperature. It could not be
said that the failure to develop at 37° C. was directly related to
a condition of degeneration, for many vibrios which failed to
liquefy gelatine grew at 37°, and vice versa; many which liquefied
gelatine failed to grow at 37° C. The cultures on potato failed
to be of use as a means of diagnosis. With regard to the
nitroso-indol reaction, all the vibrios isolated from water were
able to produce indol, but very few of them at the same time
were able to reduce nitrates to nitrites. As regards pathogenicity,
only four out of the thirty-two microbes were found to be patho-
genic when injected intra-peritoneally into guinea-pigs; but
many of them, though they did not cause immediate lethal
effects, produced a condition of gradual wasting from which the
animals eventually died. It was also found that if 1 c.c. of a
sterilised culture of B. coli was injected into the peritoneum of
a guinea-pig, and a small quantity of these vibrios at the same
time into the pleura of the animal, the vibrios multiplied and
caused the death of the animal. Sanarelli then endeavoured to
find out whether there was any reciprocal reaction between
animals vaccinated with the four pathogenic water vibrios and
two other vibrios derived from undoubted cases of cholera.
Guinea-pigs were vaccinated with cultures grown for eight to
ten days in peptone-gelatine and then sterilised at 120°. A
successful vaccination was proved by the animal resisting a fatal
dose of the active culture. The results of the experiments are
shown in the following table. The sign + indicates that the
guinea-pig, vaccinated against a given variety, survived after
inoculation with a different variety ; the sign — indicates the
death of the vaccinated animal. Sanarelli does not state in his
paper whether the cultures used for vaccination were sterilised
at 120° C. or 120° F. If the former temperature was used, the
cultures would not be likely to have much protective influence.

L>52 BACTERIOLOGICAL EXAMINATION OF WATER.

Inoculation.

Guiue.i-pigs vaccinate 1 zigainst vibrios coming from

Courbevoie
(choleiM

c:su).

Angers
(cholera
case).

Toint-du-
Jonr
(water).

S. lint-
Clou:!

(water).

Genne-
villiers

(water).

Versailles
(water).

Guine:i--iii<i's
vaccinated with

Courbevoie .

__

+

+

+

Angers .

—

—

—

—

—

Point-du-Jour

+

—

+

+

+

Saint-Cloud .

+

—

—

+

+

Gennevilliers

+

—

—

—

Versailles

+

+

+

+

This table shows that there is no constant vaccinal reciprocity
between the vibrios derived from different sources. Even in the
case of the true cholera vibrios the spirillum from Courbevoie did
not confer immunity against the spirillum from Angers, and
vice versa.

Sanarelli also made experiments to see the effect of placing
the Vibrio Saint-Cloud under conditions closely resembling those
found in nature. Two flasks were filled with water from the
Seine at Clichy and then sterilised at 120° C. Some drops of a
culture of the Vibrio Saint-Cloud were then added to each flask,
one of which was kept at the room temperature (20°— 24° C.)
and the other placed in ice at 5°-10° C. After a month the
vibrio was again isolated from the flasks, and it was found that
the vibrio from the first flask showed a feeble cholera-red
reaction, as compared with the original culture, and in the case
of the vibrio from the second flask the cholera-red reaction
was almost imperceptible. Both vibrios had also completely
lost their virulence, intra-peritoneal injections into guinea-pigs
of large doses of the microbes producing no effect. A specimen
of the water at Saint Cloud, from which the Vibrio Saint-Cloud
was isolated, was kept at the room temperature for three months.
Vibrios were then isolated from the water, but the new specimens
gave no indol reaction, liquefied gelatine much more slowly than
before, and had no pathogenic action on animals.

Kohlbrugge quite recently investigated water from an arm
of the Rhine at Utrecht (free from cholera for seven years),
and found vibrios present during every month in the year ; but
whereas the vibrios isolated in the summer months gave a
marked cholera-red colour, those isolated in the winter failed to

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 253

give any trace of this reaction. As a result of his comparisons
of the water- vibrios with true cholera spirilla, he concluded that
the two groups could not be distinguished by cultural reactions?
and that a diagnosis of cholera could only be made when the
bacteriological examination was confirmed by epidemiological
facts and the appearance of disease.

METHODS PROPOSED FOR THE ISOLATION OF THE CHOLERA
SPIRILLUM FROM WATER.

The .simplest method is to convert a considerable bulk of the
specimen of water into a 1 per cent, peptone and 0*5 per cent,
salt solution. This is easily done by keeping a stock solution con-
taining 10 per cent, peptone and 5 per cent, salt ; 10 c.c. of the
stock solution are added to 90 c.c. of the water placed in a
sterile flask, which is then incubated at 37° C. After twelve to
twenty-four hours a faint pellicle forms on the surface if cholera
spirilla are present. Loopfuls must then be removed from the
surface and plated out in gelatine and rubbed over the surface
of agar solidified in Petri dishes. If pure cultures of spirilla
are obtained the colonies must be fished and planted out on the
following media :

(1) Peptone and Salt Solution. — After twenty-four hours
incubation at 37° C. the tube will become turbid, and on the
addition of pure sulphuric acid the cholera-red reaction will be
obtained.

(2) Gelatine Plates. — The characteristic colonies with irregular
margins will appear.

(3) Gelatine-stab. — The typical air-bubble liquefaction will be
obtained.

(4) Agar-slope. — The growth which appears in twenty-four
hours must be tested for agglutination and for Pfeiffer's reaction
with anti-cholera serum.

A portion of the colony should be examined in a hanging-
drop for the characteristic appearance and motility of Koch's
vibrio. If a spirillum conforms to all the tests it may be classed
with the true cholera spirilla.

Metchmkoff' recommended the following method: A series of
flasks are prepared, and into each the following solution is
poured, i.e., water 50 c.c., peptone 2 grammes, salt

ii;H BACTERIOLOGICAL EXAMINATION OF WATER.

gelatine 4 grammes, and a solution of soda sufficient to give a
slight alkalinity. The flasks are sterilised in the autoclave, and
then 150 c.c. of the suspected water are added to each of them.
The flasks and contents are incubated at 37° C. ; if a film appears
on the surface after eight to twelve hours, loopfuls are removed
and treated as above described.

Sanarelli employed a similar method. He placed 200 c.c. of
the water in a sterilised flask sufficiently large to allow a con-
siderable surface of the water to be exposed to the air, and then
added 8 c.c. of the following nutritive mixture :

Gelatine .......' 20 grammes.

Dried peptone ... . . 10 „

Sodium chloride ....... 10 „

Potassium nitrate . . . . . 1 gramme.

This gelatine, prepared beforehand, is kept in sterilised tubes.
Bv adding about 8 c.c. of the nutritive mixture the water is con-
verted into a nutritive solution having the following composition :

Gelatine . . . . . . 2 grammes.

Peptone . . . > . .1 gramme.

Sodium chloride ..-.'... 1 „

Potassium nitrate . . . . . 0-10 „

Water ........ 100 grammes.

In these flasks the vibrios develop so rapidly that after twelve
hours incubation at 37° C. they appear at the surface in the
form of a thin pellicle. In order to obtain pure cultures of the
vibrios, Sanarelli either plated out loopfuls of the pellicle in
gelatine, or made several passages through the gelatine and
peptone water, re-inoculations being made every six hours. In
this way absolutely pure cultures were obtained, the vibrios
appearing before the water bacteria had time to develop.

THE VITALITY OF THE CHOLERA SPIRILLUM IN WATER, &c.

A large number of experiments have been made to ascertain
the vitality of the cholera spirillum in water. Kraus studied
the behaviour of the spirillum in three specimens of water
which had the following chemical composition :

T?.tf1 Chlorine. Xitr.ites. Nitrites. Ammonia.
(All in parts per 100,000.)

No. 1. Pipe-supply . 27'0 0'41 0'2-i 0 0 0'036

No. II. Well-water . .V.H5 2-<»7 «;•«)<; 0 0 0'304

No. III. Well-water . 5G-0 2 •:>("> 0'26 0 trace 0-020

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 25,>

Koch^s vibrio was introduced into these three waters, and the

following results were obtained :

\umber of days alter the water was inoculated.

1

2

4

8

INo. I. .

10,000

0

0

0

No. II. .

-

8,700

0

0

0

No. III. .

9,420

0

0

0

6» 1^0. I. . ..

es « «'

•=-•.2

30

400

70,000

1,400,000

j.sf KNO. ii. . _ ..

80

900

85,000

innumerable

£|°JNo. III. .

250

2,000

100,000

These results showed that Koch's vibrio could not be detected
forty-eight hours after inoculation, although No. II. specimen
contained large quantities of nitrates, and No. III. specimen
contained a good deal of organic matter. In these experiments
special methods were not used to isolate the vibrios, so the
results must be accepted with considerable reservation.

Nicati and Rietsch found cholera vibrios in

Sterilised distilled water after
Marseilles canal-water after
Sea- water after
Harbour-water
Bilge- water .

20 days.

38 ,",
64 „
81 „
32

De Giaxa found cholera vibrios in unsterilised sea-water after
two to four days ; in sterilised sea-water the vibrios appeared to
multiply, and he detected them after several weeks. Koch
found vibrios alive after thirty days existence in well-water ;
but in a canal-water at Berlin he only found them up to six or
seven days; and when the water was mixed with faeces, the vibrios
seemed to disappear after twenty-seven hours. Babes found
that cholera vibrios lived for seven days in Berlin tap- water.
Wolff hugel and Riedel stated that cholera spirilla did not mul-
tiply in sterilised water kept at 16° to 20° C. ; and in un-
sterilised water kept at the same temperature they found the
vibrios disappeared in a few days. Hueppe, working with
sterile distilled water containing common salt, found the spirilla

256 BACTERIOLOGICAL EXAMINATION OF WATER.

alive after twenty to sixty days ; he considered that the presence
of common salt exercised a favourable influence on the vitality
of the spirilla. Hochstetter detected cholera spirilla in steri-
lised Berlin tap-water, kept at 30° C., after three hundred and
ninetv-two days ; although in some cases the water was found
not to be sterile, other organisms having gained access to it,
owing to the length of time covered by the experiment. The
temperature at which the vibrios are kept seems to influence
the result. In some experiments the spirilla appeared to live
longer at a temperature of 16° C. than at a temperature of 11° C.,
which corresponds to that usually found in well-waters. At
Altona, however, Koch found that cholera vibrios rapidly
disappeared from the water in a well, but in a litre of the same
water, removed from the well and kept at 3° to 5° C., he de-
tected the vibrios after eighteen days. Uffelmann, working
with a very impure harbour-water from Rostock, found cholera
vibrios at a room temperature only up to the third day.
These experiments, which are in harmony with the outbreaks
of cholera at Nietleben and Altona in winter, show that
cholera spirilla live longer in cold than warm water ; probably
because they are not crowded out by the water organisms which
cannot multiply at a low temperature. The observations at
Nietleben also showed that cholera vibrios could exist in ice-
cold water under snow and ice. Uffelmann allowed water
containing cholera vibrios to freeze, and every day melted just
as much of the ice as he required for his experiments. Al-
though the temperature at night fell to — 24'8° C., living cholera
spirilla were isolated from the melted ice-water up to the fifth
day. Trenkmann made some remarkable experiments on the
influence of salts on the growth of cholera spirilla. Test tubes
were carefully cleaned and filled with 10 c.c. of a well-water ; then
into each tube one, two, or three drops (25-27 drops = 1 gramme)
of a 10 per cent, solution of sodium chloride, sodium nitrite,
sodium nitrate, sodium carbonate, and di-sodium phosphate
were added by means of a small pipette. The test-tubes were
sterilised, and then each of them inoculated with one loopful of
a twenty-four hours broth culture of the cholera spirillum.
The inoculated glasses were kept at 21°-24° C., and after
twenty-four hours a loopful was taken from each of them with

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 2.57

the same platinum needle and plated out in gelatine,
following results were obtained :

The

(1)

10 c.c.

sterile well- water

Plates after twenty- Plates after
four hours. eight d::,ys.
580 -JS

(2)

do.

+ 1 drop 10

percent, sodium chloride

6,120

12,480

(3)

do.

+ 2 drops

do.

do.

9,240

19,560

(-0

do.

+ 3 do.

do.

do.

15,000

10,440

(5)

do.

+ 1 drop

do.

sodium nitrite

1,740

10,920

(6)

do.

+ 2 drops

do.

do.

6,600

1,460

(7)

do.

+ 3 do.

do.

do.

17,160

2,260

(3)

do.

+ 1 drop

do.

sodium nitrate

8,040

4,040

(9)

do.

+ 2 drops

do.

do.

6,660

14,760

(10)

do.

+ 3 do.

do.

do.

20,940

16,080

A control plate, made immediately after the inoculation,
showed 1440 colonies.

(1) 10 c.c. sterile well-water

(2) do. + 1 drop 10 per cent, di-sodium phosphate

(3) do. + 2 drops do. do.

(4) do. + 2 do. do. do.

(1) 10 c.c. sterile well-water

Plate after twenty -
four hours.*

520
3,360
7,560
6,540

Plate after twenty-
four hours.

580
7,440
28,680
31,560

54,720

(2) do. + 1 drop 10 per cent, sodium carbonate

(3) do. +2 drops do. do.

(4) do. +3 do. do. do.

(5) do. + 2 drops 10 per cent, sodium chloride, and)

1 drop sodium carbonate /

A control plate, made immediately after the inoculation,
showed 1800 colonies. From these experiments Trenkmann
concluded that in waters containing no sodium chloride or
other sodium salts, the cholera bacilli diminished in numbers,
whilst in others they multiplied almost in proportion to the
:salt added. The greatest multiplication was seen with a com-
bination of sodium chloride and sodium carbonate. Trenk-
mann also found that potassium salts gave rise to a similar
multiplication of the cholera spirilla.

(1) 10 c.c. sterile well-water

(2) do.

+ 1 drop 10 per cent, potassium chloride

(3) do.

+ 2 drops do. do.

(4) do.

+ 3 do. do. do.

(5) do.

+ 1 drop do. potassium nitrate

((>) do.

+ 2 drops do. do.

(7) do.

+ 3 do. do. do.

Plate after twenty-
four hours.

1,020

4,200
38,520
43,320

9,000

19,860

22,740

R

258 BACTERIOLOGICAL EXAMINATION OF WATER.

The following experiments were then made with unsterilised
water :

Plate after twenty-four hours.
Cholera bucilli. Water bacteria.

(1) 10 c c. unsterilised well-water 216

19,300

(2) do. + 1 drop 10 per cent, sodium sulphide 0

2,370

(3) do. + 1 do. do. sodium chloride 54

30,000

(4) do. 4- 2 drops do. do. 1,940

32,400

(5) do. + 3 do. do. do. 10,000

32,400

(6) do. + do. do. + 1 drop 10 per)

294A

cent, sodium sulphide/

do. + do. do. + 2 drops do. 17,100 540

do. + do. do. + 3 drops do. 28,100 2,160

The test tubes were kept in the incubator at 21°-24° C.

It therefore appears that :

(1) In unsterilised water the cholera bacilli quickly diminish,
whilst the water bacteria quickly multiply.

(2) The addition of common salt at first has little or no
action on the cholera vibrios, but, with an increasing amount of
salt, the vibrios undergo considerable multiplication ; the water
bacteria multiply rapidly on the addition of the smallest amount
of salt.

(3) After the addition of sodium sulphide the vibrios rapidly
disappear.

(4) After the addition of sodium sulphide and sodium
chloride the cholera vibrios undergo rapid multiplication.

When sodium chloride and sodium carbonate were added
to the unsterilised well-water, the water bacteria rapidly
multiplied, but the cholera vibrios could not be isolated after
the fourth day. When, however, sodium sulphide, in addition
to sodium chloride and sodium carbonate, was added to the
water, the cholera vibrios underwent considerable multiplication
and could be isolated up to the seventh day. At a lower-
temperature of 12*5°— 16° C. the addition of sodium chloride,
di-sodium phosphate, and sodium sulphide enabled the spirilla
to be isolated up to the ninth day. As a result of all his
experiments, Trenkmann states that the addition of sodium
chloride and sodium sulphide to a water causes a rapid disap-
pearance of many species of water bacteria, and very often
only a few species, sometimes only one, remain in conjunction
with the cholera vibrios. He also points out that water fouled

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 251)

with urine, faeces and refuse from manufacturing processes, may
contain sodium chloride and sodium sulphide, and so present
conditions favourable to the duration of life of cholera vibrios,
Trenkmann's experiments certainly explain the prolonged
vitality of the cholera vibrios in certain epidemics, such as
that at Hamburg, where the water was distinctly saline
owing to the high percentage of salt discharged from salt
works into the Elbe.

Summarising the results of the various experiments which
have been made on the vitality of cholera spirilla, Tiemann
and Gartner give the following conclusions : " Cholera spirilla
behave very differently in water. Most commonly they dis-
appear in a few days ; at 0° C. they can exist for several days,
at 10° C. for weeks,' and at 20° C. for months." The rapid dis-
appearance of cholera spirilla from an infected water cannot be
safely assumed, as most of the earlier experiments which appeared
to demonstrate this point were made without attempting to
isolate the spirilla by the special methods now in use. Gruber,
employing special methods, has shown that when cholera spirilla
are mixed with putrefactive bacteria, although the latter may
gain the ascendency for some time, yet the vitality of the former
is not extinguished, for if the struggle between the two be suffi-
ciently prolonged, the presence of the cholera bacilli may be again
demonstrated by cultivation.

The vitality of cholera spirilla in water containing carbonic
acid gas is an important practical question. Hochstetter has
studied this subject with great care. In eight experimenls with
fifty-six flasks of soda-water, he found that the cholera spirilla
could be isolated in a living state after three hours exposure to
the influence of carbonic acid gas, but after twenty-four hours
not a single spirillum could be detected. The rapid disappear-
ance of the spirilla was found to be due to the carbonic acid gas
and not to the chemical constituents of the water. In soda-
water which had been heated so as to drive off* the carbonic acid
gas cholera spirilla were found alive up to the eighteenth day.
A pressure of two atmospheres appeared to have little influence
in destroying cholera spirilla ; carbonic acid gas simply passed
through the water killed the spirilla in the shortest time.

The vitality of the cholera bacillus in crude sewage has quite

260 BACTERIOLOGICAL EXAMINATION OF WATER.

recently been investigated by Houston. Ten c.c. of Crossness
sewage were poured into a sterile test tube, and then inoculated
with a loopful of a recent agar culture of the cholera vibrio.
The tube was kept in the dark at the room temperature, and
from time to time a loopful of the contents was transferred to a
peptone and salt solution tube. The peptone tube was incu-
bated at 37° C., and in less than twenty-four hours a loopful
was taken from the surface and a cover-glass preparation made.
The results obtained were as follows : Immediately after the
inoculation the cholera vibrios were present in great abundance,
and almost pure cultures were obtained. On the following day
the vibrios were demonstrated with difficulty. On the third
and fourth days, however, the vibrios were present in abundance.
From this date up to the eleventh day the vibrios were demon-
strated with difficulty. On the sixteenth day a few typical
vibrios were found, but on the twenty-fifth day many vibrios
were isolated which were longer and thicker than the normal
microbes. On the forty-second day the vibrios completely
disappeared.

The experiment was repeated, but now a whole agar-slope of
the spirillum, instead of one loopful, was added to the sewage.
Again, the same difficulty was experienced in isolating the
vibrios the day after the inoculation. On the eighth dav they
were found without difficulty ; but after the fourteenth day the
result was negative. From these experiments it appears that
cholera spirilla may lose their vitality in less than a fortnight,4
or remain viable for nearly four weeks, in crude sewage.

THE RELATION OF CHOLERA VIBRIOS TO OTHER PATHOGENIC AND
NON-PATHOGENIC MICRO-ORGANISMS.

Many experiments have been made on this important subject.
Garre grew B. fluorescens putridus on agar, and having removed
the growth of this organism planted out Sp. cholerae on the
medium. He found that the products of B. fluorescens putridus
were only slightly antagonistic to the growth of the cholera
microbe. Freudenreich filtered a broth culture of B. pyocyaneus
through a Chamberland bougie, and then planted out Sp.
cholerae in the filtrate ; the organism, however, refused to grow.
He also found that broth, in which the cholera spirillum

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 261

had been grown, was not a favourable medium for many
bacteria.

Kitasato made numerous experiments with cholera vibrios and
pathogenic and non-pathogenic bacteria. The methods em-
ployed were as follows :

(1) The cholera vibrio and the organism selected were planted
out in parallel streaks, and in the form of a cross, on gelatine
plates.

(2) The cholera vibrio and the organism were planted out at
the same time on an agar-slope, and incubated at room or blood-
temperature.

(3) The vibrio and the organism .were planted out in slightly
alkaline broth and incubated as above.

(4) The organism was inoculated into a recent broth culture
of the cholera vibrio.

(5) Cholera vibrios were planted out in a recent broth culture
of the organism.

(6) Ten hours after the inoculation made under (2), (3), (4),
and (5), and subsequently from time to time, a loopful of the
culture was removed and plated out in gelatine.

(7) Care was also taken to plant out the above mixed cultures
on fresh media, and at different temperatures, in order to make
sure that the culture which had disappeared was not able to
develop under more favourable circumstances.

Kitasato obtained the following results :

PATHOGENIC ORGANISMS.

Cholera Vibrios and- Anthrax Bacilli. — The cholera vibrios
were not in any way injured by admixture with anthrax bacilli ;
on the other hand, the anthrax bacilli were quite destroyed at
the end of two weeks.

Cholera Vibrios and Typhoid Bacilli. — The typhoid bacilli
were not in any way inj ured by the vibrios. At first the former
were slightly overgrown by the latter, but after a few days both
microbes grew well and were found alive at the end of three
months.

Cholera Vibrios and Friedlander's Pneumo-bacilU. — The
pneumo-bacilli disappeared at the end of a month, whilst the
cholera vibrios were still alive at the end of three months.

262 BACTERIOLOGICAL EXAMINATION OF WATER.

Cholera Vibrios and the Bacilli of Green Pus. — The bacilli
overgrew the vibrios ; at the end of three months the vibrios
disappeared, but the bacilli could still be found after thirteen
months. Cholera vibrios planted out in a filtered broth of
B. pyocyaneus were quite unable to develop and rapidly dis-
appeared.

Cholera Vibrios and Briegei'^s Bacilli. — The bacilli were at
first overgrown by the vibrios ; later, they developed well. At
the end of three months both were found alive in a mixed
culture.

Cholera Vibrios and Emmerich's Naples Bacillus. — The relation
was the same as in the case of Brieger's bacilli.

Cholera Vibrios and Staphylococcus Pyogenes Aureus. — The
staphylococcus was at first overgrown by the vibrios, but both
were found alive at the end of three months. The same result
was obtained with Staphylococcus pyogenes albus.

Cholera Vibrios and Streptococcus (Erysipelas). — The strepto-
cocci had no action on the vibrios, but were rapidly destroyed.

NON-PATHOGENIC ORGANISMS.

Cholera Vibrios and B. Prodigiosus. — On mixed plates the B.
prodigiosus grew so rapidly that the cholera vibrios could not
be observed ; so the mixed cultures were planted out on
potatoes and incubated at 22 C. and 37° C. At 37° C. the
cholera colonies were seen, at 22° C. the red growth of prodi-
giosus appeared. On agar plates at 37° C. the cholera vibrios
could be detected after two months. The B. prodigiosus was
found alive after thirteen months.

Cholera Vibrios and B. Indicus. — The B. indicus overgrew the
vibrios ; both disappeared in three months.

Chulera Vibrios and B. Fluorescens. — The fluorescent bacteria
had no influence on the vibrios and died out before the latter in
mixed cultures.

Cholera Vibrios and B. Violaceus. — The B. violaceus was over-
grown by the vibrios and disappeared in fourteen days.

Cholera Vibrios and B. Subtilis. — The B. subtil is grew with the
cholera vibrios without any unfavourable influence. The same
result was obtained with the B. mycoides, B. mesentericus,
Proteus species, and B. butyricus (Hueppe).

QUALITATIVE BACTERIOLOGICAL ANALYSIS. 26\S

Cholera Vibrios and B. Acidi Lactici. — When grown in a
medium free from sugar the B. acidi lactici had no influence on
the vibrios ; but in milk, owing to the production of lactic acid,
the vibrios quickly died out.

Cholera Vibrios and B. Cyanogenus. — There was no unfavour-
able influence ; both bacteria grew well together.

Cholera Vibrios and Red, White, and Orange Cocci from
Air. — The vibrios gained the upper hand and the cocci
disappeared in a month. The yellow sarcina behaved in the
same manner. Three yeasts were also destroyed in a short
time.

Bacilli isolated from faeces, singly and combined, were found
to have no prejudicial influence on the cholera vibrios.

Kitasato also investigated the growth of cholera vibrios in
sterilised cultures of a large number of bacteria, and found that
if the cultures were recent the vibrios grew well, but the older
the cultures the more feeble was the development of the vibrios.
In pyocyaneus cultures, however, the vibrios always died out.

THE BEHAVIOUR. OF THE SPIRILLUM OF ASIATIC: CHOLERA ix
NUTRIENT MEDIA CONTAINING ACIDS AND ALKALIES.

This subject has been studied by Kitasato, who employed the
same methods as were used for his studies of the growth of the
B. typhosus in acid and alkaline media. The results obtained
are shown on the following page :

Kitasato found that in acid media the growth of the cholera
vibrios was dependent on the temperature. When he added
to neutral agar media 0'225 per cent, lactic acid, and then
inoculated the media with cholera vibrios, no growth occurred
if the plates were kept at 20° to 22° C., but if the plates were
incubated at 37° C. colonies developed in three days.

The same result is seen when cholera vibrios are planted out
on slightly acid potatoes ; no growth occurs at 22° C., but at
37° C. the characteristic development takes place.

Liborius found that 0*0246 per cent, caustic lime destroyed
cholera vibrios. Kitasato^s higher figure was due to the fact
that he employed undiluted broth, and some of the lime was
expended in neutralising the phosphates contained in the
broth.

BACTERIOLOGICAL EXAMINATION OF WATER.

The sign + indicates a good growth, + restrained growth, — no growth.

Acids.

+

+

—

per cent.

per cent.

per cent.

Sulphuric acid

0-02

0-032

0-049

Hydrochloric acid . «

0-052

0-08

0-132

Nitric acid . . . .

0.052

0-08

0-132

Sulphurous acid
Phosphoric acid . . .

0-07
0-076

0*09

0-135

0-148

0-183

Carbolic acid ... jy

0-088

0-12

0-2

Acetic acid

o-l

0-153

0-2

Formic acid . .

0-11

0-167

022

Oxalic acid ....

0-U6

0-23

0 285

Lactic acid . . .

0-18

0-225

0-27

Tartar! c acid . . . .1

Citric acid ]-

0-2

0-24

0-3

Malic acid . . . . J

Tannic acid . . V*

0-9

1-3

Boric acid . ' .- .

0-48

0-8

1 -33

Alkalies.

+

+

—

per cent.

per cent.

per cent.

Caustic lime . ;

0 0922

0*0866

0-1004

Caustic potash : . ..' . \
Caustic soda . . '.. ./

0-14

0-18

0-237

Ammonia

0-2

0-24

0-338

Lithium carbonate .

0*6

0'666

0-72

Potassium carbonate

0-74

0-81

l-o

Ammonium carbonate

0-8

1-0

1*3

Barium hydrate

0-83

1-0

1-42

Sodium carbonate .

2-2

2-47

3-45

Salts.

+

±

—

per cent.

per cent.

per cent.

Potassium iodide

6-66

8-0

9-23

Potassium bromide .

8-0

9-23

10-37

Potassium chloride .

<>-l

10-6

12-0

CHAPTER XV.

THE MODE OP ACTION AND UTILITY OP THE

PASTEUR-CHAMBERLAND AND BERKEFELD

FILTERS.

THESE filters have a very trifling influence on the chemical
composition of water which is passed through them. They
are supposed, however, to prevent the passage of bacteria by
exercising a molecular influence on the microbes during the
transit of the fluid containing them through the materials of
the filtering candles. Numerous investigations have been made
on the efficacy of these filters. In 1890 Kiibler tested the
Pasteur-Cham berland filter and arrived at the conclusion that
this filter only delivered a germ-free filtrate for a period of four
days. In his first series of experiments water, containing from
6160 to 126,720 bacteria per c.c., was passed intermittently
through the filter, suction being applied by the mouth. Up
to the fourth day the filtrate was sterile ; on the fifth day
twenty-four bacteria per c.c. were counted, and on the eighth
day the bacteria were innumerable. In the second series of
experiments, employing continuous filtration, a sterile filtrate
was only obtained up to the third day. In the third series
of experiments the surface of the filter was brushed every day,
but a sterile filtrate was only obtained up to the end of the
first day. Kiibler noticed that the first bacteria to appear in
the filtrate were the fluorescent motile bacteria ; two days later*
other species appeared. He accordingly came to the conclusion
that the bacteria grew through the filtering candle owing to
their power of multiplying in water, and this idea was supported
by the fact that when the filtration was maintained at 0° C.,
a temperature unfavourable to the multiplication of micro-
organisms, a permanently sterile filtrate was obtained.

266 BACTERIOLOGICAL EXAMINATION OF WATER.

In 1892 Freudenreich re-investigated the possibility of the
passage of bacteria through the Pasteur-Chamberland filter.
A bougie was sterilised, and the case having been screwed on to
a tap, Seine water was passed through the bougie under a
pressure of one-third of an atmosphere ; 830 c.c. of the filtrate
were collected in a flask containing 500 c.c. of broth ; the flask
was then incubated at 30-35° C. The filter was maintained in
use for three days, water being allowed to pass through drop by
drop; 760 c.c were then collected as before. After another
three days 610 c.c. were again collected. All the flasks were
found to be free from bacteria after incubation for twelve days
at 35-37° C. A similar experiment was made with Ourcq water,
and these flasks also remained sterile. The results of the ex-
periments appeared to show that a sterile filtrate was delivered
from the filter for a period of six days. As the results were
quite at variance with the experiments of Kubler, Freudenreich
then devised an experiment to test the possibility of micro-
organisms going through the walls of the filtering candle. A
porcelain candle of known structure was taken, and a pipette,
with a globular enlargement above was passed through the
opening to the bottom of the bougie ; the mouth of the pipette
was plugged with cotton-wool and the whole sterilised in the
autoclave. After sterilisation the junction of the pipette to
the bougie was closed with melted paraffin ; the bougie was
then placed in a glass vessel filled with water and kept at a
known temperature. At various intervals some of the water
which had filtered through the bougie was aspirated into the
expansion on the pipette and then inoculated in broth.

The same apparatus was next used to see if typhoid bacilli
were able to grow through the walls of the filtering candle. Broth
placed in the glass vessel was inoculated with a B. typhosus cul-
ture and the apparatus was kept at the room temperature. Next
day there was a marked growth in the broth. At various intervals,
up to the end of fourteen days, portions of the filtered broth
were aspirated through the pipette but were found quite clear.
On the fourteenth day 3 c.c. were drawn through the pipette-
and placed in 200 c.c. of sterile broth ; this, however, remained
quit? sterile. In three other experiments the filter was kept at
35 C., and portions of the filtrate examined after twelve, fifteen.

ACTION AND UTILITY OF FILTERS. 2ti7

and twenty-two days were found quite sterile. It appeared,
therefore, that, under the conditions of the experiments, the
B. typhosus was not able to grow through the walls of the
filtering candle. Freudenreich believed that the failure was
due to the products secreted by the B. typhosus passing through
the bougie and exerting a negative chemiotaxis on the typhoid
bacilli, which were thus prevented from growing through the
walls. This explanation appeared probable, as it was found
that when some of the filtrate was inoculated with a fresh
culture of B. typhosus the bacilli were unable to grow in it.

Thirty experiments were made with water filtration, fifteen
at a temperature of 35° C., when the water filtrate remained
sterile after periods of filtration varying from two to fourteen
days, and fifteen at a temperature of 22° C.,in the latter experi-
ments the filtrate was sterile after from seven to eighteen
days filtration. Experiments were also made at 15-18° C. ; at
these temperatures the filtrate remained sterile after fifteen to
twenty-one days. These results also showed that water
bacteria were able to grow through the walls of the filtering
candle, but the rate of growth depended on the temperature.
Freudenreich stated that when typhoid bacilli were present in a
water supply, negative chemiotaxis would not be exercised by
the products present in the filtrate ; but in any case the typhoid
bacilli would never grow through the filtering candle more
quickly than water bacteria ; consequently, at ordinary room
temperatures, the Pasteur-Chamberland could be depended upon
to produce a germ-free filtrate for about eight days.

With regard to the Berkefeld filter, the results of the various
experiments have caused bacteriologists to express diametrically
opposite opinions as to the usefulness of the filter. Nordtmeyer
and Bitter considered that the Berkefeld filter gave a germ-free
filtrate for a longer time than the Pasteur-Chamberland filter.

O

Prochnick found the filtrate from the Berkefeld filter to be
germ free after the filter had been in continuous use for thirty-
eight days. Weyl, as the result of his experiments in Berlin,
stated that the Berkefeld filter produced a germ-free filtrate for
three days, and when the candles were brushed the filtrate
remained sterile for six days. In 1893, Kirchner made a large
number of experiments with the Berkefeld filter, and arrived at

i>6'8 BACTERIOLOGICAL EXAMINATION OF WATER.

the conclusion that the filtrate was not completely germ free
for a longer period than one day. He also tested the possibility
of pathogenic organisms passing through the filter. Broth
cultures of cholera vibrios, typhoid bacilli, and staphylococcus
pvogenes aureus were filtered through the bougie into a glass
vessel, in which there was a negative pressure of 500-600 mm. of
mercury ; one hour after the commencement of the experiment,
and then every twenty-four hours, a drop of the filtrate was
plated out in gelatine. Cholera vibrios and staphylococci were
found in large numbers in the filtrate received after twenty-four
hours, the typhoid bacilli, however, did not appear until after
seventy -two hours. A further experiment was then made by
adding 1.00 c.c. of a broth culture of cholera vibrios to one
litre of water which contained 30,000 bacteria per c.c., and
then filtering the mixture through a Berkefeld candle ; cholera
vibrios were again found in the filtrate at the end of twenty-
four hours. A similar experiment was made with B. typhosus,
and this organism was detected in the filtrate at the end of
forty-eight hours. Kirchner then concluded that the Berkefeld
filter operated in exactly the same manner on pathogenic as, on
non-pathogenic bacteria. Gruber criticised Kirchner's work,
and pointed out that, in the filtration of water, only those
bacteria could grow through a filter which found sufficient
nutriment in water to enable them to multiply in the pores of
the filter material. He considered that ordinary drinking-
water did not contain sufficient nutriment to enable pathogenic
bacteria to multiply in this manner. In Kirchner's experi-
ments pure broth cultures, or broth cultures diluted ten-fold
with water, were employed, and Gruber contends that in this
manner an amount of nutriment was supplied which enabled
the pathogenic organisms to multiply in the filter; such an
amount of nutriment he believes would never be found in an
ordinary drinking water. Gruber also considers that the
counting of the bacteria in the filtered and unfiltered water is
not a good test of a filter, and recommends the following points
for consideration when examining the power of a bougie to
prevent the passage of bacteria. The direct passage of
microbes must be sharply differentiated from the gradual
growth of bacteria through the material composing the filtering

ACTION AND UTILITY OF FILTERS. 26<)

candle. The direct passage points to some imperfection in the
filter, but the gradual growth through the material is common
to all filters, and does not indicate danger of infective disease.
He considers that the best way to test a filter is to pass through
it some easily recognisable special organism, and if this cannot
be detected in the filtrate, the filter may be considered to give
security against the passage of pathogenic organisms.

Schofer's experiments, which were made in 1893, strongly
supported Gruber^s contention that typhoid bacilli, when
inoculated into pure and impure natural waters, will not grow
through the walls of a filtering candle owing to a
deficient supply of nutriment. Schofer employed a Berkefeld
bougie placed in a glass cylinder. Sterile distilled water and
sterile well water were filtered through the bougie for sixteen
and twelve days respectively without any sign of the B. typhosus
appearing in the filtrate, although a culture of this organism
was inoculated into the contents of the glass cylinder, not onlv
on the first day but about every second day throughout the
experiment. A considerable quantity of water was passed
through the filter, and the temperature during the experiment
varied between 19° and 26° C. A similar experiment was made
with the spirillum of cholera, but no traces of this vibrio
could be detected in the filtrate. Schofer then tried the effect of
filtering a polluted sterile well water inoculated with typhoid
bacilli. The well water contained in one litre : 1205 mgms. of
total solids, 38 mgms. of organic material, marked ammonia,
traces of nitrous acid, 251 mgrns. of nitric acid, 120 mgms. of
chlorine, 158 mgms. of sulphuric acid, 245 mgms. of calcium
oxide, and 28 mgms. of magnesium oxide. The experiment
was continued for twenty-four days ; during this time a culture
of B. typhosus was added twelve times to the contents of the
glass cylinder, and the temperature was maintained between
18° and 25'2° C. The B. typhosus, however, was never detected
in the filtrate. Other experiments were made with a canal
water, which were of special interest, as they appeared to show
that the only cause of the non-appearance of the B. typhosus
in the filtrate was a deficiency of nutrient material. The
canal water contained in one litre : 215 mgms. of solids, 47
mgms. of suspended material, 8 mgms. of chlorine, and 2 mgms.

270 BACTERIOLOGICAL EXAMINATION OF WATER.

of ammonia. The water was sterilised, and filtered through a
new bougie for thirteen days. B. typhosus cultures were added
five times daring this period, but the organism could not be
detected in the filtrate. Five c.c. of broth were then added
to the water in the glass cylinder, and two days later typhoid
bacilli were found in the filtrate. Schofer concluded from his
experiments that only those bacteria were able to grow through
the filter which found sufficient nutrient material in the water
to enable them to grow out and multiply.

Sims Woodhead and Cartwright Wood in 1897 repeated and
extended the experiments made by Schofer. These observers
first tested the filtration of the New River Company water
supply through Pasteur-Chamberland and Berkefeld filters
attached to taps and used as pressure filters. In the case of the
Pasteur-Chamberland filter it was found that only one of the
four micro-organisms present in the water appeared in the
filtrate, and this microbe was so devitalised that its power of
liquefying gelatine and producing a yellow pigment was much
delayed ; it required several weeks cultivation before it recovered
its usual characteristics. In the case of the Berkefeld filter the
same organism appeared in the filtrate on the second or third
day of filtration, and was accompanied by the B. fluorescens
liquefaciens. These organisms, however, did not display any
evidence of the devitalising influence which was found in the
experiments with the Pasteur-Chamberland filter. Experiments
were then made with chromogenic organisms placed in the
reservoir of the filters before these were attached to the tap.
The Micrococcous prodigiosus, Staphylococcus pyogenes aureus
and Bacillus violaceus were employed for the tests ; but none of
these organisms could be discovered in the filtrate, although
the experiments were carried on for several weeks and large
quantities of the filtrate were added to concentrated broth and
incubated for some time. The test organisms at the end of the
experiments were easily recovered from the reservoir of the
filter, showing that their non-appearance in the filtrate was not
due to the death of the micro-organisms.

The next tests were made with pathogenic organisms, the
filters being fed day by day with emulsions of cholera and typhoid
bacilli in sterilised water. After the lapse of six and eight weeks

ACTION AND UTILITY OF FILTERS. 271

sterile filtrates were obtained from both filters, although inocu-
lations from the outer surface of the candles showed that living
pathogenic germs still remained. The addition of a little broth,
one-fiftieth of the bulk of the water, caused the pathogenic
organisms to appear in the filtrate. Typhoid and cholera bacilli
were then added to Thames water, taken near Waterloo bridge
at low tide, and sterilised before use. Negative results were
obtained with both filters, showing that these pathogenic
organisms, even when present in a highly polluted water, do not
pass through into the filtrate. "As a final and conclusive test
as to the possibility of disease germs traversing the walls of
these filters when fed with a water highly contaminated with
organic matter, an experiment was carried out with undiluted
sewage." A cholera culture was added to sterilised London
sewage, which was then filtered through a sterilised candle ;
spirilla could not be found in the filtrate although the experi-
ment was continued for eighteen days. A mixture of cholera
vibrios and sterile sewage was also passed through a sterilised
candle coated with slime resulting from previous filtration ; the
results in this experiment were negative. Lastly, a culture of
cholera was added to unsterilised sewage and then filtered
through an unsterilised candle coated with slime ; spirilla could
not be found in the filtrate. Experiments on similar lines were
also made with B. tvphosus, but these again gave an absolutely
negative result. All these results support the authors' contention
that waters naturally highly contaminated, and even sewage
itself, do not appear to supply pathogenic organisms with a
sufficiently favourable pabulum to enable them to traverse the
walls of Pasteur-Chamberland and Berkefeld candles.

In 1897 Lunt made an investigation of the capabilities of the
Berkefeld filter, in order to obtain an independent confirmation
of the results obtained by Schofer and others, as to the power of
this filter to intercept pathogenic bacteria — especially typhoid-
even when water bacteria are able to pass through. Lunfs-
experiments appeared to justify the following conclusions :

(1) When the Berkefeld filters are efficiently sterilised and used
in an approved manner, they give on the first day of using an
absolutely sterile filtrate. This is the case with all the forms of
filtering candles examined, not only with natural water but

272 BACTERIOLOGICAL EXAMINATION OF WATER.

also with water artificially infected with enormous numbers of
bacteria from cultures, and also with broth cultures teeming
with bacilli.

(2) Water bacteria, however, begin to appear on the second
or third day of using, so that to obtain a continuously sterile
filtrate it is necessary to re-sterilise the filter daily.

(3) A long series of experiments, in which London tap-water
was strongly infected with B. coli communis, showed that this
organism could not be detected in the filtered water over a
period of thirty-nine days. During this period the B. coli
communis could be detected on the outside of the filtering
candle as late as the thirty-ninth day.

(4) Exhaustive experiments showed that with London tap-
water strongly infected artificially with the typhoid bacillus, not a
single typhoid bacillus was detected in the filtered water over n
period of twenty-six days, though it was detected in the
unfiltered water up to the sixth day. As is well known, typhoid
bacilli ultimately perish in water which contains the water
bacteria which are always present in natural water, but during
the whole time the typhoid bacilli were demonstrated to be alive
in the unfiltered water, they were never once detected in the
water which had passed through the filter.

(5) London tap-water strongly infected artificially with the
cholera bacillus was found, after filtration through the filter, to
be absolutely sterile, and therefore entirely free from the cholera
organism.

(6) The method of using filters with a combined suction and
force pump, causing as it does sudden and violent changes of
pressure on the outside of the filter, gives very bad results with
filtering candles, which produce an absolutely sterile filtrate when
used by a better method. The principle of the filters in which
atmospheric pressure forces the filtered water into an exhausted
receiver is free from the above objection, and when used by
this method absolutely sterile filtrates are obtained, even when
identically the same filtering candles are used which previously
passed large numbers of bacteria from the very first, when used
by the suction and force-pump method.

(7) Seeing that the bacteria which first appear in the filtered
water are harmless water bacteria, and that such organisms as

ACTION AND UTILITY OF FILTERS. 273

B. coli communis and the typhoid bacillus do not accompany
them, even when they are present in large numbers in the un-
filtered water, it seems that in all cases in which a drinking-
water is not under suspicion of being contaminated, a semi-
weekly or weekly cleansing and sterilisation of the filtering
candle is all that is necessary as a precautionary measure
against water-borne disease. Nevertheless, in all cases of
contaminated water supplies, or supplies known to be infected
with disease organisms, a daily sterilisation of the filter-
ing candle is advisable, thus ensuring an absolutely sterile
nitrate.

All the experiments hitherto described appear to prove con-
clusively that typhoid bacilli will not be able to grow through
either a Pasteur-Cham berland or Berkefeld filtering candle.
Unfortunately, my own work has led me to a different conclu-
sion in regard to the Berkefeld candles. The experiments I
made were as follows :

Experiment I. — A Berkefeld candle (No. 12 in the catalogue)
was placed in a mantle and firmly fixed in position by means of a
binding screw working on a thread on the metal pipe delivering
the filtered water. This metal pipe was connected, by means of
a short piece of stout india-rubber tubing previously fitted
with a pinch-cock, to a short glass tube which just passed
through an india-rubber cork fitting tightly into a Kitasato's
flask. The lateral arm of the flask, and the mouth of the
mantle surrounding the filtering candle, were tightly plugged
with sterile wool. The whole apparatus was then placed in the
autoclave and exposed to a temperature of 115° C. for half an
hour. The apparatus was allowed to cool, and the filtering
candle and mantle connected in the upright position to an
ordinary burette stand. The pinch-cock being placed in
position on the india-rubber tubing so as to separate the
mantle completely from the flask, all the joints of the appa-
ratus were thickly covered with melted paraffin. The mantle
was then filled with broth so as to cover the bougie completely,
but not touch the cotton \vool. Fifty cc. of broth were placed
in the flask and, the bung being replaced, the joint was made
tight with melted paraffin. These manipulations were done
with every precaution, and carried out as quickly as possible,

27-1 BACTERIOLOGICAL EXAMINATION OF WATER.

so as to avoid contaminating the broth. Bv opening the
pinch-cock 10 c.c. of the broth from the mantle were now
allowed to filter through the bougie into the flask. The filtra-
tion occurred simply as the result of atmospheric pressure, and
the bougie worked under conditions which would allow its
molecular influence full play. The apparatus was then placed
in an incubator, the temperature of which averaged 25° C.
After twenty-four hours exposure to this temperature the broth
in the mantle and flask was found perfectly clear, showing that
the manipulations had been performed without introducing any
contamination. The broth in the mantle was then inoculated
with a loopful of a twenty-four hours agar culture of B. typho-
sus. Next day the broth in the mantle was found uniformly
turbid, but the broth in the flask was perfectly clear; the
pinch-cock was then opened and 10 c.c. allowed to filter through
into the flask. On the following day the broth in the flask
being still perfectly sterile, 10 c.c. were again filtered through
from the mantle. The same filtration process was continued
day by day, and the amount of broth filtered each day from the
mantle was replaced by fresh sterile broth, the manipulation
being carried out with every precaution. The broth in the
flask remained perfectly clear for four days, showing that there
was no imperfection in the connections of the filtering bougie
with the Kitasato flask. On the fifth day of filtration, however,
the broth in the flask was found slightly turbid. A ioopfu]
was then withdrawn from the contents of the flask by means of

B. typhosus in the mantle.

Organism in the filtrate.

Agglutination with 1

Complete 1-1000

Complete 1-1000

Berne serum . ./! Marked 1-10,000

Marked 1-10,000

Potato . . . Thin colourless growth

Thin colourless growth

Milk ....

Unchanged

Unchanged

Proskauer and Capaldi, I.

No growth

No growth

» » „ II.

Growth, medium acid

Growth, medium acid

Peptone solution . <

No indol reaction

No indol reaction

Litmus- whey, seven days j J Acidit>r less ^ 6 Per

Acidity less than »> per
N

at37°C. .

\ cent- io

cent. 10

Glucose-gelatine . . \
Lactose-gelatine . . J

No gas formation

No gas formation

Gelatine-stab and plates
M utility

Very slow growth
Very motile

Very slow growth
Very motile

Stained by Gram .

No

No

ACTION AND UTILITY OF FILTERS. 275

a bent platinum wira passed through the lateral arm, and
planted out on an agar-slope. The thin greyish growth which
resulted was then sub-cultured and tested for agglutination.
The B. typhosus used for the inoculation of the mantle was
also tested at the same time. The results are given in the
Table on the previous page.

The organism which appeared in the Kitasato flask was evi-
dently the B. typhosus which had taken five days to grow
through the walls of the No. 12 filtering candle. This result is
in harmony with the results of Schofer and Sims Woodhead.

Experiment II. — The same form of apparatus as in Experi-
ment I. was employed, a new Berkefeld bougie, size No. 12,
being fixed in the mantle. Fifty c.c. of sterile broth having
been placed in the Kitasato flask, and all the joints made tight
with melted paraffin, the mantle was filled with a sample of
sterile barrack sewage. This sewage on chemical analysis gave
the following results, expressed as parts per 100,000 : Total
solids 610 parts, chlorine 52 parts, free ammonia 1*44 parts,
albuminoid ammonia 1*846 parts, oxygen absorbed (Tidy, four
hours) 8 parts. By releasing the pinch -cock 10 c.c. of the
sterile sewage were allowed to filter through into the broth ;
the apparatus was then incubated at 25° C. Next day the
broth in the flask being perfectly clear, the sewage in the
mantle was inoculated with a sub-culture of the B. typhosus
employed for the first experiment. Ten c.c. of the sewage were
then filtered into the broth every twenty -four hours, the amount
filtered being replaced by sterile sewage. The broth remained
perfectly clear until the fifth day, when a distinct turbidity
was noticed. A loopful of the turbid broth in the flask was
withdrawn through the lateral arm and planted out on agar.
The growth which appeared in twenty-four hours was tested
with Berne serum for agglutination and sub-cultured in the
usual media. The following results were obtained :

The bacillus was completely agglutinated by the Berne serum
diluted 1-1000, and partially agglutinated by the serum diluted
1-10,000. On potato, there was a moist transparent growth. In
peptone solution, no indol was produced. Mil kwas unchanged. In
glucose-gelatine there was no gas formation. In litmus- whey, after

seven days at 37° C. the acidity equalled 4 per cent. ^ alkali.

276 BACTERIOLOGICAL EXAMINATION OF WATER.

Proskauer and Capaldi's media, No. I. showed no growth, No. II.
was rendered acid. In gelatine plates, growth was slow, the
colonies were small, and resembled those of B. typhosus. Micro-
scopical examination showed a small highly motile bacillus,
decolorised by Gram.

The tests showed that the bacillus which appeared in the
filtrate was identical with the B. typhosus planted out in the
sewage in the mantle. As the fittings were absolutely tight,
and the bacillus did not appear in the filtrate until the fifth day,
the experiment appeared to show that the sewage, just like the
broth, had supplied sufficient nutriment to enable the typhoid
bacillus to grow through the walls of the filtering candle.

Experiment III. — The same form of apparatus was employed
as in the two previous experiments. The mantle, however, was
filled with sterile barrack sewage, largely diluted with distilled
water. The dilute sewage was found to have the following
composition : Total solids, 7 parts ; chlorine, 0'3 part ; oxygen,
absorbed in four hours, 1-1 parts; free ammonia, (H248 part;
albuminoid ammonia,0'2016 part; nitrites, absent; nitrates (N()3),
0*926 part; all expressed per 100,000. The dilute sewage was
inoculated with a sub-culture of the same typhoid bacillus and
filtration carried on daily ; the amount filtered was replaced by
sterile sewage, so that the filtering candle was never allowed to
become dry or exposed to the air. The broth in the Kitasato
flask remained perfectly clear for nine days, but on the morning
of the tenth day of filtration a slight turbidity was noticed. A
loopful of the turbid broth was removed through the arm and
planted out on agar. The growth which resulted was tested in
the same manner as before, and found to be a pure culture of
B. typhosus. The results of the agglutination and cultural tests
exactly corresponded to those of the B. typhosus introduced into
the mantle. In this experiment the typhoid bacillus was not
supplied with as much nutriment as in the two former experi-
ments, and it appeared to require nine days to grow through the
walls of the filtering candle.

Experiment IV. — The same form of apparatus was employed
as before, a perfectly new No. 12 candle being fixed in the
mantle. Surface water from a ditch was sterilised and placed in
the mantle. This water on analysis gave the following results,

ACTION AND UTILITY OF FILTERS. 277

expressed as parts per 100,000 : Total solids, 44 parts ; volatile
solids, 16 parts ; chlorine, 6'2 parts ; free ammonia, 0'526 part ;
albuminoid ammonia, 0'60 part ; oxygen absorbed in four hours,
2'2 parts; nitrites, absent; nitrates (NO3), 1'005 part. The
water in the mantle was inoculated with a loopful of a twenty-four
hours agar culture of B. typhosus, and filtration carried on daily
as in the previous experiments. The broth in the flask remained
perfectly clear for eight days, but on the ninth day it was found
to be turbid. A loopful of the turbid broth, when examined as
before, was found to contain a pure culture of B. typhosus. In
this experiment the nutriment supplied was greater than in
Experiment III., but not so great as in Experiments I and II.
Corresponding apparently to the nutriment supplied, the time
taken by the typhoid bacillus to grow through the candle was
eight days as compared with four days in Experiments I. and
II., and nine days in Experiment III.

Experiment V. — In this experiment the mantle was filled with
a specimen of sterile barrack sewage, which contained a large
quantity of organic matter. The results of the analysis expressed
as parts per 100,000 were as follows : Total solids, 55 parts ;
volatile solids, 15 parts ; chlorine, 9*5 parts ; free ammonia,
8'28 parts ; albuminoid ammonia, 5*07 parts ; oxygen absorbed
in four hours, 6'4 parts ; nitrites, absent ; nitrates, absent. The
sewage was inoculated with a twenty-four hours agar culture of
B. typhosus, and filtration carried on in the same manner as in
the previous experiments. The broth in the flask remained
quite clear for four days, but was found turbid on the fifth day.
The turbidity was caused by B. typhosus. In this experiment
the available organic matter was much greater than in Experi-
ments III. and IV., and consequently the bacillus only required
four days to grow through the walls of the candle.

Experiment VI. — The same form of apparatus was used, a
new No. 12 Berkefeld candle being fitted into the mantle. A
specimen of water collected from a reservoir at Netley was
sterilised and then analysed. The following results, expressed as
parts per 100,000, were obtained : Total solids, 18 parts ; volatile
solids, 5 parts ; chlorine, 3'3 parts ; free ammonia, 0'00816
part; albuminoid ammonia, 0*0158 part; oxygen absorbed in
four hours, 0'5 part ; nitrites, absent ; nitrates (NO3), 0*8 part.

278 BACTERIOLOGICAL EXAMINATION OF WATER.

This water was placed in the mantle, inoculated with B.
typhosus, and filtration carried out as in the previous experi-
ments. The broth in the flask remained perfectly sterile until
the ninth day, when faint turbidity was noticed. A loopful
withdrawn through the lateral arm was planted out on agar,
and tested in the usual manner. The growth was completely
agglutinated by Berne serum diluted 1—1000, and when planted
out on the various media, the results obtained were identical
with those given by a true culture of B. typhosus.

Experiment VII. — The same water was used for this experi-
ment, which was conducted in exactly the same manner as
Experiment VI., a fresh No. 12 candle, however, being employed.
The B. typhosus again appeared in the filtrate on the ninth
day.

Experiment VIII. — The reservoir water was again employed,
but a large candle, No. 10, was used and fitted into a mantle
holding about 500 c.c. of water. The water in the mantle was
inoculated on three occasions with a twenty-four hours agar culture
of B. typhosus. The broth in the flask remained perfectly sterile
for ten days, but on the eleventh day turbidity was noticed, which
was found to be due to a pure culture of the typhoid bacillus.

Experiment IX. — A large candle, size No. 10, was employed,
and the mantle filled with sterile barrack sewage. A twenty-
four hours agar culture of the typhoid bacillus was planted out
in the sterile sewage, and filtration carried out in the same
manner as before. On the fifth day of filtration the typhoid
bacillus was found in the broth placed in the flask.

Experiment X. — The bougie used in Experiment VI. was
removed fiom the mantle and thoroughly cleaned with a stiff'
brush, the metal delivery pipe was also thoroughly washed and
flamed. The bougie was then replaced in a sterile mantle and
the apparatus fitted up in the usual manner. Sterile reservoir
water was then filtered through the candle, and next day the
broth in the flask was found to be turbid. The typhoid bacillus
was easily isolated from the turbid broth. This result appeared
to show that carefully cleaning a bougie would not remove the
B. typhosus which had evidently penetrated into the interior of
the material and was washed through into the filtrate by the
passage of the sterile water.

ACTION AND UTILITY OF FILTERS. 219

Experiment XI. — In this experiment the bougie used in
Experiment VII. was sterilised in an autoclave and then fitted up
in a sterile apparatus. Sterile reservoir water was filtered through
into the flask for eleven days without the slightest trace of
turbidity appearing in the broth.

Experiments on the same lines were then made with Pasteur-
Cham berland candles, fitted into glass mantles by means of
india-rubber bungs. The delivery pipe of the candle in each
case was connected by a short piece of india-rubber tubing, fitted
with a pinch-cock, to a Kitasato flask containing sterile broth.
As in the previous experiments, all the joints were made tight
with melted paraffin. In the first trial, the mantle was filled
with sterile broth, which was then inoculated with a twenty-four
hours culture of B. typhosus. Next day the broth in the mantle
was extremely turbid ; filtration was then commenced, 10 c.c. of
broth being filtered into the flask every day and 10 c.c. of sterile
broth being put into the mantle. After filtration for three
weeks, the broth in the flask was found quite sterile. One loop-
ful of the broth from the container, however, contained a large
number of typhoid bacilli. Trials were then made with the
other candles ; sterile barrack sewage, sterile sewage effluents,
polluted ditch water and reservoir water, repeatedly inoculated
with large doses of B. typhosus, were filtered for three weeks,
but no traces of B. typhosus could be detected in the broth
placed in the Kitasato flasks. The failure to pass through the
filtering candles was not due to the absence of B. typhosus
from the fluid in the mantles, as in each experiment one loopful
of the fluid in the mantle at the end of the trial was found to
contain large numbers of living, vigorous, typhoid bacilli.

The following conclusions appear to be ju^tilied .

(1) The B. typhosus is not able to grow through the walls of a
Pasteur-Cham berland candle, and, if proper care be taken to pre-
vent 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.

(2) 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.
The large size of the lacunar spaces in the Berkefeld candle,

280 BACTERIOLOGICAL EXAMINATION OF WATER.

which cannot be avoided if a fair delivery is to be obtained,
appears to militate against the immobilising and devitalising
influences which operate so strongly in filters made with very
narrow lacunar spaces. It is well known that the typhoid
bacillus can grow out into long thread-like forms, and if
this takes place in the lacunar spaces of the Berkefeld candle
the bacillus may find its way into the cavity in the interior
of the candle. It is also probable that, owing to deficient
immobilisation, the bacilli are gradually washed through the
filtering candles. This often takes place when filters are worked
with a pump bearing directly on the bougie ; if, however, an air
cavity is introduced between the pump and the filtering candle,
the pressure is equalised, and this source of danger is removed.

(3) 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 in order to obtain
protection from water-borne disease, when these filters are
employed, it is necessary to sterilise the candles every third day.
This cannot be done by simply brushing the candles, they must
be sterilised by boiling water or by exposure to saturated steam.

METHODS OF TESTING WATER FILTERS.

It is clear from what has been said, that when examining
filters as to their powers of removing bacteria from water, it is
necessary to distinguish between the direct passage of micro-
organisms through the filter, due to some imperfection in the
filtering material itself or in its connection with the receptacle for
filtered water, and the indirect passage caused by the growth of
the micro-organisms through the walls of the filtering material.
The direct passage of bacteria may be determined in the case of
non-pressure filters by placing a rich emulsion of some special
organism, such as the B. prodigiosus, in the water to be filtered
and examining the filtrate for this organism by means of plate
cultures. If the special organism be found in the filtrate, this
may be caused by some flaw in the filtering material or defects
in the fitting of the material to the delivery pipe. If the
filtering material be in the form of a bougie it should be placed
in a vessel containing an emulsion of the special organism and
then be connected to a sterile partially exhausted flask. It is

ACTION AND UTILITY OF FILTERS. 281

important to feed the vessel containing the emulsion with water
so as to prevent the bougie from becoming dry. The filtrate
which collects in the exhausted flask is then examined in the
usual manner. In the case of filters which require considerable
pressure, the receptacle for the unfiltered water must be filled
with the special emulsion of a non -pathogenic organism, and
then connected to a service supply or to a tank raised sufficiently
to give the required head of pressure. When the delivery of
filtered water is very rapid, it is necessary to collect a considerable
amount of the filtrate, especially during the first few minutes of
working, as Gruber has shown that this is the critical time of the
experiment. I have found that under these conditions it is best
to pump several litres of the filtrate through a Pasteur-Cham ber-
land candle and diffuse the deposit in 5 c.c. of sterile water. The
whole of this mixture should then be plated out in gelatine,
gradually increasing amounts being used for each plate.

In order to test the indirect passage of micro-organisms
through the filter the plan already described under the experi-
ments made with the Berkefeld candles is very convenient. If,
however, there is no mantle supplied with the filtering candle
this must be connected to a partially exhausted sterile Kitasato
flask by means of stout rubber fitted with a pinch-cock. The
candle is then placed in a narrow-necked jar, containing either
an emulsion of B. typhosus or Spirillum Cholera? Asiaticse, in
sewage or polluted water. The filtering portion of the candle
only must be placed in the fluid, and the neck of the jar is then
plugged with sterile wool. By releasing the pinch-cock a few
c.c. are allowed to filter through the candle every day, taking
care to replace the amount of fluid filtered so that the surface
of the candle may not become dry. The filtrate in the flask
in the case of the B. typhosus should be mixed with a large
quantity of broth, and then incubated at 37° C. ; if a turbidity
occurs, loopfuls should be plated out and tested in the usual
manner. In the case of the Spirillum of cholera the filtrate
should be converted into 1 per cent, peptone and | per cent, salt
solution by adding a sufficient quantity of the stock peptone
and salt solution. The flask is then incubated at 37° C., and
loopfuls removed from the surface after ten, eighteen, and
twenty-four hours are examined in the usual manner.

CHAPTER XVI.

SUMMARY OP THE PROCEDURE RECOMMENDED

FOR THE BACTERIOLOGICAL EXAMINATION

OP WATER AND PREPARATION OP

NUTRIENT MEDIA.

A. Quantitative Analysis :

Shake a portion of the sample gently, but thoroughly, so as
to break down clumps of bacteria. Ascertain roughly the
probable number of micro-organisms present by means of a
cover-glass preparation, and if necessary dilute the specimen
with sterile tap-water. Make three gelatine plates with an
amount of the water which will not produce more than 300
colonies in each plate. As a rule, TV c.e.? J c.c., and | c.c. will
be found suitable amounts for the plates. Incubate the plates
at 22 C. for as long as possible. Count the colonies day by
day, and record the average number which develop per c.c.

B. Qualitative Analysis :

(1) If the specimen does not contain many bacteria, pump
one or two litres through a Berkefeld or Pasteur-Chamberland
candle, and diffuse the deposit in 5 to 10 c.c. of sterile water.
Treat this as follows :

(a) Add 1 c.c. of the concentrated water to 15 c.c. of sterile
whole milk, heat to 80° C. for ten minutes, cool, and then
incubate at 37' C. in a Buchner tube. Note if any changes
characteristic of B. enteritidis sporogenes are produced within
twenty-four to thirty-six hours.

(b) Plate in carbolic acid gelatine (0*05 per cent, carbolic-
acid) from O'l to 1 c.c. of the concentrated water.

(c) Add from O'l to 1 c.c. of the concentrated water to broth
tubes containing 0*05 per cent, carbolic acid, and incubate the
tubes at 37 C. ; or add the same amounts of the concentrated

PROCEDURE FOR EXAMINATION OF WATER. 283

water to glucose-formate broth tubes, and then incubate in
Buchner's tubes at 42° C. (Fakes' method). Plate out all the
tubes which show any turbidity after twenty-four hours
incubation in gelatine and on alkaline glucose-litmus-agar.

(d) Examine all the plates for B. coli and other sewage
organisms.

Sub-culture the coliform colonies in :

Glucose-gelatine shake, for gas formation.

Milk, for acidity and coagulation of casein.

Peptone (Witte) water, for the indol reaction.

Agar-slope, for agglutination test.

If B. typhosus be suspected, all the tests given under the
description of this organism must be applied.

(2) If the specimen of water contains a very large number
of micro-organisms, it is useless to plate out the concentrated
sample directly in carbol-gelatine.

The best plan is to cultivate the concentrated water in amounts
varying from 0*1 to 1 c.c. in glucose-formate broth, according
to Pakes' method, and in carbolic acid broth containing not
more than 1*0 per cent, carbolic acid. The growths which
result must be plated out on alkaline glucose-litmus-agar.

(3) If the water is suspected to contain the cholera spirillum,
to 90 c.c. of the sample add 10 c.c. of the stock peptone
solution (10 per cent, peptone, 20 per cent, gelatine, and 5 per
<,'ent. salt) and cultivate the mixture at 37° C. After twelve
hours incubation remove loopfuls from the surface, and examine
in a hanging-drop for spirilla ; if these are not in pure culture
inoculate a loopful into a second flask, containing 1 per cent,
peptone and | per cent. salt. Plate out loopfuls from the
surface of the pure culture in gelatine and on agar. Then
inoculate peptone tubes for the indol reaction, and stab-
gelatined for the typical liquefaction ; also plant out a loopful
on an agar-slope, and employ the growth, which results in
twenty-four hours, for Pfeitfer's reaction and the agglutination
test.

284 BACTERIOLOGICAL EXAMINATION OF WATER

PREPARATION OF CULTURE MEDIA.

Nutrient Broth (Peptone Bouillon). — One pound of beef free
from fat must be finely minced, then infused in a litre of cold
distilled water and allowed to stand in a cold place for twenty-
four hours. The whole mass is then strained through a cloth
and distilled water added to the filtrate so as to make up the
volume of fluid to one litre. Ten grammes of peptone and five
grammes of common salt are now added to the litre of fluid,
which is then boiled in the steam steriliser for one hour. The
medium is next made slightly alkaline with a 1 percent, solution
of sodium carbonate, litmus being used as the indicator, and
again boiled in the steamer for half an hour. Coagulated
material is then removed by filtration through moistened
" Chardin " paper. The medium is finally sterilised in the
steamer on three successive days for ten to fifteen minutes each
day. Sometimes it is impossible to obtain good beef; Liebig's
extract may then be employed with advantage. The broth
obtained is perfectly reliable for ordinary purposes, and though
perhaps not so nourishing as that made with the best beef, it
has the advantage of being more uniform in quality. Five
grammes of Liebig's extract are added to a litre of tap-water
containing ten grammes of Witters peptone and five grammes
of common salt. The mixture is then boiled in the steam
steriliser or over a Bunsen flame, made faintly alkaline with
sodium carbonate and filtered into a sterile flask. The flask is-
placed in the steam steriliser for three-quarters of an hour and
coagulated material removed by filtration through moistened
" Chardin " paper. The medium may then be used to prepare
nutrient gelatine or agar-agar ; if not required for this purpose,
it is poured into a series of sterile test-tubes, 10 c.c. in each tube,,
which are then sterilised on three successive days.

Another method which is largely used in some laboratories is-
as follows : Take one pound of beef free irom fat, mince finely,
and diffuse in a litre of tap-wrater, boil for twenty minutes and
allow to stand all night in an ice chest. Next day strain the
fluid, add ten grammes of Witters peptone, five grammes of
salt, and make up the bulk to one litre, dissolve by heat and
then carefully neutralise with a 28'6 per cent, solution of sodium

PREPARATION OF CULTURE MEDIA. 285

carbonate. Next sterilise the fluid in the autoclave at 115° C.
for fifteen minutes, and then filter through moistened " Chardin"
paper. Place in tubes or a sterile flask and again sterilise at
115° C. for half an hour.

Nutrient Gelatine. — To broth prepared as above add 15 per
cent. " gold label " gelatine, and place in the steam steriliser
until all the gelatine is dissolved. Then make faintly alkaline
with the sodium carbonate solution, cool to 50° C., and add the
white of an egg dissolved in 50 c.c. of distilled water. Boil for
an hour or until the fluid is clear, and then filter through
moistened " Chard in " paper placed in a hot-water filter.
Divide the filtrate into sterile tubes, 10 c.c. in each, and sterilise
in the steam steriliser at 100° C. on three successive- days for
half an hour each day.

Agar-Agar. — Broth prepared as above is placed in a flask
and rendered distinctly alkaline. Two per cent, of agar power
is then added to the alkaline fluid ; if agar is added to an
acid medium it is converted into galactose. The flask is heated
in the steam steriliser until the agar is dissolved ; the contents
are then cooled to 50° C., and the white of an egg dissolved in
50 c.c of distilled water added. The flask is now heated in
the autoclave at 115° C. for three-quarters of an hour, and the
contents are then filtered through moistened " Chardin " paper
placed in a hot-water funnel. The filtrate is divided amongst
sterile test-tubes, 7 c.c. in each, which are then sterilised in
the autoclave at 115° C. for half an hour.

Glucose-broth, glucose-gelatine , and glucose- agar are made by
adding 2 per cent, of glucose to each of the media prepared as
above, but before sterilisation has been effected. Sugar media
should not be sterilised at a temperature above 100° C.

Cane-sugar-gelatine and lactose-gelatine are prepared in the
same manner as the glucose media.

Peptone Solution. — Witters peptone, 1 percent., and common
salt 0*5 per cent., are dissolved in water with the aid of
heat (as a rule the fluid does not require the addition of an
alkali, the peptone being sufficiently alkaline). The mixture
is then boiled, filtered, and divided amongst sterile tubes,
which are then sterilised in the autoclave at 115° C. for twenty
minutes.

286' BACTERIOLOGICAL EXAMINATION OF WATER.

Potato. — A large potato is well washed and scrubbed with a
brush. A cylinder is then bored from its interior and cut
obliquely. The brown peel is cut oft' from the ends, and the
wedges so obtained are allowed to soak overnight in distilled
water so as to get rid of the excess of starch. Each wedge is
then placed in a potato tube previously fitted with a pad of
wool at the bottom and filled for about an inch in depth with
distilled water. The tubes are plugged with cotton wool
and then sterilised in Koch's steamer for one hour. Before
use the potato tube should be incubated at 37° C. for twenty-
four hours, in order to make sure that sterilisation has been
affected. Sometimes it is necessary to autoclave the medium,
but this should be avoided if possible. If sterilisation for
one hour at 100° C. renders the potato too soft, discontinuous
sterilisation for twenty minutes on three successive days should
be practised.

Milk. — Fresh milk is steamed for fifteen minutes in the Koch's
steriliser, and placed in a cool place over-night to facilitate
separation of the cream. The milk is then siphoned off from
beneath the cream and placed in sterile test-tubes. The tubes
are plugged and sterilisation effected by steaming at 100° C. for
twenty minutes on three successive days. For cultivation of the
B. enteritidis sporogenes whole milk should be placed in sterile
tubes and then sterilised as above.

Litmus-whey. — This medium was suggested by Petruschky
for the study of the amount of acid produced by the fermenta-
tion of lactose. Fresh milk is warmed, and the casein coagu-
lated by means of a little hydrochloric acid. The separated
casein is then filtered off' and the clear whey neutralised with
dilute sodium hydrate solution. The ffuid is then steamed
for two hours ; a little acid albumen usually separates and is
filtered off*. The filtrate should now be clear, colourless, and
neutral in reaction. To the filtrate 5 per cent, of a saturated
alcoholic solution of litmus is added, and the medium placed in
tubes and sterilised. For accurate work it is better to add
neutral litmus to the neutral whey, and then sterilise as usual.

Lactose Litmus Solution. — This medium is made as follows :
Dissolve two grammes of lactose in 40 c.c. of broth and add
water to make up to 100 c.c. Carefully neutralise and boil in

PREPARATION OF CULTURE MEDIA. 287

the steamer for twenty minutes and filter. Then add neutral
litmus so as to produce a reddish-purple colour. Divide into
test-tubes and sterilise in the steamer on three successive days.

Neutral Litmus. — This solution is prepared by adding 300 c.c.
of rectified spirit to two ounces of commercial litmus. After
standing for a fortnight the spirit is poured off and a second
300 c.c. added, and allowed to stand for a second fortnight.
The alcohol is then poured off, and all traces of the spirit
removed by exhausting the flask with an air-pump. The litmus.
is now added to 800 c.c. of water placed in a flask and allowed
to stand for twenty-four hours. The watery extract so obtained
is filtered into a flask, and a few drops of pure sulphuric acid
added until the solution has an acid reaction. Baryta solution
is then added in excess, and the fluid left until it has an alkaline
reaction. The solution is then filtered, and carbon-dioxide
passed through it until an acid reaction is again obtained. The
Barium carbonate is filtered off, and the solution placed in a
test-tube. Sterilisation is effected in steam steriliser on three
successive days.

Potato-gelatine. — Crush 500 grammes of peeled potato and
macerate in a litre of water for three or four hours. Sift and
allow to stand for twenty-four hours, then decant, and make up
the volume to 1000 c.c. Dissolve 150 to 200 grammes of
gelatine by heat, and boil for a few minutes. Then add a solu-
tion of soda until the reaction is feebly but distinctly acid.
Heat to 115° C. for five minutes, and filter through " Chardin "
paper placed in a hot funnel. Divide into tubes and sterilise at
112°-115° C. Just before use Eisner recommends the addi-
tion of 1 per cent, of potassium iodide to the gelatine previously
melted in the " plate bath/"

Glucose-formate-broth. — For the study of bacteria under
anaerobic conditions Kitasato recommended the addition of 0*4*
gramme of sodium formate and two grammes of grape sugar to
ordinary broth. Fakes'* medium for the isolation of B. coli and
B. typhosus is made as follows : " To ordinary meat iniusion
are added 1 per cent, peptone, 0*5 per cent, sodium chloride, 2
per cent, glucose, and 0'4 per cent, sodium formate. When
these have all been dissolved by heating the medium is neutral-
ised, and after neutralisation 4 c.c. of normal soda solution per

288 BACTERIOLOGICAL EXAMINATION OF WATER.

litre are added. The broth is then boiled in the steam steriliser
for twenty minutes, filtered, poured into the test-tubes, and
sterilised for twenty minutes on the day it is made and the two
succeeding days.'"

Nitrate-broth. — This medium is prepared by adding 0*1 per
cent, of potassium nitrate to nutrient broth, diluted with sterile
water, so that the medium contains only 5 per cent, broth, viz.,
KNO3 O'l gramme, broth 5 c.c., sterile water 95 c.c. If pure
broth is used the medium turns almost black on adding Nessler
solution, so that it is impossible accurately to determine the
presence of ammonia.

BIBLIOGRAPHY.

ADAMETZ, "Die Bakterien der Trink und Nutzwasser."

BECO, "Value of the agglutination test as a means of diagnosis between \
the B. typhosus and Coliform races," Annales de Vlnstitut Pasteur,
1899. BESSON, " Technique micro-biologique," Paris, 1898. BLEISCH,
" Ueber f ehler-quellen bei cholera-roth-reaction," Zeitscli. f. Hygiene,
vol. xiv. p. 103. BIBERSTEIN, " Beitrage zur sero-diagnostik des
abdominal typhus," Zeitscli. f. Hygiene, vol. xxvii., 1898. BABES,
"Variability and varieties of B. typhosus," Zeitscli. f. Hygiene,
vol. xvii. BORDET, Annales de Vlnstitut Pasteur, September 1899.
BLACHSTEIN, " Contribution a 1'etude microbique de 1'eau," Annales de
VInstitut Pasteur, 1883, p. 689. BUJWID, "Die Bakterien in Hagel-
Kornern," CentralUatt f. Bald, vol. iii., 1888. BOLTON, " Ueber das
Yerhalten verschiedener, Bakterien im Trinkwasser," Zeitsch. f.
Hygiene, vol. i. BUCHXER, Archiv.f. Hygiene, vol. xvii.

CASSEDEBAT, Annales de VInstitut Pasteur, 1890. CUNNINGHAM, Report* •
of the Sanitary Commissioner -with the Government of India. CRAMEU,
" Die wasserversorgung von Zurich." CLOWES and HOUSTON, " On
the bacterial treatment of sewage," Reports to the London County
Council.

DURHAM, " On the serum diagnosis of typhoid fever, with special reference v
to the bacillus of Gartner and its allies," Lancet, Jan. 15, 1898 ; also
"On a special action of the serum of highly immunised animals,"
Journal of Pathology and Bacteriology, 1897, vol. iv. DUNBAR, "Zur
differential diagnose zwischen den choleravibrionen und anderen
vibrionen," Zeitscli. f. Hygiene, vol. xxi. p. 298. DIEUDONNE, Arbeiten
aus dem Kaiser/. Gesundheitsamt, 1894. DUCLAUX, Comptes Rendus
1885.

EISENBERG, Bakteriologische Diagnostik, 1891. ELSNER, " Untersuchungen
iiber electives wachsthum der bacterium coli-arten und des typhus
bacillus," Zeitscli. f. Hygiene, vol. xxi. p. 25.

FRANKEL, " Ueber den bakteriengehalt des eisens," Zeitsch. f. Hygiene,
vol. i. 1886. FREUDENREICH, "De Fantagonisme des bacteries."
Annales de VInstitut Pasteur, 1888 ; also " Ueber du Durchlassigkeit
der Chamberland'schen filter fur bakterien," Centralb.f. Bald, vol. xii.
p. 240. FRANKLAND, "Micro-organisms in water," London, 1894.
FRANKEL und C. PIEFKE, " Versuche iiber die Leistungen der sand-
filtration," Zeitsch. f. Hygiene, vol. viii., 1890, p. 1. FULLER,

T

2,90 BACTERIOLOGICAL EXAMINATION OF WATER.

"Differentiation of the bacillus of typhoid fever," Report of State

Board of Health, Massachusetts, 1891. FISCHER and FLATAU>

" Typhus bacillen in wasserprobe," Centralblatt f. Bald., vol. xxix.

p. 329. FODOR and RIGLER, Centralb.f. Bakt., vol. xxiii. p. 930.

GARRE, "Ueber antagonisten under den bakterien," Correspondetizblatt

f. Schweizer aertze, 1887. GILBERT and LION, " Contribution a 1'etude

des bacteries intestinales," Semaine Medicale, 1893. GORDON, " Some

varieties of B. coli coramunis," Journal of Pathology and Bacteriology,

vol. iv., 1897. GRUBER, " Gesichtspunkte fiir die Priifung und

Beurteilung von Wasserfiltern," Central]). /. Bakt., vol. xiv. p. 488.

GLOBIG, "Ueber Bakterien wachsthum bei 50° bis 70°," Zeitsch f.

Hygiene, vol. iii. 1888. GENERSICH, Centralb.f. Bakt., 1900, p. 241.

HANKIX, "On the detection of the B. typhi abdominalis in water and

other substances," Centralb. f. Bakt., vol. xxvi., 1898. HILBKRT,

" On the value of Hankin's method for the detection of the B. typhosus

in water," Centralb. f. Bakt., vol. xxvii., 1899. HOLZ, " Untersuch-

ungen iiber den Nachweis der Typhus bacillen," Zeitsch.f. Hygiene,

vol. viii. p. 143. HOUSTON, " B. typhosus simulans," Centralb. f.

Bakt., No. 14, 1898 ; also " The bacterioscopic examination of

drinking water," " The chemical and bacteriological f xamination of

the washings of soils," " On the inoculation of soils with particular

microbes," Supplement to the twenty-eighth Report of the Local

Government Board, 1898-1899. HORROCKS, "On the behaviour of

,the B. typhosus in sewage," Journal of the Sanitary Institute, 1899 :

/ also " The value of the agglutination test as a means of diagnosis

"\ between the B. typhosus and coliform organisms," British, Medical

^Journal, vol. i., 1900. HER^US, Zeitsch.f. Hygiene, vol. i., 1886.

HOCHSTETTER, Arbeiten aus dem Kaiserl. Gesundheitsante, vol. ii.

JANOWSKI, "Zur Biologic der Typhus Bacillus," Centralb. f. Bakt.,

vol.xviii., 1895. JORDAN, " Bacteria observed in sewage," Massachusetts

/State Board of Health : Purification of Water and Sewage, vol. ii.

/ JATTA, "Experimental investigations on the agglutination of the

\ B. typhosus and micro-organisms of the coli group," Zeitsch.f. Hygiene,

vol. xxiii., 1900.

KIRCHNER, " Untersuchungen iiber die Brauchbarkeit der Berkefeld
Filter," Zeitschrift f. Hygiene, vol. xiv. p. 299. KITASATO, " Ueber
das verhalten der cholera-bacterien," Zeitsch. f. Hygiene, vol. vi.,
1889 ; also " Ueber das verhalten der Typhus und Cholera bacillen
zu satire oder alkalihalten nahrboden," Zeitsch. f. Hygiene, vol. iii.,
1888. KOCH, "Der augenblichliche stand der cholera diagnose,"
Zeitsch.f. Hygiene, vol. xiv. KUBLER, " Uber die Brauchbarkeit der
Filter sans pression, systeme Chamberland-Pasteur," Zeitsch.f. H>/</;<',/<>.
vol. viii. KLIE, " Investigations on the growth of B. typhosus and
B. coli in nutritive media containing different percentages of gelatine
and at different temperatures," Centralb.f. Bakt.,vol.xx., 1896. KLEIN,
" Micro-organisms and disease " ; also " B. enteritidis sporogenes,''

BIBLIOGRAPHY. 291

Supplement to the Local Government Board Report, 1897-98 ; also
" Report on the fate of the pathogenic and other infective microbes
in the dead animal body," Supplement to the Local Government Board
Report for 1899-1900 ; also " On modifications of the cholera vibrio
artificially induced," Supplement to the Local Government Board Report
for 1896-97. KUBLER and NEUFELD, Zeitsch. f. Hygiene, vol. xxxi.,
1899.

LEE SMITH, " Coli bacilli in the stools of suckling children," Centralb.f.
Bald. No. 20, 1899. LEVY and BRUNS, " Zur Hygiene des Wassers,"
Archiv. f. Hygiene, vol. xxxvi. LORRAIN SMITH, "A study of the
epidemic of typhoid fever in Belfast." British Medical Journal, 1899.
LAWS and ANDREWES, " Report on the micro-organisms of sewage,"
London County Council, 1894. LUSTIG, " Diagnostik der Bakterien
des Wassers, Jena, 1893.

MASCHEK, " Bakt. Untersuchungen derZeitmeritzer Trinkwasser. MIQUEL,
''Manuel pratique d'analyse bacteriologique des eaux," Paris, 1891.
MAYER, "On Piorkowski's method for the diagnosis of the B.typhosus,"
Centralb.f. Bakt. vol. xxviii., 1900. MARTIN, "On the growth of
the typhoid bacillus in soil," Supplement to Local Government Board
Reports, 1898, 1899, and 1899-19CO. MACFADYEN and BLAXALL,
•k Thermophylic bacteria," Journal of Pathology and Bacteriology,
1894 ; also Transactions of the Jenner Institute, 1899. METCHNIKOFF,
" Recherches sur le cholera et les vibrious," Annales de Vlnstitut
Pasteur, 1893.

NICATI and REITSCH, "Experiences sur la vitalite du bacille virgule
cholerigene," Revue tf Hygiene, 1885.

PARES, "Bacteriology and Public Health," Public Health, 1900. PET-
RUSCHKY, "B. faecalis alkaligenes," Centralb. f. Bakt., No. 6/7,
vol. xix. p. 187. PIEFKE, " Aphorismen ueber wasserversorgung,"
Zeitsch. /. Hygiene, vol. viii., 1890. R. PFEIFFER, "Differential
diagnose der Yibrionen der Cholera Asiatica," Zeitsch. f. Hygiene,
1895. PROSKAUER, Zeitsch. f. Hygiene, vol. ix., 1890. PROSKAUER
and CAPALDI, Zeitcsh.f. Hygiene, vol. xxiii., 1896.

RAVITSCH-STCHERBA, " Nachweiss der typhus bacillen im wasser und
faces," Hygienische Rundschaw, 1893. BEMLINGER and SCHNEIDER
" Contributions to the study of the B. typhosus," Annales Vlnstitut
Pasteur, January 1897. REIDEL, "Die Yermehrung der Bakterien
im Wasser," Arbeiten aus dem Kaiserlichen Gesundheitsamte, vol. i.
REMY, "Contribution to the study of typhoid fever and its bacillus,"
Annales de Vlnstitut Pasteur, 1900. RODET, "On the agglutination
of the bacillus of Eberth and the B. coli by the serum of
immunised animals," Journal of Physiology and General Pathology,
January 1900. RABINOWITSCH, "Ueber die Thermophylic bacteria,"
Zeitsch. f. Hygiene, vol. xx., 1895. Roux, "Precis d' Analyse Micro-
biologique des eaux," Paris, 1892.
SANARELLI, "Les vibrious des eaux et 1'etiologie du cholera," Aimales de

292 BACTERIOLOGICAL EXAMINATION OF WATER.

rinstitut Pasteur, 1893. C. STERNBERG, " On the value of agglutina-
tion for the diagnosis of the B. typhosus," Zeitsch. f. Hygiene,
vol. xxxiv., 1900. SCHOFER, " Ueber das verhalten von pathogenen
keimen in kleinfiltern," Centralblatt f. Bakt., vol. xiv., 1893.
SIROTININ, Zeitsch. f. Hygiene, vol. iv., 1888. SCHEFFLER, Centralblatt
f. Bakt., vol. xxviii. p. 199.

TIEM ANN-GARTNER, uDie chemische und mikroskopisch-bacteriologische
Untersuchungen des Wassers," Braunschweig, 1895. TRENKMANN,
" On the biology of the comma bacillus," Centralblatt f. Bakt., vol. xiii.,
1893. TILS, " Bacteriol untersuchung der Freiburger Leitungs
wasser," Zeitsch. f. Hygiene, vol. ix., 1890.

UPPELMANN, "Weitere Beitrage zur biologic der cholera bacillus,"
Berliner Klinische Wochenschrift, 1893.

WARD, "Action of light upon bacteria," Proc. Roy. Soc. Loud., 1893.
WATHELET, " Bacteriological researches on the dejecta in typhoid
fever, Annales de rinstitut Pasteur, 1895. WEISSENFELD, " Ueber
hygienische beurtheilung des Wassers," Zeitsch. f. Hygiene, 1900.
WITTICH, Centralb. f. Bakt., vol. xxvii., 1900. WEIBEL, "Untersuch-
ungen iiber vibrionen," Centralb. f. Bakt., vol. iv., 1888. WOLFFHUGEL,
" Wasser versorgung," Handbuch der Hygiene, Leipzig, 1882. WOOD-
HEAD and WOOD, " Bacterial efficiency of water niters," Brit. Ned.
Journ., January 1898.

ZIMMERMANN, " Die Bakterien unserer Trink und Nutzwasser," Chemnitz,
1890.

INDEX.

ABBA, method of isolating B. coli, 102

Acidi lactici, B., 129

Action of electricity on micro-organisms, 24

Agar-agar, preparation of, 285

Agilis, M., 74

Agglutination test, 167

„ „ general conclusions, 180

„ „ methods for the performance of, 180

Albus, B., 71
Amethystinus, B., 69

„ mobilis, B., 69

Analysis, quantitative bacteriological, 6
Aquatilis communis, B., 55
„ radiatus, B., 56
„ sulcatus, B., 50

B. (Weichselbaum), 49
B., 63

B. (Frankland), 63
M., 75

„ V., 244
Arborescens, B., 61
Aurantiacus, B., 65
Aureus, B., 65
Aureus, V., 78
Aurora, M., 141

BACTEEIAL contents of ice, 26
hail, 27

„ „ lakes, 28

„ „ rain, 27

„ „ rivers, 27

snow, 26

„ „ sources of supply, 26

springs, 30

„ „ wells, 29

Beco, agglutination experiments, 171
Berolinensis, V., 245

Bleisch, solution for cholera-red reaction, 240
Bolton, experiments on the sedimentation of micro-organisms, 18
Breslaviensis, B., 194
Broth, preparation of, 284
Brunneus, B., 72

294 INDEX.

Buchner, experiments on the action of light on B. typhosus, &c., 15
Butyricus, B. (Botkin), 112

CADAVEBIS, B., Ill

Candicans, M., 77

Capillareus, B., 150

Carneus, B., 59

Cavicida, B., 96

Cholerse asiaticse, Sp., 236

„ „ „ diagnosis of, from allied forms, 242

„ „ „ effect of acids and alkalies on, 264

„ „ „ methods of isolation from water, 253

„ „ ,, relation to other micro-organisms, 260

„ „ „ vitality of, in water, sewage, &c., 254

Choleroides (a), B., 248
(0),B.,249

Chrysogloia, B., 66

Cladothrix dichotoma. 133

Classification of water bacteria, 42

Cloacae, B., 135

,, fluorescens, B., 142

Cceruleub, B. (Smith), 69
„ B. (Voges), 70

Coli communis, B., 88

„ „ „ Gordon's classification of, 92

„ „ „ methods of isolating from water, 98

„ „ „ value of, as a sign of sewage contamination. 103

Collection of water, 4

Coloides rnbescem, B., 98
„ virescens, B., 97

Concentricum, Sp., 80

DANUBICUS, V., 246
Deeleman, varieties of B. coli, 97
Delicatulus, B., 139
Deneke, Sp., 244
Denitrificans I., B., 81
II., B., 86

„ B. (Jensen), 87

Denitrifying micro-organisms, 81
Devorans, B., 56

Downes and Blunt, experiments on the action of light on bacteria, 14
Duration of life of micro-organisms, 21
Durham, agglutination experiments with cholera vibrios, 239

ELECTRICITY, action of on micro-organisms, 24
Enteritidis sporogenes B. (Klein), 108

„ B. (Gartner}, 193

Erythrosporus, B., 131
Eisner and Holz, methods of isolating B. coli and B. typhosus, 101

INDEX. 295

aligenes, B. (Petruschky), 192
„ ,, B. (Deeleman), 98

Fervidosus, M., 75
Filters, methods of testing, 280
Finkler-Prior, Sp., 242
Fischer and Flatau — Isolation of B. typhosus from a well at Kellin-

gen, 224

Flavus desidens, M., 76
„ liquefaciens, M., 74
„ tardigradus, M., 75
Fluorescens aureus, B., 47
,, crassus, B., 47

,, liquefaciens, B., 42

„ longus, B., 45

„ non-liquefaciens, B., 44

putridus, B., 130
„ stercoralis, B., 143

tenuis, B., 46
Frankland and Ward, experiments on the action of light on anthrax

spores, 16
Frankland, experiments on nitrification, 83

„ ,, „ sedimentation of bacteria, 20

Freudenreich, experiments with Pasteur-Chamberland filter, 266

„ „ on influence of pathogenic and non-patho-

genic bacteria on B. typhosus, 185
Friedebergensis, B., 194
Frondosus, B., 146
Fuscus, B., 73

„ hydrophilus. B., 73
„ limbatus, B., 73
Fusiformis, B., 147

GARRE, antagonism of micro-organisms, 14, 60

Gelatine, preparation of, 285

Genersich, isolation of B. typhosus from water at Pecs, 223

Gilbert and Lion, varieties of B. coli, 91

Gindha, V., 246

Glucose-formate-broth, preparation of, 287

Gordon, table of B. coli, 92, 93

HAIL, bacterial contents of, 138

Hankin, methoJ of isolating B. typhosus from w.tter, 199

Helvolus, B., 65

Heraeus, experiments on nitrification, 81

Hilbert, investigation of Hankin's method, 202

Hochstetter, action of caroonic acid gas on bacteria, 21, 258

Houston, sewage bacteria, 145

,, B. typhosus simulans varieties, 190

„ sewage streptococci, 120
Hyalinus, B., 138
Hydrophylus fuscus, B., 73

296 INDEX.

ICE, bacterial contents of, 26

Indigoferus, B., 70

Indigonaceus, B., 70

Influence of light on micro-organisms, 14

„ ,, chemical composition of a water on the number of micro-
organisms, 22

,, „ movement and rest on micro-organism?, 17
Tvanoff, V., 246

JANOWSKI, influence of light on the B. typhosus, 15

Janthinus, B., 67

Jatta, agglutination experiments, 172

KIRCHNEK, experiments with Berkefeld filter, 267
Kitasato, influence of acids and alkalies on Sp. choleraa, 263

„ action of pathogenic and non-pathogenic bacteria on Sp.

choleras, 260

„ influence of acids and alkalies on B. typhosus, 187
Klein, experiments on life of B. typhosus in water and sewage, 210
,, B. enteritidis sporogenes, 108

,, experiments on artificial modification of cholera vibrios, 241
„ outbreak of typhoid fever at Worthing, 221

Klein and Houston, value of B. coli and B. enteritidis sporogenes, 113
Ktibler, experiments with Pasteur-Chamberland filter, 266
Kiibler and Neufeld, isolation of B. typhosus from a well water, 222

LACTERICIUS, B., 60
Lactis aerogenes, B., 95

,, cyanogenus, B., 128

„ erythrogenes, B., 59
Lakes, bacterial contents of, 28

Laws and Andrewes, duration of life of B. typhosus in sewage, 210
Liodermus, B., 52
Liquefaciens, B., 54
Litmus-whey, preparation of, 286

„ neutral, preparation of, 287
Lividus. B., 68

Lunt, experiments with Berkefeld filter, 271
Luteum, B., 67
Luteus, M., 75

MACFADYKN and Blaxall, therrnophylic bacteria, 156

Malignant oedema, B. of, 111

Massachusetts State Board of Health, diagnosis of B. coli, 101

,, „ „ experiments on filtration through

sand, 34
,, ,, methods for isolation of B. coli,

100
,, „ „ presence of B. coli in filtered

water, 89

Massovvah, V., 245
Meg;iterium, B., 128
Meuibraneus patulus-, B., 149

INDEX. 297

Mesentericus, B. (sewage varieties), 151-2

„ fuscus, B., 51

„ ruber, B., 52

„ vulgatus, B., 51

Metchnikoff, method of isolating Sp. choleras from water, 253
Metchnikovi, V., 243

Methods for isolation of Sp. cholerse from water, 253
Methods of testing water filters, 280
Milk, preparation of, 286
Miquel, multiplication of bacteria in water, 12

„ B. thermophylus, 156
Multiplication of micro-organisms in water, 12
Mycoides, B., 53

NEAPOLITANUS, B., 97
Nitrate-broth, preparation of, 288
Nitrifying organisms, 81
Nitrobacter, 85
Nitrosococcus, 85
Nitrosomonas, 85
Nubilus, B.. 62

OCHRACEUS, B., 64

Orange-red water, B., 67

FAKES, method of isolating B. coli from water, 99

Parietti, method of isolating B. coli and B. typhosus fiom water, 99

Pasteur-Chamberland and Berkefeld filters, mode of action and utility

of, 265

Peptone solution, preparation of, 285
Pfeiffer's reaction, 184, 237
Phosphorescens, V., 246

Piorkowski, method of isolating B. typhosus from water, 203
Potato-gelatine, preparation of, 287
Potato, preparation of, 285
Preparation of media, 7, 284

,, water-plates, 9

Prodigiosus, B., 57

Proskauer and Capaldi, media as a test for B. typhosus, 166
Proskauer, chemical composition of water in relation to bacteria, 22
Proteus cloacinus, 140

mirabilis, 126
„ vulgaris, 125

Zenkeri, 126

Zopfi, 127
Psittacosis, B., 196
Punctatus, B., 55
Pyocyaneus, B., 132

QUALITATIVE bacteriological analysis, 41
Quantitative bacteriological analysis, 6

298 INDEX.

RABINOWITSCH, thermophylic bacteria, 157

Radiatus, M., 76

Rain, bacterial contents of, 27

Relation of cholera vibrios to other bacteria, 260

„ quantitative analysis to filtration through sand, 33

Remlinger and Schneider, isolation of B. typhosus from water, 221
Remy, method of isolating B. typhosus from water, 207

„ on antagonism of B. coli and B. typhosus, 216
Rivers, bacterial contents of, 27
Rubefaciens, B., 60
Ruber aquatilis, B., 59

„ Balticus, B., 59

„ Berolinensis, B., 59

„ Indicus, B., 58
Rubescens, B., 60, 137
Rubidus, B., 61
Rubrum, Sp., 79
Rugula, Sp., 79

SACQUEPEE, varieties of Eberthiform bacilli, 176
Sanarelli, method of isolating Sp. choleree from water, 253

„ vibrios isolated by, 249
Sarcina alba, 78

,, aurantiaca, 78
„ lutea, 78

Schofer, experiments with Berkefeld filter, 269
Sedimentation of micro-organisms, 18
Serpens, Sp., 80
Sewage streptococcus (Houston), 122

„ proteus (Houston), 145

Sims Woodhead and Wood, experiments with filters, 270
Snow, bacterial contents of, 26
Springs, bacterial contents of, 30

Spirillum Cholerse Asiaticse, behaviour in acid and alkaline media, 263
,, ,, ,, characteristics of, 236

•» » M diagnosis of, from allied forms, 242

.» 5J ,, relation to other bacteria, 260

;> » )5 vitality of, in water, etc., 254

Deneke, 244
Finkler-Prior, 242
„ Metchnikovi, 243
Sternberg, C., coliform organisms, 197
Stoddart, method of i>olating B. typhosus from water, 199
Streptococci, value as a sign of sewage contamination, 124

„ varieties found in sewage, 116

Streptococcus albus, 123

„ coli gracilis, 123

„ mirabili?, 122

„ vermiformis, 123

Subflavus, B., 66
Subtilis, B., 52

„ B. (sewage varieties), 154, 155

INDEX. 299

Subtilissimus, B., 148

Sulcatus liquefaciens, B., 50

Summary of procedure for bacteriological examination of water, 282

Superficial!*, B., 136

Symptomatic anthrax, B. of, 112

TENUB, Sp., 80
Therrnopbylic bacteria, 155

„ ,, (Macfadyen and Blaxall), 158

„ ,, (Rabinowitsch), 157

Thermophylus, B. (Miquel), 156
Tholoeideum, B., 96

Tiemann and Gartner, experiments on the sedimentation of bacteria, 19
Tremelloides, B., 64

Trenkmann, experiments on vitality of Sp. cholera in saline media, 256
Typhosus, B., cases in which isolated from water supplies, 220
„ ,, characteristics of, 162
„ „ methods for isolation of, from water, 197

„ ,, micro-organisms which resemble, 188

„ reaction of, with anti-typhoid serum, 166
„ „ relation to acids and alkalies in nutrient media, 187

„ „ relation to other pathogenic and non-pathogenic bacteria,

185

„ „ researches on the duration of life of, 209

„ ,, simulans (Houston), 190

UBIQUITUS, B., 71
Umbilicatus, B., 72
Ureaa, B., 130
Urese, M., 133
„ liquefaciens, M., 133

VAN DE VELDE, agglutination experiments, 169
Yersicolor, M., 77
Vesicle, B., 98
Vibrio aquatilis, 244
„ aureus. 78
„ Berolinensis, 245

Danubicus, 246
„ flavescens, 79
„ flavus, 79
„ Gindha, 246
Ivanoff, 246
,, Massowah, 245
,, phosphorescent, 246
Violaceus, B., 67

„ Berolinensis, B., 68
„ Laurentius, B., 68
„ Lutetiensis, B., 69
Yitality of Sp. choleras in water, 254

'300 INDEX.

Viticulosus, M., 77
Volutans, Sp.; 80

WELLS, bacterial contents of, 29
Widal, procedures for agglutination test, 180
Winogradsky, experiments on nitrification, 83
Wright, method of performing agglutination test, 181

YELLOW bacillus, 64

Printed by BAI.I.ANTVNK, HANSON & Co.
London &•» Edinburgh

No. 3.

London, 7, Great Marlborough Street ',
May, 1901.

A SELECTION

FROM

J. & A. CHURCHILL'S CATALOGUE,

COMPRISING

MOST OF THE RECENT WORKS PUBLISHED BY THEM.

N.B.—J. 6° A. Churchill's larger Catalogue, which contains over 600 works, with
plete Index to their Subjects, will be sent on application.

Com-

Human Anatomy :

A Treatise by various Authors. Edited
by HENRY MORRIS, M. A., M.B. Lond.,
F.R.C.S., Senior Surgeon to, and
Lecturer on Surgery at, the Middlesex
Hospital. Second Edition. Roy. 8vo, with
790 Illustrations, nearly all original, and
many of them in several colours, 365.

Heath's Practical Anatomy:

A Manual of Dissections. Eighth
Edition. Edited by WILLIAM ANDERSON,
F.R.C.S., Surgeon and Lecturer on
Anatomy at St. Thomas's Hospital,
Examiner in Anatomy for R.C.P. and S.
Crown 8vo, with 329 Engravings, 153.

Wilson's Anatomist's Vade-

Mecum. Eleventh Edition. By HENRY
E. CLARK, M.R.C.S. Eng., F.F.P.S.
Glasg., Examiner in Anatomy, F.P.S.,
and Professor of Surgery in St. Mungo's
College, Glasgow. Crown 8vo, with 492
Engravings and 26 Coloured Plates, i8s.

An Atlas of Human Anatomy.

By RlCKMAN J. GODLEE, M.S.,

F.R.C.S., Surgeon and late Demon-
strator of Anatomy, University College
Hospital. With 48 Imp. 4to Plates (112
figures), and a volume of Explanatory
Text. 8vo, £4. 145. 6d.

Human Osteology.

By LUTHER HOLDEN, Consulting Sur-
geon to St. Bartholomew's Hospital.
Eighth Edition, edited by CHARLES
STEWART, F.R.S., Conservator of the
Museum R.C.S., and ROBERT W. REID,
M.D., F.R.C.S., Regius Professor of
Anatomy in the University of Aberdeen.
8vo, with 59 Lithographic Plates and 74
Engravings. i6s.

Also.

Landmarks, Medical and Surgi-
cal. Fourth Edition. 8vo, 35. 6d.

The Essentials of Regional
Anatomy. By RICHARD J. BERRY,
M.D., F.R.S., F.R.C.S., (Edin.), Lec-
turer on Anatomy in the New School of
Medicine, Edinburgh. Post Svo, inter-
leaved, i os. net.

Anatomy of the Joints of Man.

By HENRY MORRIS, Senior Surgeon to
the Middlesex Hospital. With 44 Plates
(several coloured). Svo, i6s.

A Manual of General Pathology,
for Students and Practitioners.
By W. S. LAZARUS-BARLOW, B.A.,
M.D., Pathologist and Lecturer on Path-
ology, Westminster Hospital. Svo, 2 Is.

Pathological Anatomy of Dis-
eases. Arranged according to the no-
menclature of the R.C.P. Lond. " By
NORMAN MooRE,M.D.,F.R.C.P., Assist-
ant Physician to St. Bartholomew's Hos-
pital. Fcap. Svo, with in Engravings,
8s. 6d.

A Manual of Clinical and Prac-
tical Pathology. By W. E.
WYNTER, M.D., M. R.C.P., Assistant
Physician to the Middlesex Hospital, and
F. J. WETHERED, M.D., F. R.C.P.,
Assistant Physician to the Consumption
Hospital, Brompton. With 4 Coloured
Plates and 67 Engravings. Svo, I2s. 6d.

General Pathology:

An Introduction to. By JOHN BLAND
SUTTON, F.R.C.S., Assistant Surgeon
to the Middlesex Hospital. Svo, with
149 Engravings, 145.

The Pathologist's Handbook :

A Manual of the Post-mortem Room. By
T. N. KELYNACK, M.D., late Demon-
strator in Morbid Anatomy, Owens Col-
lege, Manchester. With 126 Illustrations,
fcap. Svo, pegamoid, 43. 6d.

A Contribution to the History
of the Respiration of Man ; being
the Croonian Lectures delivered before
the Royal College of Physicians in 1895,
with supplementary considerations of the
methods of inquiry and analytical results.
By WILLIAM MARCET, M.D., F.R.C.P.,
F.R.S. With Diagrams, imp. Svo, 55. 6d.

LONDON: 7, GREAT MARLBOROUGH STREET.

J. 4r A. CHURCHILL 'S RECENT WORKS.

Selected Researches in Pathol-
ogy. By A. G. AULD, M.D., M.R.C.P.

With 14 Illustrations, 8vo, 6s.

The Human Brain :

Histological and Coarse Methods of Re-
search. By W. BEVAN LEWIS, Medical
Superintendent, West Riding Asylum,
8vo, with Engravings and Photographs, 8s.

The Physiology and the Path-
ology of the Cerebral Circulation :
an Experimental Research. By
LEONARD HILL, M.B., Hunterian Pro-
fessor, R.C.S. With 41 Illustrations,
Royal 8vo, I2s.

Elements of Human Physiology.

By ERNEST H. STARLING, M.D.,
F.R.C.P., F.R.S., Jodrell Professor of
Physiology in University College, Lon-
don. Fourth Edition, much enlarged.
With 317 Engravings, 8vo, I2s. 6d.

Manual of Physiology :

For Students of Medicine. By GERALD
F. YEO, M.D., F.R.S. Third Edition.
Crown 8vo, with 254 Engravings and
Plate of Spectra, 145.

A Class Book of Practical Phy-
siology. By DE BURGH BIRCH, M.D.,
Professor of Physiology in the Yorkshire
College of the Victoria University. With
62 Illustrations, crown 8vo, 6s. 6d.

Practical Lessons in Elementary

Biology, for Junior Students. By
PEYTON T. B. BEALE, F.R.C.S., Lec-
turer on Elementary Biology and De-
monstrator in Physiology in King's
College, London. Crown 8vo, 35. 6d.

Medical Jurisprudence :

Its Principles and Practice. By ALFRED j
S. TAYLOR, M.D., F.R.C.P., F.R.S.
Fourth Edition, by THOMAS STEVENSON,
M.D., F.R.C.P., Lecturer on Medical
Jurisprudence at Guy's Hospital. 2 vols,
8vo, with 189 Engravings, 315. 6d.

Lectures on Medical Jurispru-
dence and Toxicology. By FRED.
J. SMITH, M.D., F.R.C.P., Lecturer
on Forensic Medicine and Toxicology at
the London Hospital. Crown 8vo, 7s. 6d.

Hygiene and Public Health.

A Treatise by various Authors. Edited
by THOMAS STEVENSON, M.D.,F.R.C. P.,
Lecturer on Chemistry and Medical Juris-
prudence at Guy's Hospital ; Official
Analyst to the Home Office ; and SHIRLEY
F. MURPHY, Medical Officer of Health
of the County of London. In 3 vols.,
royal 8vo, fully Illustrated. Vol. I., 285. ;
Vol. II., 32s. ; Vol. III., 20s.

The Theory and Practice of
Hygiene. By J. LANE XOTTER, M.D.,
Examiner in Hygiene and Public Health
in the University of Cambridge and the
English Conjoint Board, Professor of
Hygiene in the Army Medical School ;
and W. H. HORROCKS, M.B., B.Sc.,
Assistant Professor of Hygiene in the
Army Medical School. Second Edition.
With 15 Plates and 134 other Illustra
tions, Royal 8vo, 255.

A Manual of Practical Hygiene.

By the late E. A. PARKES, M.D., F.R.S.
Eighth Edition, by J. LANE NOTTER,
A.M., M.D. 8vo, with 10 Plates and
103 Engravings, i8s.

A Handbook of Hygiene and

Sanitary Science. By GEO. WILSON,
M.A., M.D., LL.D., F.R.S.E., D.P.H.
Camb., Medical Officer of Health for
Mid-Warwickshire. Eighth Edition.
Post 8vo, with Engravings, I2s. 6d.

A Simple Method of Water

Analysis, especially designed for the
use of Medical Officers of Health. By
JOHN C. THRESH, M.D.Vic., D.Sc.
Lond., D.P.H. Camb., Medical Officer
of Health for the County of Essex.
Second Edition. Fcap. 8vo, 2s. 6d.

Elements of Health : an Intro-
duction to the Study of Hygiene.
By Louis C. PARKES, M.D., D.P.H.
Lond., Lecturer on Public Health at
St. George's Hospital. Post 8vo, with
27 Engravings, 3s. 6d.

Diet and Food considered in
relation to Strength and Power
of Kndurance, Training and
Athletics. By ALEXANDER HAIG,
M.D., F.R.C.P. Third Edition. Crown
8vo, 2s.

Effects of Borax and Boracic
Acid on the Human System. By
Dr. OSCAR LIEBREICH, Professor in the
University of Berlin. With Plates, post
4to, 2s.

The Prevention of Epidemics
and the Construction and Man-
agement of Isolation Hospitals.

By ROGER McNEiLL, M.D. Edin.,
D.P.H. Camb., Medical Officer of
Health for the County of Argyll. Svo.
With several Hospital Plans, los. 6cl.

Hospitals and Asylums of the

World ; their Origin, History, Construc-
tion, Administration, Management, and
Legislation. By Sir HENRY BURDETTV
K.C.B. In 4 vols., with Portfolio. Com-
plete, i68s. net. Vols. I. and II. — Asy-
lums : 905. net. Vols. III. and IV. —
Hospitals, &c., with Portfolio of Plans,
I2os. net.

LONDON : 7, GREAT MARLBOROUGH STREET.

J.

A. CHURCHILL'S RECENT WORKS.

A Manual of Bacteriology, Clini-
cal and Applied. With an Appendix
on Bacterial Remedies, &c. By RICHARD
T. HEWLETT, M.D., M.R.C.P., D.P.H.
Lond., Assistant in the Bacteriological
Department, Jenner Institute of Preven-
tive Medicine. With 75 Illustrations,
post 8vo, i os. 6d.

Mental Diseases :

Clinical Lectures. By T. S. CLOUSTON,
M.D., F.R.C.P. Edin., Lecturer on
Mental Diseases in the University of
Edinburgh. Fifth Edition, Crown 8vo,
with 19 Plates, 143.

A Text - Book on Mental
Diseases, for Students and Prac-
titioners of Medicine. By THEO-
DORE H. KELLOGG, M.D., late Medical
Superintendent of Willard State Hospital,
U.S.A. With Illustrations, 8vo., 255.

Mental Physiology, especially
in its Relation to Mental Dis-
orders. By THEO. B. HYSLOP, M.D.,
Resident Physician and Medical Super-
intendent, Bethlem Royal Hospital ;
Lecturer on Mental Diseases, St. Mary's
Hospital Medical School. 8vo, i8s.

The Insane and the Law : a
Plain Guide for Medical Men,
Solicitors, and Others as to the
Detention and Treatment, Maintenance,
Responsibility, and Capacity either to
give evidence or make a will of Persons
Mentally Afflicted. With Hints to
Medical Witnesses and to Cross-Exami-
ning Counsel. By G. PITT-LEWIS, Q.C.,
R. PERCY SMITH, M.D., F.R.C.P., late
Resident Physician, Bethlem Hospital,
and J. A. HAWKE, B.A., Barrister-at-
Law. 8vo, 143.

A Dictionary of Psychological

Medicine, giving the Definition, Ety-
mology, and Synonyms of the Terms
used in Medical Psychology; with the
Symptoms, Treatment, and Pathology of
Insanity ; and THE LAW OF LUNACY IN
GREAT BRITAIN AND IRELAND. Edited
by D. HACK TUKE, M.D., LL.D., as-
sisted by nearly 130 Contributors. 2 vols.,
1,500 pages, royal 8vo, 423.

The Mental Affections of Chil-
dren, Idiocy, Imbecility, and
Insanity. By WM. W. IRELAND,
M.D. Edin., formerly Medical Superin-
tendent of the Scottish Institution for the
Education of Imbecile Children. Second
Edition. With 21 Plates, 8vo, 145.

Mental Affections of Childhood

and Youth (Lettsomian Lectures for
1887, &c.). By J. LANGDON-DOWN,
M.D., F.R.C.P., Consulting Physician to
the London Hospital. 8vo, 6s.

The Journal of Mental Science.

Published Quarterly, by Authority of the
Medico-Psychological Association. 8vo,5s.

Manual of Midwifery :

Including all that is likely to be required
by Students and Practitioners. By A. L.
GALABIN, M.D., F.R.C.P., Obstetric
Physician to Guy's Hospital. Fifth Edi-
tion. Post 8vo, with 298 Engravings, 155.

The Practice of Midwifery : A

Guide for Practitioners and Students.
By D. LLOYD ROBERTS, M.D., F.R.C.P.,
Consulting Obstetric Physician to the
Manchester Royal Infirmary, Physician
to St. Mary's Hospital. Fourth Edition.
. Crown 8vo, with 2 Coloured Plates and
226 Woodcuts, I os. 6d.

A Short Practice of Midwifery,
embodying the Treatment adop-
ted in the Rotunda Hospital,
Dublin. By HENRY JELLETT, M.D.,
B.A.O. Dub., late Assistant Master,
Rotunda Hospital. Second Edition.
With 57 Illustrations, crown 8vo, 6s.

Obstetric Aphorisms :

For the Use of Students commencing
Midwifery Practice. By JOSEPH G.
S WAYNE, M.D. Tenth Edition. Fcap.
8vo, with 20 Engravings, 35. 6d.

Economics, Anaesthetics, and
Antiseptics in the Practice of
Midwifery. By HAYDN BROWN,
L.R.C P., L.R.C.S. Fcap. 8vo, 2s. 6d.

Lectures on Obstetric Opera-
tions : A Guide to the Management
of Difficult Labour. By ROBERT BARNES,
M.D., F.R.C.P. Fourth Edition. 8vo,
with 121 Engravings, I2s. 6d.
By the same Author.

A Clinical History of Medical
and Surgical Diseases of
Women. Second Edition. 8vo, with
181 Engravings, 283.

Manual of the Diseases peculiar
to Women. By JAMES OLIVER,
M.D., F.R.S. Edin., M.R.C.P. Lond.,
Physician to the Hospital for Diseases
of Women, London. Fcap. 8vo, 33. 6d.
By the same Author.

Abdominal Tumours and Ab-
dominal Dropsy in Women.
Crown 8vo, 7s- 6d.

Gynaecological Operations :

(Handbook of). ByALBANH.G.DoRAN,
F.R.C.S., Surgeon to the Samaritan Hos-
pital. 8vo, with 167 Engravings, 153.

The Student's Guide to the
Diseases of Women. By ALFRED
L. GALABIN, M.A., M.D., F.R.C.P.,
Obstetric Physician to Guy's Hospita'.
Fifth Edition. Fcap. 8vo, With 142
Engravings, 8s. 6d.

A Short Practice of Gynaecology.

By HENRY JELLETT, M.D., B.A.O.
Dub., late Assistant Master, Rotunca
Hospital, Dublin. With 135 Illustra-
tions, crown 8vo, 73. 6d.

LONDON: 7, GREAT MARLBOROUGH STREET.

/. 4- A. CHURCH1LUS RECENT WORKS.

A First Series of Fifty-four
Consecutive Ovariotomies, with
Fifty-three Recoveries. By A. C.
BUTLER-SMYTHE, F.R.C.P. Edin., Sur-
geon to the Samaritan Free Hospital,
Senior Surgeon to the Grosvenor Hospital
for Women and Children. 8vo, 6s. 6d.

Sterility.

By ROBERT BELL, M.D., F.F.P. & S.,
Senior Physician to the Glasgow Hospital
for Diseases peculiar to Women, Svo, 55.

• Notes on Gynaecological Nurs-
ing. By JOHN BENJAMIN HELLIER,
M.D.,M.'R.C.S., Surgeon to the Hospital
for Women, &c., Leeds. Cr. 8vo, is. 6d.

A Manual for Hospital Nurses

and others engaged in Attending on the
Sick, with a Glossary. By EDWARD J.
DOMVILLE, Surgeon to the Devon and
Exeter Hospital. Eighth Edition. Crown
8vo, 2s. 6d.

A Short Manual for Monthly

Nurses. By CHARLES J. CULLING-
WORTH, M.D., F.R.C.P., Obstetric
Physician to St. Thomas's Hospital.
Revised by M. A. ATKINSON. Fourth
Edition. Fcap. 8vo, is. 6d.

Lectures on Medicine to Nurses.
By HERBERT E. CUFF, M.D., F.R.C.S.,
Superintendent, North Eastern Fever
Hospital, London. Third Edition. With
29 Illustrations, 35. 6d.

Antiseptic Principles for Nurses.
By C. E. RICHMOND, F.R.C.S. 8vo, is.

The Diseases of Children.

By JAS. F. GOODHART, M.D., F.R.C.P.,
Consulting Physician to the Evelina
Hospital for Children. Sixth Edition,
with the assistance of G. F. STILL, M.D.,
Medical Registrar to the Gt. Ormond St.
Hospital for Children. Crown 8vo, IDS. 6d.

A Practical Treatise on Disease
in Children. By EUSTACE SMITH,
M.D., F.R.C.P., Physician to the King
of the Belgians, and to the East London
Hospital for Children, &c. Second
Edition. 8vo, 22s.

By the same Author.

Clinical Studies of Disease in

Children. Second Edition. Post 8vo,
75. 6d.

Also.

The Wasting Diseases of Infants

and Children. Sixth (Cheap) Edition.
Post 8vo, 6s.

On the Natural and Artificial
Methods of Feeding Infants and
Young Children. By EDMUND
CAUTLEY, M.D., Physician to the Bel-
grave Hospital for Children. Crown
Svo, 75. 6d.

Chloroform :

A Manual for Students and Practitioners.
By EDWARD LAWRIE, M.B. Edin., Lt.-
Col. I. M.S., Residency Surgeon, Hyder-
abad. With Illustrations. Crown 4to,
55. net.

Materia Medica,

Pharmacy, Pharmacology, and Thera-
peutics. By W. HALE WHITE, M.D.,
F.R.C.P., Physician to, and Lecturer
on Pharmacology and Therapeutics at,
Guy's Hospital. Sixth Edition, based
upon the B.P. of 1898, and on the Indian
and Colonial Addendum. Fcap. 8vo,
7s. 6d.

An Introduction to the Study
of Materia Medica, designed
for Students of Pharmacy and
Medicine. By HENRY G. GREENISH,
F.I C., F.L.S., Professor of Materia
Medica to the Pharmaceutical Society.
With 213 Illustrations. 8vo, 155.

Materia Medica

And Therapeutics. By CHARLES D. F.
PHILLIPS, M.D., F.R.S. Edin.

Vegetable Kingdom — Organic Com-
pounds— Animal Kingdom. 8vo,25s.
Inorganic Substances. Second Edition.
8vo, 2 is.

Practical Pharmacy :

An Account of the Methods of Manu-
facturing and Dispensing Pharmaceutical
Preparations ; with a chapter on the
Analysis of Urine. By E. W. LUCAS,
F.C.S., Examiner at the Pharmaceutical
Society. With 283 Illustrations. Royal
8vo, I2s. 6cl.

Galenic Pharmacy:

A Practical Handbook to the Processes
of the British Pharmacopoeia. By R. A.
CRIPPS, M.P.S. 8vo, with 76 Engrav-
ings, 8s. 6d.

Practical Pharmacy.

By BARNARD S. PROCTOR, formerly
Lecturer on Pharmacy at the College of
Medicine, Newcastle-on-Tyne, Third
Edition. Svo, with 44 Wood Engravings
and 32 Lithograph Fac-Simile Prescrip-
tions, 145.

A Companion to the British

Pharmacopoeia. By PETER SQUIRE.
Revised by PETER WYATT SQUIRE.
Seventeenth Edition. Svo, I2s. 6d.

By the same Authors.

The Pharmacopoeias of thirty of
the London Hospitals, arranged
in Groups for Comparison. Seventh
Edition. Fcap. Svo, 6s.

LONDON: 7, GREAT MARLBOROUGH STREET.

/. & A. CHURCHILL* S RECENT WORKS.

5

The Galenical Preparations of
the British Pharmacopoeia : A

Handbook for Medical and Pharma-
ceutical Students. By CHARLES O.
HAWTHORNE, M.D., C.M., late Lecturer
on Materia Medica, Queen Margaret
College, Glasgow. 8vo, 45. 6d.

The Pharmaceutical Formulary;

a Synopsis of the British and Foreign
Pharmacopoeias. By HENRY BEASLEY.
Twelfth Edition (B.P. 1898), by J. OLD-
HAM BRAITHWAITE. i8mo, 6s. 6d.

By the same Author.

The Druggist's General Receipt-
Book. Tenth Edition. iSmo, 6s. 6d.

Also.

The Book of Prescriptions :

Containing upwards of 3,000 Prescriptions
from the Practice of the most eminent
Physicians and Surgeons, English and
Foreign. Seventh Edition. iSmo, 6s. 6d.

Selecta & Prescriptis :

Containing Terms, Phrases, Contractions
and Abbreviations used in Prescriptions,
with Explanatory Notes, &c. Also, a
Series of Abbreviated Prescriptions with
Translations and Key. By J. PEREIRA,
M.D., F.R.S. Eighteenth Edition, by
JOSEPH INCE, F.C.S., F.L.S. 24mo, 53.

Year-Book of Pharmacy :

Containing the Transactions of the British
Pharmaceutical Conference. Annually.
8vo, IDS.

Southall's Organic Materia

Medica : a Handbook treating of some
of the more important of the Animal and
Vegetable Drugs made use of in Medicine,
including the whole of those contained in
the B.P. Sixth and Enlarged Edition.
By JOHN BARCLAY, B.Sc. Lond. 8vo,
75. 6d.

Manual of Botany.

ByJ. REYNOLDS GREEN, Sc.D., M.A.,
F.R.S. , Professor of Botany to the Phar-
maceutical Society. Two Vols. Cr. 8vo.
Vol. I. — Morphology and Anatomy.
Second Edition. With 778 En-
gravings, 7s- 6d.

„ II. — Classification and Physiology.
With 417 Engravings, los.

By the same Author.

An Introduction to Vegetable

Physiology. With 184 Illustrations,
8vo, i os. 6d.

Medicinal Plants :

Being descriptions, with original figures,
of the Principal Plants employed in
Medicine, and an account of their Pro-
perties and Uses. By Prof. BENTLEY and
Dr. H. TRIMEN, F.R.S. In 4 vols., large
8vo, with 306 Coloured Plates, bound in
Half Morocco, Gilt Edges, £11 us.

The Student's Guide to Syste-
matic Botany, including the Classi-
fication of Plants and Descriptive Botany.
By ROBERT BENTLEY. P'cap. 8vo, with
350 Engravings, 33. 6d.

Therapeutic Electricity and
Practical Muscle Testing, By
W. S. HEDLEY, M.D., in charge of the
Electro-therapeutic Department of the
London Hospital. With no Illustrations.
Royal 8vo, 8s. 6d.

Practical Therapeutics :

A Manual. By EDWARD J. WARING,
C.I.E., M.D., F.R.C.P., and DUDLEY
W. BUXTON, M.D., B.S. Lond. Fourth
Edition. Crown Svo, 145.
By the same Author.

Bazaar Medicines of India,

And Common Medical Plants : With Full
Index of Diseases, indicating their Treat-
ment by these and other Agents procur-
able throughout India, &c. Fifth Edition.
Fcap. Svo, 55.

Climate and Fevers of India,

with a series of Cases (Croonian Lec-
tures, 1882). By Sir JOSEPH FAYRER,
K. C.S.I., M.D. Svo, with 17 Tem-
perature Charts, I2s.

Psilosis or "Sprue," its Nature
and Treatment ; with Observations
on various Forms of Diarrhoea acquired
in the Tropics. By GEORGE THIN,
M.D. Second and enlarged Edition,
with Illustrations, Svo, los.

A Manual of Family Medicine
and Hygiene for India. Published
under the Authority of the Government
of India. By Sir WILLIAM J. MOORE,
K.C.I.E., M.D., late Surgeon-General
with the Government of Bombay. Sixth
Edition. Post Svo, with 71 Engravings,

I2S.

By the same Author.

A Manual of the Diseases of

India : With a Compendium of Diseases
generally. Second Edition. Post 8vc,
I os.

The Prevention of Disease in
Tropical and Sub-Tropical Cam-
paigns. (Parkes Memorial Prize for
1886.) By Lieut.-Col. ANDREW DUNCAN,
M.D., B.S. Lond., F.R.C.S., Indian
Medical Service. Svo, I2s. 6d.

A Commentary on the Diseases
of India. By NORMAN CHEVERS,
C.I.E., M.D., F.R.C.S., Deputy Sur-
geon-General H.M. Indian Army. 8vo,

24S.

Hooper's Physicians' Vade-

Mecum. A Manual of the Principles
and Practice of Physic. Tenth Edition.
By W. A. GUY, F.R.C.P., F.R.S., and
J. HARLEY, M.D., F.R.C.P. With 118
Engravings. Fcap. Svo, I2s. 6d.

LONDON.- 7, GREAT MARLBOROUGH STREET.

J.

A. CHURCHILL'S RECENT WORKS.

Ihe Principles and Practice of

Medicine. (Text-book.) By the late
C. HILTON FAGGE, M.D., and P. H.
PYE-SMITH, M.D., F.R.S., F.R.C.P.,
Physician to, and Lecturer on Medicine
in, Guy's Hospital. Third Edition. 2
vols. 8vo, cloth, 405. ; Half Leather, 465.

Manual of the Practice of Medi-
cine. By FREDERICK TAYLOR, M.D.,
F.R.C.P., Physician to, and Lecturer
on Medicine at, Guy's Hospital. Fifth
Edition. 8vo, with Engravings, i6s.

The Practice of Medicine.

By M. CHARTERIS, M.D., Professor of
Therapeutics and Materia Medica in the
University of Glasgow. Eighth Edi-
tion. Edited by F. j. CHARTERIS, M.B.,
Ch.B. Crown 8vo, with Engravings on
Copper and Wood, los.

A Dictionary of Practical Medi-
cine. By various writers. Edited by
IAS. KINGSTON FOWLER, M.A., M.D.,
F.R.C.P., Physician to Middlesex Hos-
pital and the Hospital for Consumption.
8vo, cloth, 2is. ; half calf, 2^s.

A Text-Book of Bacteriology,

for Medical Students and Practitioners.
Dy G. M. STERNBERG, M.D., Surgeon-
General, U.S. Army. With 9 Plates and
200 Figures in the Text. 8vo, 245.

How to Examine the Chest :

A Practical Guide for the use of Students.
By SAMUEL WEST, M.D., F.R.C.P.,
Assistant Physician to St. Bartholomew's
Hospital. Third Edition. \Vith 46 En-
gravings. Fcap. 8vo, 55.

An Atlas of the Pathological ;

Anatomy of the Lungs. By the late I
WILSON Fox, M.D., F.R.S., F.R.C.P., |
Physician to H.M. the Queen. With
45 Plates (mostly Coloured) and Engrav-
ings. 410, half-bound in Calf, 705.
By the same Author.

A Treatise on Diseases of the

Lungs and Pleura. Edited by
SIDNEY COUPLAND, M.D., F.R.C.P.,
Physician to Middlesex Hospital. Roy.
8vo, with Engravings ; also Portrait and
Memoir of the Author, 365.

The Student's Guide to Dis-
eases of the Chest. By VINCENT
D. HARRIS, M.D. Lend., F.R.C.P.,
Physician to the City of London Hospital
for Diseases of the Chest, Victoria Park.
Fcap. 8vo, with 55 Illustrations (some
Coloured), 75. 6d.

Uric Acid

as a Factor in the Causation of Disease.
By ALEXANDER HAIG, M.D., F.R.C.P.,
Physician to the Metropolitan Hospital
and the Royal Hospital for Children and
Women. Fifth Edition. With 75 Illus-
trations, 8vo, 145.

Text-Book of Medical Treat-
ment (Diseases and Symptoms). By
NESTOR I. C. TIRARD, M.D., F.R.C.P.,
Professor of the Principles and Practice
of Medicine, King's College, London.
8vo, 155.

Medical Diagnosis (Student's
Guide Series). By SAMUEL FENWICK,
M.D., F.R.C.P., and W. SOLTAU
FENWICK, M.D.,B.S. Eighth Edition.
Crown Svo, with 135 Engravings, 93,
By the same Authors.

Outlines of Medical Treatment.

Fourth Edition. Crown Svo, with 35
Engravings, los.

Also.

Ulcer of the Stomach and Duo-
denum. With 55 Illustrations. Royal
Svo, IDS. 6d.

Also, by Dr. SAMUEL FENWICK.
Clinical Lectures on Some Ob-
scure Diseases of the Abdomen.
Delivered at the London Hospital. Svo,
with Engravings, 75. 6d.
Also.

The Saliva as a Test for

Functional Diseases of the Liver.
Crown Svo, 2s.

The Liver.

By LIONEL S. BEALE, M.B., F.R.S.,
Consulting Physician to King's College
Hospital. With 24 Plates (85 Figures).
Svo, 53.

By the same Author,

On Slight Ailments :

And on Treating Disease. Fourth Edition.
Svo, 55.

The Blood : How to Examine
and Diagnose its Diseases. By
ALFRED C. COLES, M.D., B.Sc. With
6 Coloured Plates. Svo, los. 6d.

The Physiology of the Carbo-
hydrates j their Application as Food and
Relation to Diabetes. By F. W. PAVY,
M.D., LL.D., F.R.S., F.R.C.P., Con-
sulting Physician to Guy's Hospital.
Royal Svo, with Plates and Engravings,
los. 6d.

Medical Lectures and Essays.

By Sir G. JOHNSON, M.D., F.R.C.P.,
F.R.S. Svo, with 46 Engravings, 255.

By the same Author.

An Essay on Asphyxia (Apncea).

Svo, 35.

Bronchial Asthma :

Its Pathology and Treatment. By J. B.
BERK ART, M.D., late Physician to the
City of London Hospital for Diseases of
the Chest. Second Edition, with 7
Plates (35 Figures). Svo, IDS. 6d.

LONDON: 7, GREAT MARLBOROUGH STREET.

J. $ A. CHURCH1LDS RECENT WORKS.

Treatment of Some of the Forms
of Valvular Disease of the Heart.
By A. E. SANSOM, M.D., F.R.C.P.,

Physician to the London Hospital.
Second Edition. Fcap. 8vo, with 26
Engravings, 45. 6d.

The Schott Methods of the
Treatment of Chronic Diseases
of the Heart, with an account of the
Nauheim Baths and of the Therapeutic
Exercises. By W. BEZLY THORNE,
M.D., M.R.C.P. Third Edition. 8vo,
with Illustrations, 6s.

Guy's Hospital Reports.

By the Medical and Surgical Staft'. Vol.
XXXIX. Third Series. 8vo, IDS. 6d.

St. Thomas's Hospital Reports.

By the Medical and Surgical Staff. Vol.
XXVII. New Series. 8vo, 8s. 6d.

Westminster Hospital Reports.

'By the Medical and Surgical Staff. Vol.
XI. 8vo, 8s.
Medical Ophthalmoscopy :

A Manual and Atlas. By SIR WILLIAM
R. COWERS, M.D., F.R.C.P., F.R.S.
Third Edition. Edited with the assis-
tance of MARCUS GUNN, M.B., F.R.C.S.,
Surgeon to the Royal London Ophthalmic
Hospital. With Coloured Plates and
Woodcuts. 8vo, 1 6s.

By the same Author.

A Manual of Diseases of the

Nervous System. Roy. 8vo.

Vol. I. Nerves and Spinal Cord. Third
Edition by the Author and JAMES
TAYLOR, M.D., F.R.C.P. With 192
Engravings, 155.

Vol. II. Brain and Cranial Nerves :
General and Functional Diseases. Second
Edition. With 182 Engravings, 2Os.
Also.

Clinical Lectures on Diseases of

the Nervous System. 8vo, 75. 6d.

Also.

Diagnosis of Diseases of the

Brain. Second Edition. 8vo, with
Engravings, 7s. 6d.
Also.

Syphilis and the Nervous

System. Lettsomian Lectures for 1890.
Delivered before the Medical Society
of London. 8vo, 45.

The Nervous System,

Diseases of. By J. A. ORMEROD, M.D.,
F.R.C.P., Physician to the National
Hospital for the Paralysed and Epileptic.
With 66 Illustrations. Fcap. 8vo, 8s. 6d.

Text-Book of Nervous
Diseases, for Students and Prac-
titioners of Medicine. By CHARLES
L. DANA, M.D., Professor of Nervous
and Mental Diseases in Bellevue Hospital
Medical College, New York. Fourth
Edition, with 246 Illustrations, 8vo, 2os.

Handbook of the Diseases of

the Nervous System. By TAMES
Ross, M.D., F.R.C.P., Professor of
Medicine in the Victoria University, and
Physician to the Royal Infirmary, Man-
chester. Roy. 8vo, with 184 Engravings,
1 8s.

Diseases of the Nervous System.

Lectures delivered at Guy's Hospital. By
SIR SAMUEL WILKS, BART., M.D.,
F.R.S. Second Edition. 8vo, i8s.

Stammering:

Its Causes, Treatment, and Cure. By
A. G. BERNARD, M.R.C.S., L.R.C.P.
Crown 8vo, 2s.

Secondary Degenerations of the
Spinal Cord. By HOWARD H. TOOTH,
M.D., F.R.C.P., Assistant Physician to
the National Hospital for the Paralysed
and Epileptic. With Plates and Engrav-
ings. 8vo, 35. 6d.

Diseases of the Nervous System.

Clinical Lectures. By THOMAS BUZZARD,
M. D. , F. R. C. P. , Physician to the National
Hospital for the Paralysed and Epileptic.
With Engravings, 8vo. 155.

By the same Autlior.

Some Forms of Paralysis from

Peripheral Neuritis : of Gouty,
Alcoholic, Diphtheritic, and other origin.
Crown 8vo, 55.

Also.

On the Simulation of Hysteria

by Organic Disease of the Nervous
System. Crown 8vo, 45. 6d.

On the Typhoid Bacillus and
Typhoid Fever, being the Goul-
stonian Lectures delivered before the
Royal College of Physicians in March,
1900. By P. HORTON-SMITH, M.D.,
F.R.C.P. With Illustrations, 8vo,
2s. 6d.

Gout in its Clinical Aspects.

By J. MORTIMER GRANVILLE, M.D.
Crown 8vo, 6s.

Diseases of the Liver:

With and without Jaundice By GEORGE
HARLEY, M.D., F.R.C.P., F.R.S. 8vo,
with 2 Plates and 36 Engravings, 21 s.

Rheumatic Diseases,

(Differentiation in). By HUGH LANE,
Surgeon to the Royal Mineral Water
Hospital, Bath, and Hon. Medical Officer
to the Royal United Hospital, Bath.
Second Edition, much Enlarged, with 8
Plates. Crown Svo, 35. 6d.

Diseases of the Abdomen,

Comprising those of the Stomach and other
parts of the Alimentary Canal, CEsopha-
gus, Caecum, Intestines, and Peritoneum.
By S. O. HABERSHON, M.D., F.R.C.P.
Fourth Edition. Svo, with 5 Plates, 2 is.

LONDON: 7, GREAT MARLBOROUGH STREET.

J. 8? A. CHURCHILL'S RECENT WORKS.

On Gallstones, or Cholelithiasis.
By E. M. BROCKBANK, M.D. Viet.,
M.R.C.P. Lond., Honorary Physician to
the Ancoats Hospital, Manchester. Crown
8vo, 7s.

On the Relief of Excessive and

Dangerous Tympanites by Punc-
turing the Abdomen. By JOHN W.
OGLE, M.D., Consulting Physician to St.
George's Hospital. 8vo, 53. 6d.

Headaches :

Their Nature, Causes, and Treatment.
By W. H. DAY, M.D., Physician to the
Samaritan Hospital. Fourth Edition.
Crown 8vo, with Engravings, 7s. 6d.

A Handbook of Medical Clima-
tology ; embodying its Principles and
Therapeutic Application, with Scientific
Data of the Chief Health Resorts of the
World. By S. EDWIN SOLLY, M.D.,
M.R.C.S.,late President of the American
Climatological Association. With En-
gravings and Coloured Plates, 8vo., i6s.

The Mineral Waters of France

And its Wintering Stations (Medical
Guide to). With a Special Map. By A.
VINTKAS, M.D., Physician to the French
Embassy, and to the French Hospital,
London. Second Edition. Crown 8vo, 8s.

IllustratedAmbulance Lectures :

To which is added a NURSING LECTURE.
ByJOHNM. H. MARTiN,M.D.,F.R.C.S.,
Honorary Surgeon to the Blackburn In-
firmary. Fourth Edition. Crown 8vo,
with 60 Engravings, 2s.

Surgery : its Theory and Prac-
tice. By WILLIAM J. WALSHAM,
F.R.C.S., Surgeon to, and Lecturer on
Anatomy at, St. Bartholomew's Hospital.
Seventh Edition. Post 8vo, with 483
Engravings, 155.

A Synopsis of Surgery.

By R. F. TOBIX, Surgeon to St. Vincent's
Hospital, Dublin. Crown 8vo, inter-
leaved, leather binding, 6s. 6d.

The Surgeon's Vade-Mecum :

A Manual of Modern Surgery. By R.
DRUITT, F.R.C.S. Twelfth Edition.
By STANLEY BOYD, M.B., F.R.C.S.
Crown 8vo, with 373 Engravings, i6s.

Operations on the Brain (A
Ouide to). By ALEC FRASER, Pro-
fessor of Anatomy, Royal College of
Surgeons in Ireland. Illustrated by 42
life-size Plates in Autotype, and 2 Wood-
cuts in the text. Folio, 635.

The Operations of Surgery :

Intended for Use on the Dead and Living
Subject alike. By W. H. A. JACOBSON,
M.A., M.B., M.Ch. Oxon., F.R.C.S.,
Assistant Surgeon to, and Lecturer on
Anatomy at, Guy's Hospital. Third
Edition. 8vo, with 401 Illustrations, 343.

A Course of Operative Surgery.

By CHRISTOPHER HEATH, Surgeon to
University College Hospital. Second
Edition. With 20 coloured Plates (180
figures) from Nature, by M. LEVEILLE,
and several Woodcuts. Large 8vo, 305.

By the same Author.

The Student's Guide to Surgical

Diagnosis. Second Edition. Fcap.
8vo, 6s. 6d.

Also.

Manual of Minor Surgery and
Bandaging. For the use of House*
Surgeons, Dressers, and Junior Practi-
tioners. Eleventh Edition. Fcap. 8vo,
with 176 Engravings, 6s.

Also.

Injuries and Diseases of the
Jaws. Fourth Edition. By HENRY
PERCY DEAN, M.S., F.R.C.S., Assistant
Surgeon to the London Hospital. 8vo,
with 187 Wood Engravings, 145.

Surgery.

By C. W. MANSELL MOULLIN, M.A.,
M.D., Oxon., F.R.C.S., Surgeon and
Lecturer on Physiology to the London
Hospital. Large 8vo, with 497 Engrav-
ings, 345.

The Practice of Surgery :

A Manual. By THOMAS BRYANT,
Consulting Surgeon to Guy's Hospital.
Fourth Edition. 2 vols. crown 8vo, with
750 Engravings (many being coloured),
and including 6 chromo plates, 325.

Surgical Emergencies :

Together with the Emergencies attendant
on Parturition and the Treatment of
Poisoning. By PAUL SWAIN, F.R.C.S.,
Surgeon to the South Devon and East Corn-
wall Hospital. Fifth Edition. Crown
8vo, with 149 Engravings, 6s.

Diseases of Bones and Joints.

By CHARLES MACNAMARA, F.R.C.S.,
Surgeon to, and Lecturer on Surgery at,
the Westminster Hospital. 8vo, with
Plates and Engravings, I2s.

Abdominal Surgery.

By J. GREIG SMITH, M.A., F.R.S.E.
Sixth Edition. Edited by JAMES SWAIN,
M.S., M.D. Lond., F.R.C.S. Eng.,
Assistant Surgeon to the Bristol Royal
Infirmary, Professor of Surgery, University
College, Bristol. Two Vols., 8vo, with
224 Engravings, 363.

LONDON: 7, GREAT MARLBOROUGH STREET.

A. CHURCHILL'S RECENT WORKS.

The Surgery of the Alimentary
Canal. By ALFRED ERNEST MAY-
I.ARD, M.B. Lond. and B.S., Surgeon to
the Victoria Infirmary, Glasgow. With
27 Swantype Plates and 89 Figures in
the text, 8vo, 255.

By the same Author.

A Student's Handbook of the
Surgery of the Alimentary Canal.
With 97 Illustrations. Crown 8vo, 8s. 6d.

Ovariotomy and Abdominal Sur-
gery. By HARRISON CRITPS, F.R.C.S.,
Surgical Staff, St. Bartholomew's Hos-
pital. With numerous Plates, Svo, 253.

The Physiology of Death from
Traumatic Fever ; A Study in
Abdominal Surgery. By JOHN D. MAL-
COLM, M.B., C.M., F.R.C.S.E., Sur-
§eon to the Samaritan Free Hospital,
vo, 35. 6d.

Surgical Pathology and Morbid
Anatomy. By ANTHONY A. BOWLBY,
F.R.C.S., Assistant Surgeon to St.
Bartholomew's Hospital. Fourth Edition.
Crown Svo, with 186 Engravings,
IDS. 6d.

By the same Author.

Injuries and Diseases of Nerves

and their Surgical Treatment.
Svo, with 20 Plates, 145.

Diseases of the Thyroid Gland
and their Surgical Treatment.
By JAMES BERRY, B.S. Lond., F.R.C.S.,
Surgeon to the Royal Free Hospital
and Lecturer on Surgery at the London
School of Medicine for Women. With
121 Illustrations, Svo, 143.

Short Manual of Orthopaedy.

By HEATHER BIGG, F.R.C.S. Ed.
Part I. Deformities and Deficiencies of
the Head and Neck. Svo, 2s. 6d.

The Human Foot :

Its Form and Structure, Functions and
Clothing. By THOMAS S. ELLIS, Con-
sulting Surgeon to the Gloucester In-
firmary. With 7 Plates and Engravings
(50 Figures). Svo, 7s. 6d.

Face and Foot Deformities.

By FREDERICK CHURCHILL, C.M. Svo,
with Plates and Illustrations, ios. 6d.

Royal London Ophthalmic Hos-
pital Reports. By the Medical and
Surgical Staff. Vol. XIV., Part 2. Svo, 53.

Ophthalmological Society

of the United Kingdom. Transactions,
Vol. XX. Svo, I2s. 6d.

Manual of Ophthalmic Surgery
and Medicine. By W. H. H.

JESSOP, M.A., F.R.C.S., Ophthalmic
Surgeon to St. Bartholomew's Hospital.
With 5 Coloured Plates and no Wood-
cuts, crown Svo, 95. 6d.

Nettleship's Diseases of the

Eye : A Manual for Students. Sixth
Edition, revised and edited by W. T.
HOLMES SPICER, M.B., F.R.C.S.,
Ophthalmic Surgeon to St. Bartholomew's-
Hospital and the Victoria Hospital for
Children. With 161 Engravings and a
Coloured Plate illustrating Colour-Blind-
ness, crown Svo, 8s. 6d.

Diseases and Refraction of the
Eye. ByN. C. MACNAMARA, F.R.C.S.,.
Surgeon to Westminster Hospital, and
GUSTAVUS HARTRIDGE, F.R.C.S., Sur-
geon to the Royal Westminster Ophthalmic
Hospital. Fifth Edition. Crown Svo,.
with Plate, 156 Engravings, also Test-
types, ios. 6d.

Diseases of the Eye : a Practical
Handbook for General Prac-
titioners and Students. By CECIL
EDWARD SHAW, M.D., M.Ch., Oph-
thalmic Surgeon to the Ulster Hospital
for Children and Women, Belfast. With,
a Test-Card for Colour - Blindness.
Crown Svo, 35. 6d.

On Diseases and Injuries of the

Eye : A Course of Systematic and-
Clinical Lectures to Students and Medical
Practitioners. By J. R. WOLFE, M.D.,
F.R.C.S.E., Lecturer on Ophthalmic
Medicine and Surgery in Anderson's Col-
lege, Glasgow. With 10 Coloured Plates,
and 157 Wood Engravings. Svo, £i is.

Normal and Pathological His-
tology of the Human Eye and
Eyelids. By C. FRED. POLLOCK,
M.D., F.R.C.S. and F.R.S.E., Surgeon
for Diseases of the Eye to Anderson's.
College Dispensary, Glasgow. Crown,
Svo, with joo Plates (230 drawings), 155..

Convergent Strabismus, and its

Treatment ; an Essay. By EDWIN
HOLTHOUSE, M.A., F.R.C.S., Surgeon,
to the Western Ophthalmic Hospital.
Svo, 6s.

Refraction of the Eye :

A Manual for Students. By GUSTAVUS
HARTRIDGE, F.R.C.S., Surgeon to the
Royal Westminster Ophthalmic Hospital.
Tenth Edition. Crown Svo, with 104.
Illustrations, also Test-types, &c., 6s.

By the same Author.

The Ophthalmoscope : A Manual

for Students. Third Edition. Crown Svo,
with 68 Illustrations and 4 Plates. 43. 6d_

Methods of Operating for Cata-
ract and Secondary Impairments
of Vision, with the results of 500 cases.
By Major G. H. FINK, H.M. Indian
Medical Service. Crown Svo, with 15
Engravings, 55.

LONDON: 7, GREAT MARLBOROUGH STREET.

10

J. $ A. CHURCHILUS RECENT WORKS.

Atlas of Ophthalmoscopy.

Composed of 12 Chromo - lithographic
Plates (59 Figures drawn from nature)
and Explanatory Text. By RICHARD
LIEBREICH, M.R.C.S. Translated by
H. ROSBOROUGH SWANZY, M.B. Third
Edition, 410, 405.

•Glaucoma :

Its Pathology and Treatment. By j
PRIESTLEY SMITH, Ophthalmic Surgeon
to, and Clinical Lecturer on Ophthalmo-
logy at, the Queen's Hospital, Birming-
ham. 8vo, with 64 Engravings and 12
Zinco-photographs, 73. 6d.

Eyestrain

(commonly called Asthenopia). By
ERNEST CLARKE, M.D., B.S Lond.,
Surgeon to the Central London Ophthal-
mic Hospital, Surgeon and Ophthalmic
Surgeon to the Miller Hospital. Second
Edition. 8vo, with 22 Illustrations, 5s.

Diseases and Injuries of the

Ear. By Sir WILLIAM B. DALEY,
F.R.C.S., M.B., Consulting Aural Sur-
• georl to St. George's Hospital. Fourth
Edition. Crown 8vo, with 8 Coloured
Plates and 38 Wood Engravings. IDS. 6d. j

By the saint. Author.

^Short Contributions to Aural
Surgery, between 1875 and 1896.

Third Edition. 8vo, with Engravings, 55.

Diseases of the Ear,

Including the Anatomy and Physiology of
the Organ, together with the Treatment of
the Affections of the Nose and Pharynx,
which conduce to Aural Disease. By
T. MARK HOVELL, Aural Surgeon to
the London Hospital, and Lecturer on
Diseases of the Throat in the College.
Second Edition. 8vo, with 128 Engrav-
ings, 2 is.

-A System of Dental Surgery.
By Sir JOHN TOMES, F.R S., and C. S.
TOMES, M.A., F.R.S. Fourth Edition.
Post 8vo, with 289 Engravings, i6s.

Dental Anatomy, Human and

Comparative: A Manual. By CHARLES
S. TOMES, M.A., F.R.S. Fifth Edition.
Post 8vo, with 263 Engravings, 145.

Dental Caries : an Investigation
into its Cause and Prevention.
By J. SIM WALLACE, M.D., B.Sc.,
L.D.S.R.C.S. 8vo, 55.

.Dental Materia Medica, Phar-
macology and Therapeutics. By
CHARLES W. GLASSINGTON, M.R.C.S.,
L.D.S. Edin. ; Senior Dental Surgeon,
Westminster Hospital ; Dental Surgeon,
National Dental Hospital; and Lecturer on
Dental Materia Medica and Therapeutics
to the College. Crown 8vo, 6s.

Dental Medicine :

A Manual of Dental Materia Medica and
Therapeutics. By FERDINAND [. S.
GORGAS, M.D., D.D.S., Professor of
the Principles of Dental Science in the
University of Maryland. Sixth Edition.
8vo, 1 8s.

A Manual of Dental Metallurgy.

By ERNEST A. SMITH, F.I.C., Assistant
Instructor in Metallurgy, Royal College
of Science, London. With 37 Illustra-
tions. Crown Svo, 6s. 6d.

A Practical Treatise on Mecha-
nical Dentistry. By JOSEPH RICH-
ARDSON, M.D., D.D.S. Seventh Edition
revised and Edited by GEORGE W.
WARREN, D.D.S. Roy. Svo, with 690
Engravings, 22s.

A Manual of Nitrous Oxide
Anaesthesia, for the use of Stu-
dents and General Practitioners.

By J. FREDERICK W. SILK, M.D. Lond.,
M.R.C.S., Anaesthetist to the Royal
Free Hospital, Dental" School of Guy's
Hospital, and National Epileptic Hospital.
Svo, with 26 Engravings, 55.

Skin Diseases of Children. By
GEO. H. Fox, M.D., Clinical Professor
of Diseases of the Skin, , College of
Physicians and Surgeons, New York.
With 1 2 Photogravure and Chromographic
Plates, and 60 Illustrations in the Text,
Roy. Svo, 1 2s. 6d.

A Handbook on Leprosy.

By S. P. IMPEY, M.D.,M.C, late Chief
and Medical Superintendent, Robben
Island Leper and Lunatic Asylums, Cape
Colony. With 38 Plates and Map, Svo, 12s.

Diseases of the Skin

(Introduction to the Study of). By
P. II. PYE- SMITH, M. D., F.R.S.,
F.R.C.P., Physician to, and Lecturer on
Medicine in, Guy's Hospital. Crown
Svo, with 26 Engravings. 73. 6d.

A Manual of Diseases of the

Skin, with an Analysis of 20,000 con-
secutive Cases and a Formulary. By
DUNCAN L. BULKLEY, M.I)., New
York. Fourth Edition, Roy. i6mo, 6s. 6d.

Cancerous Affections of the Skin.

(Epithelioma and Rodent Ulcer.) By
GEORGE THIN, M.D. Post 8vo, with
8 Engravings, 55.

By the same Author.

Pathology and Treatment of

Ringworm. Svo, with 21 Engravings,

55.

LONDON : 7, GREAT MARLBOROUGH STREET.

J. $ A. CHURCHILL'S RECENT WORKS.

11

Tne Operative Surgery of Malig-
nant Disease. By HENRY T. BUTLIN,
F.R.C.S., Surgeon to St. Bartholomew's
Hospital. Second Edition, with 12 En-
gravings. 8vo, 145.

By the same Author.

Malignant Disease (Sarcoma
and Carcinoma) of the Larynx.
8vo, with 5 Engravings, 55.

Also.

Sarcoma and Carcinoma :

Their Pathology, Diagnosis, and Treat-
ment. 8vo, with 4 Plates, 8s.

Cancers and the Cancer Pro-
cess : a Treatise, Practical and Theoretic.
By HERBERT L. SNOW, M.D., Surgeon
to the Cancer Hospital, Brompton. 8vo,
with 15 Lithographic Pl-ates. 155.

By the same Author.

The Re-appearance (Recurrence)
of Cancer after apparent Extir-
pation. Svo, 55. 6d.

Also.

The Palliative Treatment of
Incurable Cancer. Crown 8vo,2s.6d.

Ringworm and some other
Scalp Affections : their Cause and
Cure. By HAYDN BROWN, L.R.C.P.Ed.

8vo, 5s.

The Diagnosis and Treatment of

Syphilis. By TOM ROBINSON, M.D.,
Physician to the Western Skin Hospital.
Second Edition, Crown Svo, 35. 6d.

By the same Author.

The Diagnosis and Treatment

of Eczema. Second Edition, Crown
Svo, 35. 6d.

Also.

Illustrations of Diseases of the
Skin and Syphilis, with Re-
marks. Fasc. I. with 3 Plates. Imp.
4to, 55.

Selected Papers on Stone, Pros-
tate, and other Urinary Dis-
orders. By REGINALD HARRISON,
F.R.C.S., Surgeon to St. Peter's Hos-
pital. Svo, with 15 Illustrations. 55.

Chemistry of Urine ;

A Practical Guide to the Analytical
Examination of Diabetic, Albuminous,
and Gouty Urine. By ALFRED H.
ALLEN, F.I.C., F.C.S. With Engrav-
ings, Svo, 75. 6d.

Clinical Chemistry of Urine
(Outlines of the). By C. A. MAC
MUNN, M.A., M.D. Svo, with 64
Engravings and Plate of Spectra, 95.

BY SIR HENRY THOMPSON, Bart., F.R.CS.
Diseases of the Urinary Organs.

Clinical Lectures. Eighth Edition.
Svo, with 121 Engravings, IDS. 6d.

Practical Lithotomy and Litho-

trity; or, An Inquiry into the Best Modes
of Removing Stone from the Bladder.
Third Edition. Svo, with 87 Engravings,

IOS.

The Preventive Treatment of

Calculous Disease, and the Use of
Solvent Remedies. Third Edition.
Crown Svo, 2s. 6d.

Tumours of the Bladder :

Their Nature, Symptoms, and Surgical
Treatment. 8vo, with numerous Illustra-
tions, 55.

Stricture of the Urethra, and Uri-
nary Fistulae : their Pathology and
Treatment. Fourth Edition. Svo, with
74 Engravings, 6s.

The Suprapubic Operation of
Opening the Bladder for the.
Stone and for Tumours. Svo,

with 14 Engravings, 35. 6d.

The Clinical Examination of

Urine, with an Atlas of Urinary De-
posits. By LlNDLEY SCOTT, M. A., M.D.
With 41 Original Plates (mostly in
Colours). Crown 410, 155. net.

Electric Illumination of the

Bladder and Urethra, as a Means
of Diagnosis of Obscure Vesico-UrethraJ
Diseases. By E. HURRY FENWICK,
F.R.C.S., Surgeon to London Hospital
and St. Peter's Hospital for Stone. Second
Edition. Svo, with 54 Engravings, 6s. 6d.
By the same Author.

Tumours of the Urinary Blad-
der. The Jacksonian Prize Essay of
1887, rewritten with 200 additional cases,
in four Fasciculi. Fas. I. Royal Svo. 55.

A(so.

Ulceration of the Bladder,
Simple, Tuberculous, and Malig-
nant : A Clinical Study. Illustrated,
Svo, 55.

Also.

The Cardinal Symptoms of
Urinary Diseases : their Diagnostic
Significance and Treatment. Svo, with
36 Illustrations. 8s. 6d. y« • A

Atlas of Electric Cystoscopy.

By Dr. EMIL BURCKHARDT, late of the
Surgical Clinique of the University of
Bale, and E. HURRY FENWICK, F.R.C.S.,
Surgeon to the London Hospital and St.
Peter's Hospital. Royal Svo, with 34 Col-
oured Plates, embracing 83 Figures. 2 is.

LONDON : 7, GREAT MARLBOROUGH STREET.

12

.7. 4- A. CHURCHILL'S RECENT WORKS.

Urinary and Renal Derange-
ments and Galculous Disorders.
By LIONEL S. BEALE, F.R.C.P.,
F.R.S., Consulting Physician to King's
College Hospital. Svo, 55.

Male Organs of Generation
( Diseases of). By W. H. A. JACOBSON,
M.Ch. Oxon., F.R.C.S., Assistant
Surgeon to Guy's Hospital. Svo, with
88 Engravings. 22s.

The Surgical Diseases of the
Genito - Urinary Organs, in-
cluding Syphilis. By E. L. KEYES,
M.D., Professor in Bellevue Hospital
Medical College, New York (a revision
of VAN BUREN and KEYES' Text-book).
Roy. Svo, with 114 Engravings, 2is.

Diseases of the Rectum and
Anus. By ALFRED COOPER, F.R.C.S.,
Senior Surgeon to the St. Mark's Hos-
pital for Fistula ; and F. S WIN FORD
EDWARDS, F.R.C.S., Senior Assistant
Surgeon to St. Mark's Hospital. Second
Edition, with Illustrations. Svo, 12s.

Diseases of the Rectum and
Anus. By HARRISON CRIPPS, F.R.C.S.,
Assistant Surgeon to St. Bartholomew's
Hospital, &c. Second Edition. Svo,
with 13 Lithographic Plates and numer-
ous Wood Engravings, I2s. 6d.

By the same Author.

Cancer of the Rectum.

Especially considered with regard to its
Surgical Treatment. Jacksonian Prize
Essay. Svo, with 13 Plates and several
Wood Engravings, 6s.

Also.

The Passage of Air and Faeces
from the Urethra. Svo, 35. 6d.

Syphilis.

By ALFRED COOPER, F.R.C.S., Con-
sulting Surgeon to the West London and
the Lock Hospitals. Second Edition.
Edited by EDWARD COTTER ELL,
F.R.C.S., Surgeon (out-patients) to the
London Lock Hospital. Svo, with 24
Full -page Plates (12 coloured), iSs.

On Maternal Syphilis, including

the presence and recognition of Syphilitic
Pelvic Disease in Women. By JOHN A.
SHAW-MACKENZIE, M.D. Svo, with
Coloured Plates, IDS. 6d.

A Medical Vocabulary:

An Explanation of all Terms and Phrases
used in the various Departments of Medical
Science and Practice, their Derivation,
Meaning, Application, and Pronunciation.
By R. G. MAYNE, M.D., LL.D. Sixth
Edition by W. W. WAGSTAFFE, B.A.,
F.R.C.S. Crown Svo, IDS. 6d.

A Short Dictionary of Medical

Terms. Being an Abridgment of
Mayne's Vocabulary. 64mo, 2s. 6d.

Dunglison's Dictionary of

Medical Science : Containing a full
Explanation of its various Subjects and
Terms, with their Pronunciation, Accentu-
ation, and Derivation. Twenty-second
Edition, much enlarged. By RICHARD
J. DUNGLISON, A.M., M.D. Super-
royal Svo, 305.

Terminologia Medica Poly-

glotta : a Concise International Die
tionary of Medical Terms (French, Latin,
English, German, Italian, Spanish, and
Russian). By THEODORE MAXWELL,
M.D., B.Sc., F.R.C.S. Edin. Royal
Svo, 1 6s.

A German-English Dictionary of
Medical Terms. By Sir FREDERICK
TREVES, K.C.V.O., Surgeon to the Lon-
don Hospital; and HUGO LANG, B.A.
Crown Svo, half-Persian calf, 125.

A Handbook of Physics and

Chemistry, adapted to the require-
ments of the first Examination of the
Conjoint Board and for general use. By
HERBERT E. CORBIN, B.Sc.Lond., and
ARCHIBALD M. STEWART, B.Sc. Lond.
With 120 Illustrations. Crown Svo, 6s. 6d.

A Manual of Chemistry, Theo-
retical and Practical. By WILLIAM
A. TILDEN, D.Sc., F.R.S., Professor of
Chemistry in the Royal College of Science,
London ; Examiner in Chemistry to the
Department of Science and Art. With
2 Plates and 143 Woodcuts, crown Svo,

10S.

Chemistry,

Inorganic and Organic. With Experi-
ments. By CHARLES L. BLOXAM..
Eighth Edition, by JOHN MILLAR
THOMSON, F.R.S., Professor of Chemistry
in King's College, London, and ARTHUR
G. BLOXAM, Head of the Chemistry
Department, The Goldsmiths' Institute,.
New Cross. Svo, with nearly 300 Illustra-
tions, iSs. 6d.

By the same Author.

Laboratory Teaching;

Or, Progressive Exercises in Practical
Chemistry. Sixth Edition. By ARTHUR
G. BLOXAM. Crown Svo, with 80
Engravings, 6s. 6d.

Watts' Organic Chemistry.

Edited by WILLIAM A. TILDEN, D.Sc.,
F.R.S., Professor of Chemistry, Royal
College of Science, London. Second
Edition. Crown Svo, with Engravings,
los.

LONDON: 7, GREAT MARLBOROUGH STREET.

J. & A. CHURCHILL'S RECENT WORKS.

Practical Chemistry

And Qualitative Analysis. By FRANK
CLOWES, D.Sc. Lond., Emeritus Pro-
fessor of Chemistry in the University
College, Nottingham. Seventh Edition.
Post 8vo, with 101 Engravings and
Frontispiece, 8s. 6d.

Quantitative Analysis.

By FRANK CLOWES, D.Sc. Lond.,
Emeritus Professor of Chemistry in the
University College, Nottingham, and J.
BERNARD COLEMAN, Assoc. R. C. Sci.
Dublin ; Professor of Chemistry, South-
West London Polytechnic. Fifth Edi-
tion. Post 8vo, with 122 Engravings, IDS.
By the same Authors.

Elementary Quantitative Ana-
lysis, Post 8vo, with 62 Engravings,
45. 6d.

Also.

Elementary Practical Chemistry
and Qualitative Analysis. Third
Edition. Post 8vo, with 68 Engravings,
35. 6d.

Qualitative Analysis.

By R. FRESENIUS. Translated by
CHARLES E. GROVES, F.R.S. Tenth
Edition. 8vo, with Coloured Plate of
Spectra and 46 Engravings, 153.
By the same Author.

Quantitative Analysis.

Seventh Edition.

Vol. I., Translated by A. VACHER.

8vo, with 106 Engravings, 155.
Vol. II., Translated by C. E. GROVES,

F.R.S. 8vo, with 143 Engravings,

20S.

Inorganic Chemistry.

By SIR EDWARD FRANKLAND, K.C.B.,
D.C.L., LL.D., F.R.S., and FRANCIS
R. JAPP, M.A., Ph.D., F.I.C., F.R.S.,
Professor of Chemistry in the University
of Aberdeen. 8vo, with numerous Illus-
trations on Stone and Wood, 245.

Inorganic Chemistry

(A System of). By WILLIAM RAMSAY,
Ph.D., F.R.S., Professor of Chemistry in
University College, London. 8vo, with
Engravings, 155.

By the same Author.

Elementary Systematic Chemis-
try for the Use of Schools and
Colleges. With Engravings. Crown
8vo, 4s. 6d. ; Interleaved, 55. 6d.

Valentin's Practical Chemistry
and Qualitative and Quantita-
tive Analysis. Edited by W. R.
HODGKIMSON, Ph.D., F.R.S. E., Pro-
fessor of Chemistry and Physics in the
Royal Military Academy, and Artillery
College, Woolwich. Ninth Edition. 8vo,
with Engravings and Map of Spectra,
9s. [The Tables separately, 2s. 6d.]

Practical Chemistry, Part I.
Qualitative Exercises and Analy-
tical Tables for Students. By J.
CAMPBELL BROWN, Professor of Chemis
try in Victoria University and University
College, Liverpool. Fourth Edition.
8vo, 2s. 6d.

The Analyst's Laboratory Com-
panion ; a Collection of Tables and
Data for Chemists and Students. By
ALFRED E. JOHNSON, A. R. C.S.I.,
F.I.C. Second Edition, enlarged, crown
8vo., cloth, 55., leather, 6s. 6d.

Volumetric Analysis :

Or the Quantitative Estimation of Chemi-
cal Substances by Measure, applied to
Liquids, Solids, and Gases. By FRANCIS
SOTTON, F.C.S., F.I.C. Eighth Edi-
tion. 8vo, with Il6 Engravings, 2Os.

Commercial Organic Analysis:

a Treatise on the Properties, Modes of
Assaying, Proximate Analytical Examina-
tion, &c., of the various Organic Chemi-
cals and Products employed in the Arts,
Manufactures, Medicine, &c. By ALFRED
H. ALLEN, F.I.C., F.C.S. Svo.

Vol. 1. — Introduction, Alcohols, Neu-
tral Alcoholic Derivatives, &c.,
Ethers, Vegetable Acids, Starch
and its Isomers, Sugars, &c. Third
Edition. i8s.

Vol. II.— Part I. Fixed Oils and
Fats, Glycerin, Nitro - Glycerin,
Dynamites and Smokeless Pow-
ders, Wool- Fats, Degras, etc.
Third Edition. 145.

Vol. II.— Part II. Hydrocarbons,
Petroleum and Coal -Tar Products,
Asphalt, Phenols and Creosotes,
&c. Third Edition. 145.

Vol. JL— Part III. Terpenes, Essen-
tial Oils, Resins and Camphors,
Acid Derivatives of Phenols, Aro-
matic Acids. [In preparation.

Vol. III.— Part I. Tannins, Dyes and
Colouring Matters, Inks. Third
Edition. i8s.

VoL III.— Part II. The Amines and
Ammonium Bases, Hydrazines and
Derivatives, Bases from Tar, The
Antipyretics, &c., Vegetable Alka-
loids, Tea, Coffee, Cocoa, Kola,
Cocaine, Opium, &c. Second
Edition. 1 8s.

Vol. III.— Part III. Vegetable Alka-
loids, Non- Basic Vegetable Bitter
Principles, Animal Bases, Animal
Acids, Cyanogen and its Deriva-
tives, &c. Second Edition. i6s.

Vol. IV.— The Proteids and Albu-
minous Principles, Proteoi'ds or
Albuminoids. Second Edition. i8s.

LONDON: 7, GREAT MARLBOROUGH STREET.

.7.

A. CHURCHILL'S RECENT WORKS.

Cooley's Cyclopaedia

of Practical Receipts, and Collateral In-
formation in the Arts, Manufactures, Pro-
fessions, and Trades : Including Medicine,
Pharmacy, Hygiene and Domestic Eco-
nomy. Seventh Edition, by W. NORTH,
M.A. Camb., F.C.S. 2 Vols., Roy. Svo,
with 371 Engravings, 423.

Chemical Technology:

A Manual. By RUDOLF VON WAGNER.
Translated and Edited by SIR WILLIAM
CROOKES, F.R.S., from the Thirteenth
Enlarged German Edition as remodelled
by Dr. FERDINAND FISCHER. 8vo, with
596 Engravings, 325.

Chemical Technology ;

Or, Chemistry in its Applications to Arts
and Manufactures. Edited by CHARLES
E. GROVES, F.R.S., and WILLIAM
THORP, B.Sc.

Vol. I. — FUEL AND ITS APPLICA-
TIONS. By E. J. MILLS, D.Sc.,
F.R.S., and F. J. ROWAN, C.E.
Royal 8vo, with 606 Engravings, 303.
Vol. II. — LIGHTING BY CANDLES
AND OIL. By W. Y. DENT, J.
MCARTHUR, L. FIELD and F. A.
FIELD, BOVERTON REDWOOD, and
D. A. Louis. Royal 8vo, with 358
Engravings and Map, 2Os.
Vol. III. — GAS LIGHTING. By
CHARLES HUNT. With 2 Plates
and 292 Engravings. Royal 8vo, i8s.

Technological Handbooks.

EDITED BY JOHN GARDNER, F.I.C.,
F.C.S., and JAMES CAMERON, F.I.C.
BREWING, DISTILLING, AND WINE

MANUFACTURE. Crown Svo, with

Engravings, 6s. 6d.
BLEACHING, DYEING, AND CALICO

PRINTING. With Formulae. Crown

Svo, with Engravings, 55.
OILS, RESINS, AND VARNISHES.

Crown Svo, with Engravings. 7s. 6d.
SOAPS AND CANDLES. Crown Svo,

with 54 Engravings* 7s.

Chemistry an Exact Mechanical
Philosophy. By FRED. G. EDWARDS.
Illustrated. Svo, 35. 6d.

The Microscope and its Revela-
tions. By the late WILLIAM B. CAR-
PENTER, C.B., M.D., LL.D., F.R.S.
Seventh Edition, by the Rev. W. II.
DALLINGER, LL.D., F.R.S. With 21
Plates and 800 Wood Engravings. Svo,
26s. Half Calf, 303.

Methods and Formulae

Used in the Preparation of Animal and
Vegetable Tissues for Microscopical Ex-
amination, including the Staining of
Bacteria. By PETER WYATT SQUIRE,
F.L.S. Crown Svo, 35. 6d.

Encyclopaedia Medica, under
WATSON, M.B., M.R.C.P.E. About
now ready.

The Quarterly Journal of Micro-
scopical Science. Edited by E. RAY
LANKESTER, M.A., LL.D., F.R.S. ; with
the co-operation of ADAM SEDGWICK,
M.A., F.R.S., W. F. R. WELDON, M.A.,
F.R.S., and SYDNEY J. HICKSON, M.A.,
F.R.S. Each Number, IDS.

TheMicrotomist'sVade-Mecum:

A Handbook of the Methods of Micro-
scopic Anatomy. By ARTHUR BOLLES
LEE, Assistant in the Russian Laboratory
of Zoology at Villefranche-sur-Mer (Nice).
Fifth Edition. 8vo, 155.

Photo-Micrography

(Guide to the Science of). By EDWARD
C. BOUSFIELD, L.R.C.P. Lond. 8vo»
with 34 Engravings and Frontispiece, 6s.

A Treatise on Physics.

By ANDREW GRAY, LL.D., F.R.S., Pro-
fessor of Natural Philosophy in the
University of Glasgow. Vol. I. Dyna-
mics and Properties of Matter. With
350 Illustrations. Svo, 153.

An Introduction to Physical

Measurements, with Appendices on
Absolute Electrical Measurements, &c.
By Dr. F. KoHLRAUSCH, Professor at
the University of Strassburg. Third
English Edition, by T. H. WALLER,
B.A., B.Sc., and H. R. PROCTER, F.I.C.,
F.C.S. Svo, with 91 Illustrations, I2s. 6d.

Tuson's Veterinary Pharma-
copoeia, including the Outlines of
Materia Medica and Therapeutics.
Fifth Edition. Edited by JAMES BAYNE,
F.C.S., Professor of Chemistry and
Toxicology in the Royal Veterinary
College. Crown Svo, 7s. 6d.

The Veterinarian's Pocket Re-
membrancer : being Concise Direc-
tions for the Treatment of Urgent or
Rare Cases. By GEORGE ARMATAGE,
M.R.C.V.S. Second Edition. Post Svo,
3s.

Chauveau's Comparative Anat-
omy of the Domesticated Ani-
mals, revised and enlarged, with the Co-
operation of S. ARLOING, Director of the
Lyons Veterinary School. Edited by GEO.
FLEMING, C.B., LL.D., F.R.C.V.S.,
late Principal Veterinary Surgeon of the
British Army. Second English Edition.
Svo, with 585 Engravings, 313. 6d.

Human Nature, its Principles
and the Principles of Physio-
gnomy. By PHYSICIST. Part I.,
fmp. i6mo, 2s. Part II. (completing the
work), 2s. 6d.

the general editorship of CHALMERS
12 Volumes, 205. each net. Vols. I. to VI.

LONDON: 7. GREAT MARLBOROUGH STREET.

INDEX TO J. & A. CHURCHILL'S LIST.

Allen's Chemistry of Urine, n

Commercial Organic Analysis, 13

Armatage's Veterinary Pocket Remembrancer, 14
Auld's Pathol. Researches, 2

Barnes (R.) on Obstetric Operations, 3

— on Diseases of Women, 3

Beale (L. S.) on Liver, 6

Slight Ailments, 6

___ Urinary and Renal Derangements, 12

Beale (P. T. B.) on Elementary Biology, 2

Beasley's Book of Prescriptions, 5

Druggists' General Receipt Book, 5

Pharmaceutical Formulary, 5

Bell on Sterility, 4

Bentley and Trimen's Medicinal Plants, 5
Bentley's Systematic Botany, 5
Berkart's Bronchial Asthma, 6
Bernard on Stammering, 7
Berry's (Jas.) Thyroid Gland, 9

(R. J.) Regional Anatomy, i

Bigg's Short Manual of Orthopaedy, 9

Birch's Practical Physiology, 2

Bloxam's Chemistry, \z

— — — Laboratory Teaching, 12

Bousfield's Photo- Micrography, 14

Bowlby's Injuries and Diseases of Nerves, 9

Surgical Pathology and Morbid Anatomy, 9

Brockbank on Gallstones, 8

Brown's (Haydn) Midwifery, 3
Ringworm, n

(Campbell) Practical Chemistry, 13

Bryant's Practice of Surgery, 8

Bulkley on Skin Diseases, 10

Burckhardt's (E.) and Fenvvick's (E. H.) Atlas o

Cystoscopy, n

Burdett's Hospitals and Asylums of the World, 2
Butler-Smythe's Ovariotomies, 4
Butlin's Operative Surgery of Malignant Disease, u
Malignant Disease of the Larynx, n

Sarcoma and Carcinoma, 1 1

Buzzard's Diseases of the Nervous System, 7
Peripheral Neuritis, 7

Simulation of Hysteria, 7

Cameron's Oils, Resins, and Varnishes, 14

Soaps and Candles, 14

Carpenter and Dallinger on the Microscope, 14
Cautley's Infant Feeding, 4
Charteris' Practice of Medicine, 6
Chauveau's Comparative Anatomy, 14
Chevers' Diseases of India, 5
Churchill's Face and Foot Deformities, 9
Clarke's Eyestrain, 10
Clouston's Lectures on Mental Diseases, 3
Clowes and Coleman's Quantitative Analysis, 13

Elmntry Practical Chemistry, 13

Clowes' Practical Chemistry, 13

Coles on Blood, 6

Cooley's Cyclopaedia of Practical Receipts, 14

Cooper on Syphilis, 12

Cooper and Edwards' Diseases of the Rectum, 12

Corbin and Stewart's Physics and Chemistry, 12

Cripps' (H.) Ovariotomy and Abdominal Surgery, 9

Cancer of the Rectum, 12

Diseases of the Rectum and Anus, 12

Air and Faeces in Urethra, 12

Cripps' (R. A.) Galenic Pharmacy, 4
Cuffs Lectures to Nurses, 4

Cullingworth's Short Manual for Monthly Nurses, 4
Dalby's Diseases and Injuries of the Ear, 10

Short Contributions, 10

Dana on Nervous Diseases, 7

Day on Headaches, 8

Domville's Manual for Nurses, 4

Doran's Gynaecological Operations, 3

Druitt's Surgeon's Vade-Mecum, 8

Duncan (A.), on Prevention of Disease in Tropics, 5

Dunglison's Dictionary of Medical Science, 12

Edwards' Chemistry, 14

Ellis's(T. S.) Human Foot, 9

Encyclopaedia Medica, 14

Fagge's Principles and Practice of Medicine, 6

Fayrer's Climate and Fevers of India, 5

Fenwick (E. H.), Electric Illumination of Bladder, n

Tumours of Urinary Bladder, n

— Ulceration of Bladder, n

Fenwick (E. H.), Symptoms of Urinary Diseases, ir
Fenwick's(S.) Medical Diagnosis, 6

Obscure Diseases of the Abdomen, 6

Outlines of Medical Treatment, 6

Ulcer of the Stomach and Duodenum, 6~

• The Saliva as a Test, 6

Fink's Operating for Cataract, 9

Fowler's Dictionary of Practical Medicine, 6

Fox (G. H.) on Skin Diseases of Children, 10

Fox (Wilson), Atlas of Pathological Anatomy of Lungs, 6

Treatise on Diseases of the Lungs, 6

Frankland and Japp's Inorganic Chemistry, 13
Fraser's Operations on the Brain, 8
Fresenius Qualitative Analysis, 13
Quantitative Analysis, 13

Galabin's Diseases of Women. 3

Manual of Midwifery, 3

Gardner's Bleaching, Dyeing, and Calico Printing, 14
Brewing, Distilling, and Wine Manuf., 14

Glassington's Dental Materia Medica, 10
Godlee's Atlas of Human Anatomy, i
Goodhart's Diseases of Children, 4
Gorgas' Dental Medicine, 10
Gowers' Diagnosis of Diseases of the Brain, 7

Manual of Diseases of Nervous System, 7

Clinical Lectures, 7

Medical Ophthalmoscopy, 7

Syphilis and the Nervous System, 7

Granville on Gout, 7

Gray's Treatise on Physics, 14

Green's Manual of Botany, 5

Vegetable Physiology, 5

Greenish's Materia Medica, 4

Groves' and Thorp's Chemical Technology, 14

Guy's Hospital Reports, 7

Habershon's Diseases of the Abdomen, 7

Haig's Uric Acid, 6

Diet and Food, 2

Harley on Diseases of the Liver, 7
Harris's (V. D.) Diseases of Chest, 6
Harrison's Urinary Organs, n
Hartridge's Refraction of the Eye, 9
Ophthalmoscope, 9

Hawthorne's Galenical Preparations of B.P., 5
Heath's [njuries and Diseases of the Jaws, 8

Minor Surgery and Bandaging, 8

Operative Surgery, 8

Practical Anatomy, i

Surgical Diagnosis, 8

Hedley's Therapeutic Electricity, 5
Hellier's Notes on Gynaecological Nursing, 4
Hewlett's Bacteriology, 3
Hill on Cerebral Circulation, 2
Holden's Human Osteology, i

Landmarks, i

Holthouse on Strabismus, 9
Hooper's Physicians' Vade-Mecum, 5
Horton-Smith on Typhoid, 7
Hovell's Diseases of the Ear, 10
Human Nature and Physiognomy, 14
Hyslop's Mental Physiology, 3
Impey on Leprosy, 10

Ireland on Mental Affections of Children, 3
Jacobson's Male Organs of Generation, 12

Operations of Surgery, 8

Jellett's Practice of Midwifery, 3

Gynaecology, 3

Jessop's Ophthalmic Surgery and Medicine, o

Johnson's (Sir G.) Asphyxia, 6

Medical Lectures and Essays, 6

(A. E.) Analyst's Companion, 13

Journal of Mental Science, 3

Kellogg on Mental Diseases, 3

Kelynack's Pathologist's Handbook, i

Keyes' Genito-Urinary Organs and Syphilis, 12

Kohlrausch's Physical Measurements, 14

Lancereaux's Atlas of Pathological Anatomy, 2

Lane's Rheumatic Diseases, 7

Langdon-Down's Mental Affections of Childhood, 3

Lawrie on Chloroform, 4

Lazarus-Barlow's General Pathology, i

Lee's Microtomist's Vade Mecum, 14

Lewis (Bevan) on the Human Brain, 2

Liebreich (O.) on Borax and Boracic Acid, 2

Liebreich's (R). Atlas of Ophthalmoscopy, 10

[Continued on the next page.

LONDON: 7, GREAT MARLBOROUGH STREET.

INDEX TO J. & A. CHURCHILL'S LIST — continued

Lucas's Practical Pharmacy, 4

MacMunn's Clinical Chemistry of Urine, n

Macnamara's Diseases and Refraction of the Eye, 9

Macnamara's Diseases of Bones and Joints, 8

McNeill's Epidemics and Isolation Hospitals, 2

Malcolm's Physiology of Death, 9

Marcet on Respiration, i

Martin's Ambulance Lectures, 8

Maxwell's Terminologia Medica Polyglotta, 12

Maylard's Surgery of Alimentary Canal, 9

Mayne's Medical Vocabulary, 12

Microscopical Journal, 14

Mills and Rowan's Fuel and its Applications, 14

Moore's (N.) Pathological Anatomy of Diseases, i

Moore's (Sir W. J.) Family Medicine for India, 5

Manual of the Diseases of India, 5

Morris's Human Anatomy, i

Anatomy of Joints, i

Moullin's (Mansell) Surgery, 8
Nettleship's Diseases of the Eye, 9
Notter's Hygiene, 2
Ogle on Tympanites, 8
Oliver's Abdominal Tumours, 3

Diseases of Women, 3

Ophthalmic (Royal London) Hospital Reports, 9

Ophthalmological Society's Transactions, 9

Ormerod's Diseases of the Nervous System, 7

Parkes' (E.A.) Practical Hygiene, 2

Parkes' (L.C.) Elements of Health, 2

Pavy.'s Carbohydrates, 6

Pereira's Selecta e Prescriptis, 5

Phillips' Materia Medica and Therapeutics, 4

Pitt- Lewis's Insane and the Law, 3

Pollock's Histology of the Eye and Eyelids, 9

Proctor's Practical Pharmacy, 4

Pye-Smith's Diseases of the Skin 10

Ramsay's Elementary Systematic Chemistry, 13

Inorganic Chemistry, 13

Richardson's Mechanical Dentistry, 10
Richmond's Antiseptic Principles for Nurses, 4
Roberts' (D. Lloyd) Practice of Midwifery, 3
Robinson's (Tom) Eczema, n

Illustrations of Skin Diseases, n

Syphilis, ii

Ross's Diseases of the Nervous System, 7
St. Thomas's Hospital Reports, 7
Sansom's Valvular Disease of the Heart, 7
Scott's Atlas of Urinary Deposits, n
Shaw's Diseases of the Eye, 9
Shaw- Mackenzie on Maternal Syphilis, 12
Short Dictionary of Medical Terms, 12
Silk's Manual of Nitrous Oxide, 10
Smith's (Ernest A.) Dental Metallurgy, 10
Smith's (Eustace) Clinical Studies, 4

Disease in Children, 4

Wasting Diseasesof Infants and Children, 4
Smith's (F. J.) Medical Jurisprudence, 2

bmith's (J. Greig) Abdominal Surgery, 8
Smith's (Priestley) Glaucoma, 10
Snow's Cancer and the Cancer Process, n
- Palliative Treatment of Cancer, n
— — — Reappearance of Cancer, n
Solly's Medical Climatology, 8
Southall's Materia Medica, 5
Squire's (P.) Companion to the Pharmacopoeia, 4
-- London Hospitals Pharmacopoeias, 4
-- Methods and Formulae, 14
Starling's Elements of Human Physiology, 2
Sternberg's Bacteriology, 6
Stevenson and Murphy's Hygiene, 2
Sutton's (J. B.), General Pathology, i
Sutton's(F.) Volumetric Analysis, 13
Swain's Surgical Emergencies, 8
Swayne's Obstetric Aphorisms, 3

Pathology and Treatment of Ringworm, 10

- on Psilosis or " Sprue," 5
Thompson's (Sir H.) Calculous Disease, n

- Diseases of theUrinaryOrgans,!!
--- Lithotomy and Lithotrity, n
-- Stricture of the Urethra, n
--- Suprapubic Operation, n

- -- ; -- Tumours of the Bladder, n
Thome's Diseases of the Heart, 7

Thresh' s Water Analysis, 2

Tilden's Manual of Chemistry, 12

Tirard's Medical Treatment, 6

Tobm's Surgery, 8

Tomes' (C. S.) Dental Anatomy, 10

Tomes' (J. and C. S.) Dental Surgery, 10

Tooth's Spinal Cord, 7

Treves and Lang's German-English Dictionary, 12

Tuke's Dictionary of Psychological Medicine, 3

Tuson's Veterinary Pharmacopoeia, 14

Valentin and Hodgkinson's Practical Chemistry, 13

Vintras on the Mineral Waters, &c., of France, 8

Wagner's Chemical Technology, 14

Wallace on Dental Caries, 10

Walsham's Surgery : its Theory and Practice,

Waring's Indian Bazaar Medicines, 5

Practical Therapeutics, 5
Watts' Organic Chemistry, 12
West's (S.) How to Examine the Chest, 6
Westminster Hospital Reports, 7
White's (Hale) Materia Medica, Pharmacy, &c., 4
Wilks' Diseases of the Nervous System, 7
Wilson's (Sir E.) Anatomists' Vade-Mecum, i
Wilson's (G.) Handbook of Hygiene, 2
Wolfe's Diseases and Injuries of the Eye, 9
Wynter and Wethered's Practical Pathology, i
Year- Book of Pharmacy, 5
Yeo's (G. F.) Manual of Physiology, 2

N.B.—J. Sf A. Churchill's larger Catalogue of about 600 works on Anatomy,
Physiology, Hygiene, Midwifery, Materia Medica, Medicine, Surgery, Chemistry,
Botany, $c. $c., 'with a complete Index to their Subjects, for easy reference,
will be forwarded post free on application.

AMERICA. — /. fy A. Churchill being in constant communication with
various publishing houses in America are able to conduct negotiations
Ja'vourable to English Authors.

LOADONs i,

BOROUGH STREET.

I MVKRSITY OF CALIFORNIA LIBRARY

95829

>LOGY

THE UNIVERSITY OF CALIFORNIA LIBRARY

^liifiwS' :Sv ^;^:;>r
'li-: