DAIRY BACTERIOLOGY

DAIRY
BACTERIOLOGY

BY

ORLA-JENSEN

DR. PHIL.

PROFESSOR OF TECHNICAL BIOCHEMISTRY IN THE POLYTECHNIC

COLLEGE, COPENHAGEN.
FORMERLY DIRECTOR OF THE SWISS EXPERIMENTAL DAIRY STATION

TRANSLATED

FROM THE SECOND DANISH EDITION, WITH ADDITIONS
AND REVISIONS

BY

P. S. ARUP

B.SC. (LOND.), F.I.C.

CHIEF CHEMIST TO ENGLISH MARGARINE WORKS (1919) LIMITED

WITH 70 ILLUSTRATIONS

PHILADELPHIA
P. BLAKISTON'S SON & GO.

1012 WALNUT STREET
1921

0

Printed in Great Britain.

TRANSLATOR'S PREFACE

THE success of Professor Orla- Jensen's " Dairy Bacteriology,"
in the original as well as in the German, Dutch, and Finnish
translations, has led me to hope that an English translation might
find a welcome. The author has had the unique experience of an
intimate knowledge of the highly-evolved dairy industries of two
countries so widely different as Denmark and Switzerland, and,
as the reader will find, he is keenly interested in the valuable
work which has been carried out in English-speaking countries,
particularly in America.

The translation has been based on the second Danish edition
of 1916, which is the standard text-book on its subject for Danish
students of dairying. As Professor Orla-Jensen has spared him-
self no pains in correcting and adding to the text in order to bring
it thoroughly up-to-date, the present edition may be regarded as
an entirely new one. Certain portions of purely local interest
have been omitted, though the number of omissions on this
account have been very few ; the scientific character of the
work as a whole renders it of international interest.

At the present time, when the question of pure milk is beginning
to attract the attention which it deserves in the United Kingdom,
it is hoped that the work may offer some useful suggestions to
those engaged on the difficult problems which the whole question
involves. In spite of the modesty of the author's claims I
consider that his book should also convey something to those who
can read between the lines, and on this account I venture to hope
that not only students, but also dairymen and all others who
have to deal with milk and dairy products, whether from the
medical, veterinary or analytical side, will find something of
interest in its pages.

In conclusion, I wish to express my thanks to Mr. D. R. Wood,
Public Analyst for the County of Somerset, for kindly reading
the MS. and offering useful hints.

PAUL S. ARUP.

Liverpool.

524744

AUTHOR'S PREFACE

IN order to avoid the inclusion of superfluous matter in this
work on Dairy Bacteriology, I have, on the one hand, omitted
from the teaching of dairy practice everything which is not
exactly of bacteriological interest, and, on the other hand, from
the bacteriological side everything which is not of interest in dairy
practice. It is, therefore, assumed that the reader will have
obtained from other sources a knowledge of dairy practice, and
that through the study of this book he will obtain some guidance
in the bacteriological technique.

This work is not a treatise, but only a text-book, for which
reason one must not expect to find in it mention of every single
microorganism to which is ascribed the power of affecting milk
or dairy products in one way or another. I have purposely
scrutinised the literature on the subject as closely as possible,
and through my own investigations I have sought to form an
opinion as to the actual conditions. The book is in all details
built up on my own experiences, culled from twenty -five years
of research work, and I, therefore, feel that I can confidently
recommend it to my pupils.

ORLA-JENSEN.
Copenhagen.

CONTENTS

TRANSLATOR'S PREFACE
AUTHOR'S PREFACE .
LIST OF ILLUSTRATIONS

PAGE
v

vii
xi

PART I.— GENERAL

CHAPTER I
MICROORGANISMS AND FERMENTATIONS

General Description of Microorganisms, their Nutrition and
Growth — Fermentations — Enzymes — Variability — Methods
of Culture and Examination.

CHAPTER II

BACTERIA .

Classification — The Lactic Acid Fermentation and the Classi-
fication of the Lactic Acid Bacteria — The Propionic Acid
Fermentation — The Butyric Acid Fermentation — The Putre-
faction Process.

CHAPTER III

YEASTS AND MOULDS

The Species commonly found in Milk and Dairy Products and
their Action.

26

49

PART II

CHAPTER I
CLEANING AND THE PROCUREMENT OF MILK ....

Cleaning and Disinfecting — The Procurement of Milk — Clean-
liness in Milking, etc. — Microorganisms in Milk, their Origin.

CHAPTER II
THE NORMAL AND ABNORMAL MICROFLORA OF MILK .

The Normal Flora : Microorganisms in Milk, their Growth in
and Influence on Milk under various Conditions — Abnormal
Flora : Primary Milk Defects — Diseases of Cows, their Influence
on Milk — Secondary Milk Defects — Infections with Disease
Organisms — Abnormal Behaviour, Appearance, Taste and
Smell.

54

63

x CONTENTS

CHAPTER III

PAGE

THE PRESERVATION OF MILK AND ITS TREATMENT FOR DIRECT CON-
SUMPTION 77

Cooling - - Sterilisation - Pasteurisation - - Bacteria in
Pasteurised Milk — Condensed Milk — Dried Milk — Preservatives
— Milk for Town Supplies — Cleaning, Pasteurising and Cooling
— Milk for Infant Feeding.

CHAPTER IV

THE APPLICATIONS OF THE LACTIC ACID FERMENTATION IN THE

DAIRY INDUSTRY 100

Dietetic Preparations — The Souring of Cream — Lactic Acid
Cultures and Starters — Treatment of Starter Milk — Buttermilk
— Separated Milk — Whey for Preserving Manure.

CHAPTER V
THE NORMAL AND ABNORMAL MICROFLORA OF BUTTER . .120

The Normal Flora : Butter from Unpasteurised and Pasteurised
Cream — Changes in Butter on Keeping — Fat Hydrolysis —
Rancidity — The Abnormal Flora : Original Butter Defects —
Secondary Defects.

CHAPTER VI
THE RIPENING PROCESSES OF THE DIFFERENT CHEESES . .128

Action of Rennet and Lactic Acid in making the Curd and in
Ripening — Types of Ripening Processes — Changes undergone
by Constituents of Cheese in Ripening — Hard Rennet Curd
Cheeses — Moulds in Cheese Ripening — Soft Rennet Curd
Cheeses — Smeared and Mouldy Cheeses- -Acid Curd Cheeses.

CHAPTER VII

DEFECTS OF CHEESE ..... .151

Origin of Defects — Sponginess — Defects in Colour, Taste and
Smell.

CHAPTER VIII
THE GRADING OF MILK .157

Taste and Smell — Estimation of Dirt — Leucocyte Test —
Catalase Test — Rennet Test — Acidity — Fermentation Test —
Reductase Test — Combined Reductase and Fermenting Test,
with special Reference to the Interpretation of Results.

INDEX • 175

LIST OF ILLUSTRATIONS

FIG. PAGE

1. Acetic Acid Bacteria (Bacterium aceti). (After Hansen) . . 2

2. Brewers' Yeast (SaccJiaromyces cerevisice). (After Hansen) . 2

3. Penicillium glaucum ........ 2

4. Growing Top Yeast. (After Mitscherlich) .... 3

5. Clostridia. (After Prazmowski.) Spore Formation in the

Common Butyric Acid Bacteria ..... 4

6. Plectridia. (After Migula.) Spore Formation in the Tetanus

Bacterium ......... 4

7. Spore Formation in Wine Yeast. (After Hansen} ... 4
SA. Mucor Mucedo. (After Kerner) ...... 5

8s. Transverse Section of a Single Sporangium. (After Brefeld) . 5

9. Penicillium glaucum. (After Brefeld) ..... 6

10. Oidium lactis. (After de Bary) ...... 6

11. Mucor racemosus with Chlamydospores. (After Brefeld) . . 6

12. Zoogloea of " Leuconostoc," a Coccus which forms slimy lumps

in Cane Sugar Solutions. (After Zopt) .... 7

13. The Bacterium of "Long Milk." Capsule Stage. Milk not yet

slimy .......... 7

14. The Bacterium of " Long Milk." Slimy Stage ... 8

15. The Bacterium of " Long Milk." Milk less slimy, but beginning

to go thick owing to acid formation .... 8

16. Stab Cultures. A. The Anthrax Bacterium (aerobic). B. The

Swine Erysipelas Bacterium (facultative anaerobic). (After
Migula.) C. The Tetanus Bacillus (obligate anaerobe).

(After Ball) 10

17. Involution Forms of Bacterium aceti produced by cultivation at

39° to 41° C. (After Hansen) . . ' . . .16

18. Freudenreich Flask 17

19. PetruschJcy Flask 17

20. Incubator with Petri Dishes, Plugged Test Tubes, Apparatus

for Measurement of Gas Production, etc. . . .18

21. Desiccator for Anaerobic Cultivation . . . . .19

22. Stribolt's Method for the Cultivation of Anaerobic Bacteria.

(After Salomonsen) ....... 22

23. J or gens en' s Moist Chamber ....... 23

24. A. Cornet's Forceps. B. Kuhne's Forceps .... 24

25. Bacterium pyocyaneum, monotrich. (After Migula) . . 26

26. Bacterium syncyaneum, lofotrich. (After Migula) ... 26

27. Bacterium typhosum, peritrich. (After Migula) ... 27

28. Bacterium vulgare, peritrich. (After Migula) .... 27

29. Various Cocci. (After Flugge) 27

30. Vibrio cTiolerce. (After Migula) 28

31. Various Screw-shaped Bacteria. (After Flugge) ... 28

32. Actinomyces bovis. (After Bostr'om) 29

xii LIST OF ILLUSTRATIONS

FIG. PAGE

33. Thermobacterium bulgaricum. Grown in sterile milk. Stained

with methylene blue. The grains are round and dark
blue 30

34. Thermobacterium bulgaricum. Grown in milk pasteurised by

heating previously to 80° C. for half an hour. Stained
with methylene blue. The grains are long-drawn and red.
Capsule clearly shown . . . . . . .31

35. Thermobacterium helveticum from Emmental Cheese. (After

Freudenreich) . . . . . . . .34

36. Streptobacterium casei from Danish Dairy Cheese ... 34

37. Streptococcus cremoris (Streptococcus laclicus) .... 35

38. Streptococcus cremoris (Storch's No. 18) . . . . .36

39. Streptococcus thermophilus ....... 36

40. Streptococcus lactis ........ 37

41. Betacoccus from thin juice from Nakskov Sugar Factory . . 37

42. Streptococcus liquefaciens (Freudenreich' s Micrococcus casei

amari) ......... 37

43. Streptococcus liquefaciens (Escherich's Streptococcus coli gracilis) 37

44. Various Betacocci in Stab Cultures in Cane-sugar Gelatine . 38

45. The Coli Bacterium which spoilt the Milk and Butter at Duelund

Dairy in 1888. (After C. O. Jensen) .... 40

46. Bacterium acidi propionici (a) . . . . . .41

47. Bacterium acidi propionici (b). Grown at 39° C. ,. . .41

48. Bacillus subtilis. (After Migula.) a. Active rods before spore

formation, b. A piece of film. c. Kods with spores.

d. Stained to show the flagellse 46

49. Mycoderma cerevisice. (After Holm) ..... 50

50. M onilia nigra. (After Burri and Staub) . . . .51

51. Two Varieties of Penicillium brevicaule 52

52. Kotting Swede containing numerous Pectin -fermenting Plec-

tridia 57

53. Milking Eoom at Fauerholm ...... 59

54. Gurler's and Stadtmiiller' s Milk Pails. (After Conn) . . 61

55. Cooling and Straining of Milk at Fauerholm. In the foreground

Ice is being put into the bottom of BuscJc's Milk Pail . 78

56. Jonas Nielsen's Steriliser 80

57. Heine's Cleaning Separator . . . . . . .91

58. Laval's Cleaning Separator 92

59. Hygienic Stopper 93

60. Orla-Jensen's Household Pasteurising Apparatus ... 97

61. Bacterium bifidum, distinctly branched. From faeces of bottle-

fed infant . . . . . . . . 101

62. Kefir Grains, natural size. (After Kern] . . . .103

63. Section through a Kefir Grain 104

64. Leucocyte Sediment from the Milk of a Cow suffering from

Streptococcic Mastitis . . . . . . .159

65. Catalase Test Apparatus 160

66. Gelatinous Types ...... .164

67. Blown Types ...... .164

68. Spongy Types . . . 165

69. Cheesy Types . . . . . . . . . 165

70. Apparatus for Reductase and Fermenting Test to take 200

Samples . 168

DAIRY BACTERIOLOGY

PART I
GENERAL

Chapter I
Microorganisms and Fermentations

BY microorganisms or microbes, we understand all organisms
which are too small to be seen by the naked eye ; they remained
unknown until it had become possible to construct strong magni-
fying glasses or the combination of lenses which make up the
microscope. Bacteria were first observed by the Dutch optician
Leeuwenhoek in 1675, but no real knowledge of their nature was
gained until the latter third of the last century, when Pasteur
carried out his classical researches. Bacteriology is thus a
comparatively new science ; nevertheless it has already revolu-
tionised both medical science and the technology of the fermenta-
tion industries, among which may be included the dairying
industry.

As far as the fermentation industries are concerned, only three
groups of microorganisms come into consideration, Bacteria,
Yeasts and Moulds. Bacteria and yeasts proper are unicellular,
while moulds are generally multicellular . Unicellular organisms
may be united in chains, but the individual cells of the chain show
no differentiation as regards structure or function, excepting such
modifications as may be due to differences in age or nutrition.
In the moulds, on the other hand, it is possible to distinguish
between two groups of cells ; first, the mycelium, which is con-
cerned with the nutrition of the organism, forming a tangled
network in the nutrient medium, and second, the fine thread-like
shoots which bear the reproductive organs and generally project
out of the medium so that the spores may be carried away by air
currents (see Fig. 8).

DAIRY BACTERIOLOGY

The size of microorganism is given in micromillimetres, or
microns, designated by the Greek letter /z, being one-thousandth
part of a millimetre. Bacterial cells generally measure from
0-5 to 2 JJL in thickness. Some idea of their minuteness may be
formed on consideration of
the fact that the space of
a cubic millimetre will hold
one thousand million bac-
teria of average size. From
Figs. 1, 2 and 3 it will be
seen that the cells of yeasts
and moulds are from five
to ten times as large as
those of the bacteria. V^ \ \W V V

FIG. I.— Acetic Acid Bacteria (Bac-
terium aceli). (After Hansen.)

FIG. 2. — Brewers' Yeast (Saccharomyces
cerevisice). (After Hansen.)

All magnified 1,000 times (x 1,000).

FIG. 3. — Penicillium glancum.

GROWTH AND REPRODUCTION

Under normal conditions a bacterial cell will soon grow to its
maximum size, often becoming more or less elongated in form,
when a cross partition appears dividing it transversely into two

MICROORGANISMS AND FERMENTATIONS 3

separate daughter cells of equal size, which in their turn grow
longer and divide, the process being repeated indefinitely. This
mode of reproduction is known as fission. Yeast cells, on the
other hand, reproduce by budding, or gemmation ; small round
outgrowths appear and continue to develop until they attain the
size of the mother cell (see Fig. 2).

In the moulds we generally find growing points as in the higher
plants ; only the outer cells are capable of reproduction, which
mostly takes place by fission and only rarely by budding.

Spore Formation. — As the higher plants form seeds, so many of
the microorganisms form spores, the function of which is to
preserve the species under adverse conditions. Spores are
accordingly more resistant to the influence of desiccation, poisons
and high temperatures than the ordinary cells. Two distinct

\i

FIG. 4. — Growing Top Yeast. (After Mitscherlich.) I. At 7 p.m. II. Next
morning, 8 a.m. III. 9 a.m. IV. 10.15 a.m. V. Noon. VI. 3.30 p.m.
VIT. 8 p.m. VIII. 9 p.m. IX. 10 p.m. X. 11 p.m.

types are found : endogenous and exogenous spores. It is com-
paratively seldom that the cell becomes completely changed into
an Arthrospore, absorbing reserve food material freely and
thickening its wall1. Endogenous spores are formed inside a cell
while exogenous spores arise as constrictions on the end of a cell.
Bacteria only form spores of the first mentioned type, only one
spore appearing in each cell. If the spore causes the cell to bulge
out locally, drumstick (Plectridium) or club-shaped (Clostridium)
formations arise according as the spore lies at the end or in the
middle of the cell. Germination in a direction approximately at
right angles to the length of the cell is said to be lateral, and
germination from the end polar. The yeasts likewise form
endospores only, but several spores, up to ten in number, may be

1 In this connection it may be mentioned that, according to Preisz'
investigations ( " Centralblatt f. Bakt.," 1 Abt., 1918, Bd. LXXXIL, p. 321),
the spore proper (the refractile body) is always only condensed reserve food
material ; outside this may be demonstrated a mass of protoplasm from
which growth proceeds.

1—2

4 DAIRY BACTERIOLOGY

formed in a cell. As a rule, two, three or four spores are found
together (Fig. 7), and they generally grow by budding like the
ordinary yeast cells.

When moulds form endospores (e.g., the various Mucors) this
takes place in specially shaped cells known as sporangia, which
may contain several hundred spores (Figs. 8 A and SB). As a rule,
however, the moulds form exogenous spores ; these are known as
Conidia when formed as constrictions on specialised spore bearing
members (Conidiophores) (see Figs. 3 and 9), and as Oidia when
formed on the ordinary branches of the mycelium (Fig. 10).
Considered as spores, the oidia are less characteristic than the
conidia ; the Chlamydospores, however, are exceptional, being
unusually thick walled oidia, which may be formed anywhere in
the mycelium (Fig. 11). In a chain of conidia, the outer members
are generally the oldest, and the inner ones the youngest.

STRUCTURE AND CHEMICAL COMPOSITION OF CELLS

The main constituent of the cell is a viscous jelly-like trans-
parent solution of proteins, known as protoplasm, which is the
living substance of the cell. It is an extremely complex mixture of
unstable compounds, which are largely built up of combinations

FIG. 5.— Clostridia. (After Praz- FIG. 6.— Plectridia (After

mowski.) Spore Formation
in the Common Butyric Acid
Bacteria.

Migula.) Spore Forma-
tion in the Tetanus
Bacterium.

All x 1,000.

FIG. 7. — Spore
Formation in
Wine Yeast.
(After Hans en.)

of a considerable number of different amino acids. The feeble
acid and basic properties of these acids are also characteristic of
the proteins, a circumstance which enables the latter to combine
loosely with a vast number of other substances, thus fulfilling a
condition necessary for the vital processes. Oil drops and other
reserve foodstuffs are often found in the protoplasm, and as a rule

MICROORGANISMS AND FERMENTATIONS 5

there are one or more spaces containing cell sap (vacuoles) . Young
and vigorous cells are nearly filled with protoplasm ; old or
starved cells have only a thin layer on the cell wall. All cells
possess a nucleus which, although similar to the rest of the proto-
plasm in composition, is of a still more complex nature. When
reproduction takes place, the nucleus is divided between the two
new cells.

The protoplasm is bounded by a cell wall which grows thicker
as the cell ages. It is composed not of cellulose, as in the case of

FIG. 8 A. — Mucor Mucedo. (After Ker-ner.)

FIG. SB. — Transverse Section of a
Single Sporangium. (After Bre-
feld.)

the higher plants, but of an allied carbohydrate which has been
named hemicellulose ; it becomes coated with a nitrogenous
substance identical with or very similar to chitin, the chief con-
stituent of the shells or exoskeletons of the Crustacea and the
insects. Mucin, a substance which forms a sticky mass with water,
may also be present. As these substances are typical of many of
the lower forms of animal life, it may be surmised that the bacteria
form a sort of link between animal and plant life, and this theory
derives some support from a consideration of their mode of life.
The cell wall may sometimes swell up considerably ; in cases where

DAIRY BACTERIOLOGY

FIG. 9. — PenicilUum glaucum. (After Brefeld.)

FIG. 10. — Oidium lactis. (After de Bary.)

FIG. 11. — Mucor race-
mosus with Chlamy-
dospores. (After
Brefeld.)

MICROORGANISMS AND FERMENTATIONS 7

it retains its sharp contour, it is known as a capsule. Numbers
of cells may thus form large jelly-like masses, known as zoogloea
Occasionally the cell walls will become assimilated with the
surrounding liquid, which then becomes slimy or ropy throughout
(see Figs. 13, 14 and 15).

FIG. 12. — Zoogloea of " Leuronostoc," a Coccus which forms slimy lumps in Cane
Sugar Solutions. (After Zopt.) x 1,200.

FIG. 13. — The Bacterium of " Long Milk." Capsule Stage.

X 1,000.

Milk not yet slimy.

NUTRITION

The following twelve elements enter into the composition of all
organisms : — Oxygen, Hydrogen, Carbon, Nitrogen, Sulphur,
Phosphorus, Chlorine, Sodium, Potassium, Calcium, Magnesium
and Iron. It follows that substances composed of these elements
are necessary for the normal growth of microorganisms. Some of
them, e.g., sulphur, calcium and iron, are only used in such minute

DAIRY BACTERIOLOGY

FIG. 14.— The Bacterium of " Long Milk/' Slimy Stage, x 1,000.

FIG. 15.— The Bacterium of " Long Milk/' Milk less slimy, but beginning to
go thick owing to acid formation, x 1,000.

quantities that it is unnecessary to add them specially to the
nutrient media if ordinary tap water is used. Chlorine and

MICROORGANISMS AND FERMENTATIONS 9

sodium are only required in appreciable amounts by organisms
normally living in salt liquids, i.e., marine and many pathogenic
organisms. Like all other organisms, the microorganisms are
able to derive the necessary sulphur, phosphorus and metallic
elements from inorganic salts. These salts are found in the
required proportions in plant ashes, and formerly it was the
practice to add them to nutrient media in this form. If micro-
organisms are cultivated in milk or the extracts of meat or plants
they will generally be well supplied with the requisite inorganic
nutriment.

Nitrogen. — Microorganisms may be divided into two main
groups according to their ability or inability to assimilate all the
nitrogen they require from inorganic sources. A few species
belonging to the former group can assimilate nitrogen from the
air, and thus incidentally improve the soil for plant growth. The
great majority of these, however, require their nitrogen in the
form of ammonia or nitrates ; they are represented by the typical
water bacteria, acetic acid bacteria, and many yeasts and moulds.
Belonging to the group which cannot grow in the absence of
proteins or the immediate decomposition products of proteins
are many of the putrefactive bacteria and the true lactic acid
bacteria.

Carbon. — Microorganisms fall into two groups with respect to
carbon assimilation. A few species of soil bacteria which oxidise
ammonia to nitrates resemble the higher plants in being able to
utilise atmospheric carbon dioxide as their sole source of carbon.
Unlike the higher plants, however, they are best able to carry out
this process in darkness — no bacteria tolerate direct sunlight.
Most microorganisms require organic sources of carbon. Some
putrefactive bacteria can obtain all the carbon they need from
proteins, thus presenting a broad analogy to the carnivorous
animals. As a rule, however, microorganisms require special
sources of carbon such as carbohydrates, alcohols and organic
acids, and often show marked preferences for certain substances ;
thus some species will only grow in presence of certain sugars, a
circumstance which is put to advantage in distinguishing closely
related species from one another.

Oxygen. — Oxygen and Hydrogen are assimilated from most of
the nutrient substances and from water. Water must be regarded
as the most important of the nutrient substances ; it constitutes
about four-fifths of the total substance of the microorganism,
besides which it is absolutely essential to life as a solvent and
distributing medium for all the other nutrient substances. Like
animal and plants, most microorganisms are capable of assimilating
oxygen direct from the atmosphere. The analogy is only super-

10

DAIRY BACTERIOLOGY

ficial, for nearly all microorganisms can live without atmospheric
oxygen for shorter or longer periods, provided that other sources
of energy are available, and towards some species oxygen acts as
a poison. Organisms which cannot live without atmospheric
oxygen are defined as aerobic, and those which can do without it
as anaerobic. The latter group is subdivided into facultative and
obligate anaerobes, which are respectively helped or hindered in
their growth by the presence of atmospheric oxygen. As a rule

FIG. 16.— Stab Cultures. A. The Anthrax Bacterium (aerobic). B. The Swine
Erysipelas Bacterium (facultative anaerobic). (After Mifjula.) C. The
Tetanus Bacillus (obligate anaerobe). (After Ball.)

aerobic organisms will, sooner or later, form a film on the surface
of the culture solution or spread over the surface of solid media,
only penetrating slightly below the surface. Facultative anae-
robes will grow equally well at all .depths or on the surface,
while obligate anaerobes only thrive below a certain depth.
These relations are illustrated by the accompanying figures.
The true lactic acid bacteria generally tolerate air, but thrive
best in its absence ; in stab cultures, therefore, they will not
spread over the surface, but penetrate the medium evenly from
all points.

MICROORGANISMS AND FERMENTATIONS U

FERMENTATION PROCESSES

The assimilated nutrient substances are employed, partly in
building up the cells during growth and reproduction, and partly
as sources of energy for these and other vital processes. All
substances are not equally suited for both purposes, and distinction
may be made between the nutrient substances proper and those
which merely function as sources of energy. As long as active
reproduction is taking place, a considerable proportion of the
latter will be pressed into service as cell-building material, later
on to be decomposed into simpler products, when the energy thus
liberated will be utilised by the microorganisms. That portion of
the period occupied by the reproductive process during which no
appreciable amounts of decomposition products are separated is
known as the incubation period. The decomposition products, which
bear a certain general comparison with the substances contained in
the urine of animals, are known as fermentation products, and fer-
mentations are to be regarded as decompositions brought about
by microorganisms. Exceptions occur when the organic sub-
stances undergo complete oxidation to carbon dioxide and water ;
such changes are looked on as respiration processes analogous to
those observed among the higher organisms. Fermentation thus
implies only a partial decomposition through the agency of micro-
organisms. It is evident that such a process entails a certain loss
of efficiency, and consequently a larger consumption of material.
Complete decomposition is only effected by aerobic microorganisms,
more particularly by the moulds which, as previously mentioned,
are able to form aerial shoots. Moulds only function as fermenta-
tion organisms when excluded from access to air, under which cir-
cumstances many of them produce alcohol, like the yeasts. It
will now be understood that the most typical of the fermentation
organisms must be facultative or obligate anaerobes. Fermenta-
tions are also known which are simple oxidations, such as the
acetic acid fermentation. Processes of this nature require a
plentiful supply of air, and can, therefore, only be carried out by
aerobic organisms.

Fermentations and fermentation organisms are often named
after their most characteristic products, e.g., alcohol and butyric
acid fermentations, lactic acid or butyric acid bacteria. Some-
times the name is derived from the substance attacked, e.g.,
cellulose fermentation or cellulose bacteria. The former system
is the more logical. Soil formation is the natural bacterial
decomposition of plant remains into products, which, as a
rule, are not stinking. Putrefaction is the decomposition of
animal remains into products which are usually stinking. As

12 DAIRY BACTERIOLOGY

animal tissues consist for the most part of proteins, the term putre-
faction has become synonymous with the bacterial decomposition
of proteins. Formerly, when nothing was understood of the real
nature of the different fermentations, they were generally looked
on as chemical changes which apparently originated spontaneously,
and which might improve a natural product. In cases where the
natural product was spoilt instead of being improved the process
was known as putrefaction. As alcohol fermentation is the most
widely-known example, it is often supposed that the evolution
of gas is characteristic of fermentations in general ; but in its
modern sense the term fermentation includes a number of
processes, such as the lactic acid and acetic acid fermentations,
in which no gas is produced.

ENZYMES

The manifold chemical changes due to microorganisms are
carried out through the agency of certain special substances,
known as enzymes. These are bodies of unknown composition,
possibly more complex than the proteins, which may be separated
more or less easily from the protoplasm, and which are able in
small amounts to induce certain chemical changes in relatively
large amounts of material. Enzyme action may be extracellular
or intracellular, according as the enzyme acts outside (exoenzyme)
or inside (endoenzyme) the cell in which it has been formed. To
the former class belong the digestive enzymes of animals ; as the
function of these is to prepare the food for assimilation, they
must necessarily act outside the assimilating cells. Many micro-
organisms form enzymes which are analogous to these. On the
other hand, the typical fermentation enzymes decompose or
oxidise the assimilated food material inside the cell, and must
therefore act inside the cell, so that the energy liberated in the
process may be directly available to the cell.

Distinction was formerly made between unorganised ferments,
i.e., enzymes such as are contained in the digestive juices, and
organised ferments, by which were understood the microorganisms.
It was not until 1897, when Buchner succeeded in separating from
yeast the enzyme which brings about alcoholic fermentation
(zymase), that it became clear that all fermentations are ultimately
due to enzyme action.

The activity of enzymes increases with the temperature, but
above certain limits the enzymes, like the proteins, become de-
natured and lose their special properties. The optimum for most
enzymes lies between 35° and 65° C. In aqueous solution they
are generally rendered completely inactive at temperatures

MICROORGANISMS AND FERMENTATIONS 13

between 65° and 85° C. In the dry state, some enzymes will
stand temperatures of about 150° C.

Enzymes resemble living organisms in being destroyed by heat ;
on the other hand they are unaffected by many substances which
act as poisons towards living protoplasm, such as toluene, ether,
chloroform, ethereal oils, phenol, salicylic acid and benzoic acid.
Some such substance, usually toluene, may therefore be added to an
enzyme solution to keep it sterile ; boric acid is added to preparations
of rennet. Most enzymes act best in feebly acid, all but neutral,
solutions ; pepsin is exceptional in acting in presence of the
appreciable proportions of acid contained in the gastric juice.

As a rule, enzymes are named after the substances on which they
act, thus maltase acting on maltose, lactase acting on lactose, urease
on urea, etc . In some cases, however, previous usage has established
other names. Diastase or, better, Amylase (amylum = starch) con-
verts starch into dextrin and maltose ; it occurs plentifully in malt,
and plays an important pa;rt in the brewing and distilling industries ;
during the mashing process it converts the starch into fermentable
products. It also occurs in the saliva and the pancreatic juice of
mammals ; as it only makes its appearance some time after birth,
newly-born animals are unable to assimilate starchy foods.
Invertase hydrolyses saccharose, i.e., cane or beet sugar, into dex-
trose (d. glucose) and Isevulose (d. fructose), a mixture which is
found in honey and many fruits. Invertase, maltase, lactase and
amylase are the chief carbohydrate splitting enzymes. Enzymes
which hydrolyse fats into glycerol and fatty acids are known as
lipases, and those which split proteins as proteolytic enzymes.
Proteins may be hydrolysed by stages, each stage yielding a
simpler product, thus : proteins to metaproteins to proteoses to
peptones to polypep tides to amino acids. The first decomposition
product of casein is paracasein, which, according to Hammarsten,
differs from natural casein in being precipitated by the small
amounts of calcium salts found in solution in normal cow's milk1.
A microorganism which secretes proteolytic enzymes will always
coagulate milk, and then gradually redissolve, i.e., peptonise, the
precipitated paracasein ; unless the organism in question also
belongs to the acid-producing group, the process of peptonisation
will tend to produce an alkaline reaction in the milk.

Peptonising organisms generally liquefy gelatine more or

1 The coagulation of casein by rennet appears to be analogous to the
coagulation of other proteins by heat. See Orla Jensen, " Kemiske Under-
sogelser over Maelkens Koagulering og Koaglets Oploselighed i Saltvand."
Det kgl. danske Videnskabernes Selskabs Oversigter, 1914, No. 4.
Chemische Untersuchimgen iiber die Gerinnung der Milch und iiber die
Loslichkeit des Ger. in Salzwasser. Zeitschr. f. Physiol. Ohem.. 1914,
Bd. XCIII., p. 283.

14 DAIRY BACTERIOLOGY

less readily, being usually referred to as liquefying organisms.
The proteolytic enzymes of the gastric juice, chymosin (rennet)
and pepsin, only carry the cleavage of proteins as far as the peptone
stage ; the trypsin of the pancreas effects complete hydrolysis to
amino acids. The erepsin of the intestinal juice is particularly
thorough in its action, but only attacks proteins which have
already been partially broken down by other enzymes. Casein
is exceptional in being directly attacked by erepsin1.

Oxidising and reducing enzymes are known as oxidases and
reductases ; they are of prime importance in connection with the
breathing of animals and the reduction of carbohydrates to fats.
Storch's well-known reaction, by which it is possible to ascertain
whether milk has been heated to over 80° C. or not, depends
on the presence in milk of an oxidase which is destroyed
at 80° C., and can transfer the loosely bound oxygen of
hydrogen peroxide to paraphenylene diamine or certain other
colourless substances giving coloured products. Paraphenylene
diamine gives a violet or, in the presence of casein, a blue
colour. Many vegetable products, such as potatoes, fruits or
fungi, contain oxidases as well as substances which yield coloured
products on oxidation ; hence the darkening in colour which
takes place when they are cut into pieces and left exposed to
air ; if the material has been boiled, the oxidase will have been
destroyed, and no darkening occurs. Reductases, on the other
hand, are generally recognised by taking advantage of the fact that
many substances, such as methylene blue, are decolorised on
reduction. Catalase occupies a position intermediate between
oxidases and reductases ; it decomposes hydrogen peroxide, but
the oxygen so liberated will not act on paraphenylene diamine or
similar oxidisable substances. Catalase only occurs in small
quantities in milk, but plentifully in blood ; as constituents of
the blood always pass into milk, which is drawn from diseased
udders, an abnormally copious evolution of oxygen in the catalase
test (see p. 159) is to be regarded as a bad sign.

Certain poisons, easily destroyed by heat, known as toxins, are
closely related to the enzymes. They are found in a few plants,
in poisonous reptiles and other animals, and are frequently
secreted by microorganisms. The various toxins are generally
the active agents in diseases due to pathogenic organisms. Healthy
living tissues are highly resistant towards enzyme action, and are
provided with special poisons, bactericidal substances, which repel

1 It follows that microorganisms which secrete no other enzymes than
erepsin attack casein, but do not liquefy gelatine. Conversely, the author
has found that micrococci which liquefy alkaline, but not neutral or acid
gelatine, do not peptonise casein.

MICROORGANISMS AND FERMENTATIONS 15

bacterial invasion unless they have previously been weakened by
the action of toxins. To counteract this, the higher organisms
secrete substances, known as anti-toxins, which play an important
part in recovery from disease and the subsequent more or less
permanent period of immunity from the disease in question.

VARIABILITY

Microorganisms are classified together when they resemble one
another in structure and habits of living and growth. Many
microorganisms, however, are very liable to vary in appearance
and in their biological characteristics, so that we may often have
to deal with dissimilar forms which nevertheless belong to the
same species. As we have really no sure guide enabling us to
distinguish between essential and non-essential characteristics, it
may often be difficult to determine whether we are dealing with
different species or only variants of the same species. Undoubtedly,
the biological properties, such as those which relate to nutrition,
growth and energy, are more important in this respect than out-
ward form, and, again, the means by which energy is produced are
of more importance than the particular raw materials from which
the energy is derived. Though it may be granted that an organism
which derives its energy from an alcoholic fermentation instead of
a complete oxidation to carbon dioxide and water would hardly be
classified among the higher organisms, yet it is often observed that
closely related organisms are able to utilise different nutrient
materials. If a lactic acid organism undergoes variation, it does not
follow that it will start producing alcohol or butyric acid instead
of lactic acid, but it may easily lose its power to ferment a poly-
saccharide, such as milk sugar, which requires a special enzyme,
lactase, to convert it into fermentable material. In other words*
the intracellular enzymes are much more characteristic of a given
microorganism than the extracellular enzymes. It may be assumed
that the chief products of a fermentation process will always be
the same under similar conditions, but the by-products may vary
considerably according to the condition of the organism. The
ability to produce certain substances which affect the taste, smell
or colour of the medium is especially variable. The nature of the
cell wall may also vary, so that an organism may appear sometimes
with and sometimes without a capsule, or the cell wall may dis-
integrate into a mucilaginous mass. Bacteria are particularly
liable to undergo this form of degeneration — especially the lactic
acid organisms— generally as the result of over-nutrition ; an
analogous instance is seen in fatty degeneration in animals.

Temperature is an important factor in determining the form of

16

DAIRY BACTERIOLOGY

microorganisms ; on cultivation at temperatures approaching the
maximum, acetic acid bacteria may be induced to undergo a
striking transformation. In the case of acid-producing bacteria,
the high concentrations of acid produced in the medium will also
tend to cause the appearance of abnormal forms, known as
involution forms, and in old cultures of lactic acid bacteria strange
elongated, swollen or even branched cells may often be observed.
Prolonged cultivation under adverse conditions is the best method
of producing new varieties ; thus Emil Christian Hansen induced
yeasts permanently to lose their capacity of forming spores by

FIG. 17. — Involution Forms of Bacterium aceti produced by cultivation at
39° to 41° 0. (After Hansen.) x 1,000.*

cultivating for several generations at temperatures above the
maximum for sporulation.

METHODS OF CULTURE

It is not possible to compound a universal nutrient medium, for
some substances may be essential for the well-being of certain
organisms and poisonous to others. While certain water bacteria
will only thrive in very dilute solutions of organic nutrient matter,
yeasts and moulds thrive best in wort and fruit juice, milk bacteria
in milk, and pathogenic bacteria in meat broth or blood serum.
Such natural media are of variable and partly unknown composi-
tion, and, therefore, do not lend themselves so well to the study of
fermentations from the chemical point of view. For this purpose
artificial media containing as far as possible only substances of
known composition in definite proportions are the best ; for
bacteria which require proteins, a preparation of peptonised fibrin
known as Witte peptone is used ; unfortunately this involves the

MICROORGANISMS AND FERMENTATIONS 17

introduction of a substance of which the composition is not quite
constant ; only the best brands, prepared under standard condi-
tions, should be used. Most milk bacteria thrive in the following
solution : — Tap water, 1 litre ; sodium chloride, 2 grams ; dipo-
tassium phosphate, 2 grams ; magnesium sulphate, 1 gram ;
dextrose, 20 grams ; peptone, 20 grams.

This solution has an alkaline reaction, and phosphoric acid
should be added until it only just turns litmus paper blue. For
the cultivation of water bacteria and most pathogenic organisms,
the medium should be distinctly alkaline ; for yeasts and moulds
it should be acid. The media are distributed in flasks or test
tubes closed with plugs of non-absorbent cotton wool and sterilised
by heating in an autoclave for a quarter of an hour at 1 10° to 120° C.

The Freudenreich flask is the most convenient vessel to use, as
the contents are not so easily dried up or infected as in test tubes.
If the nutrient solution is to be used for the investigation of acid

FIG. 18.— -Freudenreich Flask. FIG. \<d.—Petruschlcy Flask.

formation, then 10 c.c. are introduced into each test tube, and the
inoculated solution is titrated when the maximum acidity is
certain to have been reached, say, after fourteen days at 30° C.
The amount of acid produced by lactic acid bacteria increases
with the amount of nitrogenous nutrient material in the medium ;
these organisms thrive better on casein peptone l than on Witte's
peptone, but most of the rod-shaped lactic acid bacteria thrive
best on an extract of autolysed yeast 2.

In order that the cultures may be kept at definite temperatures
the bacteriological laboratory must be equipped with several
incubators heated by gas or electricity, and, if necessary, cooled
by water circulation, in such a way that the temperature is
automatically kept constant.

In order to examine microorganisms, to obtain them as pure
cultures or to determine their numbers, they must first be isolated.
This may be accomplished by Koch's method of plating, which
consists in distributing a definite small quantity of the liquid
containing the organisms in a solid transparent medium which
J, 2 See footnotes at end of this Chapter.

DB. 2

FIG. 20. — Incubator with Petri Irishes, Plugged Test Tubes, Apparatus for Measurement of

Gas Production, etc.

MICROORGANISMS AND FERMENTATIONS

19

has previously been melted, and then allowing this to set in a thin
layer. The isolated germs, fixed in position, will after some time
multiply into colonies visible to the naked eye. From the number
of colonies, the number of organisms per cubic centimetre of the
original liquid can be calculated. Care must be taken that the tem-
perature of the melted medium is not high enough to cause injury to
the organisms. For milk bacteria the best media are whey or the
solution recommended above, with the addition of 12 per cent,
of gelatine or 1J per cent, of agar. Gelatine may be regarded as
a protein, whereas agar, a product obtained from certain seaweeds,
is composed of carbohydrates known as pectins. Gelatine is to
be preferred, as it can be made to
yield clearer media than agar, but as
it melts slightly over 20° C. or lower
if it has been heated too frequently
or at too high a temperature, it
must be replaced by agar when deal-
ing with organisms which only thrive
at higher temperatures. If litmus
or chalk be added, acid-producing
organisms may at once be recognised.
Solid media are usually allowed to
set in Petri dishes, flat dishes with
lids, which have been sterilised dry
at 150° to 170° C. Organisms which
do not grow in presence of oxygen
are best cultivated in Burri's tubes,
which are closed at one end by a
rubber stopper and at the other by a
plug of cotton wool ; on removing
the stopper, the agar column may
be shaken out and examined for

colonies. Some obligate anaerobes form characteristic surface
colonies ; to obtain these, the Petri dish must be kept in an
atmosphere of hydrogen or in a vacuum. The oxygen which
gradually leaks in may be absorbed by alkaline pyrogallol solu-
tion. The simplest plan is to use a vacuum desiccator, which
is tilted during exhaustion in such a way that the chemicals
(5 grams of pyrogallol and 50 c.c. of 10 per cent, caustic potash)
in the sulphuric acid container are only mixed when the desic-
cator is stood straight after evacuation.

As a rule, the liquids to be examined contain so many germs that
they must be diluted with sterilised water before plating, in order
to avoid overgrowth. The number of colonies to be aimed a't is
40 to 200 per plate, and it will generally be necessary to make

FIG. 21. — Desiccator for Anae-
robic Cultivation.

20 DAIRY BACTERIOLOGY

several dilutions to achieve the desired result. A number of
100 c.c. Freudenreich flasks sterilised in the autoclave with 50 c.c.
of water should be prepared in readiness, and also some dry
sterilised plugged pipettes, graduated in quarters of a cubic
centimetre. When making counts in milk, the following dilutions
may be made : —

Dilution 1. 1 c.c. milk in 49 c.c. water, of which 0-5 c.c. in Petri

dish No. 1.
„ 2. 0-5 c.c. of Dilution 1 in 49*5 c.c. water, of which

0-5 c.c. in Petri dish No. 2.
„ 3. 0-5 c.c. of Dilution 2 in 49'5 c.c. ^water, of which

0-5 c.c. in Petri dish No. 3.

The number of organisms per cubic centimetre is obtained
by multiplying the number of colonies on plates 1, 2 or 3 by
100, 10,000 or 1,000,000, respectively. Butter is measured by
means of a rounded platinum spoon, holding exactly 0-25 c.c.
The spoon and the knife used for scraping off the surface of the
butter must be sterilised by passing through a Bunsen flame
before use. If the water used for dilution is warmed to about
40° C., the butter will readily melt from the spoon and become
distributed on shaking. Cheese gives more trouble, as it must be
weighed and comminuted ; about 1 gram of cheese may be
accurately weighed into a Freudenreich flask containing 50 c.c.
of water and ground up by repeatedly pressing it out in the water
by means of a flamed glass rod. An alternative method, resembling
that for determining the soluble nitrogenous matter, is to grind
10 grams of the cheese with sterile water at 40° to 50° C. in a
sterile mortar, pouring the emulsion into a sterilised 250-c.c.
measuring flask and making up to the mark with sterile water.
This method is to be preferred as a more representative sample
is obtained, a matter of some importance as the organisms are
usually very unevenly distributed in cheese. Further, the cheese
is more easily ground up in the mortar than in the flask, while the
number of organisms introduced by infection from the air will be
negligible compared with the large number present in the cheese.

In all cases uniform working methods should be adopted in
order that the results may be comparable ; the colonies should
be counted after incubating for a definite time, say seven days at
20° C. The counting is facilitated by placing the plate on a
squared glass plate resting on a black surface.

For use when travelling, Petruschky flasks (Fig. 19), sterilised
with the proper amount of nutrient gelatine, are more convenient
than Petri dishes ; they are generally used in water examinations,
which are best carried out on the spot, as an increase in the

MICROORGANISMS AND FERMENTATIONS 21

number of microorganisms generally takes place during transit.
Bacterial multiplication may, however, be avoided to a con-
siderable extent if the samples are packed in ice. For counts of
water samples the peptone gelatine recommended above may
be used, modified as follows : the sugar is omitted, and the
neutralised medium is treated with 15 c.c. of 10 per cent, soda
solution per litre. In these, as in all bacterial counts, it is im-
possible to get all the germs present to grow on the same medium ;
as many typical water bacteria, for example, the thread-forming
and sulphur organisms, will not grow on gelatine at all, the counts
must not be regarded as representing the actual numbers of
organisms present, though they furnish useful indications as to
the relative purity of samples if the same working conditions are
adhered to throughout. It must also be remembered that long
chains of bacteria or large pieces of mould mycelia only yield
single colonies. The method is thus subject to considerable
error, a fact which may easily be demonstrated by comparing the
results with those obtained by direct microscopical counts.
According to Barthel's investigations, 2 to 200 times as many
organisms are found by direct counts as by plating according to
Skar's method 1. Pasteurised milk cannot be examined by the
direct method as there are no means of distinguishing by micro-
scopic inspection between living and dead bacteria 2.

Furthermore, one cannot be certain that any particular colony
may not have arisen from several different species which may
have adhered together or become intertwined in the original
liquid. If, therefore, a pure culture is desired, it will be necessary
to sow plates from a well-isolated colony, and only when the
resulting colonies are found to consist exclusively of the desired
organism is it possible to be sure that the individual colonies are
pure. It is still safer to make a single cell the starting point, a
method which is described under the next heading.

Pure cultures are best preserved in Freudenreich flasks on
nutrient agar. As a rule stab cultures are made by piercing the
medium with a platinum wire on the point of which a trace of the
culture has been picked up. Streak cultures are made of organisms
which require free access to air ; the agar is allowed to solidify
in a slanting position so as to expose a large surface, and the
infected wire is drawn lightly across the surface. In order to
protect stab cultures of anaerobic bacteria from the air, the
cotton wool plug is flamed and pushed down the test tube till it

1 " Milcliwirtschaftliches Zeiitralblatt," 1912, p. 455.

2 H. W. Conn (U.S. Public Health Keport, 1915, No. 295) gives some
valuable information concerning the limits of accuracy in^the bacterio-
logical analysis of milk.

22

DAIRY BACTERIOLOGY

nearly reaches the agar surface. A plug of absorbent cotton wool
soaked in alkaline pyrogallol is then inserted a short distance
above the first plug, and the tube is closed by a rubber stopper or
a cotton wool plug soaked in melted paraffin wax (see Fig. 22).
As soon as the cultures have made growth they must be kept at
a low temperature or they will quickly die. The acid-producing
bacteria keep better, the smaller the amount of sugar contained
in the agar medium. Lactic acid bacteria may be preserved
alive for years if not more than 0-25 per cent, of dextrose is
employed, but it is safer to sow into fresh media every month.
If it is desired to keep lactic acid bacteria in perfect condition
with respect to their action on milk, they
a should always be cultivated in milk, and

reinoculated into fresh sterilised milk as fre-
" quently as possible.

Inoculation is carried out by means of a
platinum wire sealed into a glass rod or fixed
into a screwed aluminium holder ; the end of
G the wire may be shaped as desired ; when

dealing with liquids it is looped. The wire is
always flamed immediately before use.
d In order to demonstrate the presence of

certain organisms which may be greatly out-
numbered by other species, the method of
plating fails ; it will be necessary to use the
enrichment method of inoculating a small
quantity of the material into a medium which
Method for the Cul- f avours the growth of the organism in question
tivation of Anae- to the disadvantage of others ; the cumulative
(After Salomons™*) effect of several reinoculations into the same
medium will often be the production of a pure
culture of the desired organism, or at any rate a culture in which
the organism can easily be recognised. Such methods are
employed, among others, for demonstrating the presence of
pathogenic organisms.

METHODS OF EXAMINATION

Of all the methods in use for the determination of the species
of microorganisms, microscopic examination is one of the most
important, for until the shape and size of the cells have been
noted it will not even be possible to say definitely whether one is
dealing with a yeast or a bacterium. The more highly developed
the organism, the completer will be the information to be gained
by microscopic examination, especially if the various stages of

MICROORGANISMS AND FERMENTATIONS

23

development are studied. In this way the moulds may readily
be identified. Greater difficulties are encountered in dealing
with the two other groups, which are much simpler in morpho-
logical structure ; in these groups cultural and especially bio-
chemical characteristics are more important as means of identifica-
tion. It has already been mentioned that the appearance of a
stab culture will reveal the character of the organism with respect
to atmospheric oxygen, and as to whether it produces proteolytic
enzymes or coloured products. Sometimes even the shape and
general macroscopic appearance of the colony may be so charac-
teristic as to afford in itself a means of identification. By varying
the composition of the nutrient medium characteristic preferences
may be discovered — for example, an investigation may be made
to determine which sugars are fermented with the production of
acid or acid and gas. Satisfactory results, however, can only be

FIG. 23. — Jorgensen's Moist Chamber.

got by finding out what fermentation products are formed under
certain conditions.

For detailed descriptions of the microscope, works on micro-
scopy or optics may be consulted. It need only be mentioned
here that high magnification is achieved as the product of the
magnifications due to the two systems of lenses, the eye-piece
(ocular) and the objective. The object to be examined is placed
on a glass slide and covered with a square or round cover slip of
thin glass. For high magnifications, a drop of cedar- wood oil of
the same refractive index as glass is placed between the cover
slip and the special immersion objective, an arrangement which
entails less loss of light than when air intervenes, and which is
essential when magnifications of 1,000 diameters are required.

In order to determine whether an organism is motile or non-
motile, it is observed in a hanging drop adhering to the under side
of a cover slip placed over a hollow slide ; in this way the growth
and development of the organism may also be observed. In
examining moulds, which require a plentiful supply of oxygen,
the cover slip is separated from the slide by a glass ring, and the
whole arrangement, known as a moist chamber, is sealed together

24 DAIRY BACTERIOLOGY

with vaseline ; a drop of water is placed on the slide to keep the
air in the chamber moist, and to prevent evaporation of the drop
on the cover slip. The moist chamber is also employed in making
single cell cultures as follows : a miniature gelatine plate is made
on a special cover glass marked with numbered squares, and with
the aid of the microscope the most isolated cells are sought out
and their positions noted with respect to the markings on the
cover glass (see Fig. 23). The colonies which grow in these
positions may then be sown separately into suitable media. The
first yeast culture of undoubted purity was prepared in this vay
by Emil Christian Hansen, an achievement of far-reaching im-
portance in the brewing industry. The method is not adapted

FIG. 24. — A. Cornet's Forceps. B. Kilhne's Forceps.

for the preparation of pure cultures of the bacteria, as these are so
much smaller than the yeasts.

Bacteria may be seen more clearly if they have been stained
after fixation on the cover slip. Fixation is accomplished by
picking up the cover slip by the edges, between the thumb and
first finger, and twice drawing it slowly through a Bunsen flame ;
as long as the fingers are not burnt, the preparations will not be
overheated ; the bacteria are killed, and "on this account stain
better. If the culture medium is whole milk, the preparation
should be freed from fat by immersion in chloroform before
staining. The cover glass is immersed for a few minutes in the
staining solution in a small dish or watch glass, washed with pure
water, laid on the slide and dried on the surface by means of filter
paper ; it should be handled by means of the Kuhne's forceps
(Fig. 24). The stain may also be dropped on to the cover slip
held clamped by means of the Cornet forceps (Fig. 24). The

MICROORGANISMS AND FERMENTATIONS 25

stains most used are alcoholic solutions of methylene blue or
fuchsine. The former, which must not be diluted too much with
water, is specially suited for preparations from milk, as it does
not stain the casein strongly ; if fuchsine is used, a red patch may
be formed in which it is impossible to distinguish the bacteria. As
fuchsine stains very^deeply, it is used in dilute solution ; its
staining capacity is increased by the addition of phenol (carbol
fuchsine). Tubercle bacteria, which are coated with wax, and
intracellular spores, which are very resistant to staining, may be
stained by warming with carbol fuchsine, and when thus stained
they retain the colour so persistently that mineral acids fail to
remove it. Methylene blue is also suited better than fuchsine for
staining broth cultures. Acid cultures should, however, first be
neutralised, unless the methylene blue solution has been treated
with a little alkali.

Gram's method of staining is especially useful in the identification
of bacteria ; it depends on the fact that microorganisms after
treatment with gentian violet are stained a violet black by a
solution of iodine in potassium iodide, and retain this stain more
or less completely on treatment with absolute alcohol l. Bacteria
which retain the stain are known as Gram- positive, and those
which lose it as Gram-negative. Most yeasts and moulds and all
true lactic acid bacteria are Gram-positive, while most of the
harmful milk bacteria are Gram-negative. The Gram method
lends itself well to the examination of lactic acid cultures and
sour milk preparations in general, as the casein is completely
Gram- negative.

Instead of staining, Bum's Indian ink method may be adopted ;
a drop of the liquid to be examined is placed on a cover slip, mixed
with a little sterile liquid Indian ink, allowed to dry, mounted in
water, and examined. If the liquid is acid it should be neutra-
lised, as the colloidal Indian ink is coagulated by acid. Treated
in this way, the organisms show up white on a black background.
Permanent preparations may be made by mounting in Canada
balsam instead of water.

1 V. Jensen (" Hospitalstidende," 1912, No. 20) recommends a 0-5 per
cent, solution of methyl violet in water, and a solution of 1 gram of iodine
and 2 grams of potassium iodide in 100 c.c. of water.

Notes to page 17.

1 100 gm. sugar free casein (precipitated by acid) were digested for a
week at blood heat with a litre of water containing 4-6 per cent. HOI and
2 gm. pepsin. The resulting solution contained, after neutralisation,
sterilisation and filtration, about 1 per cent. N and 1-2 per cent. NaCl.

2 At 50° C. the digestion is completed in 24 hours. The sugar dis-
appears at the same time. The highly acid solution contains about 2 per
cent. N.

Chapter II
Bacteria

MANY bacteria differ from the yeasts and moulds in being
capable of independent motion, which is accomplished by means
of fine threads, known as flagellce, growing from the protoplasm.
The monotricha have a single flagellum growing from one end,
and the lofotricha a tuft of flagellae similarly situated ; the peri-
tricha have flagellse all round the cell. Motility may be restricted
to certain stages of development, chiefly the period of active
growth ; on the other hand, a large number of bacteria appear to
be non-motile at all stages, and on this account it may be presumed
that they are without flagellse ; unfortunately these extremely
fine thread-like structures can only be rendered visible by means
of complicated methods of staining which sometimes fail. There
is, however, no doubt that the disposition of the flagellse affords

,

FIG. 25. — Racterinm pyocyaneum, monotrich. FIG. 26. — Bacterium syncyaneum, lo/o
(After Migula.) ' X 1,000. trich. (After Migula.) X 1,000.

the most important of the morphological methods of distinction,
and one on which the classification of the bacteria should primarily
be based. Thus the monotricha and the lofotricha require com-
paratively simple nutrient material, being chiefly water bacteria,
while the peri trich bacteria are typical fermentation organisms l.
According to their shape the bacteria are divided into sphere,
rod or screw forms. As the first-mentioned have no axis of length,
division is possible in three dimensions, and, as a matter of fact,
cases are known where division takes place in either one, two or

1 Orla Jensen. " Hovedlinierne i det naturlige Bacteriesystem." Viden-
skabernes Selskabs Oversigter, 1908, No. 5. Die Hauptlinien des Natiir-
lichen Bakteriensystems. Verlag von Gustav Fischer, Jena, 1909.

BACTERIA

27

three dimensions, resulting in the Streptococcus, Micrococcus and
Sarcina forms respectively, provided that the cocci remain united
after cell division has taken place. Streptococci may be com-
pared with strings of beads, Micrococci form tetrads, i.e., squares
composed of four spheres, and Sarcinse form bundles of cells.

IG. 27. — Bacterium typhosum, peritrich.
(After Migula.) X 1,000.

FIG. 28. — Bacterium vulgar e, peritrich.
(After Migula.) X 1,000. •

Gocci which adhere together in pairs are known as Diplococci,
d those forming irregular aggregates resembling clusters of
grapes are named Staphylococci ; these names, however, have no
systematic significance. Cocci which divide in more than one
direction do not as a rule stretch before division, but form two
hemispheres immediately after division. Many streptococci

a.

I

.;•
* '

FIG. 29.— Various Cocci. (After Flugge.} x 1,000.

behave similarly and resemble chains of spheroidal links, flattened
as though compressed along the direction of growth. Other
streptococci become distinctly elongated, forming oval or even
rod-shaped links ; an example is seen in Streptococcus lactis, the
commonest of the lactic acid bacteria. With the exception of a
few sarcinae, the cocci have never been found to form spores.

The rod-shaped bacteria are divided into the genera Bacterium
and Bacillus. Originally the term bacillus indicated long rods,

28

DAIRY BACTERIOLOGY

later, according to Migula, motile rods, while now it generally
denotes spore-forming rods.

The screw-shaped bacteria, which always have polar flagellse,
are classified into the genera Vibrio and Spirillum. The vibrios
are monotrich and form single curves, being comma- shaped ; the
spirilla are lofotrich and more or less screw- shaped. Certain
spirilla are said to be able to form spores.

Fiu. 30. — Vibrio cholerce. (After Migula.)

FIG. 31.— Various Screw-shaped Bacteria, (After Fliiyye.) X 1,000.

The above remarks may be condensed into the following tabular
classification into families and genera :—

Family 1. — Coccaceae, or Sphere Forms.

Genus Streptococcus, dividing in one direction.
Genus Micrococcus, dividing in two directions.
Genus Sarcina, dividing in three directions.

Family 2. — Bacteriaceae, or Rod Forms.

Genus Bacterium, without spores.
Genus Bacillus, with spores.

Family 3. — Spirillaceae, or Screw Forms.
Genus Vibrio, monotrich.
Genus Spirillum, lofotrich.

BACTERIA

29

As bacteria may lose the power to form spores, the distinction
between the bacterium and the bacillus groups is not sharp.
The distinctions between the micrococci and the sarcinae,
and between the vibrios and other monotrich rod forms, are
still more ill-defined. On the whole, the above system of
classification, due to Lehmann and Neumann, is unsatisfactory,
as it groups together organisms which have but little in
common and separates others which should be grouped
together. It is only mentioned here as being that which is

As far as the true lactic acid bacteria are
to use generic terms of more essential

may be classified the Sulphur
Bacteria and the Ray Fungi. Sulphur

FIG. 32. — Aetinomyces boris. (After
Bostrom.) X 1,000.

most commonly used,
concerned, I propose
significance.

Together with the bacteria
Bacteria, the Thread
bacteria are always present
where hydrogen sulphide is pro-
duced, for example in rotting
seaweed, where they transform
the poisonous gas into sulphates
necessary for plant growth.
Colourless and red species are
known in forms of almost end-
less variety. The thread bac-
teria, like the sulphur bacteria,
are typical water organisms ;
they are found attached to solid
objects, and their outer cell walls
form sheaths in which may be accumulated large amounts of iron.
They may cause stoppages in water pipes, and contribute towards
the formation of bog-ore. The ray fungi, or Actinomycetes, may
be regarded as being intermediate between the bacteria and the
moulds, especially the Mycomycetse (Mucorinse) ; on the one hand
their cells are as slender as those of the bacteria, on the other
hand they ramify and form oidia by constriction. They are of
very common occurrence in soil, contributing to its characteristic
smell. They may develop in butter, in which they produce a
similar smell 1. One variety, Aetinomyces bovis (Fig. 32), is the
cause of actinomycosis in cattle. The organisms of diphtheria,
and especially those of tuberculosis, are closely related to this
group.

The Spirilla are also typical water bacteria, being difficult to
cultivate artificially. As representing the vibrios, only the well-
known cholera vibrio need be mentioned (Fig. 30). In dairy

1 Orla Jensen, " Centralblatt fur Bacteriologie," 2 Abt., 1902, Bd. VIII.,
p. 250.

30

DAIRY BACTERIOLOGY

practice we have to deal almost exclusively with cocci and rod-
shaped organisms ; special attention will therefore be paid to
these groups in this work.

THE LACTIC ACID FERMENTATION

This is the most important fermentation in dairy practice. In
the true lactic acid fermentation, lactic acid is practically the sole
product arising from the fermented sugar. The chief by-products
are carbonic and acetic acids. The production of carbonic acid

FIG. 33. — Thermobacierium bulgaricum. Grown in sterile milk. Stained with
methylene blue. The grains are round and dark blue, x 1,000.

in recognisable amounts is a sign that the lactic acid bacteria are
in full vigour ; acetic acid, on the other hand, only appears in
appreciable quantities in stale cultures, or when the bacteria are
given a plentiful supply of air, either by aerating the liquid during
fermentation or by allowing it to expose a large surface to the air.
Pseudo lactic acid fermentations are characterised by the formation
of considerable quantities of volatile acids, gases and other by-
products, such as succinic acid and alcohol. In view of the fact
that in dairy practice only the true lactic acid fermentation can
be turned to economic advantage, while the pseudo lactic acid
fermentations generally cause trouble, the terms " true " and

BACTERIA 31

"pseudo" may be taken as corresponding with the concepts of
useful and harmful lactic acid bacteria.

Lactic acid was first studied systematically by the Swedish
chemist Scheele in 1782 1. It is a syrupy liquid, entirely soluble in
water, alcohol and ether. The most convenient method of
preparing it pure is to evaporate sour whey to a syrup, and to
extract this with ether. If chalk is stirred into the fermenting
liquid from time to time the acid will be neutralised, and thus
prevented from weakening the bacteria, which will then be able to
ferment the whole of the sugar. Lactic acid may be prepared on

FIG. 34. — Thermobacterium bulgaricum. Grown in milk pasteurised by heat-
ing previously to 80° C. for half an hour. Stained with methylene blue.
The grains are long-drawn and red. Capsule clearly shown X 1,000.

a large scale from a 10 to 20 per cent, solution of maltose, made
by saccharifying starch with extract of malt containing active
diastase. The solution is treated with chalk, sterilised and fer-
mented by a pure culture for a week or two at 50° C., the culture
being kept pure at this temperature. As calcium lactate is formed,
the solution gradually sets to a pasty mass of crystals, which is
pressed and decomposed with sulphuric acid ; after filtering off
the calcium sulphate the liquid is evaporated, preferably in vacuo,
till it contains 40 to 80 per cent, of lactic acid. The yield is about
75 per cent. Lactic acid and its salts are largely used as mordants
in the dyeing and tanning industries. Before the War, Germany

1 " Kgl. Vetenskaps Academiens nya Handlingar," Bd. III., p. 120.

32 DAIRY BACTERIOLOGY

produced 1,000 tons annually. Polarimetric observation reveals
three forms of lactic acid, dextro and Isevorotatory, and inactive.
The last- mentioned is a mixture of the two optically active acids
in equal proportions, and differs from these in the solubility of its
salts as well as in its optical properties 1. In order to identify a
lactic acid bacterium, it is necessary to know what form of lactic
acid it produces under certain conditions.

In 1857, Pasteur discovered that the lactic acid fermentation
was due to the action of certain bacteria 2, and soon after the
discovery of the method of making gelatine plate cultures, Hueppe,
in 1884, succeeded in isolating one of these bacteria3. This
organism, which was named Bacillus acidi lactici, is not one of the
true lactic acid bacteria, but belongs to the aerogenes group, which
will be dealt with later. The organism which plays the principal
part in the self -souring of milk had already been correctly described
by Lister, in 1878, as an oval diplococcus, and given the name
Bacterium lactis 4. A similar form was isolated by Grotenfeld in
1879, and named Streptococcus acidi lactici5. The importance of
this organism was, however, only recognised by £,eichmann in
18946, and by Gunther and Thierfelder7 in 1895, who demonstrated
that it produced dextro lactic acid in milk. Leichmann named it
very appropriately the bacterium of sour milk, Bacterium lactis
acidi. The large number of lactic acid bacteria now known really
only resemble one another in not forming spores, and therefore
being comparatively easily destroyed by heat.

The True Lactic Acid Bacteria. — These bacteria
ferment carbohydrates and higher alcohols to lactic acid ; they
only grow in presence of proteins or complexes of amino acids, and
not in presence of ammonium salts or single amino acids. They are
Gram-positive, non-motile, non-sporing rod or sphere forms. It
will only be possible here to mention the more important character-
istics by which nearly related forms are best distinguished. The
following is a brief summary of the results of the author's recent
researches on this subject 8 :—

1 The lactic acids are best identified by conversion into zinc salts. The
active zinc lactates rotate the plane of polarised light in senses opposite to
those due to the free acids, and they crystallise with two molecules of water,
corresponding to 12-9 per cent. H20, which is not driven off below 140° C.
The inactive^zinc lactate is much less soluble, and crystallises with three
molecules of water (18-2 per cent. HZ0), which is more readily driven off
on heating.

' Comptes rendues," tome 45, p. 913.

8

rfc.

Mitt. Kais. Ges.-Amt.," Bd. II., p. 309.

Trans. Path. Soc. of Condon," vol. 29, p. 425.

Fortschr. Medizin," Bd. VII., p. 121.

Milchzeitung," Bd. XXIII., p. 523.

Archiv. fur Hyg.," Bd. XXV., p. 164.

The Lactic Acid Bacteria." Monograph published in English by the

Danish Academy of Sciences, Copenhagen, 1919.

BACTERIA 33

(A) No catalase formation, nitrate reduction nor surface growth.

(a) Forming only traces of by-products in addition to lactic, acid.

Rod forms . . Genera Thermobacterium and Strepto-

bacterium.
Sphere forms . . Genus Streptococcus.

(b) Generally forming appreciable amounts of gas and other by-

products in addition to lactic acid.

Rod forms . . Betabacterium.
Sphere forms . . Betacoccus.

(B) Usually forming catalase, reducing nitrates and showing
surface growth.

Rod forms . . Microbacterium.
Sphere forms . . Tetracoccus 1.

Group B has been included because it forms a limiting group to
the lactic acid bacteria on the one side, just as the Coli and
Aerogenes group do on the other, though none of these can be
reckoned as true lactic acid bacteria. The last-mentioned group
includes group A, and also Bacterium bifidum, which, on account
of its branched form and obligate anaerobic character, occupies a
unique position.

Group A. — Although we may well suppose that the genera
Streptobacterium and Streptococcus and the genera Betabacterium
and Betacoccus are respectively more closely related to one another
than the various rod forms to one another, and the various sphere
forms to one another, we will, however, seeing that the rod forms
as a whole are stronger acid producers than the sphere forms, keep
up the old tradition and treat the rod and sphere forms separately.

(1) Rod Forms (named Lactobacilli by Beijerinck). — Short or long,
straight or curved rods which may grow out into long threads.
Sometimes they contain granules which stain more readily with
methylene blue than the rest of the protoplasm. They may
occur in pairs or chains of varying lengths and regularity of form.
To this group belong the most typical of the lactic acid bacteria,
which produce, and are able to stand, greater quantities of lactic
acid than the other lactic acid organisms. Some of them may
form as much as 3 per cent, of lactic acid in milk. The members
of the Thermobacterium genus are long rods, which thrive at 40° to
50° C. or even higher temperatures, but not below 22° C. ; they
will generally obtain predominance in milk kept above 40° C. They
form Isevo or inactive lactic acid, and with one exception, Tbm.

1 Common term for acid-forming Micrococci and Sarcmae.

D.B. 3

34 DAIRY BACTERIOLOGY

cereale (Bacillus Delbrucki), they attack casein to a considerable
extent, and thus come to play an important part in the ripening
of strongly scalded cheeses such as Emmental or Gruyere. To
this genus belong the strongest acid producers, e.g., Tbm. helve-
ticum (Bacterium casei e), which may produce over 2-7 per cent,
of inactive lactic acid, and which plays an important part in
the ripening of Emmental cheese. Tbm. bulgaricum (Bacillus
bulgaricus) forms up to 1-7 per cent, of Isevo acid, and Tbm.
jugurt forms as much inactive acid as Tbm. helveticum, and grows
in peculiar feathery- shaped colonies. These two last-mentioned
organisms occur in Bulgarian sour milk.

Organisms belonging to the Streptobacterium genus are short or
long chains of short or long rods, which as a rule have a maximum

FIG. 35.—Thermobacterium helveticum from FIG. 36. — Streptobacterium casei from
Emmental Cheese. (After Freudenreich.) Danish Dairy Cheese, x 1,000.

X 1,000.

of 35° to 40° C. ; they gradually come to predominate in dairy
products which are kept at temperatures below 35° C., and are
therefore frequently found in cheese. They form inactive or
dextro acid. Sbm. casei (Bacterium casei a) hydrolyses casein,
while Sbm. plantarum does not do so.

The Betabacteria almost always form inactive acid, and when
in a freshly isolated state, perceptible amounts of by-products ;
they have no action on casein, and as a rule do not grow well
in milk. As examples may be named Bbm. breve and longum
(Bacterium casei y and 8 respectively). The former ferments
arabinose strongly, and frequently also xylose ; it has a maximum
temperature of 38° C. The latter never ferments arabinose, but
frequently ferments xylose and raffinose ; its maximum tempera-
ture is 45° C. Bbm. caucasicum is the chief constituent of Kefir

BACTERIA

35

grains ; it forms appreciable amounts of acid together with yeast,
and its optimum temperature lies below 30° C.

(2) Sphere Forms. — The Streptococci are related to the Strepto-
bacteria, the points of similarity having been mentioned above.
They usually grow out into long chains when cultivated in broth,
but in milk and on solid media they may present varied appear-
ances. They grow well in milk, and always form dextro lactic
acid, though very seldom more than J to f per cent., i.e., not much
more than is required to coagulate milk. They are better able to
grow on the surface of solid media than the rod forms, being less
strictly anaerobic in character, but they do not form spreading

FIG. 37. — Streptococcus cremoris. (Streptococcus lacticus.}

colonies. With the exception mentioned below, the Streptococci
show very little tendency to hydrolyse casein, and completely lose
the power to do so if not cultivated in milk. Their optimum
temperature is as a rule 30° C., and many strains do not grow above
37° C. Sc. thermophilus, however, grows best at 40° C. ; it is irregular
in shape, and is easily isolated from milk which has been kept at a
fairly high temperature. The greatest interest attaches to Sc.
cremoris, which is used for ripening cream in butter making ; it
forms long chains in milk and grows best at 25° to 30° C., but not
at blood heat. Slime-producing strains are represented by the
" bacterium of long milk," and Sc. Hollandicus. The former of
these was first described by Gerda Troili-Petersson under the name

3—2

36

DAIRY BACTERIOLOGY

of Bact. lactis longi1, and the latter by Weigmann2 ; this organism
was to be found in the ropy whey formerly used in the manufacture
of Dutch (Edam) cheese. Sc. mastitidis (Sc. agalactice), which
produces notable quantities of lactic acid in milk, is the cause of
mastitis or inflammation of the udder in cows. It may be recog-
nised by the orange colouring matter which it produces after some
time in agar or broth with casein peptone and soluble starch.
Unlike Sc. cremoris, it ferments saccharose, maltose and dextrin. Sc.
pyogenes is a general term for a number of pathogenic streptococci
which do not coagulate milk ; they cause boils and many other
similar diseases in animals and human beings.

Belonging to this group are also the common bacteria of sour
milk, Sc. lactis (Bact. lactis acidi, Leichmann), which always obtain

FIG. 38. — Streptococcus
No. 18). >

•rmioris (Starch's
1,000.

FIG. 39. — Streptococcus thermophilus.

predominance in milk which is kept at ordinary room temperature,
appearing mostly as diplococci in milk. It ferments dextrin, but
not saccharose. Sc. fcecium is also a typical diplococaus form,
which grows even at 50° C., and is very common in the manure of
mammals. Other related organisms appear both as diplococci,
and as longer chains in the same culture. They are distinguished
according to their power to ferment a number of different sub-
stances such as glycerine, sorbitol, maltose, dextrin and salicin.
and generally also pentoses and saccharose ; their maximum
temperature is about 45° C. As an example may be mentioned Sc.
liquefaciens 3, which liquefies gelatine and produces a bitter taste
in cheese.

1 "Zeitschr. fiir Hyg. und Infectionskrankheiten," 1899, Bd. XXXII.,
p. 361.

2 " Milchzeitung," 1899, Bd. XVIII., p. 18.

3 Freudenreich originally named this organism Micrococcus casei amari
("Landwirt. Jahrbuch der Schweiz," 1894, p. 136).

BACTERIA

37

The Betacocci have been so named by the author because
they are generally found in sugar and other beets, swedes
and mangold wurzels, especially when these are in a state of
decomposition. In countries where such roots are largely used
as cattle fodder, the Betacocci are of very common occurrence

FIG. 40. — Streptococcus lactis.

FIG. 41. — Betacoccus from thin juice
from Nakskov Sugar Factory.

FIG. 42. — Streptococcus Uquefacien-s (Freu-
denreicKs Micrococcus casei amari).

FIG. 43. — Streptococcus liquefaciens
(Escherich's Streptococcus coli gracilis).

in milk and in the cheese made therefrom. The Betacocci
occur as diplococci or short chains, which are not to be dis-
tinguished microscopically from the streptococci which have
been dealt with above. They generally form Isevo lactic acid,
gas and other by-products, and render saccharose broth more
or less slimy. The slime formation may best be observed in

38 DAIRY BACTERIOLOGY

stab cultures in saccharose gelatine. Some species liquefy this
medium after some time, though they do not decompose casein.
The Leuconostocs (Fig. 12, p. 7), which may give a deal of
trouble in beet sugar manufacture, are Betacocci. They grow
at temperatures as low as 5° C., and some species thrive better
at room temperature than at 30° C. The sour cabbage bac-
terium, Sc. brassicce, also belongs to this group.

FIG. 44. — Various Betacocci in Stab Cultures in Cane-sugar Gelatine.

Group B. — The bacteria belonging to this group are not true
lactic acid bacteria, differing from the forms hitherto described
in forming catalase, reducing nitrates and generally growing well
on the surface of solid media.

The microbacteria are very small rods, usually only 0-3 to 0-4 /z
thick, which stand fairly high temperatures, and are therefore to
be found in pasteurised milk. They form lactic acid, and some of
them (Bacillus acidophilus) are common intestinal organisms. To
these bacteria are related a group of small rod bacteria which
liquefy gelatine, but only produce traces of acid.

The tetracocci include the acid-producing forms of the Micro-
cocci and Sarcince. Division into these two groups is not feasible,
and confusion may even occur with the Streptococcus group, as
many of the Tetracocci appear as diplococci ; they may, however,
readily be distinguished by their power to decompose hydrogen
peroxide, for, as mentioned above, they differ from the true lactic
acid bacteria in producing catalase. They produce less lactic acid

BACTERIA 39

than the Streptococci, while they form notable amounts of acetic
acid besides. They are more aerobic in character than the
Streptococci, forming as a rule large surface colonies which are
often coloured yellow, orange or pink. Most of them liquefy
gelatine, though generally slowly, and they will therefore also
coagulate milk, though the quantity of acid formed is seldom
sufficient to accomplish this. Several of the liquefying species
probably play some part in the ripening of certain cheeses, for
example Tetracoccus liquefaciens1 , which forms white colonies
and produces dextro lactic acid. The Tetracocci are found in
great numbers in cow dung and earth, whence they find their way
into dust ; they are well able to stand desiccation and common
salt, and many of them will also stand heating to over 70° C.
They are common skin bacteria, some giving rise to pustules
and others to inflammations, which are generally not of a
dangerous character. According to Beijerinck, certain lactic-
acid-producing sarcinae which occur in soil, but are of no import-
ance in dairy practice, produce large amounts of carbon dioxide
and hydrogen.

The Pseudo Lactic Acid Bacteria. — These are motile
or non-motile, Gram-negative short rods with rounded ends,
which seldom form chains or threads of any length. They do not
require organic nitrogen, and in stab cultures show profuse surface
growth. As a rule they do not liquefy gelatine, and are mainly
intestinal and excremental bacteria. The gas which they produce
from sugars consists of carbon dioxide and hydrogen in widely
varying proportions ; according to certain American investigators,
the composition of the gas formed affords a basis for the classifica-
tion of these organisms 2.

The aerogenes bacteria are non-motile rods which produce large
amounts of gas, being able to convert most of the sugar present
into gas, especially if the acid which is produced is neutralised.
The gas may contain up to three times as much carbon dioxide as
hydrogen, while aerogenes forms are known which may even
produce carbon dioxide alone. In dairy bacteriology two types
may specially-be distinguished. The one forms much slime which
gives rise to outstanding colonies with the shiny appearance of
porcelain ; appreciable amounts of alcohol are formed, but not
acid enough to coagulate milk. If the acid is neutralised by chalk
as fast as it is formed, the milk will gradually be converted into a
thick slime. To this type belong certain pathogenic bacteria, such

1 The author ("Landwirt. Jahrbuch der Schweiz," 1904, pp. 349 and 369)
has described this organism under the name Micrococcus casei liquefaciens.

2 L. A. Rogers, W. Mansfield Clark, Brooke J. Davis and Alice C. Evans
(" Journ. of Infectious Diseases," vols. 14, 15 and 17).

40 DAIRY BACTERIOLOGY

as Bacterium pneumonice, and some bacteria which, according to
Gillebeau, cause inflammation of the udder. The second type, to
which belongs Bacterium lactis aerogenes, produces less slime, and
its colonies on gelatine are often but little larger than those of the
lactic acid streptococci. They generally coagulate milk by the
production of succinic acid and Isevo lactic acid. The aerogenes
bacteria can convert the citric acid of the milk into acetic and
carbonic acids. On corn and flour aerogenes bacteria are found
which produce a yellow colouring matter. Bacterium cloacce is a
liquefying aerogenes bacterium. Motile forms related to this
organism may be regarded as a link writh the Proteus bacteria.
In cheese such forms are often found, but these form spores in one

end of the cell, and must there-
fore be designated as aerobic
Plectridia.

The coli bacteria are as a
rule peritrich, though often
they are only slightly motile.
They differ from the aerogenes
bacteria in producing about
equal quantities of carbon
dioxide and hydrogen, and also
^n their Power to decompose
casein. In the latter respect
they show a certain resem-
blance to the true lactic acid
FIG. 45.— The Coli Bacterium which spoilt bacteria, but they carry the

the Milk and Butter at Duelund Dairy . .

in 1888. (After c. o. Jensen.) x 1,000. decomposition further than the

ammo acid stage, producing

evil-smelling bodies, a circumstance which links them with the
putrefactive" bacteria. The kind of lactic acid formed by the
coli bacteria is influenced to some extent by the particular
sugar fermented, and also by the nature of the nitrogenous
nutrient matter present. They commonly form large amounts
of succinic and acetic acids. Their surface colonies may re-
semble those of the aerogenes bacteria in exceptional cases, but
as a general rule, they spread out in thin, skinny layers. Numerous
species of coli bacteria are known, many being pathogenic, giving
rise to intestinal and other diseases such as typhoid fever and
dysentery. Bacterium typhosum (Fig. 27) distinguishes itself by
being very actively motile, while the dysentery bacteria are non-
motile. None of these form gas. The commonest cases of meat
poisoning are also due to certain coli bacteria, Bacterium enteriditis.
The very dangerous coli bacteria which are dealt with here can
easily be distinguished from the ordinary coli bacteria by the fact

BACTERIA

41

that they do not ferment lactose to any appreciable extent. Even
though they may make milk slightly acid to begin with, they
always turn it alkaline at last.

THE PROPIONIC ACID FERMENTATION

The second fermentation process to be described is that in
which sugar or calcium lactate is converted into propionic, acetic
and carbonic acids. This fermentation is of particular interest in
dairy practice in connection with the formation of cavities or
" eyes " in cheese. The author's investigations on the ripening
of Emmental cheese have shown that the above-mentioned
products mostly appear at the stage at which the eyes are formed l ;

FIG. 46. — Bacterium acidi propionici (a).
X 1,000.

FIG. 47. — Bacterium acidi propionici (6).
Grown at 39° C. x 1,000.

and Clark has shown that normal eyes contain nothing but carbon
dioxide 2. The presence of the organisms in question can only be
demonstrated after several successive inoculations into peptone
broth containing calcium ' lactate instead of sugar. If such a
medium be innoculated with almost any kind of cheese, active
propionic acid fermentation will be found to have set in after
systematic cultivation for a week or two.

Bacterium acidi propionici was first isolated by Freudenreich
and the author3. These bacteria are non-sporing, non-motile and
Gram-positive. Several species are known which differ consider-
ably in appearance. The species (a), which is mainly instrumental

1 " Studier over de flygtige Syrer i Ost." Doctoral. thesis, published by
J. Gjellerup, Copenhagen.

2 U.S. Dept. of Agriculture, Bureau of Animal Industry, Bull. 151. 1912.

3 "Uber die im Emmenthalerkase stattfindende Propionsauregarung "
("Landwirt. Jahrbuch der Schweiz," 1906, p. 320.)

42 DAIRY BACTERIOLOGY

in forming eyes in Emmental cheese, resembles the streptococci,
but it does not coagulate milk. To this group belong various colour-
producing species which may form red and brown spots in Emmen-
tal cheese 1. Other strains (b) are stout irregular rods, some of
which become either branched or club-shaped, and stain so un-
evenly that, like the diphtheria bacteria, they present a segmented
appearance. These species produce sufficient acid (not lactic
acid) from milk sugar to coagulate milk. The propionic acid
bacteria do not decompose casein. Some of them are anaerobic,
but may be accustomed to aerobic conditions ; when this occurs,
their fermentative powers are gradually weakened. The aerobic
varieties form outstanding colonies in stab cultures, and are on
the whole as prone to slime production as the aerogenes type.
The propionic acid bacteria grow at temperatures between 15° and
40° C. According to Bum 2, they occur in appreciable numbers in
cow dung, and probably they find their way into milk from this
source.

THE BUTYRIC ACID FERMENTATION

In this process carbohydrates or lactates are converted into a
number of different products, among which butyric acid, carbon
dioxide and hydrogen have attracted particular notice. In
addition to these, notable amounts of lactic, acetic, propionic and
formic acids are formed, and sometimes also various alcohols, so
that the process is an extremely complicated one.

The first butyric acid bacterium was described by Pasteur in
1861, a discovery which for the first time revealed the existence of
obligate anaerobic organisms ; the butyric acid bacteria will under
no circumstances grow in presence of atmospheric oxygen. They
are fairly large rods which form spores, assuming either the
Clostridium or Plectridium forms (Figs. 5 and 6) ; the spores
exhibit polar germination. In the butyric acid bacteria it is often
possible to demonstrate reserve food material which stains blue or
violet with iodine ; this is most noticeable just before sporing if
grown on starchy material such as potato slices. The young
bacteria are generally Gram-positive. The true butyric acid
bacteria can subsist on inorganic nitrogen, and in conjunction
with aerobic bacteria they can assimilate atmospheric nitrogen3.
They do not attack proteins ; in this respect they differ from a

1 Thoni and Allemann, " Centralblatt fiir Bacteriologie," 2 Abt., 1910,
Bd. XXV., p. 8. According to the author's investigations, stab cultures of
white varieties of the propionic acid bacteria are often red below the surface
of the medium.

" Landwirtschaftliches Jahrbuch der Schweiz," 1912, p. 481.

3 Winogradsky, " Archives des Sciences hiologiques," 1895, tome 3,
p. 295, and Bredemann, " Zentralblatt f. Bact.," 2 Abt., 1909, Bd. XXIIL,
p. 385.

BACTERIA 43

number of nearly related obligate anaerobic bacteria to be described
in connection with the putrefactive process. As all the non-
sporing bacteria in a liquid can be killed by heating for a few
minutes at 90° C., the sporing organisms may readily be isolated by
this means. If the liquid is rich in butyric acid bacteria, a small
quantity may be inoculated after heating, into a tube containing
a deep layer of sugar agar, but if only a few of these organisms are
present their number must be increased by adding the heated
liquid to sterile milk which all but fills a bottle securely closed by
a spring stopper ; the fermentation should not be allowed to
proceed too far before the bottle is opened, or the pressure of the
accumulated gases may easily cause a burst. The butyric acid
bacteria ferment most sugars (not mannitol), and are said to
be able to ferment starch. Butyric acid is the chief product
formed in milk. They occur principally in soil, manure and
flour.

According to Grassberger and Schattenfroh1, distinction is to be
made between the motile and the non-motile forms, Bacillus
butyricus mobilis and B. b. immobilis. The former is peritrich,
and readily forms spores which are killed after only three minutes'
boiling ; it does not ferment calcium lactate. It is responsible for
the large amounts of butyric acid found in certain sour milk
cheeses 2. The latter is somewhat larger and forms spores less
readily, but on the other hand the spores are decidedly more
resistant to heat, for they will survive one and a half hours' boiling.
According to Barthel 3, it is far more common in milk than the
motile form, which accounts for its regular occurrence in the human
and animal intestines. As a gas-producing organism, it may give
rise to the formation of cavities in cheese. It may sometimes be
pathogenic, and is supposed by some to be merely a degene-
rated variety of the organism of the cattle disease, blackleg or
quarter evil (Bacillus Chauvoei), which is normally peritrich and
attacks proteins. The butyric acid bacteria grow well at 16° to
40° C.

While the butyric acid bacteria do not attack cellulose, a number
of closely related Plectridium forms carry out a cellulose fermenta-
tion which is of prime importance in the digestive process of herbi-
vora and in soil formation. Other allied forms ferment the
pectins which cement together the vegetable cells, and come to
play an important part in the retting of flax and hemp.

1 "Archiv. f. Hygieine," XXXVII., XLII. and XLVIII.

2 Freudenreich and Orla- Jensen, " Landwirtschaftliches Jahrbuch der
Schweiz," 1905, p. 312.

3 " Obligat anaerobe Bakterien in Milch und Molkereiprodukten "
("Centralblatt 1 Bacteriologie/' 2 Abt., 1910, Bd. XXVI. , p. 1).

44 DAIRY BACTERIOLOGY

THE PUTREFACTIVE PROCESS

The preceding sections have been devoted to certain carbo-
hydrate fermentations. We now pass on to the fermentations of
the proteins, which, according to our previous definition, come
under the heading of putrefaction. In this process two distinct
phases may be recognised : first, the protein hydrolysis or peptoni-
sation, due exclusively to the proteolytic enzymes, in which the
proteins are split up into soluble amino acids, and second, the
decomposition of these acids by other enzymes (amidases, oxidases
and reductases), with the formation of ammonia and various evil-
smelling substances. While the final products of carbohydrate
fermentation are always acid, those of protein fermentation are
always alkaline. As the action of proteolytic enzymes arid also
the growth of the organisms which secrete them are inhibited by
small quantities of free acid, it is easy to understand why the
presence of carbohydrates in appreciable amounts will prevent
putrefaction ; it is only when the carbohydrates and their acid
products have been destroyed or neutralised that the process of
putrefaction will obtain a proper start. As all bacteria which
secrete proteolytic enzymes may take part in a putrefactive
process, we have to deal with a large number of types. We will
first deal with the aerobic putrefactive bacteria, and then with
those which are obligate anaerobes, for the former initiate the
process and, consuming the available atmospheric oxygen, they
prepare the ground for the latter group, which then carry the
decomposition further ; it is especially during this second phase
that the evil-smelling products, *so characteristic of putrefaction,
are formed.

A. The Aerobic Putrefactive Bacteria. 1. The
Fluorescent Bacteria. — These are non-sporing motile rods with
polar flagellse, producing a fluorescent green colouring matter
(insoluble in chloroform) on neutral or alkaline media. The
commonest species are monotrich and Gram-negative. They do
not ferment lactose, and some are liquefying, others non -liquefy ing.
Several, especially the non-liquefying species, are denitrifying
organisms, i.e., they reduce nitrates to free nitrogen, and thus rob
plants of their nitrogenous nutrient matter. They are very
widely distributed in soil and water, and can as a rule grow at
temperatures only slightly above 0° C. A species which is parti-
cularly active in liquefying gelatine, Bacterium fluorescens lique-
faciens, hydrolyses fats, and may therefore play an important part
in turning butter rancid. Bacterium pyocyaneum (Fig. 25) is a
nearly related form which also liquefies gelatine and hydrolyses
fats, but it grows so slowly at ordinary temperatures that it does

BACTERIA 45

not spoil butter under normal conditions. In addition to a
fluorescence, this organism produces a blue colour (which is soluble
in chloroform and is turned red by acids). It is often found as
one of the active organisms in inflammations, when it imparts a
blue or green colour, to the pus. A similar but Gram-positive,
lofotrich and non-liquefying species is the bacterium of blue milk,
Bacterium syncyaneum (Fig. 26), which in addition to a fluorescence
produces a grey colour which is turned blue by acids ; it can
therefore only colour milk blue in the presence of lactic acid
bacteria. A peptonising rod form which colours neutral milk blue
was isolated by Carl Lind, working in the author's laboratory.
In all cases the blue coloration starts on the surface of the milk.
The last- mentioned organism is perhaps identical with the water
bacterium, B.cyaneofuscus, isolated by Beijerinck 1, which produces
blue spots in cheese.

The group most nearly allied to the fluorescent bacteria are the
marine phosphorescent bacteria, which cause dead1 fish to become
luminous. Being marine organisms, they require a nutrient
medium containing sodium chloride.

2. Peptonising Cocci. — The liquefying micrococci and sarcina
forms, especially those which do not produce acid, generally take
part in putrefactive processes. Several of the colour-producing
varieties thrive well in the outer paste which becomes the rind of
many cheeses.

3. Coli Bacteria. — As mentioned above, these are intestinal
organisms which find their way into milk from the excreta. Being
acid-producers, they will generally retard rather than promote
putrefaction, as carbohydrates are present in sufficient quantity
both in the intestine and in milk. In the absence of carbo-
hydrates, however, they act as typical putrefactive bacteria,
decomposing amino acids.

4. The Proteus Bacteria. — As their name implies, these rod
forms may assume many different shapes and sizes, sometimes
growing out into long threads which may form a tangled network.
They form no spores, and do not show any uniformity in their
behaviour towards Gram's stain ; as a rule they are actively
motile, being plentifully supplied with flagellse all over the cell.
The proteus bacteria mostly ferment sugars, especially dextrose,
with the production of succinic and acetic acids and various gases,
and they generally liquefy gelatine. A non-liquefying form,
Bacterium Zopfii, whose colonies ramify so freely on gelatine that
it might easily be mistaken for a mould, is frequently found in
milk. The most typical of the aerobic putrefactive bacteria is

1 "Botaiiische Zeituiig," 1891, Bd. XLIX., p. 704.

46

DAIRY BACTERIOLOGY

Bacterium vulgar e (Fig. 28), which coagulates and peptonises milk ;
it can grow at low temperatures. Many closely related forms are
pathogenic. The ptomaine poisons produced in meat are generally
due to certain proteus and coli bacteria. The proteus group
includes several colour-producing bacteria, such as the Bacterium
synxantum of yellow milk and Bacterium erythrogenes of red milk.
The latter, however, differs from the other proteus bacteria in
being non-motile. Bacterium prodigiosum, which produces a fine
red colour, especially when grown on starchy media, is far commoner
of occurrence than the two last- mentioned organisms ; this

FIG. 48. — Bacillus subtilis. (After Miyula.) a. Active rods before spore
formation, d. Stained to show the flagellse. c. Rods with spores. All
these X 1,000. b. A piece of film. X 100.

phenomenon was formerly taken for the appearance of blood on
the Host. As the red colouring matter is not soluble in water,
the bacterium cannot colour milk red, but can only form red spots
on the cream layer. It is Gram-negative, and most varieties
coagulate milk quickly by forming both acid and rennet. The
author has shown that it hydrolyses fat as actively as Bacterium
fluorescens liquefaciens.

5. The Hay and Potato Bacteria. — These organisms are distin-
guished by their ability to form spores which are more resistant to
heat than those formed by the butyric acid bacteria. They are
only killed after three to six hours' boiling, and a few species which
do not grow at ordinary temperatures form spores capable of

BACTERIA 47

resisting twenty hours' boiling. "The spores germinate laterally.
The organisms of this group are typical soil bacteria, finding their
way into the soil from hay, potatoes and other feeding stuffs, in
which they were first found. With the dust, and from the excre-
ments of the stable, they find their way into milk, rendering it
very difficult to sterilise. Being obligate aerobes, they generally
form films on the surface of the culture solution and carry the
oxidation of carbohydrates almost to completion. They contain
diastase, which enables them to hydrolyse starch ; if bread con-
tains spores of these bacteria, which have not - been killed in
baking, it becomes slimy after a few days. They peptonise
actively, are Gram-positive and peritrich. The two best-known
species, Bacillus subtilis and Bacillus mesentericus, are best dis-
tinguished from one another by cultivation on potato slices, on
which the former produces a smooth film and the latter a wrinkled
film. Bacillus mycoides is most frequently found in incompletely
sterilised milk ; it forms stellate colonies like those of B. Zopfii,
in stab cultures, before the gelatine is melted. Bacillus anthracis
(Fig. 16A), grows similarly, but is non-motile. Most of the hay
and potato bacilli are thermophile, that is they are able to grow at
high temperatures, many of them growing well at 50° C., and some
even at 60° C. Other species actually prefer temperatures of
50° to 70° C., and do not grow at ordinary temperatures ; these are
often non-motile, and anaerobic rather than aerobic. When
masses of hay or other vegetable matter become warm, this is in
the first place due to the oxidation or respiration of the vegetable
matter itself, but when the temperature reaches 50° C. the vital
processes of the plant remains cease, and further production of
heat is due to the action of the thermophile bacteria until their
maximum temperature is reached. Further rise in temperature
and spontaneous ignition may take place owing to purely chemical
oxidation, i.e., combustion.

B. Anaerobic Putrefactive Bacteria. — All the organ-
isms of this group are peritrich spore-forming rods which
closely resemble the butyric acid bacteria, from which, how-
ever, they differ in being able to attack proteins. The most
important member is Bacillus putrificus which contributes more
than any other organism to the development of the putrefactive
odour ; it is able to grow in solutions of pure proteins. The
Plectridium foetidum, isolated by Weigmann l from cheese, which
produces in milk a smell like that of Limburg cheese, is, according
to Barthel 2, identical with Bacillus putrificus. The author 3 has

1 " Centralblatt f. Bakt.," 2 Abt., 1896, Bd. II., p. 150.

2 " Centralblatt f. Bakt.," 2 Abt., 1910, Bd. XXVI., p. 1. *

3 "Centralblatt f. Bakt.," 2 Abt.. 1904, Bd. XIII.. p. 754.

48 DAIRY BACTERIOLOGY

examined a similar Plectridium form which was isolated from
cheese, but which differs from the last-mentioned organism in
being able to ferment calcium lactate and not thriving in the
absence of special sources of carbon. Further, it produces in milk
formic, butyric arid valerianic acids. Several poison -producing
pathogenic forms also belong to this group, e.g., Bacillus botulinus,
which forms poisons in sausages, and Bacillus tetani, the organism
of tetanus or lockjaw (Figs. 6 and 16c).

The subject of putrefactive fermentation has been dealt with
at some length for two reasons : first, the putrefactive organisms
are the most harmful with which we have to deal in dairy practice,
and second, the ripening of cheese is a process of protein decom-
position which takes a special course thanks to the acid which is
always formed at the outset.

The few milk bacteria which are not connected with the fermen-
tation processes already described will be dealt with in Part II.

Chapter III
Yeasts and Moulds

UNLIKE the bacteria, the yeasts and moulds prefer acid nutrient
media, their natural habitat being soft, juicy fruits. They also
thrive in sour milk and dairy products, in which they can produce
a variety of changes owing to the number of enzymes which they
secrete. They generally oxidise lactic acid, and thus render the
medium better adapted for the nutrition of bacteria ; it is only
with the co-operation of yeasts and moulds that putrefactive
fermentation can become established in sour milk. Yeasts and
moulds play an important part in the ripening of many sour milk
and soft rennet cheeses. As most moulds hydrolyse fats, they
contribute largely towards the development of rancidity in butter.
Most yeasts and moulds can grow at comparatively low tempera-
tures. While bacteria only develop in media containing at least
20 to 30 per cent, of water, moulds require only 14 per cent., and
will therefore grow on comparatively dry feeding stuffs, whence
they find their way with the stable dust into milk.

A. YEASTS

The best-known property of these organisms is the power to
ferment sugars with the production of alcohol and carbon dioxide.
They find technical application in the manufacture of beer, wine
and spirits, while pressed yeast is used in baking to raise the
dough. The sporing and the non-sporing forms are known as
Saccharomycetes and Torulce, respectively. The non-sporing
Mycodermce occupy a special position ; like the moulds, they only
produce alcohol quite exceptionally, but they are able to oxidise
this substance.

1. Saccharomycetes. — These are the most typical of the alcohol-
producing yeasts (Figs. 2, 4 and 7). As the form of the cells is
not very characteristic, and subject to considerable variation, the
various species are best distinguished by the scheme of Emil
Christian Hansen, which is based on the investigation of the spore
formation and observation of the time which elapses before spores
are formed under certain conditions. Moist plaster of Paris
blocks are particularly well adapted for this purpose. The pure
cultures used in breweries are generally round or oval yeasts, which

50

DAIRY BACTERIOLOGY

develop comparatively few spores in each cell ; certain wild
yeasts, which spoil beer, grow as elongated cells and form large
numbers of spores in each cell. Yeasts which collect at the
bottom of the liquid towards the end of the fermentation are
known as bottom yeasts ; those which rise to the surface are known
as top yeasts. Bottom yeasts are employed in slow fermenta-
tion at low temperatures, in the making of light wines and beers
of the lager type. Top yeasts, on the other hand, cause more
vigorous fermentations, and are used in the making of heavier
wines and beers and spirits. Bakers' yeast is usually a top yeast.
The yeasts used in the production of alcoholic beverages ferment
maltose and sucrose, but not lactose ; the saccharomycetes met
with in dairy practice ferment lactose and sucrose, but not maltose.

2. The Torulse. — These are smaller than the saccharomycetes

and play only a subordinate
part in the alcoholic fermen-
tation industries. On the
other hand, they are of far
more frequent occurrence in
dairy products, while the
formation of alcohol in Kefir
and similar beverages is
chiefly due to the action of
certain torulse which ferment
lactose either by themselves
or in symbiosis with cer-
tain lactic acid bacteria. A
lactose - fermenting species,
Torula amara, which lives in

sycamore leaves, will, according to Harrison, turn milk bitter in
a few hours. The torulse which do not ferment sugar are still
more widely distributed ; according to the author's investiga-
tions, they develop freely in butter which is kept for some time.
Certain species which form characteristic stellate colonies in stab
cultures hydrolyse the fat more or less vigorously, provided that
the butter is sufficiently acid. A certain torula which hydrolyses
fat will colour butter red ; the growth of this organism is inhibited
by common salt.

3. My coder mae. — While most yeasts form a film on the surface
of the liquid after the fermentation is over, the mycodermse form
a dull-looking film at the very outset. As a rule they are elongated
cells containing a few bright granules. In the making of Emmen-
tal cheese, a home-made rennet is generally employed which, if
properly made, should be a practically pure culture of Thermo -
bacterium helveticum covered by a mycoderma film which excludes

FIG. <l$.—Mycodernta cerei-isice.
(After Holm.)

YEASTS AND MOULDS

51

air, and thus secures anaerobic conditions favourable to the growth
of the lactic acid bacterium. This is an instructive example of
the way in which an organism may act beneficially by indirect
means. The mycoderma in question ferments dextrose but not
disaccharides.

B. MOULDS

It will not be necessary to enter into the classification of the
moulds, as only a few species are met with in dairy practice ;
the yeast-like forms, Monilia, Cladosporium and Oidium lactis,
and a few Penicillium species.

1. Monilia and Cladosporium. — Both of these groups reproduce
by budding, so that in young cultures they cannot be distinguished
from yeasts ; it is only at a
later stage that the cells be-
come elongated and form a
true mycelium which grows up-
wards into the air. The Monilia
group further resembles the
yeasts in being able to bring
about a true alcoholic fermen-
tation. A species which be-
comes black in old cultures,
Monilia nigra, will, according
to Burri and Staub 1, form
deeply penetrating black spots
on the rind of hard cheeses.
On the other hand, the black-
ening of sour milk cheeses and
soft cheeses is due to Cladosporium herbarum, a mould which
is fairly prevalent in dairies 2. A nearly related form, Clado-
sporium butyri, which is first white, then green, brown and
finally black, has been found by the author 3 to play an important
part in the development of rancidity in butter. Like several
of the mycodermse, it produces an ethereal fruity odour in milk,
but does not bring about alcoholic fermentation. The above-
mentioned Monilia and Cladosporium species liquefy gelatine, and
are able to grow in presence of large amounts of salt, so that their
growth in cheese is not inhibited by salting.

2. Oidium lactis. — This white mould is hardly ever absent from
any milk. It thrives better in cream than in skim milk, and

1 " Landwirtschaftliches Jahrbuch der Schweiz," 1909, p. 487.

2 This mould was first observed by Hers; (Milchzeitung, 1885). Adametz
has described several black moulds in his book, " Ueber die Ursachen und
Erreger der abnormalen Reifungsvorgange beim Kase" (Bremen, 1893).

3 " Landwirtschaftliches Jahrbuch der Schweiz," 1901, p. 367.

4—2

FIG. 50. — Monilia nigra,. (After Burri
- and Staub.)

52 DAIRY BACTERIOLOGY

better in sour than in fresh milk. It is always this mould which
forms the velvety layer on sour milk which has been kept too long.
On sugar gelatine it only shows luxuriant development of aerial
threads in places where lactic acid bacteria are also growing,
which shows that it thrives best in presence of lactic acid. It
liquefies gelatine very slowly. As already mentioned, it has no
specialised conidiophores. It is mostly found on the surface of
many soft cheeses, and contributes towards the development of
rancidity in butter. Several allied forms are known, some of
which, according to Weigmann, will produce a smell of turnips
and other feeding stuffs in milk. A red form, Oidium aurantiacum,

FIG. 61. — Two varieties of Penicittivm brcvicaule.

will, according to Adametz, produce orange or red spots on the
surface of cheese.

3. The Penicillia. — These are represented by numerous species,
difficult to distinguish from one another. Their distinguishing
feature is a sheaf of conical cells, growing out from the upper end
of the conidiophore^ from which the conidia (usually green) are
formed by constriction. The name of this group of moulds is
derived from Penicillium, a brush, owing to the characteristic
tassel or brush-like appearance presented by the arrangement of
the spores. The commonest species is Penicillium glaucum
(Figs. 3 and 9), which liquefies gelatine and produces the well-
known mouldy or musty smell. A nearly related form, which
thrives particularly well on apples, produces an ethereal fruity
odour. In Roquefort. Gorgonzola and Stilton cheeses, a special
species, Penicillium roqueforti, is found, which forms shorter
conidiophores and somewhat larger conidia. On sugar gelatine
its colonies develop an uneven and often fairly wide white rim ;

YEASTS AND MOULDS

53

they have no smell, and do not liquefy gelatine. On a fatty
medium this mould produces the characteristic smell and taste of
Roquefort cheese. Among the many similar forms which are
known, only those can be employed which form a perfectly
colourless mycelium. If they colour the nutrient medium yellow
or brown, or their conidia become brown instead of green after a
short time, they will colour the Roquefort cheese brown. Accord-
ing to Staub, a mould of this kind, Penicillium casei, will produce
on the surface of cheese yellowish brown spots which gradually
become reddish brown, and coalesce so that the whole of the rind
becomes coloured. On Camembert and other French soft cheeses,
are generally found Penicillium camembertii, which forms pale
green or greyish green conidia, and Penicillium candidum, which
forms conidia up to half a centimetre long, being perfectly
white even in the ripe state. Both these moulds liquefy gelatine
slowly. While all the species mentioned above form conidia
which are perfectly smooth and spherical when ripe, Penicillium
brevicaule forms irregular warty conidia (Fig. 51) ; this mould is
usually yellow or brown and grows on manure ; according to
Weigmann and Wolff, it produces a smell of turnips or onions in
milk and dairy products 1. According to -Thorn, varieties of this
mould are often found on Camembert cheese.

1 " Centralblatt f. Bakt.," 2 Abt., 1909, Bd. XXIL, p. 657.

PART II

Chapter I

Cleaning and the Procurement of Milk

CLEANING

Systematic Cleanliness is the golden rule in dairy practice :—
more than this, it includes all that is of prime importance in
Hygiene ; and it brings every one into actual contact with the
science of Bacteriology. Unfortunately it is only too obvious
that few people realise the true significance of cleanliness. For
example, in dusting a room, many think that they are serving
the interests of cleanliness by simply whirling up the dust with a
dry cloth from places where it is most conspicuous only to let it
settle in a thin layer all over the room after some time ; again,
many will wash a number of floors with the same pail of water,
which may at last become as black as ink, or wipe the inside of
cooking utensils with swabs which may be in a most disgusting
condition. It must also be said that most people consider the
daily brushing of their teeth with a tooth brush which is never
cleaned as the acme of cleanliness.

As the great majority of the defects of milk and dairy products
arise through lack of cleanliness, a few remarks concerning the
cleaning of the surfaces with which the milk comes into contact will
not be out of place. If the cleaning is to be effective, the dirt must
not merely be spread over a larger area, but it must be completely
removed, and not only must the microorganisms present be killed,
but care must be taken that no fresh organisms are introduced.
The main object in cleaning is to get rid of the microorganisms,
and as these are usually embedded in the dirt, the first step is to
secure the removal of the dirt ; in doing this,, the great majority
of the microorganisms will be removed as well, while the few
which remain will be prevented from multiplying owing to lack of
nutrition. Moreover, even sterile dirt will always be undesirable
as it renders the surface of the vessel rough, and even if odourless
at the start, it will acquire an unpleasant taste and smell on being
submitted to the processes necessary for the killing of the micro-
organisms. In order to remove the dirt, which in our case con-
sists of the constituents of milk, it must be dissolved or at least
loosened ; thus fat will be loosened by the use of hot water. The

CLEANING AND THE PROCUREMENT OF MILK 55

water should, however, not be used too warm to begin with, or the
proteins will be rendered insoluble. It is well-known that cheese
cloth quickly becomes stiff and useless if not rinsed in cold water
before washing in boiling water. By the use of soda or lime, the
casein is dissolved and the fat is emulsified, 'i.e., reduced to a fine
state of division. These chemicals also act as poisons towards
bacteria, i.e., disinfectants. They are most effective if used
together, but unfortunately tinned vessels do not stand this
treatment well. Soda is to be preferred to lime for the cleaning
of such vessels, for though it is somewhat more energetic in its
action on tin, it removes fat more readily. In order to prevent
the tinning on pasteurisers from being attacked, it is best to soak
the crust which has formed in a cold solution of soda, whereupon
it will easily be removed on scrubbing with clean warm water.
The plates of centrifuges may be cleaned in the same way. With
lime, the fat forms insoluble lime soaps which makes the tin dull
and rough. On the other hand, lime is to be preferred for wood
work, partly because if rubbed in as milk of lime it will remain
where it is for some time, so that its disinfecting action will be
prolonged, and partly because the lime soaps fill up the pores and
render the surface of the wood smoother and firmer. The cleaning
must be finished with several rinsings with plenty of water, or
some of the chemicals and dissolved dirt will remain behind.
The last rinsing should be with hot (sterile) water to ensure rapid
drying. Wood easily frays with too much heating, but good wood
will stand liberal quantities of pure warm water and even steam ;
churns are best cleaned by simply rinsing repeatedly with water
at 90° C. Tinned or other metal vessels should if possible be boiled
out or steamed as a final treatment, thus ensuring an extra
sterilisation and rapid drying. This last point is a very important
one, for in spite of all reasonable care absolute cleanliness and
sterility are seldom achieved, but if only the vessels are dried as
soon as possible, no new vegetation will be allowed to develop in
them. Wherever possible, make the best use of direct sunlight,
which both dries and sterilises. Wind and through draughts are
also useful provided that they bring no dust. All that has been
said regarding pails, apparatus and piping applies equally well to
the cloths and scrubbing, brushes used in cleaning ; these must be
thoroughly cleaned and finally scalded with boiling water and
dried to prevent them becoming slimy ; they may conveniently
be dried in the boiler room. Every dairyman should clearly
understand that cloths and brushes may do more harm than good
if not perfectly clean. It is well known that cleaning cannot be
effective if the vessels have inaccessible corners or rough surfaces ;
frayed woodwork or rusty pails should therefore not be tolerated.

56 DAIRY BACTERIOLOGY

While clean metal ware may be sterilised by simply prolonging
the steaming sufficiently, wood work presents a more difficult
problem, and should the dairy be troubled with an infection of
harmful organisms, the usual cleaning should be supplemented
with a washing with 1 to 2 per cent, formaldehyde made by
diluting commercial formalin with twenty to forty times its bulk
of water ; it is advisable to protect the hands from the action of
this chemical, which is best distributed by means of a garden
syringe. In dealing with churns or combined churns and workers,
the disinfecting action may be intensified by keeping them well
closed for several hours after washing, as the formaldehyde vapour
has a disinfecting action. A 2 per cent, solution of ammonium
bifluoride is also a very effective disinfectant. The vessels must
be rinsed and dried before using again. The growth of moulds is
best prevented by avoiding excessive dampness ; even in cellars
used for storing cheese, the humidity need not exceed 90 per cent.
If the walls of the cellar have become mouldy (Cladosporium
herbarum often appears only as very small black spots) they
must immediately be limed. Two per cent, of copper sulphate
may be added to the lime with advantage. In investigating
the cause of any defect in cheese, it should not be forgotten
that the cutting and stirring appliances and the cheese cloths
may be infected, and that the wooden thermometer case may
harbour bacteria.

In dealing with the subject of cleaning, it may be well to give a
more explicit definition of what is understood by infection. By
an infection in the broadest sense of the term, we understand any
admixture which may minimise the value of the product in ques-
tion. If the foreign substance is sterile, the resulting condition is
a purely chemical matter. If the foreign matter is not sterile,
we shall be dealing with what is generally known as an infection
(opposite disinfection), and this is a purely biological matter.
The latter trouble will generally be more serious in its consequences
than the former, as the effects of the biological action increase on
keeping. Milk especially is subject to rapid alteration, not only
because it is an ideal nutrient medium for many microorganisms,
but also because, unlike most other dairy products, it is a liquid.
A solid medium can only be spoilt gradually by an infection from
without spreading inwards, and then only if the medium is not too
dry to allow of the growth of microorganisms.

THE PROCUREMENT OF MILK

The most troublesome sources of milk infection are the udder
and teats of the cow, and the vessels with which the milk comes

CLEANING AND THE PROCUREMENT OF MILK 57

into contact ; compared with these, contamination from the air
usually plays quite a minor part.

An important influence on the quality of the milk is often
ascribed to the feeding, and it is well known that certain fat-
soluble colouring and odiferous matters contained in the fodder
may pass into the milk, and also that the composition of the milk
fat is largely influenced by the nature of the feed. At times the
natural acidity and the sensitiveness to the action of rennet may
be slightly influenced, but otherwise the chemical composition of
the milk is not sensibly affected by the factor in question 1. It
should, however, not be forgotten that the feeding is the most
important factor in determining the consistency of the dung, and
the thinner the dung the dirtier the cows. Moreover, the feed
will directly or indirectly
determine the nature of the
bacteria predominating in
the dung, and thus the
nature and number of the
bacteria in the milk. . It will
therefore be understood that
the most important effect of
the feed on the quality of
the milk is of a bacterio-
logical nature, and this can
partly be eliminated by keep-
ing, the cows clean ; it is,
however, difficult to do this
properly if they are suffering
from diarrhoea, and it is
therefore of prime import-
ance to avoid all risk of digestive trouble. Sudden changes
in the feeding are especially to be avoided ; in spring, the first
green fodder should be given in the stable, gradually decreasing
the proportion of dry fodder. If diarrhoea occurs in winter, the
proportion of beet must be decreased and the ration made up with
hay, while care must also be taken that the drinking water is not
too cold. Should the weather be cold and rainy in summer, the
cows should be kept in the shed at night, or supplied with covers.
Beet and turnip tops very often cause trouble, though this is rather
of a bacteriological than of a chemical nature. If the tops are not
given in large amounts, they will do no harm provided that they
have been harvested properly, but if they have lain in the field
and become wet and soiled with earth, they will develop undesir-

FIG. 52. — Rotting Swede containing numerous
Pectin- fermenting Plectridia.

Orla-Jensen, " Landwirtschaftliches Jahrbuch der Schweiz," 1905.

58 DAIRY BACTERIOLOGY

able fermentations. Diffusion slices l are generally supposed to
exert a particularly undesirable influence on milk in connection
with cheese making ; but as a matter of fact, they are harmless
in a properly soured condition ; on the other hand, if they are
given in spring or summer when they have begun to putrefy, they
will be just as dangerous as any other putrefying roots ; the author
has found that such material is the chief source of butyric acid
and aerogenes bacteria in milk.

It is obvious that cleanliness will be promoted by cutting the
hair on the udders and hindquarters of the cows as short as
possible. The condition of the ground or the covering of the stable
floor has a most important bearing on cleanliness. The most
favourable conditions are on the pastures in dry weather. In
rainy weather the cows may be badly soiled with mud which, as
has been mentioned above, is a fruitful source of sporing bacteria ;
milk obtained under these conditions is on this account difficult
to sterilise. In damp meadows or during a prolonged spell of wet
weather, microorganisms will grow abundantly on the surfaces of
plants. The floor of the cowshed must be well covered with fresh
straw, but it is better for the cows to rest on a clean cement floor
than on decaying straw or husks which give rise to dust rich in
microorganisms. Peat, especially the long fibred variety, is good,
but it must not be allowed to remain until it becomes sloppy. If
the shed is not planned so that the cows cannot lie down in their
dung, someone must be at hand to remove or cover up the dung
at once. There should be a sharp fall in the floor where the dung
is deposited. If once the udders have become badly soiled, it will
be impossible to obtain the milk in a decent condition. Rubbing
the udder with a cloth, which will very quickly be dirtied, is of
little avail ; proper cleaning of the udders will be very difficult of
accomplishment on large farms. Prevention will be found to be
far easier than cure, and if this matter were only given the atten-
tion which it justly deserves, the farmer would be rendering a
most valuable service to the cause of clean milk.

It should go without saying that everyone who touches food
should have clean hands, but it is of little avail that the milkers
wash their hands only to soil them again immediately they com-
mence their work. Of course, something will be gained if the
hands are thoroughly washed after the milking of each cow,
provided that they are not soiled again by touching the bottom
rim of the pail or the cobwebs on the beams, or by slapping the
dirty flanks of the cow to make it give room for the milking.
Clean hands and clean clothes are without doubt much to be

1 A by-product from beet sugar manufacture, which is used as fodder in
countries where this industry has been developed.

CLEANING AND THE PROCUREMENT OF MILK 59

desired, but in the first place we must insist on clean cows and
clean sheds ; when these points have been gained the rest will no
doubt follow as a matter of course. Dry milking is naturally
more hygienic than wet milking, but as experience has shown, it
is extremely difficult to carry out in practice ; it is facilitated by
smearing the teats with a little vaseline or fatty material. This
practice, which was first proposed by Gillebeau of Berne, has the
additional advantage that the dirt is taken up by the greasy matter ;
should a drop of the grease fall into the milk it will not mix readily
with the latter, and it will be removed on straining through cotton
wool. The hope that cleaner milk would be obtained by the use

FIG. 53.— Milking Room at Fauerholm.

of milking machines has hitherto been disappointed. These
machines, with their numerous corners, cavities and rubber tubes,
are so difficult to clean and sterilise that they require far more
intelligent and conscientious attention than they are likely to
receive at the hands of the average milker.

As already mentioned, the cleaned Pails should receive a final
rinsing with hot sterile water, and this is especially necessary if
the farm has not a good water supply. It has often been main-
tained that bad drinking water for the cows will give rise to bad
milk ; this, however, only applies in cases where the cows may
have contracted some disease through the water ; the direct
infections caused by rinsing the pails with bad water will have far

60 DAIRY BACTERIOLOGY

more serious consequences. To avoid the introduction of dirt
into the milk in filling or emptying the pails, cans or churns during
weighing or measuring, care must be taken that these vessels do
not become soiled during transport, and especially that they do
not become plastered with mud round the bottom rim where they
will be gripped on emptying. The cans should be covered with
tarpaulin during transport, and they should not be stood on the
ground but on clean planks.

The possibility of infection being carried by Flies is by no means
to be overlooked as microorganisms may be brought from any
place where the flies may have lodged. In cowsheds the danger
is best avoided by removing the dung as often as possible and by
hanging up, just below the ceiling, wide shallow enamelled dishes
containing skim milk to which four tablespoonfuls of formalin
have been added per litre (or quart).

To ensure fresh air during milking, the cowshed must be well
ventilated beforehand, and as many doors as possible be kept open
during the milking ; in this way, the additional advantage of a
good light will be secured. Cleaning operations of any kind or
feeding should be avoided just before or during the milking, in
order that the air may be as free from dust as possible. Naturally,
it would be an advantage if the hindquarters of the cows could be
brushed down just before milking l ; but the evil effects of the
dust raised in the process are only to be avoided if a special
milking room is available. Fig. 53 illustrates the room set aside
for this purpose at Fauerholm, near Frederiksborg. whence the city
of Copenhagen receives its highest grade milk, which is known as
" Ismaelk " (ice milk), as it is received into pails specially designed
by Busclc, which are provided with jacketed bottoms containing
a freezing mixture.

The milk is (or should be) strained on the farm, in order to
remove the coarsest of the dirt particles. By means of Ulander's
filter, in which the filtering medium is a layer of cotton wool,
most of the finer dirt particles may also be removed. Although
the renewal of the cotton wool each time it becomes choked may
entail some expense, this filter is strongly to be recommended, as
the benefit to be derived from the immediate removal of the dirt
is incomparably greater than that derived from the subsequent
cleaning which the milk may receive on arrival at the dairy, when
the soluble dirt and the bacteria contained therein will have
become distributed throughout the milk owing to the shaking of
the cans during transport. The best practice of all is to prevent
the dirt from ever entering the milk by tying a straining cloth over

1 In the procurement of the American " certified milk " the cows are
often vacuum cleaned before milking.

CLEANING AND THE PROCUREMENT OF MILK 61

the pail, as is done when using Gurler's pail (Fig. 54) ; the cloth
must of course be changed frequently in order to avoid the dirt
particles thereon being broken up and washed through by the
impact of the milk from the udder.

We need not here enter into the discussion of other refinements
calculated to further the production of clean milk, regardless of
cost, which have been adppted in some other countries, for the
public catered for in such cases is of necessity strictly limited ;
the points which do immediately concern the general public are
the necessity for improved cowsheds and for more efficient and
intelligent labour for the tending of the cows and the cooling of
the milk. As, however, these matters also involve extra expense,
the only hope of progress lies in the institution of a system of
payment according to quality.

The following figures illustrate the points discussed above : —

FIG. M.-Gurlers and Stadt matter's Milk Pails. (After Conn.)

Cow dung contains over 1,000 million organisms per gram.

Straw and earth contain up to twenty million organisms per
gram.

According to Barthel^ the air in a well-kept cowshed contains
on an average, 300,000 organisms per cubic metre during the
dinner hour, and ov-er a million during the feeding of the cattle.

According to Harrison, 20,000 organisms fall into the milking
pail per minute if the manure is removed and straw laid during
the milking, but only 1,000 if this work is done one hour before the
milking.

However clean the cows and the shed may be kept, it is im-
possible to obtain absolutely sterile milk as the udder, even when
healthy, always contains some bacteria which find their way in
through the opening in the teats 1. There will always be a fairly

1 A few investigators are of the opinion that bacteria also find their way,
into the udder through the blood.

62 DAIRY BACTERIOLOGY

large number of bacteria in the milk duct, and if high grade milk
is desired, the very first portion of the fore milk from each teat
should always be excluded. Thus the author found the following
results on milking from a washed udder and teats into sterile
bottles : —

In the fore milk from the four quarters, 16,000 organisms per

cubic centimetre.
In the middle milk from the four quarters, 480 organisms per

cubic centimetre.
In the strippings milk from the four quarters, 360 organisms

per cubic centimetre.

The number of bacteria present in the udders varies ; if the
teats are regularly washed and disinfected, and protected from
dirt between the milkings, e.g., by enclosing them in a waterproof
bag, the bacteria in the milk may be reduced to ten per cubic
centimetre. The American " certified milk " which contains less
than 10,000 organisms per cubic centimetre, keeps extremely
well ; instances have been observed of this milk keeping quite
good at 0° C. for more than a month, and only containing 1,000
organisms per cubic centimetre after a week's keeping.

According to Burri, milk fresh from the cow, obtained under
ordinary circumstances, contains 3,000 to 86,000, and on an
average, 21,000 organisms per cubic centimetre.

Chapter II

The Normal and Abnormal Microflora

of Milk

A. THE NORMAL FLORA

MILK fresh from healthy clean cows does not contain many
organisms beyond a few from the air, and those which normally
occur in and on the udder and teats. These organisms are nearly
all micrococci and sarcina forms, most of which are without
action on milk, while a few both acidify and peptonise it. Accord-
ing to Arthur Wolff 1, certain alkali producing short rod forms
(Bacterium lactis innocuum) rank next to the micrococci in point
of numbers, in fresh milk. In appearance, these organisms and
their colonies closely resemble the aerogenes bacteria, but they
do not ferment sugars, and they render milk feebly alkaline
instead of acid without causing any other change. According to
Burri and Hohl, Streptococcus liquefaciens is occasionally found as
a pure culture in the udders of healthy cows 2.

Milk which has been less carefully handled contains, in addition
to the above, coli, aerogenes, proteus and hay bacteria, ray fungi,
moulds, yeasts, fluorescent bacteria and sometimes also butyric
acid bacteria. The coli, aerogenes, proteus and butyric acid
bacteria generally come from the dung, the fluorescent bacteria
from the water used for rinsing the pails, and the others chiefly
from the bedding and the stable dust. If the cattle are on
pasture, the fluorescent and sporing bacteria may come from the
ground. Curiously enough, the typical streptococci are seldom
found in milk fresh from the cow. . According to BartheVs investi-
gations 3, they occur in cow dung with which they are distributed
over the fields, being therefore found on all cultivated plants.
From the latter they find their way back to the cow, and milk

1 Inaugural Dissertation, Zurich, 1908. Bacterium lactis innocuum is
probably identical with the organism known in the literature as Bacterium
alcaligenes.

2 " Schweitzerische Milchzeitung," 1916, Nos. 3 to 8.

3 " Landbruks-Akademiens Handlinger och Tidsskrift," 1905, p. 403. It
will, however, be necessary to revise these investigations since we have
now learnt better to differentiate between the various species of lactic acid
bacteria. Thus a large proportion of the lactic acid bacteria of the plants
are not streptococci in the narrower sense, but betacocci.

64

DAIRY BACTERIOLOGY

may thus be infected with streptococci not only direct from the
manure, but also from the bedding and the dust from the fodder.
The milk pails are equally important as sources of infection, for
they will as a rule be impregnated with lactic acid bacteria. These
bacteria may also be introduced by flies.

The bacteriological state of milk will be determined at the
outset according to the degree of care with which it has been
handled, and the nature of the bedding and the feed. If, how-
ever, milk is kept for any length of time, the temperature at which
it is kept will play an all-important part as it is the factor which
determines which groups of microorganisms shall develop in
preference to other groups. Not only the various groups of
bacteria, but also the milk itself must take part in the struggle for
existence, for milk contains bactericidal substances which, however,
are gradually weakened in their action, presumably by the bacteria
themselves ; only at low temperatures, at which bacterial develop-
ment is inhibited to a considerable extent, can these substances
retain their activity for any length of time, and at these tem-
peratures it is found that the number of organisms decreases at
first instead of increasing. In the following table are given the
numbers of organisms after twenty-four and forty-eight hours in
the same milk kept in sterile flasks at different temperatures. The
milk originally contained 84,000 organisms per cubic centimetre,
of which 2,000 were liquefying.

0°C.

12° C.

20° C.

30° C.

38° C.

45° C.

After twenty-four hours keeping.

Gelatine —
Total
Liquefying
Agar total .

52,000
18,000

8,200,000
1,600,000

163,000,000
6,000,000

*
380,000,000
200,000
460,000,000

*
17,400,000
0
20,000,000

*
12,000,000
0
100,000,000

After forty-eight hours keeping.

Gelatine — -
Total
Liquefying
Agar total .

252,000
60,000

27,000,000
1,800,000

*
350,000',000
2,000,000

*
380,000,000
0
500,000,000

*
3,000,000
0
60,000,000

*
1,200,000
0
222,000,000

* The milk was clotted.

It will be seen that at 0° C. the number of microorganisms was
lowered in twenty-four hours from 84,000 to 52,000 owing to the
action of the bactericidal constituents of the milk. At the same
time the number of liquefying bacteria had increased, and the

NORMAL AND ABNORMAL MICROFLORA OF MILK 65

examination after forty-eight hours showed the total number to
have increased. Chilling to 0° C. will therefore not prevent bacterial
development indefinitely. At temperatures above 10° C., bacterial
multiplication is rapid ; thus in the present example, after twenty-
four hours at 12°C., the number of organisms had increased one
hundred fold, and at 30° C., five thousand fold. Thirty degrees to
thirty -five degrees is the optimum for most milk bacteria ; already
at 38° C., many of them cease to grow and, moreover, the inhibitory
effect of the acid which is formed increases with the temperature,
as is shown in the table, particularly in respect to the liquefying
organisms. In order to make bacterial counts of samples kept at
the higher temperatures, it was necessary to employ agar, prefer-
ably in Bum's tubes (see p. 19) at 40° C., as organisms would be
present which would not grow at ordinary temperatures. A
comparison between gelatine and agar plates from milk kept for
forty-eight hours at 38° C. showed that the original flora was being
suppressed, while a new flora consisting entirely of rod-shaped
lactic acid bacteria was making its appearance. At 45° C., a
temperature unfavourable for the development of the common
milk bacteria, this change proceeds much more rapidly. The
lactic acid rods which develop at this temperature are mostly
thermobacteria producing Isevo lactic acid. The results of
investigations of the flora of milk at different temperatures, by
Conn and Esten 19 Arthur Wolff, Luxwolda 2 and the author, are as
follows :—

1. Under 5° C. the fluorescent bacteria predominate.

2. Between 5° C. and 10° C., in addition to fluorescent bacteria,
proteus bacteria, micrococci, alkali producing rods, and, as
Beijerinck has shown, also some rods which produce an aroma of
fruit.

3. Between 10° C. and 15° C. betacocci, streptococci, and some
species of aerogenes bacteria in addition to the above.

4. Between 15° C. and 30° C. the streptococci, especially Sc. lactis,
predominate.

5. Between 30° C. and 40° C. coli and aerogenes bacteria and lactic
acid forming rods in addition to streptococci.

6. Above 40° C. lactic acid forming rod bacteria and Saccharo-
mycetes, which ferment lactose, predominate. The streptococci
which are found are now chiefly Sc. thermophilus and Sc. fcecium.

It is well known that the lower the temperature to which milk
is cooled the better it keeps ; in this connection, some results
obtained by Kjaergaard Jensen may be quoted ; they repre-
sent the bacterial counts of the same milk kept for eighteen

1 Ann. Rep. Storr's Exp. Station, 1904.
" Centralblatt f. Bakt.," 2 Abt., 1911, Bd. XXXI., p. 129.

P,B. 5

66 DAIRY BACTERIOLOGY

hours at the temperatures which are most general in actual
practice.

Number of bacteria
per co.

Immediately after milking . . . . 1,480

After standing eighteen hours at 9° C. . . 2,100

12° C. .. 5,600

15° C. ._ 156,000

18° C. . . 550,000

21° C. . . 6,750,000

Naturally, the cleaner the milk the greater will be the benefit
derived from cooling. As already pointed out, it is not advisable
to keep milk longer than twenty-four hours, even if cooled to
0° C. ; if it is to be kept longer it must be frozen in order to avoid
the risk of the development of fluorescent and other water bacteria
which produce an unpleasant taste ; at somewhat higher tem-
peratures certain toxic proteus bacteria will also develop. For
these reasons cooled milk or cream which has stood for any
length of time are to be regarded writh suspicion, even if apparently
unchanged. As a rule milk does not become coagulated when
kept at temperatures below 10° C., but above this temperature
coagulation takes place in the course of a few days owing to the
action of rennet and acid forming bacteria. At 20° C. milk
quickly becomes coagulated, obviously owing to the formation
of acid ; at this temperature the streptococci, especially Sc. lactis,
develop so freely that after a time they will come to con-
stitute about 90 per cent, of the bacterial flora. The presence of
large amounts of lactic acid inhibits the growth of other milk
bacteria, for which reason the sour milk thus produced is harmless
as an article of food. As already mentioned, at higher tempera-
tures the streptococci give place to rod bacteria which produce
higher concentrations of lactic acid. As regards the gas forming
lactic acid bacteria, a few of these grow even below 10° C., but as
typical intestinal bacteria they have as a rule a high optimum
temperature and will most readily obtain predominance at
38° to 40° C. ; this temperature is somewhat high for the strepto-
cocci, and the slowly growing lactic acid rod bacteria will only
come to exercise an inhibitory effect on the other forms at a later
stage. For similar reasons the temperature mentioned is also the
most favourable for the development of the anaerobic sporing
bacteria which form butyric acid. On the other hand, the aerobic
sporing hay and potato bacilli, which often constitute the main
flora of pasteurised milk, are practically inert in raw milk ; their
spores do not germinate at the ordinary temperature, and at
higher temperatures the growth of their vegetative cells is inhibited

NORMAL AND ABNORMAL MICROFLORA OF MILK 67

by the lactic acid formed by other organisms ; as a group, they
are extremely sensitive to acid. There are, however, varieties of
Bacillus mycoides which can become accustomed to fairly large
amounts of acid and can even be induced to produce acid them-
selves.

As the aerobic organisms grow best near the surface of the milk,
and the anaerobic near the bottom, the pseudo lactic acid bacteria
will be more plentifully represented in the cream, and the true
lactic acid bacteria will predominate near the bottom of the
vessel l. Hence the spontaneous curdling of milk by souring
always starts from the bottom. While several of the pseudo
lactic acid bacteria produce Isevo lactic acid, the common lactic
acid streptococci produce the dextro acid only ; the result of the
combined action of the two groups is a mixture of inactive and
dextro acids. After some time, these organisms are supplanted
by lactic acid forming rods, not only at higher temperatures
(thermobacteria), but also after keeping at lower temperatures
(strepto- and betabacteria) ; as many of these rods form inactive
acid, the so-called fermentation acid will consist chiefly of this
variety.

In the spontaneous souring of milk at the ordinary temperature,
three stages may be distinguished. At the outset the original
flora of the milk develops rapidly ; as this is poor in true lactic
acid bacteria, the milk will only acquire a slightly unpleasant
smell and taste which may be described as stale. Only by degrees
do the streptococci gain the upper hand, whereupon the milk
becomes coagulated, and the acid which is formed masks the first
taints. Finally comes the third stage in which the lactic acid
rods predominate and in which the concentration of the acid may
rise from 0-6 per cent, to well over 1 per cent. Before this happens
however, a plentiful development of yeasts and moulds (especially
Torulae and Oidium lactis) will have taken place, and these
consume the acid or neutralise it by producing ammonia or other
basic products of protein hydrolysis ; in the latter case their
action is similar to that of chalk excepting that they destroy most
of the acid instead of conserving it. Through the joint action of
all these organisms the milk sugar will be fermented and the lactic
acid destroyed, so that the way is prepared for the putrefactive
organisms. As a rule, however, many weeks must elapse before
this state of affairs is brought about. The shallower the layer of
milk, the quicker will the acid disappear, provided, of course,
that the milk does not dry up in the meantime.

1 When whipped cream, i.e., cream containing numerous air bubbles, is
kept at the ordinary temperature, coli and aerogenes bacteria will obtain
predominance in it.

6— a

68 DAIRY BACTERIOLOGY

B. THE ABNORMAL FLORA

The changes brought about by the normal flora in milk cannot
be looked on as defects unless they appear at too early a stage.
Thus the fact that milk turns sour on standing is in itself not a
defect, but if the milk arrives at the dairy in a sour condition it is
certainly defective. Defects of milk we understand to be not the
normal but the abnormal changes. Defects of milk may be
spoken of as primary if present from the outset, or secondary if
they appear at a later stage in the history of the sample. In the
former case they may simply be due to changes to which the milk
is naturally subject at different stages of the period of lactation,
or they may originate from the fodder or diseases of the cows ;
in the latter case they will be due to the action of microorganisms
which find their way into the milk either during the milking or at
a later stage.

Primary Milk Defects. — It is well known that just after
calving or towards the end of the period of lactation the milk is
abnormal in composition and coagulates badly with rennet. While
colostrum has a strongly acid reaction, milk obtained towards the
end of the lactation period is rather neutral than acid. Owing
to the rapid growth of the foatus, much potash and especially
phosphoric acid are used up instead of passed into the milk. On
the other hand, the percentage of sodium chloride often increases
to such an extent that the milk tastes salt or bitter salt. It may
be pointed out that so long as the cow is not in calf, the milk may
be suitable for cheese making during two to three years 1.

The feed may have an undesirable effect on the milk by impart-
ing to it an abnormal taste and smell, the best known examples
being a taste of onions through feeding with onions, a turnip- like
taste through giving liberal amounts of turnips and swedes,
mustard and rape seed cake containing mustard, and a bitter
taste from lupins (unboiled) or large amounts of vetches. Further-
more, poisonous substances may be derived from certain plants
which, however, are generally avoided by the cows when on
pasture. Greater danger attaches to such poisons as iodine,
arsenic and mercury compounds which may be given as medicines
and thus passed into the milk. For this reason no milk should be
sent from any farm where the cows are given medicine of any
kind, without the sanction of a veterinary surgeon. Similarly,
disinfectants such as carbolic acid may pass through the blood
into the milk, or they may be absorbed direct from the air, in
which case the milk will also be unsuitable as food. Finally the

1 Orla Jensen, " Landwirtschaftliches Jahrbuch der Schweiz," 1905,
p. 542.

NORMAL AND ABNORMAL MICROFLORA OF MILK 69

milk may be poisonous on account of toxins which have passed
into it as the result of fevers or serious digestive troubles.

An indirect influence may of course be exercised by diseases of
the udder, among which inflammation of the udder or mastitis and
tuberculous udder are the most important. The alterations in the
composition of the milk are bacteriological as well as chemical.
In the first place the reaction of the milk- is altered ; frequently it
becomes alkaline rather than neutral, due to the inflammation
produced by Streptococcus mastitidis, though as a rule it eventually
becomes acid as this organism produces appreciable amounts of
lactic acid. The milk acquires a bitter, salt or other unpleasant
taste, and the percentage of lactose, the most constant factor in
the milk as long as the udder remains healthy, falls off appreciably.
Flakes and lumps of pus and casein will be seen in the milk, and
the colour changes. With streptococcic mastitis the milk becomes
yellow, and with tuberculous udder bluish. Occasionally the
milk may be coloured red by blood, and generally speaking,
increasing quantities of the constituents of the blood pass into the
milk while the normal constituents of the milk fall off. At last a
watery secretion containing pus is obtained which can no longer
be described as milk.

Inflammation of the Udder. — The most dangerous form of this
disease is that caused by Streptococcus mastitidis. It is very
infectious, being easily transmitted from one cow to another, so
that for this reason alone it is desirable that the milker's hands
should be washed after milking each cow. Inflammation of the
udder is also occasioned by Bacterium pyogenes and certain coli
and aerogenes bacteria and micrococci. B. pyogenes is a very
small rod form which may produce a very unpleasant smell in
the milk. A similar form, Bacterium minimum mammae, which
decomposes casein, but does not liquefy gelatine, and which
produces small amounts of lactic acid, has been found repeatedly
by Gorini x in the udders of cows which have not been milked
properly. Under these conditions, micrococci, possibly the same
as those which are normally found in and on the udder, may gain
predominance. This may also happen if the udder is in an
unhealthy state owing to chills. The micrococci generally only
produce light catarrhs. Bacterium pyocyaneum may also be
found in inflamed udders.

As most of the bacteria mentioned here may give rise to stomach
and intestinal diseases, milk from inflamed udders must be
regarded as dangerous.

Udder and other Tuberculosis. — Owing to the prevalence of

1 " Revue generate du Lait," 1907, Vol. VI., No. 24.

70 DAIRY BACTERIOLOGY

bovine tuberculosis in general (in Denmark 30 to 50 per cent, of
the cows are affected in one form or another), it is no wonder that
tuberculosis of the udder is often met with. Generally speaking,
large herds seem to be affected the most. As udder tuberculosis
makes rapid progress, the cows from which nursery milk is
obtained should be examined by a veterinary surgeon at least once
a fortnight. Owing to the dangerous nature of this disease, the
Danish law orders the slaughtering of cows with tuberculous
udders. Tuberculosis of the udder is, however, not the only form
of the disease which may involve the infection of the milk with
tubercle bacteria. This may also very well happen in cases of
tuberculosis of the uterus and kidney or the intestine, and even
tuberculosis of the lungs may be dangerous in this respect as the
animals swallow most of the slime which they bring up, with the
result that the bacteria pass into the manure. Every form of
open tuberculosis must therefore be regarded as a source of danger
as far as the milk is concerned. As tubercle bacteria do not grow
at temperatures much below blood heat, they will not multiply, in
milk or milk products under normal conditions, but as they are
not killed by small amounts of lactic acid they can live in butter-
milk. In the separating of milk the majority of the tubercle
bacteria are removed with the separator slime, though appreciable
numbers pass into the cream while only very few remain in the
separated milk. Raw milk is on this account less dangerous than
raw cream or butter. Tubercle bacteria can live in butter for a
much longer period than it is usually kept nowadays. In cheese
making the great majority of the bacteria as well as fat globules
are precipitated with the curd, for which reason milk and especially
fresh cheese may be far more dangerous than whey. According
to Harrison1, the hard cheeses may contain virulent tubercle
bacteria even after keeping for two months. The latest researches
have established that the organisms of human and bovine tuber-
culosis are different varieties, the latter being less dangerous to
adults than was formerly supposed though dangerous to children.
Tuberculosis is usually contracted by adults through inhaling the
dried saliva of consumptive persons. On the other hand, bovine
tuberculosis is very dangerous to calves and pigs, for which
reason a law was passed in Denmark at the suggestion of Professor
B. Bang, ordering all dairies to heat separated milk and buttermilk
to 80° C. before returning it to the farmers, who use it chiefly for
feeding pigs. It is regarded as a matter of the greatest importance
that pathogenic germs are excluded from Danish butter in a
similar manner.

1 " Landwirtschaftliches Jahrbuch der Schweiz," 1900, p. 317,

NORMAL AND ABNORMAL MICROFLORA OF MILK 71

Other Diseases of the Cow. — Just as milk from cows suffering
from tuberculosis of the uterus and intestine may easily become
infected with tubercle bacteria, so milk from cows affected by
other diseases of the sexual and digestive organs may become
infected with the corresponding organisms. Experience has
shown that the milk should not be used for cheese making until
ten days after calving, the reason being that not only is the
chemical composition abnormal, but also that during this period
the milk is especially subject to infection from the uterus, and to
taints from the disinfectants often used during calving. The
small rod form, Bacillus abortus, which causes abortion, may
occasionally be found in milk1. The consequences of ordinary
diarrhoea have already been mentioned. Chronic diarrhoea, when
the motions contain blood or are otherwise abnormal, is far more
dangerous, for in such cases we are dealing with an infectious
disease and the organisms in question (according to C. O. Jensen,
often coli bacteria of the swine fever group) may cause similar
intestinal trouble in human beings, especially in children. Mor-
tality among calves will always be a danger signal for human
beings. Finally there will always be a danger of infection through
the milk in cases of anthrax and all skin diseases such as foot and
mouth disease, cowpox, etc., more especially if the udder is
affected, and possibly also in cases of udder actinomycosis, acute
lung diseases and rabies. The chemical composition of the milk
is affected through several of these diseases.

Secondary Milk Defects. — Milk may also be infected with
pathogenic organisms from human beings, and any one suffering
from an infectious disease or living in the same house with sufferers
should be forbidden to handle or sell milk. It is generally accepted
that tuberculosis, diphtheria, scarlatina, cholera and typhoid fever
may be transmitted by milk. Epidemics of typhoid, cholera,
scarlet fever and diphtheria have undoubtedly been caused re-
peatedly through milk infections. Although Bacterium typhosum
does not ferment lactose, it grows freely in milk, and it is also
stated to be able to live for some time in butter and cheese ; like
the organisms of the other diseases mentioned above with the
exception of that of anthrax, it is killed by pasteurisation at a low
temperature.

1 Schroeder and Cotton, U.S. Dept. of Agric., Bur. of Animal Industry,
28th Annual Report, 1911 ; Alice C. Evans, " Journal of the Washington
Academy of Sciences," Vol. V., No. 4, 1915, and " Journal of Infectious
Diseases," Vol. XVIII., No. 5, 1916. According to the latter paper. Bacillus
abortus and other similar forms seem to be fairly common udder bacteria,
to be found in over 20 per cent, of the milk fresh from the cows. It appears
to be chiefly a fat-splitting variety (var. lipolyticus), which is killed after
warming only to 52° C. for thirty minutes.

72 DAIRY BACTERIOLOGY

While many of the above mentioned defects, however dangerous
they may be, will not be readily detected and indeed are only to be
demonstrated with great difficulty, the following defects will hardly
escape observation, and it is accordingly these which are commonly
alluded to in speaking of milk defects. We will begin with those
which are visible and then pass on to defects of taste and smell.

Fermenting and Gassy Milk. — Milk may be described as fer-
menting if it shows copious evolution of gas even before coagula-
tion sets in, and gassy if this only occurs after keeping for some
time at a high temperature (see the fermenting test). Both
defects are due to coli and aerogenes bacteria (more seldom to
yeast), and the difference between them is only determined by the
degree of infection. Fermentation may arise through direct
infection from fermenting fodder or when the cows are suffering
from violent diarrhoea or aerogenes mastitis. Both fermenting
and gassy milk should be avoided in cheese making. Fermenting
cream may give trouble in churning.

Prematurely Coagulating and Cheesy Milk. — Badly-cooled milk
may coagulate when only a few hours old ; as the milk will not be
appreciably sour in such cases the phenomenon must be ascribed
to an action resembling that of rennet ; it is always due to excep-
tionally active development of peptonising lactic acid bacteria
(Streptococcus liquefaciens or Tetracocci), which will probably
have started to secrete coagulating enzymes in the udder. The
well-known phenomenon of milk being specially liable to curdle
in thundery weather does not seem to be ascribable to any other
reason than the high temperature which usually precedes a
thunderstorm. Milk which has been coagulated by rennet is
generally described as cheesy ; it can easily be distinguished from
sour milk by the separation of clear whey and the contraction of
the coagulum. If gas-producing enzymes are present as well, the
curd will become flaky or lumpy owing to the disturbance of the
milk by rising bubbles. The same effect may be produced in
practice in presence of coagulating bacteria alone, if the milk is
stirred or shaken.

Slimy and Ropy Milk. — Milk may become slimy on standing, and
sometimes the effect is so pronounced that the milk may be
pulled out into threads a metre long. The change is most marked
at 18° to 20° C. At higher temperatures the slimy organisms may
be suppressed by the lactic acid, but even in pure cultures they
form more slime at the temperatures mentioned. According to
Gillebeau, one of the * commonest slime-producing organisms is
Micrococcus Freudenreichii, a large coccus1 2 ^ thick, which

1 " Landwirtschaftliches Jahrbuch der Schweiz," 1891, p. 135, and
1902, p. 342.

NORMAL AND ABNORMAL MICROFLORA OF MILK 73

liquefies gelatine and produces an appreciable effect in milk
within five hours. Like several allied forms, it may occur in bad
water. If a stable once becomes infected with these organisms,
nothing short of a very thorough disinfection will eliminate them.
The author has isolated a non-liquefying micrococcus, which, being
an obligate aerobe, only makes the surface of the milk slimy ; in
pure cultures, the sliminess is preserved for many weeks ; the
milk becomes faintly alkaline, but is not peptonised. It turns
the surface of agar brownish black. Storm has isolated a motile
rod which similarly only turns the surface of the milk slimy ; it
coagulates and peptonises the milk. As already mentioned, the
slime is derived from the outer portion of the cell membrane.
While the organisms mentioned can form slime from the protein
of the milk, the lactic acid bacteria which produce slime require
sugar in order to do so, but, on the other hand, they make the milk
slimy throughout. As has been mentioned, the most important
slime-producing lactic acid bacteria are certain varieties of
Sc. cremoris.

As far as the Swedish " long milk " or the Dutch " long whey "
are concerned, sliminess is a desirable characteristic, but otherwise
it is highly undesirable, for slimy cream gives a bad yield of butter,
and slimy whey is difficult to press out of cheese, collecting under
the rind, and, as Bum has shown, causes the cheese to crack at a
later stage1. For this reason slimy bacteria are no longer used
in making Dutch cheese, as other lactic acid bacteria have been
found to possess the same advantages without the disadvantage
in question. Further, slimy organisms easily lose their character-
istic property, especially if grown at higher temperatures ; con-
versely, the common Streptococcus cremoris is inclined to make
milk slimy if cultivated at 'lower temperatures for any length of
time. Other lactic acid bacteria, pure (certain thermobacteria),
as well as pseudo (certain aerogenes bacteria), may cause sliminess.

Coloured Milk. — In the old-fashioned process of allowing milk
to stand in order to let the cream rise, coloured spots often
appeared on the cream, or the milk would gradually become blue
or red throughout. As these defects have not been met with in
practice since the introduction of the centrifuge, they need not
detain us here. They are due to the colour-producing organisms
already mentioned (see pp. 44, 45 and 46).

Milk with an Unclean Sour Taste. — There is no defect more
commonly met with in dairies than this one, as it is in no way due
to foreign or rare milk organisms. It is simply the stale stage
prematurely reached owing to insufficient cooling.

1 " Centralblatt f. Bakteriologie," 2 Abt., 1904, Bd. XII., p. 192.

74 DAIRY BACTERIOLOGY

Milk with Stable or Grass Taste. — Formerly it was supposed that
a stable smell was exclusively due to air absorbed in the stable.
The taste, however, usually becomes worse after the milk has left
the stable, and it has been shown to be due to the bacteria which
produce the smell in the stable, i.e., the intestinal bacteria. A
heavy infection with manure will not only give the milk a taste of
manure by direct means, but will introduce a number of bacteria
which will continue the decomposition of the specific constituents
of the manure and, naturally enough, do not neglect the con-
stituents of the milk itself. In the same way. milk may acquire
an unduly strong aroma of grass in spring. In the usual course
some principles of colour and taste will always pass into the milk
from the young grass, but the strong odour of herbs, which is
appreciated by some people, will only arise on infection with the
liquid secretions of grass.

Milk with a Taste of Turnips. — As already mentioned, the milk
acquires this taste when the feed includes too great a proportion
of the pungent principles characteristic of the Cruciferae, turnips
and swedes being especially dangerous if given in a rotting con-
dition or too cold, so as to cause digestive troubles. Under these
conditions other roots may, of course, also have an undesirable
effect1. The taste-producing constituents of the roots need not
necessarily pass into the udder, but may also be introduced into
the milk with the manure, and, still worse, they will then be
accompanied by the microorganisms which vegetate on the roots,
and which, therefore, are equipped with those enzymes which are
capable of hydrolysing such substances as the glucosides of
mustard oil. According to Weigmann, the active organisms in
such cases are chiefly coli bacteria, Penicillium brevicaule and
certain species of oidium. C. 0. Jensen found in the course
of his well-known researches at Duelund dairy, which led to
the pasteurisation of cream for butter making2, that a coli
bacterium living in water could produce in the milk a very un-
pleasant turnip-like taste even when the cows had not been fed
on roots at all. Bacterium fluorescens liquefaciens also produces
a taste of turnips, and this defect is accordingly often met with in
milk which has been kept for any length of time at a low tem-
perature. Weigmann has isolated a non -liquefying fluorescing
bacterium, Bacterium carotce, which produces a strong smell of
carrots in all nutrient media. The author has cultivated this

1 According to Johannes Uolle, " Zeitsclirift f. Untersuchung d. Nahrung
u. Genussmittel," 1915, Bd. XXX., p. 361, feeding with liberal amounts of
beets may cause betain to pass into the milk. As this substance is a base,
it will delay the coagulation of milk by acid.

2 C. O. Jensen and Lunde, " Forsogslaboratoriets/' 22 Beretning,
1891.

NORMAL AND ABNORMAL MICROFLORA OF MILK* 75

organism for years, the smell of the cultures remaining as strong
as ever.

Milk with a Soapy Taste. — As is well known, this taste arises
when milk is neutralised with alkalies, and is therefore produced
by bacteria which chiefly produce ammonia. As the neutralisa-
tion of the milk is counteracted by acid production, and the
ammonia-forming bacteria grow better at low temperatures than
the acid-forming bacteria, it will readily be understood that the
defect in question is most noticeable in cooled milk. Bacterium
sapolacticum, isolated by Eichholz, can grow even at 5° C. ; it is
a non-liquefying fluorescent organism.

Bitter Taste in Milk. — This taste nearly always occurs in milk
which has not been completely pasteurised, being produced by
the peptonising sporing bacteria. Bitter substances are nearly
always formed in the first stages of protein hydrolysis. The
defect may also be due to the cow, to certain feeding stuffs,
yeasts (Torula amara) and peptonising cocci, especially Strepto-
coccus liquefaciens. According to Weigmann, Bacterium Zopfii,
Bacterium lactis innocuum and aerogenes bacteria may turn milk
bitter.

Metallic and Tallowy Taste in Milk. — Badly-tinned vessels will
always impart a metallic taste to milk which is acid or fairly warm.
In Denmark the pasteurised separated milk is sent back warm
from the co-operative dairies to the farms, and in such cases it is
difficult to prevent the cans from being attacked to a somewhat
greater extent than usual, and thus giving the milk a metallic
taste. The metallic taste must not be confused with the tallowy
taste which may arise when the cream layer is exposed to direct
sunlight, or through the action of certain microorganisms. As the
metallic taste which is caused by copper is not to be distinguished
from the tallowy taste, it may be presumed that copper salts
promote by purely catalytic means the oxidation processes owing
to which the tallowy taste arises. According to Rosengren l, it is
only milk or cream which is or has been heated which gets a
foreign taste from the copper. The same holds good for butter
made from the cream. According to Storch, certain lactic acid
bacteria may cause a tallowy taste. Most liquefying rod bacteria
(the hay bacillus, etc.), may produce a sickly, tallowy taste at
first, but the more rapidly they peptonise the milk the sooner will
the taste become bitter. The author has found that some aerobic
non-sporing rod bacteria occasionally occurring in water are
generally the cause of the tallowy taste which may arise in milk
on long standing 2.

1 " Landbrugsforsogsmeddelelse," No. 197, 1920.

2 As an instructive example it may be mentioned that a dairy which

76 DAIRY BACTERIOLOGY

Bacteria are known which form hydrogen sulphide from sulphur,
and therefore also from vulcanised rubber. Milk which has passed
through rubber hose pipes, such as in milking machines, may
therefore, on warming to a relatively high temperature, acquire a
smell of rotten eggs.

supplied Copenhagen with pasteurised cream which tasted good on arrival,
but which became tallowy before it reached the consumers, eliminated this
defect by rinsing the transport cans with boiled water. The rod bacteria
mentioned above were found both in the cream and the water from this
dairy.

Chapter III

The Preservation of Milk and its Treat-
ment for Direct Consumption

PRESERVATION

THE general methods used for the preservation of foods are,
constant cooling, short heating, concentration, and the addition
of bactericidal substances. All four methods or combinations
thereof are applied to milk.

Cooling is most used because it is the cheapest method and
because the milk suffers no chemical change in the process. The
principle of the method has been discussed above, and it is well-
known that the sooner and the more thoroughly the milk is cooled
the better will be the result. Even in cold weather the milk
cannot be cooled quickly enough by simply allowing it to give up
its heat to the air. It is absolutely necessary to stand the pails
in cold water which is frequently changed ; the level of the water
must be higher than that of the milk in the pail. The water
container should be fitted with a grid or the bottom should be
fluted so that the water may circulate freely under the pails.
The same object may also be achieved by standing the pails on a
flat surface if their lower rims are provided with holes. The water
inlet pipe should reach the bottom while the overflow should be
at the top. On large farms it is best to have a tank above the
cowshed, which is filled just before milking, and from which the
water slowly flows through wooden troughs situated on the coolest
side of the shed ; the cooling may thus be started during the
straining process. The pails are first placed in the end of the
trough nearest to the outlet and are moved by degrees towards
the inlet end, whence they are removed from the trough. The
troughs must be covered over by a lean-to roof sufficiently high to
allow a man to stand under it. The used cooling water may with
advantage be used for drinking water for the cows. When the
milk has been cooled to the temperature of the water, it may be
further cooled by standing the pails in ice water or by passing it
over a cooler through which a mixture of four parts of ice in small
lumps and one part of salt is circulated. In summer time the
cooler may be moved out into the fields, though if this is done the

78

DAIRY BACTERIOLOGY

consumption of ice will be unreasonably large. The accompany-
ing illustration shows the arrangement of the cooling apparatus
at Fauerholm dairy which supplies Copenhagen with milk. Here
the milk is cooled to 3° C. By allowing the milk warm from the
cow to flow over a cooler or a special aerating apparatus, some of
the unpleasant animal odour is eliminated, but at the same time
the milk loses carbon dioxide and takes up oxygen, with the result
that the development of the aerobic putrefactive bacteria will be
favoured at the expense of the true lactic acid bacteria. As

FIG. 55. — Cooling and Straining of Milk at Fauerholm. In the foreground
Ice is being put into the bottom of Busck's Milk Pail.

aeration will also expose the milk to air infection, it may often
do more harm than good on the whole. This will be the case
when the cows are in the stable, but here it will be easy to secure
rapid and thorough cooling by other means.' It is a different
matter when the cows are on pasture where the air will be pure,
and more than three hours may elapse before the milk arrives at
the farm to be cooled ; in this case immediate cooling on an aerating
appliance will do a great deal of good. Weigmann1 has found that
pasture milk contains on an average, when retailed, six times as
many bacteria as stable milk obtained under the same conditions

1 " Centralblatt f. Bakteriologie," 2 Abt., Bd. XLV., p. 105. •

PRESERVATION OF MILK AND ITS TREATMENT 79

of cleanliness, a fact which is undoubtedly to be attributed to the
delay in the cooling of the former. Mid-day milk can only be
satisfactorily preserved until the following morning by the use
of ice ; as this further adds to the cost of production incurred by
milking three times a day instead of twice, it is a question whether
the extra 6 to 7 per cent, of milk and milk fat which, according to
Fleischmann1, is gained in this way, is not too dearly paid for,
considering the present cost of labour. There is no doubt that
the bad state in which the mid-day milk arrives at the dairies can
only influence the dairy products for the worse.

Heating. — As already explained, cooling only serves to preserve
milk for a comparatively short time, for it would entail too great
an expense to maintain the low temperatures necessary to prevent
all bacterial development. Milk cannot be kept for any length of
time without the aid of heat. The temperature and time of
heating necessary to sterilise milk depend very largely on its
quality, and on the size of the bottles or tins which are used.
With tins, water filled autoclaves are best as the heat is transferred
more rapidly and uniformly by this method. By heating for one
hour at 105° C., or for a quarter of an hour at 115° to 120° C., milk
will usually be rendered sterile, and may then be kept for years,
especially if care be taken to close the bottle immediately after the
autoclave is opened, and while the milk is still boiling. The
bottles will be practically free from air, and as the most resistant
spores require an abundant supply of oxygen in order that they
may germinate, the keeping properties of the milk will be greatly
enhanced. In order to prevent the fat globules from collecting
as a solid lump in the neck of the bottle, the milk should first be
homogenised. As sterilised milk is brown in colour and consider-
ably altered in chemical composition, it cannot be recommended
for children. The steriliser invented by Jonas Nielsen is less
destructive in its action, the milk being momentarily heated to
130° to 135° C. by being pumped through a system of small thin
pipes heated by steam. Burning and the change in taste which
this causes, are completely avoided if only the milk is kept during
the time of heating at a pressure sufficiently high to prevent it
from boiling. According to the author's investigations, even milk
to which have been added very resistant bacterial spores, already
becomes sterile in forty seconds at 135° C., and in fifty seconds at
130°, but only after ninety seconds at 125° C. It naturally follows
that the milk must be cooled very rapidly after this drastic heating
in order that it may not become brown ; this is accomplished oy
means of a system of piping similar to that which is used for the

1 " Das Molkereiwesen," Brunswick, 1875, p. 81.

80 DAIRY BACTERIOLOGY

heating process, but which is surrounded by cooling water instead
of by steam. If the milk sterilised in this way is filled into
sterilised cans, avoiding infection from the air (Jonas Nielsen has
patented special appliances for this purpose), it may be sent any
distance. On passing through Jonas Nielsen's steriliser the milk
undergoes a partial homogenisation. Milk and cream intended
for export should, in order to meet all eventualities, be sent as
cold as possible, and in this connection it may be mentioned that
the results of cooling are in general far more noticeable in heated
than in raw milk (provided only that the milk vessel has not been

FIG. 56. — Jonas Nielsen's Steriliser.

contaminated or rinsed with unboiled water), for the bacteria
which grow best at low temperatures are the first to be destroyed
on heating while the most resistant spores only germinate at high
temperatures. In the production -of sterilised cream (export
cream) homogenisation also plays a very important part, as cream
with a certain fat percentage appears to be richer the more finely
divided the fat globules. Cream, however, like condensed milk,
cannot be homogenised at quite so high a pressure (at the most
150 atmospheres) as ordinary milk which may be treated right up
to 250 atmospheres.

If the object is only to free the milk from pathogenic germs, and
to increase its keeping powers to some extent, complete sterilisa-

PRESERVATION OF MILK AND ITS TREATMENT 81

tion is not necessary and a less drastic treatment, i.e., pasteurisation,
will suffice. The easiest method consists in heating the milk for
a short time, two minutes at the most, in a continuous pasteuriser.
It has been shown * that the keeping powers of milk may be
improved by heating even to 70° to 75° C. in this way. According
to American researches 2 a far better result is obtained at 80° C.,
and as this temperature is necessary to ensure the destruction of
the tubercle bacteria, it has been fixed as the minimum by the
Danish pasteurisation law. Very little is gained by increasing the
temperature to 85° C., and the milk or cream easily acquires a
" cooked " flavour if it cannot be cooled rapidly enough. On the
other hand, skim milk should be heated to 90° to 95° C. if, as is
the practice in Denmark, the milk is to be returned from the
co-operative dairy to the farms in the unrinsed cans ; in this
way the remains of milk in the cans will also be pasteurised.

From the bacteriological point of view, approximately the same
results may be achieved by prolonged heating at lower tempera-
tures as by short heating at higher temperatures ; in America
milk is often pasteurised for half an hour at 60° to 70° C. The
author has found that this method of treatment has a less destruc-
tive effect on the proteins 3, and is therefore to be recommended
for the treatment of milk for infants or for cheese making, where
it is of importance to preserve the natural properties of the milk.
The following table shows that on heating to over 70° C., albumin
coagulation occurs fairly rapidly, while the milk loses its power of
coagulation with rennet 4.

Percentage of albumin

Time to coagulate with rennet accord-

Time heated in

coagulated. ing to Schaffer in minutes.

70° C.

75°C.

80° C.

90°C.

Unheated.

70°C.

75° C.

80° C.

90° C.

Momentarily

_

15

39

71

13

_

14

17

26

5

13

49

88

100

15

16

19

28

28

15

18

62

91

100 15

17

22

60

60

30

30

83

100

100

15

19

32

57

57

60

37

93

100

100

14

20

34

46

46

" Forsogslaboratoriets 22 Beretning," 1891.

2 .Harding and Eoqers, New York Experimental Station, Bulletin 172,
1899.

3 " Landwirtschaftliches Jahrbuch der Schweiz," 1905.

4 It must be added that the extent of the change suffered varies greatly
with different samples of milk. In the experiments recorded in the table
the same samples .were used throughout for the same times of heating, but
different samples for different times of heating ; a maximum effect was
soon reached, which was first exceeded when the milk was heated to a
temperature sufficiently high to turn it brown.

82 DAIRY BACTERIOLOGY

The first visible £Jfpr.t of iveating^is the gradual destruction^ of
the~cTearrr line, and as the public usually judge the richness of the
milk according to the thickness of the cream layer which separates
out, this matter is not without significance in practice. The cream
line is already affected by heating to 63° C. for some time 1, and as
it happens that heating for half an hour at this low temperature
is critical for most milk bacteria 2, especially the colon bacteria,
this method of treatment, which we will refer to as low temperature
pasteurisation (the American holder process), would seem to be the
best. Moreover, the bacteria which cause foot and mouth disease,
diphtheria, cholera, dysentery, typhoid and other intestinal
diseases, are killed just as surely by low temperature pasteurisation
as by short heating to 80° to 85° C. Different opinions have been
held with regard to tubercle bacteria, but according to Barthel and
Stemstrom's researches 3 there can no longer be any doubt that
these organisms will under normal conditions be killed even by
only heating to 60° C. for ten to twenty minutes.

Whichever method of pasteurisation may be adopted, it is of
the greatest inapo^ajice to prevent the fora^tionjof s^in and
frptli> as these act as insula/tors, prevenETng the transference of
heat. This point was first demonstrated by Theobald Smith 4.
Thus, wheji pasteurising at low temperajtur^Jr^ large^vessels, the
milk must be in constant though not violent motion. Another
condition is that the milk must not contain so much acid or rennet
that the heating will precipitate the casein which may thus come
to enclose and protect the bacteria. Pasteurisation can by no
means be trusted in implicitly to cover all defects, for the result
depends largely on the state of the raw milk. Fresh and cleanly
handled milk may be practically sterilised on pasteurisation,
though the milk flowing from the pasteurisers in some dairies
frequently contains several thousand bacteria per cubic centimetre,
and the subsequent passage over the coolers will, of course, not
improve matters in this respect. The expression of the efficiency
of pasteurisation as the percentage of bacteria killed on the
number originally present is somewhat misleading ; using the
processes described above, the percentage is usually over ninety-
nine with bad milk, but it often falls as low as ninety with good
milk, although the actual number of organisms is always many
times greater in pasteurised bad milk than in pasteurised good
milk.

1 Weigmann, " Mitt, des Deutschen Milchwirtschaftliches Verieins," 1914,
Bd. 31 ; Burri, " Schweitzerische Milchzeitung," 1915, Nos. 42 and 43.

2 Ayers and Johnson, " Journ. of Agric Res.," 1915, Vol. III., No. 5.

3 " Meddelande," Nos. 117 and 118, fran Centralanstalten for forsok-
svasandet pa jordbruks omradet, 1915.

4 " Journ. of Experimental Medicine," New York, 1899, vol. 4.

PRESERVATION OF MILK AND ITS TREATMENT 83

It is obvious that the sooner the individual particles of the milk
can be heated, the shorter will be the time occupied by the process
of heating, and by spraying the milk into the pasteurising vessel,
as is done in Oscar Lobeck's Biorisator, an effect is attained by
momentary heating to 75° C , which is as efficient in killing the
bacteria and as lenient towards the milk itself as that attained by
heating to 63° C for half an hour. Biorisation may thus be regarded
as a process intermediate in character between high and low
temperature pasteurisation. Unfortunately the atomiser becomes
choked so easily that the apparatus cannot be recommended in its
present form. On investigating the Biorisation process 1) the
author obtained the following remarkable results : — Not only the
milk treated at 70° C., but also that treated at 80°, 85° and 90° C.,
kept worse than that treated at 75° C. Milk treated at 70° C.
rapidly became sour, while that treated at 80° C. became putrid,
mainly owing to the action of hay and potato bacilli. Tholstrup
Pedersen 2 repeated these tests, heating the milk as quickly as
possible in a boiling water bath. As this method of heating is
less rapid than biorisation, the best results were obtained at a
somewhat lower temperature, i.e., 70° C. The fact that hay bacilli
and other heat-resisting organisms grow quicker in milk heated to
higher temperatures can only be explained by the supposition that
the bactericidal constituents of the milk are killed at temperatures
above 70° C. ; Tholstrup Pedersen has adduced other evidence in
favour of this supposition. From this it follows that the bacterial
count of milk immediately after pasteurisation is not a perfectly
.reliable gauge of its keeping power, as the keeping power depends
just as much on the bactericidal properties of the milk as on the
count (and of course also as on the nature of the organisms
present). If the milk is kept at 10° C. or lower temperatures,
these properties are more noticeable in lightly pasteurised milk
in which the surviving organisms have become weakened by
heating than in raw milk, an appreciable diminution in the
number of organisms taking place during the first twenty-four
hours.

If milk is pasteurised at a high temperature (85° to 95° C.) or
boiled (as in Soxhlet's apparatus for pasteurising nursery milk),
only the spores survive, and milk or milk foods (e.g., chocolate)
treated in this manner will therefore not become sour on standing,
but will develop putrefactive or butyric acid fermentations. In
most cases Bacillus mycoides and other aerobic sporing bacteria
will be found. The milk soon acquires an unpleasant, sickly
taste, a particularly dangerous feature being that poisonous sub-

1 " Maelkeritidende," 1915, p. 483.

2 "Maelkeritidende," 1916, p. 231.

«— 2

84 DAIRY BACTERIOLOGY

stances may be formed before any apparent change has set in.
Immediate cooling to under 14° C., or better still to under 10° C.,
delays these changes considerably. As many of the sporing
bacilli can grow and even thrive well at high temperatures,
the slow cooling of milk pasteurised by the methods under
consideration may have disastrous results ; Tholstrup Pedersen
has shown l that milk treated thus is affected most rapidly at
70° to 60° C. At these temperatures non -motile, aerobic rods
will appear after only three hours' standing ; after four hours
the milk is so changed that it will no longer stand boiling, and
after six to eight hours it coagulates owing to the action of
fermentation acid and bacterial rennet. At 60° to 50° C., the
bacteria in question do not grow so rapidly, but in their place
anaerobic plectridium forms develop freely. It is only under 50° C.
that the common hay and potato bacilli appear, while at 40° to
30° C. the true butyric acid bacteria appear as welL The above
outline gives an idea of the possible bacterial developments in
milk pasteurised at high temperatures and allowed to cool spon-
taneously. It has an important bearing on the treatment of skim
milk in the Danish co-operative dairies ; as was mentioned above,
the skim milk is sent back warm from the pasteuriser to the farms ;
it is contained in cans holding up to 50 litres, and after three to
four hours its temperature may be above 50° C., and in summer
sometimes above 60° C.2 Formerly it was thought that the holding
of milk at high temperatures over long periods enhanced the
effect of pasteurisation, but now it is known that while this
treatment certainly tends to destroy the normal flora of the milk,
especially the lactic acid bacteria, it also encourages the growth
of another and far more dangerous group of organisms. For this
reason skim milk should be thoroughly cooled in the dairy ; there
is no reason for heating it to temperatures above 80° to 85° C., the
limit set by the Danish pasteurisation law. If this procedure is
adopted, the cans must be cleaned before receiving the cooled
milk ; the extra trouble thus involved is amply paid for by the
saving in coal which may be effected by the use of the regenerative
system of cooling ; moreover, the quality of the separated milk
will be improved both as regards taste and keeping powers, while
the tinning on the cans will last longer. It should be pointed out
that the cleaning of the cans by the dairy by no means relieves
the consumers from the obligations in this connection. The
harmful bacteria may also be suppressed by souring the milk, but

1 " Maelkeritidende," 1915, p. 817, and 1916, p. 35.

2 These investigations also point to the necessity for caution in hay-box
cookery, the food being kept for many hours at temperatures favourable
to the development of thermophilic bacteria.

PRESERVATION OF MILK AND ITS TREATMENT 85

as the resulting product cannot always replace unsoured milk,
there is no need to go to this trouble unless facilities are lacking
for cooling the milk to under 14° C. l From the results of these
investigations it may also be understood why milk or cream
which has been homogenised at 60° to 70° C., and which commonly
has to stand for some time before it can be sterilised, may acquire
a bad taste. The homogenisation should rather be accomplished
at 80° to 90° C. , after which the milk should be sterilised at once.
While high temperature pasteurisation alters the relative pro-
portions between the numbers of the good and the harmful
bacteria to the advantage of the latter group, the opposite effect
may be produced by low temperature pasteurisation. Ayers and
Johnson 2 have shown that the lactic acid bacteria have, as a
group, greater heat-resisting powers than was formerly supposed,
and that a larger relative proportion of these organisms is to be
found in milk which has been warmed to 63° C. for half an hour
than in the raw milk. This would appear to be an additional
recommendation for low temperature pasteurisation. On the other
hand, it must be mentioned that the lactic acid bacteria which
survive this process sour the milk slowly at ordinary temperatures,
and that there is yet a possibility that the milk treated in this way
may suffer undesirable changes before it becomes sour. If the
raw milk is particularly good, and therefore poor in true lactic
acid bacteria, there is a possibility that it may eventually become
just as harmful after low temperature pasteurisation as after
heating to higher temperatures. In ordinary milk the true lactic
acid bacteria are only killed off entirely by heating to 77° to 82° C.
for half an hour. In agreement with this observation, Tholstrup
Pedersen 3 found that milk which had been heated to 80° C. still
contained many lactic acid bacteria, but that milk heated to
85° C. contained none. The author has found that the lactic acid
bacteria growing at high temperatures, i.e., the thermobacteria,
and in addition the microbacteria, Streptococcus thermophilus,
Streptococcus fcecium, Streptococcus glycerinaceus, and a few micro-
cocci are the lactic acid bacteria which best survive heating.
The heat-resisting powers of the thermobacteria have been taken

1 The foregoing remarks, as well as those on the same subject on p. 70,
are of particular interest in connection with the Danish co-operative dairy
industry, in which separated milk is obtained as a by-product from butter
making and sent to the farms, where it is extensively used for feeding pigs.
The matter is, however, of general interest as an example of the great
importance of cooling after pasteurisation. In the large dairies of the
United Kingdom the separated milk is usually treated as recommended
above before sending to biscuit and other factories. — Translator.

2 U.S. Dept. of Agric. Bureau of Anim. Industry, Bull. 161, 1913 ;
" Journ. of Agric. Kesearch," 1914, vol. 2, No. 4.

a " AT r»^n^/-K».,' •*-:/! /->^-,^i^ " 1015 -p SPJ

86 DAIRY BACTERIOLOGY

advantage of in the Swiss cheese making industry for many
centuries. Pure cultures are obtained in the acid vats by filling
up with warm " Schotte " (boiled clarified whey) so that the
temperature rises to about 60° C. The thermobacteria are
of no great interest in connection with pasteurised milk, as
they develop very slowly in it unless kept at temperatures
over 35° C.

According to W eigmann's l investigations on low temperature
pasteurisation, practically the whole of the bacterial reduction
takes place during the first ten minutes, the remaining 1 or J per
cent, is halved during the second ten minutes, while practically
nothing happens during the last ten minutes. It is, however,
necessary to continue for the last ten minutes if the milk is
pasteurised in bottles, as it is considerably colder near the
bottom than at the top. Ayers and Johnson have even continued
heating for six hours without having observed any further bacterial
reduction. It follows that the few bacteria which are not killed
during the first ten to twenty minutes cannot be got rid of without
raising the temperature, so that there is nothing to be gained by
lengthening the time beyond the prescribed thirty minutes.
According to the author's investigations, the advantages of pro-
longed low temperature pasteurisation would be illusory if only
for the reason that prolonged heating (e.g., for five hours), even to as
low a temperature as 60° C., causes considerable chemical changes
in the milk, and, what is still worse, the development of the thermo-
philic putrefactive bacteria will be favoured by keeping the milk
for so long a time between 60° and 70° C.

The concentration of milk consists either in condensing in vacuo
or complete evaporation to dried milk.

Distinction is made between sweetened and unsweetened con-
densed milk, the latter also being called evaporated, milk. Both
are as a rule sold in hermetically-sealed tins, but the sweetened
milk may also be shipped in cans or barrels. As the condensation
is carried out at 50° to 60° C., the multiplication of bacteria is by
no means excluded during this process, and the milk should
therefore previously be freed from as many germs as possible by
heating to the neighbourhood of 100° C.

Fourteen to sixteen per cent, of cane sugar is added to the milk
which is to be sweetened, and the product is evaporated to a third
of its bulk, cooled quickly and stirred meanwhile so as to prevent

1 Weigmann, Wolff, Trensch and Steffen have confirmed Ayers' and
Johnson's results, and also shown that the proportion of lactic acid to other
bacteria is higher in stable milk than in pasture milk after low tempera-
ture pasteurisation, because the former is richer in lactic acid bacteria
than the latter (" Centralblatt f. Bakteriologie," 2 Abt., 1916, Bd. XLV.,
p. 63).

PRESERVATION OF MILK AND ITS TREATMENT 87

the formation of large crystals of the difficultly soluble milk sugar,
and finally filled into steamed (preferably sterile) vessels. Milk
which has been treated in this manner is by no means sterile, but
the high sugar content inhibits the development of microorganisms,
as in jam. Quite a number of orange and white micrococci can,
however, always be found in the milk, and occasionally also
yeast, red as well as white Torulse, of which latter a few ferment
saccharose even in these high concentrations, and, therefore, can
make the tins bulge out 1. If the milk has not been thoroughly
aerated after it has been in the vacuum pans, it is only in the top
end of the tins that alcohol is formed. In this end moulds may
develop in addition to yeasts, but their vitality is cut short as soon
as all the oxygen present in the tin has been used up. This stage,
however, need not necessarily mark the cessation of the harmful
effect of the moulds, as the proteolytic enzymes contained in the
moulds, after having digested the mycelium itself, may continue to
act on the surrounding medium ; Rogers, Dahlberg and Evans have
thus shown 2 that certain reddish brown lumps which are now and
again found in old condensed milk, and the origin of which has
hitherto been a mystery because no cells were found therein, are
formed by Aspergillus repens in the way described above.

As the unsweetened condensed milk is not so viscous, it must
be homogenised to prevent the separation of cream ; it must be
sterilised after having been tinned, in order that it may keep. As
highly concentrated milk coagulates to a gelatinous mass at the
high temperatures used in sterilisation, the unsweetened milk
cannot be evaporated so far as that which has been sweetened ;
the author was the first to show 3 that the soluble calcium salts
present in the milk are the cause of this phenomenon, their con-
centration naturally increasing with that of the milk ; if acid has

1 Hammer has named this yeast Torula lactis condensi (Iowa Agric.
Exp. Station, 1919, Bull. 54, pp. 211 — 220. According to the author's
researches, the micrococci as well as the yeasts come in the majority of
cases from the sugar which is used, it is therefore very necessary that the
sugar should be heated to a sufficiently high temperature after having been
dissolved. The large amount oi! cane sugar in condensed milk inhibits the
development of the ordinary cane-sugar-fermenting Saccharomycetes ; only
Zygosaccharomycetes (i.e., certain sexually differentiated Saccharomycetes)
will be able to grow in the concentration of cane sugar in question, but these
are, so far as the author is aware, never found in milk products.
" Journal of Dairy Science," 1920, vol. 3, p. 122.

3 Hoppe Seylefs " Zeitschrift f. physiol. Chemie," 1914, Bd.. 93, pp. 299,
300. In this work it is shown that the coagulation of faintly acid milk on
boiling, which is a well-known phenomenon in cookery, is not directly due
to the slight amount of acid, but to the soluble lime salts formed by it.
The conditions causing the coagulation of normal (not acid and not con-
dciised) milk on -sterilisation at 130° to 140° C. are more complicated, as in
this case a proteolytic decomposition also takes place. For further informa-
tion regarding these conditions, see the author's work in " Landwirtschaft-
liches Jahrbuch der Schweiz," 1905, p. 235.

88 DAIRY BACTERIOLOGY

been formed in the milk the concentration of soluble calcium salts
is further increased. The worst conditions occur if rennet-forming
bacteria are present in the milk in addition to acid-forming bacteria,
as the amount of calcium necessary to precipitate the paracasein
is only half that necessary to precipitate the casein. It is
therefore obvious that only milk of the very best quality can be
used for condensing, and such milk is of course also the easiest to
sterilise with satisfactory results. According to Hunziker's investi-
gations l, the bacteria which are found in incompletely sterilised
evaporated milk are not so often hay and potato bacilli as a
slender peptonising, non-acid-forming, non-sporing rod bacterium.
This organism presumably belongs to the group of microbacteria
which the author points out as being able to stand comparatively
high temperatures for non-sporing bacteria. If the tins of evaporated
milk bulge out, this is always due to anaerobic spore formers, i.e.,
butyric acid bacilli as well as anaerobic putrefactive bacteria. In
leaky tins in which the milk has been infected after sterilisation
one may, of course, find lactic acid bacteria as well as many other
stray organisms. In order that the bad tins may be picked out,
the manufacturers should always keep the condensed milk for a
short time before sending it out. Fermentation in sweetened
milk will always show itself in the course of about a week at
20° to 25° C., while the " incubation " time for sterilised un-
sweetened milk often lasts two to three weeks, even at a
temperature of 30° C.

Milk is dried on drums in vacuo (Eckenberg's method), on drums
heated to 140° C. (Hatmaker's method), or by finely dividing it
and causing it to meet a current of air at 120° C., whereby it is
instantaneously converted into powder (spray process). Of these
methods, Hatmaker's is the easiest and cheapest, but the powder
which is _ produced is difficultly soluble even if alkali has been
added. The dried milk made by the spray process is far better,
being completely soluble when freshly prepared. This is due to
the fact that no denaturisation of the proteins takes place, as the
current of warm air is cooled right down to 60° C. immediately on
meeting the sprayed milk, owing to heat being absorbed in the
process of evaporation. All the methods can of course be com-
bined with an initial condensation of the milk. Only dried milk
made by Hatmaker's process is sterile. The dried milk made by
the other processes can of course be made more or less free from
fijerms by pasteurising the milk more or less thoroughly at the

1 Otto F. Hunziker, " Condensed Milk and Milk Powder," Illinois, 1920.
This work, which is based on wide experience, is strongly recommended to
any one wishing to obtain information on these special branches of dairy
practice. See also C. Porcher, " Le Lait Desseche," Lyon, 1912.
Porcher strongly recommends dried milk for infant feeding.

PRESERVATION OF MILK AND ITS TREATMENT 89

outset. Only skimmed milk is adapted for drying, as the milk
fat is quickly oxidised owing to the large surface presented by the
powder, and thus acquires a highly unpleasant taste. For this
reason condensation must still be regarded as the best method for
preserving whole milk, at any rate when it is a question of pre-
serving it for a long time.

Preservatives. — If it is only a question of preserving the sample
for analysis, comparatively large amounts of antiseptics may be
used, e.g., copper sulphate, potassium dichromate or formaldehyde
(0-1 per cent.), but all such substances must be excluded if the
milk is to be used as food for animals or human beings. According
to Lazarus 13 the following can be added per litre of milk without
the taste being affected : 3 grams sodium carbonate or bicarbonate,
1 to 2 grams boric acid, 0-75 gram salicylic acid. Soda naturally
delays the souring, but not always the coagulation, as the growth
of the rennet-forming bacteria is promoted thereby ; it has also a
favourable action on the development of the pathogenic bacteria.
Boric acid is without effect in the above proportions, while the
salicylic acid inhibits the growth of bacteria to some extent 2.

Behring has recommended the addition of a slight amount of
formaldehyde to milk, 1 to 10,000 parts, to destroy the tubercle
bacteria. This is certainly the highest dilution which can have
any bactericidal action whatever.

The only substance which can come into consideration in the
present case is hydrogen peroxide, which is decomposed by the
catalase of the milk into water and oxygen, the latter having an
antiseptic action in the nascent state. The Danish engineer
Budde has shown that the decomposition takes place most
rapidly at 45° to 50° C., and as several species of bacteria are killed
only by heating to these temperatures, the so-called Buddisation
consists in treating the milk for several hours (stirring at first) at
52° C. with 0-35 part per thousand of hydrogen peroxide ; the
bacteria are thus subjected to the simultaneous action of poison
and heat. Commercial 3 per cent, hydrogen peroxide cannot be
used on account of the poisonous impurities which it contains, and
also the fact that it would dilute the milk'with 1 per cent, of water ;
pure hydrogen peroxide must be used, which renders the method
too costly, especially as all that is achieved is pasteurisation and
not sterilisation. Barthel has shown that milk treated by Budde's
process no longer gives Starch's reaction 3. In order to sterilise milk
by means of hydrogen peroxide, 1 to 2 parts per thousand must be

1 " Zeitschrift f. Hygieine," VIII., p. 207.

2 According to Proks, 2 per cent, of benzole acid has no greater anti-
septic action than 1 per cent of salicylic acid.

3 " Nordisk Maelkeritidende," 1903, No. 5.

90 DAIRY BACTERIOLOGY

used, in which case the milk acquires a very disagreeable taste.
Moreover, the milk is appreciably affected by the milk enzyme
galactase (see under the ripening of cheese) in the course of a few
weeks, all of which goes to show that milk cannot be preserved for
any length of time without having recourse to strong heating1.
By adding substances containing catalase, such as liver extract
(Hepin), blood serum or yeast extract, the excess of hydrogen
peroxide may be removed immediately before the milk is used ;
this process was used on a large scale for the production of milk for
infant feeding (Perhydrase milk) on the estate of Prince Ludwig of
Bavaria. According to the author's researches, ozone has no bacteri-
cidal action on milk, and only gives it a highly disagreeable taste.
In this connection, mention may be made of the suggestion to
use the ultraviolet rays of the mercury arc lamp for sterilising milk.
As is known from Finsen's researches, these rays have a bacteri-
cidal action, and water has been sterilised on a large scale in this way.
Milk, however, is somewhat impervious to the rays and, moreover,
it generally contains highly resistant spores ; it is therefore
doubtful whether the method will ever come to have any practical
significance, notwithstanding the number of appliances which have
already been devised for carrying it out. Powerful alternating
currents are said to have been applied with greater success.

TREATMENT OF MILK FOR TOWN SUPPLIES

As the general methods usually available for producing good
milk have already been discussed, the subject of the proper treat-
ment of milk which is to be retailed in towns may be dealt with quite
briefly. The difficulties to be faced are often considerable, for as
the milk has often to be sent long distances by railway, it may not
reach the consumer until it is six to thirty hours old, and even
then it may have to stand for a long time under unfavourable
conditions before it can be used up. Only milk which has been
cleanly handled from the outset can be expected to keep good
under these conditions. On arrival at the large dairies, the milk
is first graded, smelt and tasted, and frequently sampled for
further examination. After weighing or measuring, the milk is
freed from particles of dirt by filtering or centrifuging. This
treatment, at a stage when the bacteria have long since been
distributed throughout the milk, can only be justified on sesthetic
grounds ; it can have no influence on the keeping power of the
milk. Some of the bacteria will naturally be removed, but, on the
other hand, large clumps of bacteria are broken up with the result

1 As hydrogen peroxide itself has a solvent action on proteins, it will
rather promote the action of the proteolytic enzymes than inhibit it.

PRESERVATION OF MILK AND ITS TREATMENT 91

that plate counts show apparently more bacteria than before
cleaning. The best results are got with the so-called cleaning
separator (Fig. 57), which first whirls out the coarser and
heavier particles in the form of slime, and then removes the finer
particles (but also a number of fat globules if the milk is cold), by
pressing the milk through a filter cloth which is stretched out in
the form of a cylindrical bag. As the illustration shows, the bowl
of the centrifuge is comparatively large in diameter and the
filtration cylinder is supported by a truncated cone through the
bottom of which the milk enters. Laval has constructed a clean-
ing centrifuge with plates which acts only by separating the slime

FIG. 57. — Heine's Cleaning Separator.

(Fig. 58). The milk enters under the bottom plate into the
large slime chamber, and then passes in between the plates to the
outer side of the outlet pipe along which it is discharged from the
top.

After cleaning the milk is cooled to a few degrees above 0° C.
on surface coolers through which brine is circulated from the
refrigerator, after which it is run into well- cleaned cans or bottles
and kept cold. In summer it is necessary to use ice on the carts
from which the milk is retailed. Obviously, the necks of the
bottles must be free from cracks, holes or crevices in which dirt
may collect. The bottles are best closed by aluminium caps,
pressed on by machinery so that they can only be removed by
breaking them. Stoppers which can be removed and replaced at
will must have a proper seal ; they have the advantage that they

92

DAIRY BACTERIOLOGY

serve to protect the milk from flies and dust after the bottle has
been opened and the milk partty used, and they allow of the
shaking up of the milk in order to distribute any cream which may
have separated. The best stopper of this description is that
shown in Fig. 59. It consists of a fairly stout porcelain stopper
which is held firm by being simply clamped against the under side
of the collar on the bottle neck ; it is easily removed when the
bottle is opened, and is therefore easy to clean ; the joint is made
by a paraffined paper ring which should constantly be renewed ;

FIG. 58. — Laval's Cleaning Separator.

rubber rings are not so good, as they are not always free from
smell. (Another method which is largely used on grounds of
economy is to close the bottle by a paraffined cardboard disc which
fits tightly inside a groove in the neck of the bottle ; this method
is less satisfactory than those described above, for after the disc
has been removed and replaced, it easily becomes soaked through
with the milk, and the internal groove in the neck of the bottle
is objectionable on grounds of cleanliness. — Translator.)

Milk is often pasteurised', the better to overcome the difficulties
which beset the problem of keeping it good until it can be retailed.
For the reasons stated above, this should be accomplished by the
low temperature process or by short heating (flashing) to 70° to

PRESERVATION OF MILK AND ITS TREATMENT 93

75° C.1 Low temperature pasteurisation (holder process) is best
carried out in closed bottles completely immersed in water, as is
done in the Danish beer pasteurising apparatus ; treated in this
way, the milk does not lose its carbon dioxide, while the bottles
are pasteurised as well. As the milk does not give off its loosely
combined carbon dioxide at the temperatures employed, no
appreciable pressure is developed in the bottles. If the milk is
pasteurised in bulk and then bottled, the bottles must be sterilised
beforehand. Ayers and Johnson 2 propose to bottle the milk
warm in a warm room, and to let it stand for ten minutes before
cooling ; as the temperature will only fall about 5° C. at the
most during this period, the organisms which have gained access
to the milk in the bottle during
filling will generally be destroyed,
and the milk will keep as well as
if it had been pasteurised in the
bottle.

Milk which has been pasteur-
ised rationally according to the
methods suggested above still
gives a positive Starch's reaction,
and is consequently not recognised
as pasteurised according to the
Danish law, the definition being
reserved for the milk which has
been strongly heated so as to give
a negative test. In any legisla-
tion on this subject, it is highly
desirable that distinction should
be made between high and low
temperature pasteurised milk. In the case of skim milk or
buttermilk, however, the Storch test may quite well be employed
as the criterion. Milk which is to be retailed to the public should
either be raw, in which case it should be derived from herds kept
under strict veterinary supervision, or it should be pasteurised by
the low temperature process ; in addition to control by inspec-
tion of the recording thermometer charts in the dairy, the

1 Tholstrup Pedersen proposes simply to slime centrifuge the milk- after
it has been warmed to 70° C. This proposal is worthy of consideration,
especially as Prescott and Breed ("Journal of Infectious Diseases," 1910,
vol. 7) have shown that the white blood corpuscles, which carry much dirt
and many bacteria with them, are more readily separated from warm milk.
Unfortunately the cream line is affected by this treatment, as was shown in
the trials made with the first Laval cleaning centrifuge (" Deutsche milch-
wirtschaftliche Zeitung," 1912, No. 12). Milk is, therefore, now cleaned
by centrifuging at 10° to 12° C. instead of at 70° C.

2 U.S. Dept. of Agric., Bureau of Animal Industry, Bull. 240 1915.

FIG. 59.— Hygienic Stopper.

94 DAIRY BACTERIOLOGY

efficiency of the process should be checked first and foremost by
ascertaining whether the number of microorganisms is satis-
factorily low, which will be the case if it shows a long time for
reduction (at least nine hours) in the reductase test (see p. 166).
In this way a double purpose will be served, for the test will
only show satisfactorily withjreshly pasteurised milk, and this
is exactly what is required, for, unless pasteurised milk is fresh,
it will be a source of danger, and should be regarded in the
same light as an adulteration, as it purports to be better than raw
milk, whereas in reality it is worse. As low temperature pas-
teurised milk, like that pasteurised at a high temperature, cannot
be relied on to sour spontaneously, it should not be sold as raw
milk, but should be plainly marked with the date of pasteurisation.

The high temperature pasteurised milk which is at present sold
in Copenhagen is often as rich in bacteria as raw milk, and always
shows copious evolution of gas in the fermenting test on account
of its comparatively high content of butyric acid and pseudo
lactic acid bacteria. The latter probably find their way into the
milk on the open coolers and from unsterilised bottles. The
advantages offered by such milk are exceedingly doubtful, and
any one who fears the risk of infection had far better buy raw milk
and boil it.

According to the author's repeated investigations, the ice milk
supplied to Copenhagen, which was mentioned above (sold from
carts at about 9 a.m.), contains on an average 60,000 bacteria per
cubic centimetre, and the ordinary milk from large dairies, from
several hundred thousand to about two millions. Small dairies
generally sell far inferior milk, the counts sometimes running as
high as one hundred million organisms per cubic centimetre.
Better conditions can hardly be expected without further legisla-
tion. The author suggested at the International Dairy Congress
at Stockholm1 that no hard and fast rule should be laid down
once and for all as to the maximum number of bacteria and the
minimum percentage of fat in different grades of retail milk, but
that the dairies should be required to make some guarantee which
should be plainly stated on the label in such a way that it could
not be misunderstood. Competition would then bring about a
gradual improvement in the quality of the milk, and the authorities
would only have to see that the dairies did not promise more than
they fulfilled. In this way the smaller dairies which can guarantee
nothing, would be suppressed, and the trade would pass entirely
into the hands of the large and well-regulated firms 2.

1 " Maelkeritidende," 1911, p. 731.

2 The bacteriological condition of the milk retailed in the large towns of
the United Kingdom leaves quite as much to be desired. It is to be

PRESERVATION OF MILK AND ITS TREATMENT 95

The author has proposed the following regulations for the milk
supply of Copenhagen1 : — (1) The milk trade of Copenhagen
should be concentrated in the hands of a few authorised firms
possessing the proper equipment and managed by competent
experts, such dairies to be under municipal control in all details.
(2) The prices paid for milk by the dairies to the producers, as well
as the prices paid by the consumers to the dairies, should depend
on the hygienic qualities of the milk, its cleanness, keeping
qualities and fat content. (3) All persons having anything to do
with the milk up to bottling should be under medical supervision,
and the herds from which the milk is derived should be under
effective veterinary supervision. (4) At the dairies, samples from
each farmer should be taken at least once a week to be tested by
the reductase and fermenting test, and analysed for fat ; these
tests should be carried out by impartial persons appointed by the
municipality in conjunction with a representative of the dairy.
(5) According to the results of the reductase and fermenting test,
the milk should be graded into four classes (see p. 170), the
standard price to be that of second grade milk with 3-3 per cent,
of fat (the average fat percentage in Copenhagen milk) . First grade
milk to command a higher prfte, and third and fourth grade milk
lower prices, a certain bonus or deduction being calculated for
each tenth per cent, of fat over or under the average. (6) First
and second grade milk, sold for direct consumption, should be
pasteurised by the low temperature process ; third and fourth
grade milk, which should only be sold for cooking and baking,
should be pasteurised at a high temperature. (7) After pasteuri-
sation all milk should be cooled to about 4° C., and kept on ice or
in refrigerators during transit or in the shops. It should be
plainly marked with the name of the dairy, its grade and fat
percentage, as well as the date of pasteurisation ; it should not be

feared that any attempts to effect improvement of milk by stimulating
competition, as suggested by Professor Orla Jensen, would prove abortive,
owing to the absence of any real competition in many parts of the country,
and that the grading of milk by legalised standards will prove the only
remedy for the unsatisfactory state of affairs. The enhanced market value
of milk only gives emphasis to the necessity for cleaner handling and
scientific treatment, and it is most unsatisfactory from the point of view of
the public that the enormous increase in price has not been accompanied by
any improvement in quality. The best example of what may be achieved
under climatic conditions often worse than those of the United Kingdom is
to be seen in the American Bacteriological Standards for Milk (U.S. Public
Health Reports, Reprint No. 192). Here the regulation of the market
value of milk according to its bacteriological condition receives the atten-
tion which it deserves. Standards for certified special milk are often as
low as 10,000 bacteria per cubic centimetre, and in many of the large towns
the standards for raw milk are 500,000 bacteria -per cubic centimetre, or
even less. — Translator.

1 " Ugeskrift for Laeger," No. 16, 1919.

DAIRY BACTERIOLOGY

sold later than two days after pasteurisation, while first grade
milk which is to be used as nursery milk should not be allowed to
stand for more than one day after pasteurisation. (8) While
first and second grade milk should be sold exclusively in bottles,
third and fourth grade milk should only be sold in cans.

No question concerning milk, and it may even be said no
question concerning the public welfare, is of greater importance
than that of the Nutrition of Infants, and it will naturally claim
our attention here. As Bunge was the first to show, the milks of
different animals, are richer in proteins and calcium phosphate
the quicker the animals for which they are intended form flesh and
bone. To illustrate this point, the author has collated the
following table of average analyses of different milks :—

Woman

Ass.

Mare.

Cow.

Goat.

Sheep.

Sow.

Bitch.

Rabbit.

Water

88-3

90-3

90-6

87-7

86-8

78-9

82-2

80-1

69-5

Proteins

1-6

1-8

2-0

34

3-7

6-2

6-9

7-3

15-5

Lactose

64

6-2

5-8

4-8

4-6

5-0

2-2

2-8

2-0

Fat .

3-4

1-3

1-2

3-4

4-1

8-9

7-7

8-5

10-5

Ash .

0-3

0-4

0-4

0-7

0-8

1-0

1-0

1-3

2-5

K,O

0-08

0-08

0-10

0-18

0-15

0-20

0-19

' 0-14

0-25

Na20

0-03

0-03

0-02

0-rM

0-06

0-08

0-08

0-08

0-19

CaO

004

0-11

0-12

0-17

0-19

0-24

0-45

0-45

0-89

MgO

0-006

0-013

0-012

0-017

0-02

0-02

0-016

0-02

0-05

P20i

0-04

0-14

0-13

0-20

0-29

0-34

0-34

0-f)l

0-99

01. .

0-05

0-03

0-03

0-10

0-10

0-13

0-07

0-16

0-13

Number of days

!'

taken to double

180

60

47

•>~2

15

14

9

6

the weight.

-

The table shows that the milk of the ass and the mare onty
differ from human milk in containing somewhat less fat and a
little more calcium and phosphoric acid ; as they both resemble
human milk in being alkaline towards litmus, and contain at least
as much casein as albumin, they form the best substitute in cases
where infants cannot be breast-fed. Unfortunately it is nearly
always necessary to have recourse to cow's milk, as the only kind
of milk available in large quantities. In order to make cow's milk
resemble human milk, water, sugar, and, if possible, also cream,
should be added to it. Cane sugar is often used as being the
cheapest, but as it favours the development of butyric acid bacteria
in the intestine at the expense of the true lactic acid bacteria1,
milk sugar or malt extract is to be preferred. As the child grows
older, it becomes possible to depart more and more from the
composition of the mother's milk, and the additions of sugar and

1 Paul Sittler, " Die wichtigsten Bakterietype der Darmflora beim
Saugling," Wurtzburg, 1909, p. 60.

PRESERVATION OF MILK AND ITS TREATMENT 97

water to cow's milk may be gradually decreased. As the infant
begins to secrete larger amounts of diastase, oat gruel or other
starchy liquids may be substituted for sugar. Infectious germs
are destroyed by boiling the milk in a saucepan or a Soxhlet's
apparatus. The albumin, which is none too plentiful to begin
with, is thus coagulated, while the casein is altered in such a
manner that the milk coagulates more slowly and less completely
with rennet. The change does not, as was previously supposed,
solely consist in the precipitation of certain calcium salts which
are a necessary factor in coagulation — when the milk comes into
contact with the acid gastric juice the effect will be neutralised —
but, according to the author's investigations1, a real denaturing
of the casein takes place, which must mean that it becomes less
digestible, for the coagulation of casein by rennet is (see p. 13)

FIG. 60. — Orla-J 'enseri ',<? Household Pasteurising Apparatus.

only a sign that the first stage in digestion has been completed.
The fact that raw cow's milk is coagulated into lumps immediately
on entering the stomach is regarded by the author as a practical
provision against the passage of large amounts of undigested
casein through the stomach. Furthermore, the milk acquires a
cooked flavour, and its natural enzymes, bactericidal constituents
and antitoxins are destroyed. Even though opinions may be
divided regarding the practical bearing of all these changes in
connection with infant nutrition, it is certain at least that many
children cannot stand boiled milk, for which reason it is advisable
to employ low temperature pasteurisation in freeing infants' milk
from pathogenic germs. Both in order to be on the safe side as
regards tubercle bacteria and to avoid the too rapid separation of
cream — an undesirable feature in infants' milk — the lowest tem-

1 " Landwirtschaftliches Jahrbuch der Schweiz," 1905, p. 241.

D.B. 7

98 DAIRY BACTERIOLOGY

peratures should not be employed ; it will only be necessary to
see that the milk is not heated above 70° C. The author has
designed a household pasteuriser which consists simply of a water
bath large enough to retain a temperature of not under 65° C. for
half an hour after it has been heated to 70° 0. and the source of heat
has been removed. Hot and cold water may also be mixed in the
bath so as to obtain a temperature of 78° C., the filled bottles then
being placed in it for an hour ; already, after five minutes, the
milk will be at 67° C. , and the water at the same temperature. The
bottles must be fully immersed ; as they are closed during warm-
ing, the formation of foam or skin is avoided. As with Soxhlefs
bottles, the stopper may be replaced by a rubber teat. The
pasteurised milk must of course be kept cold J.

Recent investigations have shown that milk contains a fat
soluble as well as two water-soluble vitamines, which substances
are absolutely indispensable for the normal growth of young
animals. The fat-soluble and one of the water-soluble vitamines
are only destroyed by prolonged boiling or at temperatures over
100° C. The. second water-soluble vitamine, on the other hand,
already begins to be destroyed at a comparatively low tempera-
ture, a fact which more than any other warns us against heating
infants' milk more than absolutely necessary. The lack of this
last-mentioned vitamine causes scurvy, or, particularly in infants,
Barlow's disease 2. By adding lemon or orange juice to the milk,
the missing vitamine will be replaced.

The fat-soluble vitamine is onl^ sparingly distributed in nature
(besides in milk fat, it is only found in fairly large amounts in cod
liver oil), for which reason milk fat and butter cannot be com-
pletely replaced by other fats.

It is well known that milk contains a sufficiency of all the
substances necessary for the nutrition of the infant, with the
exception of iron3. At birth, the child is provided with a reserve

\ The household pasteurising apparatus is sold by Apoteker Delholm,
Vaisenhusapoteket, Copenhagen .

2 In contradistinction to Barlow's disease, rachitis is due to lack of lime.
According to P. Rohmann ("Die Chemie in Einzeldarstellunge," Berlin,
1916), the three following causes may be operative : — (1) Primary lack of
lime, owing to the mother's milk immediately after birth not containing
enough lime ; later on it will be sufficiently rich in lime, but then it is
customary to change over to farinaceous foods, which are very poor in lime.
(2) Secondary, lack of lime due to bad assimilation of lime salts, caused by
disturbances in the secretion of bile, or in the action of the intestinal glands,
so that the lime is kept back as salts of the fatty acids. (3) Disturbances
in the cells which deposit the lime in the bones ; these may be of a local
nature, or due to the thymus gland, the secretions of which have an influence
on this process.

3 A litre of milk, direct from the cow, contains, on an average, only half
a milligram of iron. If the milk contains appreciably more iron, it will be
in the form of inorganic salts from iron vessels with which the milk has

PRESERVATION OF MILK AND ITS TREATMENT 99

of iron in the liver, but when, after six to nine months, this has
been used up, food containing iron must be given. Spinach and
egg-yolk (beaten up in thin semolina gruel or buttermilk soup) are
the best media for introducing iron, and the large lecithin content
of egg-yolk no doubt also contributes to the growth and welfare
of the child, lecithin being an important constituent of nerve and
brain.

come into contact. Normal human milk contains twice to three times as
much iron as is present in cow's milk. On centrifuging, most of the natural
iron, as well as of the lecithin — the latter is probably combined with the
proteins — passes into the cream. Again, on churning, the bulk of these
constituents remains in the buttermilk, so that clear filtered butter fat is
free from lecithin. It will be seen from this that buttermilk, besides being
richer in fat than separated milk, is also richer in lecithin and organically
combined iron, both substances of very great physiological importance ; it
follows that separated milk in which fat has been emulsified cannot replace
unseparated milk. If it is found that buttermilk is more easily digested
than ordinary milk, the reason is that its casein is already so finely divided
that it can no longer form large lumps in the stomach.

Chapter IV

The Applications of the Lactic Acid
Fermentation in the Dairy Industry

IN Northern and Central Europe " long " and " thick " milks
are prepared in the household as articles of diet ; in Southern
Europe and the neighbouring regions of Asia and Africa, other
forms of sour milk are prepared which generally also develop a more
or less vigorous alcoholic fermentation. These milk products,
which are all credited with great dietetic value, appear under
different names in each country without necessarily being different
in nature. Thus, the author found the same organisms in the
Sardinian Gioddu as in the Bulgarian Yoghurt, while Kefir grains
have been found to vary in composition to such an extent that all
that they could be said to have in common was their peculiar
structure. All these products are of considerable bacteriological
interest, as in many cases they contain different organisms which
mutually benefit one another, forming genuine illustrations of
symbiosis. They may be classed in two groups according to the
completeness of the symbiosis. To the first group belong Mazun
(Armenia) and Yoghurt (Bulgaria), and to the second group,
Leben (Egypt), Kumys (Russian Steppes) and Kefir (Caucasus).

The products of the first group contain both true lactic acid
bacteria and lactose-fermenting Saccharomycetes. The latter
contribute to the aroma, but otherwise play a minor part. The
principal organisms are rapidly growing lactic acid streptococci
and strongly acidifying rod forms. The former are less anaerobic
than the latter and prepare the way for them. The rod form of
Mazun produces dextro lactic acid, while that of Yoghurt, Thermo-
bacterium bulgaricum, produces the Isevo acid. In addition to
the last -mentioned organism, another rod form may occur in
Yoghurt ; this forms feathery colonies in agar tubes and pro-
duces inactive lactic acid ; the author has named it Thermo -
bacterium Jugurt. Good Yoghurt * may be prepared by boiling
or pasteurising good milk, cooling to 50° C., inoculating with the
proper organisms (the yeast may be dispensed with), bottling
and keeping at 40° C. At higher temperatures the rods will pre-
dominate, while at lower temperatures these will be displaced by

1 This name is derived from the Turkish "Jugurt."

APPLICATIONS OF LACTIC ACID FERMENTATION 101

streptococci. As soon as the milk has coagulated, it is cooled in

running water and placed on ice ; the production of acid must

be stopped at the right moment, otherwise the Yoghurt will

become too sour. Sometimes the milk is concentrated to half

its original bulk before inoculation, in which case the product

will be more nutritious, but less refreshing than that prepared

from unconcentrated milk. It will then correspond in com-

position to Yoghurt made from Buffalo milk, which is con-

sidered particularly choice. According to Metschnikoff, the

Yoghurt rod bacteria inhibit the growth of the intestinal putrefactive

organisms to a greater extent than do the other lactic acid bacteria,

and thus prevent digestive troubles, rheumatism, calcareous

growths, and, generally speaking, prolong life. Centenarians are

said to be more numerous

in Bulgaria than in other

countries, and considering

the claims which were made

on its behalf, it is no wonder

that Yoghurt at one time

came into demand in all civil-

ised countries. People ate

not only Yoghurt, but Yog-

hurt tabloids, which fre-

quently contained no living

lactic acid bacteria, but

always Bacillus mycoides or

other sporing organisms. To

what extent Thermobacterium FIG. 61. — Bacterium bifidum, distinctly

bulgaricum Can become accli-

matised to the conditions of
the alimentary canal of man and animals is extremely doubtful 1.
The author has never succeeded in finding this organism in
the faeces of adults, even after large daily doses of Yoghurt.
Neither was it possible to find this easily recognisable rod form
in the faeces of an infant constantly fed on milk which had
been inoculated with a few drops of Yoghurt 2. However, by
inoculating a little of this faeces into milk which was kept at 45° C.,
the Yoghurt rod was obtained as a pure culture. A more rational
procedure would be to introduce into the intestine those lactic

1 Thus Hull and Eettger found that Thermobacterium bulgaricum could
not be acclimatised in the alimentary canal of the white rat (" Centralblatt
f. Bakteriologie," 1 Abt., 1914, Bd/LXXV., p. 219).

2 A slight addition of Yoghurt has an aperient action, and it is certainly
more harmless than other aperients. The thermobacteria seem to be able
to thrive in the stomach of young animals as long as they are fed on milk
alone, and where the production of hydrochloric acid is lower than later on.

branched. From faeces of bottte-fed
infant. X 1,000.

102 DAIRY BACTERIOLOGY

acid bacteria which are known to thrive there under normal
conditions, i.e., the obligate anaerobic rod form Bacterium bifidum
(Bacillus bifidus, Tissier)1, and various streptococci which the
author finds to be related to Streptococcus fceciuw, and glycerinaceus.
The common bacteria of sour milk, Sc. lactis and Sc. cremoris, do
not thrive in the alimentary canal. If the typical intestinal lactic
acid bacteria do not always obtain predominance spontaneously,
it will be owing to a faulty diet. If care be taken that the diet
contains a far greater proportion of carbohydrates (especially
starch, maltose and lactose) than of proteins, then the lactic acid
bacteria will always be able to gain the upper hand over the
putrefactive bacteria. This result may be achieved by eating
more fruit, farinaceous and milk foods than meat, eggs and ripe
peas and beans. It should also be borne in mind that man is not
purely herbivorous, and can only digest very small amounts of
cellulose. The nutrient substances enclosed in voluminous
greens are too difficult of access for our digestive system, and are
to a great extent carried out into the large intestine, where,
together with the cellulose, they set up various anaerobic fermenta-
tions. Already, in 1876, Baumann showed that intestinal putre-
faction is just as pronounced on a pure vegetarian diet as on a
meat diet ; it is most successfully avoided by the rational selection
of a mixed diet.

In the second group of sour milk products alcoholic fermentation
plays a far more important part, although the lactose fermenting
Saccharomycetes are generally replaced ' by Torulse which are
unable to ferment milk sugar directly, but only after it has been
hydrolysed to dextrose and galactose by certain lactic acid
bacteria. This is the case in Leben ; according to Rest and
Khoury 2, the lactic acid rod forms occurring in this product are
only feeble acid producers growing at the ordinary temperature.
Olav Johan Olsen Sopp 3 found a similar case of symbiosis in the
Norwegian Kaeldermaelk (cellar milk), which is made from boiled
milk by inoculating it with a special kind of ropy milk, and
keeping it as cold as possible ; on account of the large amounts of
lactic acid (up to 2-5 per cent.) and alcohol (about 0-5 per cent.),
which are eventually formed, this preparation keeps particularly
well and is not even inclined to turn mouldy ; it often forms a
substitute for fresh milk when the cows are away from the farms
on the mountain pastures, and was formerly very commonly used
by the Norwegian peasantry. Kumys is made from mare's milk,

1 " Recherches sur la flore intestinale norm ale et pathologique du
nourisson," Paris, 1900.

2 " Annales de 1'Institut Pasteur," 1902, p. 65.

3 " Centralblatt f. Bakteriologie," 2 Abt., 1912, Bd. XXXIII., p. 1.

APPLICATIONS OF LACTIC ACID FERMENTATION 103

which, on account of its high sugar content, is well adapted for
alcoholic fermentation, being capable of developing up to 3 per
cent, of alcohol. According to Rubinsky's investigations l a
yeast which ferments milk sugar and a rod bacterium which
produces over 1 per cent, of lactic acid are the organisms which
play the most important part in the production of Kumys. In
pure cultures the lactic acid rod does not grow under 23° C., but in
conjunction with the yeast it grows at the ordinary room tem-
perature. Kumys should, however, not be made at too low a
temperature. Kefir presents the greatest interest, for the so-
called Kefir grains, which resemble small cauliflowers (Fig. 62),
are produced by the symbiotic growth of the active organisms.
Sectional preparations (Fig. 63) show them to consist of a net-
work of small or large rods, in which yeast cells are enclosed. The

FIG. 62. — Kefir Grains, natural size. (After Kern.) Above, dried. Below, swollen.

rods are chiefly Betabacterium caucasicum, a lactic acid bacterium
resembling that of Leben in its chief characteristics, while the
yeast also resembles that of Leben in fermenting saccharose and
maltose, but not lactose without the help of lactic acid bacteria.
Kefir grains generally also contain strongly acidifying Strepto-
bacteria. The Betabacteria are best found by inoculating into
yeast water, which is the only medium in which they thrive well.
In this symbiosis, the lactic acid bacteria appear to be more
dependent on the yeast than conversely. In good Kefir, only rod
bacteria and yeast will be found, and the former always collect
round the latter, showing a mutual attraction. Kefir grains
imported direct from the East exhibit vigorous growth, and have
often been regarded as a miraculous remedy for many ills. Grains
obtained from several European depots wers of quite a different
nature ; all of these contained a torula which fermented lactose

1 " Studien uber den Kumiss," Inaugural Dissertation, Leipzig, 1910,
Verlag Gustav Fischer, Jena.

104 DAIBY BACTERIOLOGY

directly, and a few of them were found to contain some Gram-
negative rods in addition to the usual Gram-positive rods. The
Gram-negative rods must be classed with Bacterium cloacae or
rather with the Proteus species ; they are peritrich and non-
spbring ; they ferment practically all sugars x with copious
evolution of gas, production of alcohol and small amounts of
acid (acetic and lactic). They coagulate milk, forming a slimy
film which sticks fast to the bottom of the flask, and on shaking

FIG. 63.— Section through a Kefir Grain, x 1,000.

clots together, enclosing other microorganisms and casein. These
rods therefore doubtless play some part in the formation of the

1 In addition to the common mono and disaccharides, they ferment
raffinose and inulin, and, among the alcohols, glycerol and mannitol ; but,
like the true lactic acid bacteria, they do not ferment dulcitol. The
aerogenes bacteria mentioned by Kunze in his exhaustive work on Kefir fer-
mentation ("Centralblattf. Bakteriologie," 2 Abt., 1909, Bd. XXIV., p. 112)
are possibly identical with these. This investigator also describes certain
butyric acid bacteria which are said to play an important part in Kefir
fermentation, and, as a matter of fact, a butyric acid fermentation can
always be induced by inoculating stale Kefir grains into sterilised milk.
The butyric acid fermentation, however, only does harm, and it can only
be regarded as a defect here as in all other milk products. In order to
avoid this 'fermentation, the Kefir grains should be revived by inoculation
into milk distributed in shallow layers. Kefir grains occasionally have
certain Mycodermse on the surface. The author was formerly inclined to
regard these as organisms which normally contributed towards the growth
of the grains, but now regards their presence as a defect.

APPLICATIONS OF LACTIC ACID FERMENTATION 105

grains ; they confer on the Kefir its slimy consistency and bring
about the peptonisation of the casein, besides contributing
towards the production of alcohol. In Kefir made from these
grains, various streptococci always appear, which contribute
towards the production of acid, but for 'the most part they appear
to be casual guests. According to Freudenreich, who was one of
the first investigators of the Kefir fermentation1, certain special
strains of streptococci play a part in this process. The grains
investigated by Freudenreich appear to occupy an intermediate
position between the two varieties described above.

The preparation of Kefir from the dried grains of commerce
requires a somewhat lengthy process of revival. The grains must
first be soaked in lukewarm water and then in lukewarm milk
which is constantly to be changed before it coagulates. This
treatment must be continued till the grains produce sufficient gas
to make them rise in the liquid, which may take a whole week.
The grains which float are then added to milk in the proportion
of 2 to 3 kilos per 10 litres. If there is not a large proportion of
grains to milk, the bacteria of the milk itself may easily come to
predominate. Under favourable conditions, the weight of the
grains will increase by over 40 per cent, in fourteen days. The
milk is kept at 20° to 25° C., and frequently stirred or shaken.
When active fermentation has set in, the grains are strained off
and may, after washing with clean water, be used for a fresh
batch ; the milk is bottled, closed up with spring stoppers and
kept for one or two days at 15° to 20° C. ; at higher temperatures
the bottles may easily burst. The precipitated casein is distri-
buted by careful shaking, and the resulting product is an
effervescent, somewhat thick fluid containing about 1 to 1J per
cent, of acid, and about 0-25 per cent, of alcohol. On longer
standing the taste becomes less pleasant owing to the progress of
alcoholic fermentation ; in the course of six days, the alcohol
content may increase to over 1 per cent. Kefir is undoubtedly
of greater dietetic value than Yoghurt, for the author has found
that on the drinking of Kefir, the proportion of Gram -positive rod
bacteria rapidly increases in the faeces, while the yeast is com-
pletely digested, the faeces showing only the empty cell walls.
As extract of yeast counteracts certain illnesses, the healing
properties of Kefir must be ascribed to the yeast. The author
has found that yeast extract promotes the development of the
lactic acid rods, but inhibits the development of streptococci,
especially the pathogenic species. For people who have com-
paratively little hydrochloric acid in the stomach, Kefir, like

1 " Landwirtschaftliches Jahrbuch der Schweiz," 1896, p.,1.

106 DAIRY BACTERIOLOGY

Yoghurt, acts favourably by simple reason of its large content of
lactic acid.

As a third group of milk products may be mentioned Sparkling
Whey, and Whey Champagne, in which alcoholic fermentation is
very pronounced. In the Carpathians, sparkling whey made
from sheep's milk (Urda and Slcuta) is a favourite beverage, and
in Chile whey champagne is made by the fermentation of whey
with the addition of various saccharine and aromatic substances.

In addition to its applications in the preparation of these
national milk products, the lactic acid fermentation plays an
important part in butter making from soured cream and in the
manufacture of margarine from soured skim milk. Its application
in the manufacture of lactic acid has already been touched upon
(see p. 31). The souring of cream will now occupy our special
attention.

THE SOURING OF CREAM

Like most practical discoveries, the souring of cream for butter
making owes its origin more or less to chance. In order to avoid
churning every day, it became the custom on small farms to save
the cream for several days and sometimes even for a whole week,
the different batches being mixed together, and naturally the cream
became sour. In many cases it was sour before mixing, as the
milk was often kept in a warm place for the cream to rise. In
the same way, the 'custom of churning sour cream ("gathered
cream ") naturally arose in the American creameries where
fresh cream is not available. Originally the butter prepared from
soured cream was extremely bad, but experience soon showed
how the worst defects might be avoided, and it was found that
under certain circumstances a particular aroma would be imparted
to the butter, which could not be produced with unripened cream,
and was appreciated by those who had acquired the taste for it.
Further, the use of ripened cream results in an increased yield of
butter as the fat globules coalesce more readily during churning
when the casein has been precipitated. Finally, the butter made
from ripened cream keeps better than that from unripened cream .
For these reasons the practice of souring cream was continued
when the dairies became so large that churning became a daily
operation. Even in Central Europe, where butter was generally
churned from fresh cream, it has become the custom to use soured,
or at any rate slightly soured, cream. The only difference between
" sweet " and " sour " butter will soon be that the former is
usually not salted.

As the self-souring of cream is the oldest process, we may first
deal briefly with the changes which it involves. An actual

APPLICATIONS OF LACTIC ACID FERMENTATION 107

example may be taken from Conn and Esten's investigations1.
Although the cream was hand-skimmed, it did not show a high
bacterial count to start with. It was allowed to stand at 20° to
21° C.

From the accompanying table it is seen that the count increases
steadily during the first forty-eight hours, and even reaches to over
1,000 million bacteria per cubic centimetre, after which it decreases.
At first all the different groups of bacteria increase in numbers,
but later on the lactic acid bacteria gain the upper hand, and by
degrees suppress the other organisms. The most persistent seem
to be the pseudo lactic acid bacteria and the alkali-producing
bacteria (Bacterium lactis innocuum), the cream having been
particularly rich in the latter at the outset.

Percentages of the Bacterial Groups.

No. o; bacteria

.

Time after
skimming.

per cubic centi-
metre of
cream.

True
lactic
acid
bacteria.

Pseudo
lactic
acid
bacteria.

Alkali-
forming
bacteria.

Liquefy-
ing
bacteria.

Other
bacteria.

3 hours.

195,600

6-2

7-6

66-2

64

13-6

12

4,750,000

5-1

1-8

70-2

11-8.

114

24

54,000,000

37-4

5-1

33-8

2-1

244

36

528,000,000

90-2

5-0

4-7

0-1

0

48

1,023,000,000

94-6

3-1

2-1

0-2

0-

60

994,000,000

96-1

3-0

0-9

04

0

72

687,000,000

954

2-3

1-8

04

0

84

420,000,000

96-3

1-1

2-0

0-6

0

Among the lactic acid bacteria, only the Streptococci multiply
at first, but after the count (lactose gelatine plates) has reached
the maximum, bacteria belonging to other groups gradually
appear and begin to suppress those present. At the same time,
yeast and Oidium lactis appear, as in milk. The period of the
maximum count, which always synchronises with the thickening
of the cream, is also the period of the purest lactic acid fermenta-
tion. If the cream becomes over-sour, the most beneficial of the
lactic acid bacteria will disappear and other microorganisms will
develop which may spoil the keeping qualities of the butter, but
which may, on the other hand, impart to it a stronger aroma.
It is difficult to fix any particular degree of acidity at which the
souring should be interrupted, for, as we shall see later, the degree

1 " The Ripening of Cream," Storr's Agricultural Experiment Station
Report, 1900, p. 13.

108 DAIRY BACTERIOLOGY

of acidity is inversely proportional to the fat percentage. The
consistency is a far better guide than the acidity, especially in
view of the difficulty in measuring an exact amount of the thick
cream, and in determining the end point in the titration. As soon
as the cream has the proper consistency, the souring should be
stopped by churning and washing out the bulk of the lactose, or,
if churning cannot be proceeded with immediately, by thorough
cooling.

In order to promote the souring, a Lactic Acid Culture or Starter
may be added. In this way the first stage of the souring may be
shortened, i.e., the stage during which many different kinds of
milk bacteria develop ; although some of these may tend to
improve the aroma of the butter, most of them will in all proba-
bility do harm. The final stage of the souring will also be
shortened, especially if an over-sour starter, such as buttermilk, is
used. Buttermilk is probably the oldest form of starter used,
and if it originates from butter of good aroma, and has not been
diluted with bad water, it will impart a good aroma to the product ;
as a rule, however, the development of harmful organisms will be
encouraged and butter defects will be perpetuated by this method.
A safer method is to let extra good milk sour spontaneously, using
it as a starter when it has reached the right stage ; this method
became prevalent before the so-called pure cultures were put on
the market. The introduction of these cultures was due to the
initiative of Storch, who found that different species or strains of
lactic acid bacteria produced very different odours or tastes in
the soured products 1. Among other lactic acid bacteria, Storch
isolated a stout streptococcus, " No. 18 " (Fig. 38), which produced
a particularly fine aroma. This particular strain was not used in
actual practice to any extent, but already in the same year the
firm Blauenfeldt and Tvede put cultures on the market, and others
soon followed their example. All these cultures which are pro-
pagated in the best pasteurised milk, generally contain several
different species of lactic acid bacteria, and very probably they
owe their good qualities to this very fact, for a higher degree of
acidity is obtained by the co-operation of several species than by
the action of one species alone, while lactic acid bacteria in pure
cultures are apt to degenerate or become slimy. As a rule,
diplococci are met with among the stout streptococci (Strepto-
coccus cremoris). Boekhout2 and Storch3 have described anon-
souring or slightly souring streptococcus which appears to con-

1 " Forsogslaboratoriets," 18 de Beretning, 1890.

2 " Vereeniging tot Exploitatie eener Proefzuivelboerderij te Hoorn,"
Verslag over det jaar, 1917.

3 Researches carried out some years before his death, still unpublished.

APPLICATIONS OF LACTIC ACID FERMENTATION 109

tribute towards the aroma in the souring of cream. The author x
has isolated this bacterium from commercial starters, and finds
it to be a typical milk organism, possibly a variety of 8. cremoris
which has lost the power to form acid, but on the other hand
developed an increased power to produce aroma. Also Hammer
and Baily 2 have shown that the acid-forming bacteria proper
produce more volatile acids (which constitute part of the aromatic
substances) in symbiosis with another bacterium. Dry cultures
may be obtained as well as liquid cultures. The latter are dried on
some neutral material such as lactose or starch, and products are
obtained which are more permanent than the liquid preparations
and better suited for sending long distances. On the other hand,
dry cultures will not keep indefinitely, and it is not advisable to
buy them unless they are stamped with the date of preparation or a
date after which they must not be used. In any case they must
be revived by propagating several times before they can be used,
and they are, on account of the method of their preparation, much
oftener infected with yeasts and moulds than the liquid cultures.
Their use is therefore not to be recommended if liquid cultures
can be obtained instead.

The full benefits of the use of pure cultures are only -attained
when all the other microorganisms in the cream have previously
been killed by pasteurisation, and it was only after this practice
had been introduced that commercial cultures came into extensive
use. The pasteurisation of the cream and the subsequent use of
good cultures have more than anything else contributed towards
making Danish butter a uniform and well-keeping product.
Pasteurising above 80° C. brings the additional advantage of
killing the tubercle bacteria which pass into the cream in large
numbers on separating ; the resulting buttermilk will thus be
rendered harmless to pigs and calves, and the butter harmless for
human consumption. The legal enforcement of pasteurisation of
cream at a temperature of at least 80° C. by the law passed, in
Denmark in 1898, with the object of combating the results of
bovine tuberculosis, must therefore be signalised as an important
step forward in Danish dairy practice.

Although there can be no doubt that the pasteurisation of cream
can only be a benefit, it is still maintained by many that the
commercial cultures do not produce so strong an aroma in the
butter as is obtained by spontaneous souring. Pasteurisation
does away with the first stage of spontaneous souring, while the
development of yeasts and yeast-like moulds is avoided during the

1 " The Lactic Acid Bacteria," monograph published in English by the
Danish Academy of Sciences, Copenhagen, 1919.

2 Iowa Agricultural Experimental Station, 1919, Bulletin No. 55.

110 DAIRY BACTERIOLOGY

last stage by the use of commercial starters ; naturally both of
these factors have an influence on the aroma of the butter. The
most aromatic butter, that from Isigny, is still made by the old-
fashioned methods from very over-ripe cream in which aroma -
producing yeasts have been found. At the same time, the aroma
depends on the nature of the feed as well as on the action of micro-
organisms ; very probably it is the resultant of the combined
influence of both factors. Thus it is well known that grass butter
has a far more pronounced aroma than winter butter, and there
can hardly be any doubt that the fine aroma of Isigny butter is
largely attributable to the excellent Norman pastures. In this
connection, however, it should be pointed out that the aromatic
substances originating from the feed do not always pass unchanged
into the milk (whether through the udder or by way of the
manure) ; the most usual course of events is the secretion of
complex substances, which only give rise to aromatic products on
hydrolysis. Microorganisms which can effect these hydrolyses
will thus be aroma-producing in milk or cream which contains the
substances in question, but not necessarily in any milk or cream.
Several butler aroma bacteria (mostly belonging to the. Coli group)
have already been isolated and recommended for use in con-
junction with lactic acid bacteria in the ripening of cream, and
the fact that none of them have attained to any practical signi-
ficance is probably to be attributed, to the circumstances just
mentioned. At present we have only certain lactic acid bacteria,
producing a faint aroma, which can invariably be used with
advantage ; the above-mentioned discovery of an organism which
might be used in conjunction with these in order to increase the
aroma may be of great practical value. Experience shows,
however, that butter which has a pronounced aroma does not
generally keep well, and it will always be more to the advantage
of the producer to make sure of the keeping qualities than to
concentrate attention on the aroma, which, after all, is a matter of
local or even only temporary preference as far as the public is
concerned, and this can be influenced to a certain extent by the
large producers.

We will now pass on to the detailed discussion of the ripening
of cream by modern methods. The first question of importance
is the fat percentage of the cream. As the lactic acid bacteria do
not act on the butter -fat, they must necessarily produce the
characteristic butter aroma and acid from the other constituents
of the milk. Seeing that the aroma is produced outside the fat
globules, one might suppose that it would be washed out as easily
as the lactic acid, but this is not the case. Thus Isigny butter
has a stronger aroma than any other make, but its process of

APPLICATIONS OF LACTIC ACID FERMENTATION 111

manufacture involves a very thorough washing. The explanation
is that butter-fat is able to absorb essential oils and other odiferous
substances, both pleasant (a property which has been made use of
in perfumery) and unpleasant, and the aroma of the cream will
therefore pass into the fat globules. The greater the proportion
of fat globules present, the less aroma will be available for each
globule, and experience also shows that a high fat percentage is
not conducive to the production of an aromatic butter. Un-
fortunately there is a tendency to work with richer cream in
dairies where increasing quantities of milk have to be treated ;
thus in Denmark it was formerly considered inadvisable to churn
cream containing more than 20 per cent, of fat, but of late the
practice has generally been to increase the percentage to thirty or
even more. Fortunately difficulties in churning prevent the raising
of the latter limit to any extent.

As far as the pasteurising of the cream is concerned, it need only
be mentioned that out of consideration for the keeping properties
of the butter, the pasteurising temperature should be as high as
possible. As the cooked flavour which milk acquires on heating
does not originate in the fat, cream is not particularly sensitive to
this treatment, and if the subsequent cooling be sufficiently rapid,
no harm will be done by raising the temperature to 95° C. In
this connection we must bear in mind the well-known fact that
milk fat occurs naturally in different stages of supercooling, and
that a partial alteration of this condition must take place before
the butter can " come." As all the constituents of the fat melt
completely on pasteurisation, freshly pasteurised cream is difficult
to churn. The longer the cream has stood, the nearer will the fat
globules be to the point of crystallisation, and the quicker will the
butter come. In order to promote crystallisation, it is usual to
cool the cream as completely as possible for a short time im-
mediately after pasteurisation, and then to warm it up to the
souring temperature1. The crystallisation of fat, however, takes
time, and less is achieved by a cooling of short duration at a low
temperature than by a more prolonged cooling at a somewhat
higher temperature. Prolonged cooling, lasting many hours, is
more practicable after the souring process than before, and under
these circumstances the cream is only cooled to the souring

1 Cf. the author's theory of churning (" Maelkeritidende," 1907, p. 943)
van Dam's investigations (" Molkereizeitung," 1915, Nos. 25 and 26) clearly
show the great importance of cooling the cream to a low temperature. At
16° C. all the fat was liquid even after twenty-one hours. At 11° C. one-
half of the fat had solidified in the same time, and at 6° to 8° com-
plete crystallisation had taken place in four hours. It is, therefore,
recommended to cool the cream for some time to 6° C., especially in
autumn.

112 DAIRY BACTERIOLOGY

temperature after pasteurisation1. On the usual type of surface
cooler the cream is exposed to great risk of infection from the air,
but, on the other hand, certain evil-smelling substances may be
eliminated, while oxygen, will be absorbed. Opinion is divided as
to whether the latter is advantageous or not. Aeration tends to
inhibit the development of lactic acid bacteria, but it may possibly
promote the production of aroma ; at any rate the amount of
volatile acids is increased thereby 2. It will be best to pour the
starter into the cream tanks before the cream, so that the souring
of the cream may commence immediately after the pasteurisation ;
the surviving microorganisms (chiefly bacterial spores) will then
have no time to develop. It follows as a matter of course that the
starter must be properly distributed throughout the whole bulk
of the cream, and the cream should be stirred occasionally so that
it may become uniformly sour.

As the amount of starter which is necessary to sour the cream in
a given time depends on the souring temperature, these two
factors may be discussed simultaneously. As the harmful
bacteria which have resisted pasteurisation grow best at high
temperatures, the cream should be soured at the lowest possible
temperature. The author has found the optimum for vigorous
cultures of Streptococcus cremoris to be 25° C., so that there is no
reason why this temperature should be exceeded. On the other
hand, the temperature should not be so low that the lactic acid
bacteria do not thrive well, for then the full aroma will not be
obtained. Without doubt, 18° to 20° C. is the best temperature,
and if, as is sometimes the case in Denmark, a somewhat lower
temperature of 15° C. to 17° is frequently employed, it is either
because the dairies have not the equipment necessary to cool the
cream to a low temperature after souring, or else for the reason
that it is obviously more convenient to carry out the process in
such a way that the cream will be as nearly as possible at the churn-
ing temperature when it is ripe. In order to secure a pure
fermentation, and especially to obtain a good firm butter, Rosen-
gren has recommended in Sweden, and Weigmann in Germany,
to sour the cream at a still lower temperature, i.e., at 8° to 12° C.,
and to add 10 to 20 per cent, of starter. It will be seen that many
different methods are used in practice, and it is hardly possible to
lay down any hard and fast rule except the following : — The
fresher and cleaner the milk, the higher should be the souring tem-
perature, and the lower the percentage of starter used, and conversely,

1 See the author's article " Can a Cleaner Churning be attained by a
more thorough Cooling of the Cream." ( '' Maelkeritidende," 1910, p. 801).

2 Kayser, " Contribution a 1'etude de la fermentation lactique," Doctoral
thesis, Paris, 1894.

APPLICATIONS OF LACTIC ACID FERMENTATION 113

the older and dirtier the milk, the lower should be the temperature,
and the higher the percentage of starter. If the souring is carried
out at 18° to 20° C., the amount of starter should, nevertheless, not
be below 4 to 5 per cent., for the cream should in any case be ripe
before nine p.m., so that there will be time to cool it in order
that it may stand at a low temperature overnight. The com-
pleter the cooling, the less will be the likelihood of oversouring,
the smaller the loss on churning, and the firmer the butter.
Cooling is best accomplished by the addition of crushed ice which
is stirred into the cream. If ice is not available, it is necessary to
use ripening tanks, furnished with cooling jackets or pipes, through
which cold water can be circulated. Ahlborn's cream-ripening
tank1 is of a type which can specially be recommended ; several
imitations of it are made in Denmark, some of them being im-
provements on the original type. It is doubtful whether any
advantage is to be gained by covering the ripening tank with a
lid if it stands in a clean and dry room ; the air between the
cream surface and the lid will be saturated with moisture, thus
affording ideal conditions for the development of moulds. For
example, if two plates of milk are left to sour, one of them being
covered with another plate, a vigorous growth of mould will be
found on the surface of the covered milk after twenty -four hours,
while none will be seen on the uncovered milk.

We may now discuss the starter, which influences the quality of
the butter more than anything else. It is of great importance to
keep the starter pure, cleaning and sterilising everything with
which it comes into contact with the greatest care. Only per-
fectly fresh morning milk must be used in its preparation. By
the systematic application of the reductase and fermenting test
(see p. 166), it will be easy to pick out the farmers who supply the
cleanest milk. By passing through a clean separator, the milk
will be further purified, for, as was mentioned above, a large
proportion of the bacteria pass into the cream, and particularly
into the separator slime. As regards the pasteurisation of the
starter milk, the chief point is to carry it out in such a manner as
to secure the best conditions for the development of the lactic
acid bacteria which are to be added to the milk. Formerly the
lactic acid bacteria were supposed to thrive best in raw milk; as
this contains dissolved albumin, and from this point of view, low
temperature pasteurisation would be the best. On the other
hand, there is a good deal to be said in favour of high temperature
pasteurisation, which destroys the foreign bacteria more effectively
and also destroys the bactericidal constituents of the milk, which

1 " Maelkeritidende," 1910, p. 1094.

J 14

DAIRY BACTERIOLOGY

may inhibit the activity of the useful and the harmful organisms
alike. In order to clear up this question, the author made a
series of souring experiments, in which milk heated for half an
hour to various temperatures was used. The experiments were
all made with two grades of separated milk, one showing a low
bacterial count (A), and the other being an average retail milk (B).
The samples were inoculated with 1 per cent, of a culture of lactic
acid bacteria in milk.

B 1!

B.

11

A.

B. 1-9^ A. B.

C CL.

-i

•g-g

After 18

After 28

||

After 18

After 18

After 28 Igf After 18

After 18

After 28

1

11

hours.

hours.

|1

hours.

hours.

hours. si'js hours.

hours.

hours.

-2

lj

1

8

C «

g

8

8

11 s

8

V

a

1 s

1

>.

c

•gl

c

c

1

£ I

c

CS

I *

§

1 1

|

1

1

.-t"

If

1 1

It

1

j II 1

~

« *'

ft -o

"5

1

03 *

A
"3

ft

'S

II

ft C

p.

0

A

'f

ft 0

ft

t3

Oi

None.

gl

84

None. s*

"^

^

-*,
29

^

Raw.

None.

H'l

32

gl 36

"l

38

gl

38

30

"3

S7 3-)

60°

f

7

f 9

f

8

j

14

24

27

31

,". :J6

65°

7

7

"

7

1 1

f'

22

,'f

32

85

70°

H

7

7

8

||

10

19

1 3

gl

21

34

75D

|f

7

8

|

8

"^

6

"

12

11

f

18

ci 31

80°

7

8

7

7

7

" 8

8

f 8

85°

7

7

7

7

7

8

8

8

90°

M

7

7

|

7

7

7

8

8

14

95°

j}

7

7

7

6

7

8

8

8

Sterile. 20°

M

8

8

30° „' 8 „'

8

8 40° i^ 8

8

8

1

Raw.

None.

f

13

ci

14

Strepto- %2

34

g3

38

g3

40 Slrepto-

*2

28

c? 35

60°

,,

11

f

12

coccus

gi

25

gl

27

gl

36

coccut

27

„ 33

65°

,,

7

M

10 i cremoris.

f

15

f

21

"

36

t.hermo-

25

» 30

70°

,,

9

M

13

11

M

20

33

philus.

f

19

S3

24

75°
80°

gl

15 ci

16 ' j?i ci

17
16

||

10
10

»

11
11

f

gl

18
33

gl

19
29

:: it

85°

15

gl

17

||

10

||

11

31

29

gl 33

90°

,,

14

16

12

18

28

gl'ci

30

gici 33

95°

M

14

15

L5

33

33

„ 136

Sterile.

64°

f

8 f

8 30°

»

20

"

29

"

31

40°

gl

37

gl

40

Raw.
60°

]

Strepto-
coccus

f

.-,4
16

g3
gl sl

2? gg*l

40
34

Thermo- 83 34
bacte- „ 30

««

I
39 si ci 60
33 si 43

65°

luctis.

16

26 gl,,Sl

S3

rium f 26

35 si ci 50

70°

>» •

15

fgl

29 gi

85

helpe- „ 9 f

20 gi»i 43

. 75°

,

5J

14 gi

32

37

tit- tun. f gl 21

81 ci

36 ci 63

80"

17 „

34

39

f 12

f

16 g2 44

85°

17 „

35 „

39

» 12

26 gi 30

90°

14 ..

33 „

36

" 13 1 ;;

16 gi si 36

95°

14

33 „

37

» 13 „

16 a? 21

Sterile.

30°

" 19

"

35 „

38

40° „ 22 | „

35

„ 60

Raw.
60°

Vigorous
starter.

gl 41
, 40

gl

36

34

gl

40 ' Thermo-
39 bacte-

¥

71

gl

64
51

g3
gl

80
64

65'

, 41

35

,, 41 rium

"

16

gl ci 41

48

70°

• , 41

||

35

,, 42 bulga-

19

f 18

S52

27

75°

A 30

, 42

34

,, • 41 ricum.

19

„ 20

gl sl

28

80°

44

||

36

42

||

18

fl gl

19

27

85°

B 30

„ 40

35

.. 40

f"g.

19

ill 20

:vi

90°

>, 42

||

32

41

gl

21

„ 21

g'l

30

95"

43

38

41

24

36

39

Sterile.

„ 43

»'

35

''

42

40°

M

63

60

»

70

Spongy curdled, c = Cheesy (casein contracted and much

f = Fluid, g = Gelatinous curdled,
clear whey), fg = Semi-solid.

See further under the " Fermentation Test."

The acidity is given in Soxhlet-Henkel degrees (see p. 161).

The original acidity of the milk was 7 ; t1]? acidity of the sterilised milk was, however, 8.

APPLICATIONS OF LACTIC ACID FERMENTATION 115

The table first shows the behaviour of the uninoculated milk on
standing ; at 20° C. it was practically unaltered after twenty-eight
hours, which is explained by the slow growth of the heat-resisting
bacteria at the ordinary temperature, whereas at 40° C. it was
affected fairly quickly. The sudden interruption of the increase
in acidity on heating the milk above 75° C. is explained by the fact
that the" heat-resisting lactic acid bacteria will survive heating to
75° C., but not to 80° C. ; microscopic examination confirmed this
view,'as streptococci were found in great numbers in the milk heated
to 75°C., but none were found in that heated to 80° C. or more1. From
the results already considered, we may draw the conclusion that
it is impossible to be sure of the exclusion of foreign lactic acid
bacteria from the starter unless the milk has been heated to at least
80° C. Microscopic examination also shows the milk pasteurised
at 75° C. to contain most hay bacilli after standing, 75° C. being the
lowest temperature at which the bactericidal substances of the
milk are completely destroyed. These points are particularly
well illustrated in the case of the milk which was kept at 64° C. ; in
this milk only those thermophile bacteria grew which were not in
the slightest degree affected by the pasteurisation, but which, on
the other hand, seem to be fairly sensitive towards the bactericidal
substances in the milk. This explains the somewhat paradoxical
result that under these conditions the milk which has been heated
to 75° C. or over is that which is most rapidly affected.

The conditions are still more complicated in the inoculated milk.
In the raw as well as in the low temperature pasteurised milk the
lactic acid fermentation is inhibited by the bactericidal substances
but encouraged by the lactic acid bacteria of the milk itself, and
in the milk heated up to 70°, 75° and 80° C., a sharp struggle for
existence takes place between the lactic acid and the hay bacteria,
which often results favourably for the latter at the outset. Finally,
it must be borne in mind that the higher the temperature to which
the milk has been heated, the less oxygen will it contain. As the
different species of lactic acid bacteria are differently affected by
the conditions under consideration, no definite rule for pasteurisa-
tion can be laid down on this head. Thus it was found that
Streptococcus cremoris and the nearly related Sc. thermophilus
develop most slowly in the milk heated to 75° C. (and the latter also
in the milk heated to 70° C.), while Sc. lactis shows a minimum
capacity for souring (though not very pronounced) in the milk
which has been pasteurised at low temperatures and retained its
bactericidal constituents. The point of special interest is that the
vigorous lactic acid culture (starter) is only slightly affected by the

1 Living micrococci are, however, still found in milk pasteurised at 80° C. ;
they grow best at 20° C.

116 DAIRY BACTERIOLOGY

previous treatment of the milk (perhaps a maximum capacity
for souring is found in that heated to 80° C.). On the other hand,
the two thermobacteria which are distinctly anaerobic in character
are greatly influenced by the previous treatment of the milk, as
is seen from the acidity and also/ in the case of Thermobacterium
bulgaricum (of Yoghurt), by direct microscopic examination (cf.
the illustrations, p. 30). Both these organisms grow best in raw
milk, where the oxygen is quickly consumed by other organisms,
and in sterile milk which is quite free from oxygen. Thermo-
bacterium helveticum also shows a maximum of souring in the milk
heated to 75° C. in which the bactericidal substances have been
destroyed, but which still contains a little dissolved albumin. Of
the lactic acid bacteria employed in these experiments, this was
the only one which was influenced by the albumin, all the others
growing better in milk which had been heated to 80° C. or above.
From what has been said we may draw the conclusion that no
harm will be done by pasteurising the starter milk thoroughly.
All that is necessary is to avoid discolouration or burning, which
may cause difficulty in judging the aroma of the starter or impart
a cooked flavour to the butter in cases where large amounts of
starter are used. Experience shows that pasteurisation for one
hour at 85° gives good results. The milk is heated in a water
bath, stirring frequently, and it is cooled as quickly as possible to
the souring temperature, stirring carefully, the pail or can being
placed in running water. Aeration by pouring the milk from
vessel to vessel is unnecessary ; the risk of infection is only
increased by exposure to the air, and especially by contact with
more vessels than is strictly necessary. It is a great mistake to
pass the pasteurised milk over a cream cooler. As in the case of
cream, undesirable fermentations are best avoided by keeping the
souring temperature low, and moreover, it is unsound as a matter
of principle to accustom the bacteria to a sensibly higher tem-
perature in the starter than that at which they are required to act
in the cream. A temperature of 22° to 23° C. is quite satisfactory
if only this is kept constant by placing the starter cans in a
sufficiently large water-bath at the right temperature. Boekhout's
and Starch's aroma bacteria, it may be noted, do not thrive well
at higher temperatures. On the other hand, it is not advisable
to lower the temperature further, as Sc. cremoris nearly always
degenerates on propagation several times at temperatures below
22° C. As cream is always soured under this temperature, we find
here a new and hitherto unknown reason why the use of butter-
milk for this process is unsatisfactory in the long run. The water-
bath may be made of wood or galvanised or tinned iron, and
covered in in the same way as the culture apparatus, so that only

APPLICATIONS OF LACTIC ACID FERMENTATION 117

the tops of the cans project. The cans are covered with steam
sterilised double covers which allow of air circulation. As nothing
is gained by keeping the starter cold overnight, there is no reason
to add more culture than is necessary to ripen the starter before
the following morning ; about \ to 1 per cent, is sufficient. Im-
mediately before use the starter is skimmed, as the top layer will
be richest in foreign germs if infection has taken place1. The
starter is given a more liquid consistency by stirring it vigorously,
after which it is poured into the cream ripening tank . As the starter
is more easily titrated than cream, its acidity should be checked
from time to time. Experience has shown that the best results
are obtained with an acidity of 90° to 100° (number of cubic
centimetres of normal soda per litre, corresponding to 18 to20c.c.
of decinormal soda per 20 c.c. of milk). This corresponds to
36 to 40 Soxhlet-Henkel degrees. If the starter becomes weak
and shows a lower acidity although it has been kept at a tempera-
ture over 22° C. all the time, the culture used is weak, while if the
acidity is too high the milk has probably been oversoured, and the
most beneficial bacteria will not be in full vigour. The culture will
thus often contain rod-shaped lactic acid bacteria, in which case
it will be best to use a new culture.

What has been said regarding the starter also applies to the
culture from which the starter is made, only in this case still
greater care is necessary, for an infection in the starter need only
degrade the quality of a single churning, while an infected culture
may cause repeated trouble. Fortunately it is easier to keep the
culture pure, and there is not the same difficulty in obtaining the
small amounts of finest grade milk required for this purpose as
may be experienced in obtaining sufficient quantities of milk in
the condition in which it should be used for the starter. Herein
lies the advantage of propagating the culture and the starter
separately instead of simply inoculating the new starter milk
with some of the old starter. It is a good plan to keep two or
three cultures going independently of one another, inoculating
daily into fresh milk, for by comparing the tastes of the different
cultures the detection of defects in taste is much facilitated and
as it would be a very exceptional misfortune if all the cultures
became bad at once, at least one good culture will always be
available if the precaution is taken of renewing any bad cultures
as soon as possible. Several forms, of culture making apparatus
are on the market ; it is important that means should be provided

1 If the milk is stirred during the souring there is no benefit to be derived
from this precaution. According to Bostrorns experiments at Alnarp
dairy, the lactic acid bacteria thrive better if the milk is stirred occasion-
ally. By this means local over- or under-souring is avoided.

118 DAIRY BACTERIOLOGY

for keeping the temperature of the milk as constant as possible
during the souring period.

Before leaving the subject of cream ripening, it may be pointed
out that the so-called souring defects need by no means be ascribed
to infections of totally foreign groups of microorganisms such as
pseudo lactic acid bacteria, yeasts and moulds ; as often as not
they may originate from the true lactic acid bacteria. Thus
Storch once isolated a lactic acid bacterium which was able to
produce a tallowy taste in cream and butter, and C. O. Jensen
found, almost simultaneously, lactic acid bacteria, some of which
gave the butter an oily taste, and others which imparted a burnt
or malty taste 1. The last-mentioned taste is often produced by
strains of Sc. lactis. From the foregoing it will be seen that the
species of lactic acid bacterium chosen for the culture is by no
means a matter of indifference.

Buttermilk is obtained as a by-product from butter making ;
its good qualities have already been mentioned. It becomes
particularly good when the cream is pasteurised and soured with
a pure starter. As the most favourable stage of the lactic acid
fermentation has always been reached before the churning, the
buttermilk quickly deteriorates in taste and throws out large
lumps of casein unless it is kept quite cold. It follows as a matter
of course that water should not be added to buttermilk which is
to be retailed in towns ; under these circumstances all washing
and rinsing of the butter should be done with ice-cooled butter-
milk from a previous churning. Separated milk is not so suitable
for this purpose, as it causes the buttermilk to curdle more quickly.
The undiluted buttermilk will always have a lower acidity than
the starter which has been used in souring the cream from which
it is made, even if the cream and the starter milk have been
soured in exactly the same way. Tholstrup Pedersen 2 has shown
that the difference is largely due to the fact that the buttermilk
loses its carbonic acid during churning ; the sour starter will also
show a lower acidity if it is shaken before titration.

The Souring of Separated Milk. — In the manufacture of mar-
garine, separated milk is soured and churned into an emulsion
with melted fat. The aroma of the soured milk is taken up by
the fat in much the same way as occurs in butter making ; the
difference between the two cases lies in the fact that while the
aroma is taken up by the fat globules in the cream during the
ripening process, this occurs in margarine only during the churn-
ing, or emulsifying process, and the ensuing operations. As
regards the actual souring of the separated milk, this subject has

1 " Forsogslaboratoriets,'' 22 de Beretning, 1891.
a " Maelkeritidende," 1916, p. 65.

APPLICATIONS OF LACTIC ACID FERMENTATION 119

been discussed at some length in dealing with the preparation of
starters for the ripening of cream, and the process does not differ
materially from that described under the souring of cream.
Blichfeldt l has devised an appliance for the continuous souring of
separated milk, consisting of a closed cylindrical vessel into which
fresh separated milk is introduced, and from which soured skim
milk is withdrawn simultaneously. The contents of the vessel
are kept stirred, and J>y regulating the temperature and the rate
of output, the acidity of the product may be kept constant. The
apparatus is worked under sterile conditions as far as the avoidance
of infection from outside is concerned, while the fresh separated
milk must be efficiently pasteurised before use, which of course is
also the case in the tank souring process. The continuous process
has the advantage that by its means a large amount of milk may
be treated in a relatively small space.

The Preservation of Stable Manure by addition of Whey. — This
process has been proposed by Barthel2, and will have economic
value where whey is plentiful and peat is unobtainable. The lactic
acid fixes the ammonia in a form easily available for the nutrition
of plants (or the nitrifying bacteria). By using 50 to 100 litres of
whey per 1,000 kilos of manure, at a cost of about Id., the increased
yield to be obtained from good soil will amount to the value of
6s. 8d. In this connection it may be mentioned that all dairy
refuse, even the washing water, has great fertilising value in virtue
of the nitrogenous matter which it contains. The washing water
should therefore, wherever possible, be used for surface irrigation
on the fields instead of, as is sometimes done, run into ditches where
it will putrefy. In this case the milk sugar, or rather the lactic
acid formed therefrom, is a drawback, but as a rule the latter
will be neutralised by the large amounts of lime used in cleaning.

1 English patent 4504, 1912. The patent covers continuous fermenta-
tions.

2 C. Barthel and Sigurd Ehodin. Meddelande Nr. 57 fran Centralanstalten
for forsoksvasanet pa jordbruksoniradanet, 1912.

Chapter V

The Normal and Abnormal Microflora
of Butter

• THE NORMAL FLORA

FRESH butter from unpasteurised cream will naturally contain
the same microorganisms as milk, and the bacterial changes which
take place on keeping will be the same as those which occur in
milk kept at the same temperature. Thus at low temperatures
water bacteria will tend to predominate, while at the ordinary
temperature the fresh butter will soon become sour owing to the
rapid development of lactic acid bacteria, and, in particular,
streptococci. Later on, lactic acid rod bacteria, yeasts and moulds
appear. As the moulds, which hydrolyse the fat, only appear on
the surface,? the keeping qualities of the butter will be greatly
enhanced by packing it in large casks instead of in small
flat slabs. Such small pieces will, unless frozen, be subject
in the course of a few days to the same changes, originating
on the surface and gradually working towards the centre, as
occur in a mouldy soft cheese. Butter from pasteurised ripened
cream will have a much simpler flora to begin with, as Strepto-
coccus cremoris, used in the ripening process, will generally be the
principal organism present. This organism, however, does not
seem to be capable of living long in butter, and is gradually
replaced by yeast, and generally also by lactic acid rod bacteria.
Moulds, which are an unavoidable infection from the air, gradually
appear on the surface. As the yeasts and bacteria only develop
in the small drops of water which constitute only about one-sixth
of the weight of the butter, it is obvious that butter will never
show such high bacterial counts as milk and cheese. The number
of microorganisms in butter will depend on the amount of nutrient
matter and antiseptic substances present in the water droplets;
i.e., on the washing, working and addition of salt, boric acid or
other preservatives 1. Properly treated butter seldom contains

1 The preserving action of salt is more pronounced the lower the per-
centage of water in the butter. Thus in butter containing 2 per cent, of
salt, the aqueous portion will contain 12-5 per cent, of salt if the water
percentage is 16, but 20 per cent, if the water percentage is 10. The
development of microorganisms is only completely inhibited when the

MICROFLORA OF BUTTER 121

more than a few million bacteria per cubic centimetre. In butter
from unripened cream, the bacterial count increases during the
first few days 1, after which it decreases. In butter from ripened
cream the count is highest when the butter is freshly made (ten
to twenty million lactic acid bacteria per cubic centimetre), and
decreases steadily in the course of a few weeks to a few 'hundred
thousand, and sometimes even to as low a figure as a few thousand,
the reason being that the original lactic acid bacteria die off at a
quicker rate than their successors develop, for the available
nutrient matter, especially lactose, is continually falling off. In
cases where fat hydrolysis is taking place, this source of carbon
will be replaced by glycerine, and the lactic acid bacteria will then
again fare better.

On keeping for any length of time, butter acquires a stronger
flavour, and it becomes rancid quicker than other fatty materials
on account of the large amounts of water and nutrient substances
which it contains. Before the introduction of margarine, com-
pound and other substitutes, it was a common custom in Central
Europe to preserve the butter fat by melting and separating off
the other constituents ; the pure butter fat kept good if preserved
in well closed stone jars in a cool place.

Dudaux 2 was the first to study the changes which take place in
butter on keeping. He found that direct sunlight promoted the
action of atmospheric oxygen, i.e., its function was similar to
that of the oxidases, causing oxygen to combine with the con-
stituents of the butter-fat, especially the olein. Dudaux also
found that butter was spoiled by various moulds to which he
ascribed an action similar to that of sunlight in oxidising the fat.
The author's investigations3 show that there is an important
difference between the action of sunlight and that of micro-
organisms. While fats are principally oxidised under the action of
sunlight, the iodine value being reduced, they are hydrolysed by the
microorganisms into fatty acids and glycerine, the acid value being
increased. Oxidation causes changes in taste which are much
more undesirable than those caused by hydrolysis ; if butter is

concentration of salt reaches 25 per cent., e.g., with 13 per cent, of water
and 3*3 per cent, of salt (tinned butter). If permissible by law, it is specially
recommended to add 0-75 per cent, of benzole acid, or 2 per cent, of sodium
benzoate, or a mixture of O5 per cent, of benzole acid and 0'5 per cent, of
sodium benzoate. As benzole acid is a physiological product (it is trans-
formed to hippuric acid), it may be regarded as one of the less objectionable
preservatives.

1 The highest count ever obtained by the author from butter made from
unripened cream was 59 millions per cubic centimetre.
" Ls lait," Paris, 1894.

3 " Studien iiber das Kanzigwerden der Butter" ( " Centralblatt f.
Bakt.," 2 Abt., 1902, Bd. VIII., p. 11). References to earlier literature given
here.

122 DAIRY BACTERIOLOGY

exposed to strong sunlight for only one hour, its surface becomes
bleached and perfectly uneatable. The taste thus produced bears
the closest resemblance to that of bad tallow, and is generally
described as " talloivy." The rancid taste proper is only produced
by the microorganisms, and is due to the lower members of the
fatty acid series, i.e., the volatile fatty acids, as well as certain
esters which have a fruity odour, such as ethyl and amyl acetate.
The glycerine, which is usually completely transformed by the
microorganisms, is, the starting substance in the formation of thes
esters as well as of the alcohols which may be formed. As most
fat hydrolysing organisms require air for their development, the
exclusion of air is the most important precaution necessary to
protect the butter from the changes under discussion. For this
reason, butter intended for use in warm climates is packed in
hermetically sealed tins. Heat, like sunlight, promotes the
oxidation of the fat. According to Ritsert 1, carbonic acid, even in
darkness, will impart a tallowy taste to butter ; this point is not
without bearing on the ripening of cheese, as considerable quantities
of carbonic acid are often produced in cheese.

The most important fat hydrolysing microorganisms are Bac-
terium fluorescens liquefaciens, Bacterium prodigiosum, Oidium
lactis, Penicillium glaucum and Cladosporium butyri. As they do
not form spores, they are all easily destroyed on heating ; of
the two bacteria, B. fluorescens liquefaciens is, as mentioned
above, very widely distributed in water, for which reason there is
a danger in letting butter come into contact with water or ice.
In many cases the advantages gained by pasteurisation are
nullified when we introduce bacteria with the washing water,
which have a worse effect on the keeping qualities of the butter
than those which were destroyed in pasteurising. It should be
made a principle to pasteurise (in a regenerative apparatus) all
the water used for rinsing the cream cans, ripening tanks and
churns, and for washing the butter. Moulds may come from the
starter (or from the milk if the cream has not been pasteurised),
from the air or from the packing material. By treating the casks
and parchment paper for twenty -four hours or longer with con-
centrated brine, nutrient substances are extracted and the mould
spores are considerably weakened. It is better to treat the
inside of the casks with melted paraffin wax immediately before
use ; they will then be sterilised and rendered airtight. If this
is done, parchment paper may be dispensed with, and the wood
will not soak up brine and increase the tare at the expense of the
nett weight 2. Although Oidium lactis and Penicillium glaucum

1 Inaugural dissertation, Berne, 1890.

2 Rogers, U.S. Dept. of Agric., Bureau of Anim. Ind., 1906, Bull. No. 89.

MICROFLORA OF BUTTER 123

are the most active in hydrolysing fat, they do not spoil the-butter
to the same extent as the above-mentioned bacteria. Not only do
the moulds hydrolyse the fat, but they also consume part of the
free fatty acids, especially the lower members of the series ; the
acids produced by the bacteria, however, are allowed to accumulate
to such an extent that the fat hydrolysing bacteria, which, curiously
enough, are fairly sensitive to acid, are destroyed. This explains
the fact that very rancid butter may occasionally be sterile some
distance from the surface. The formation of the esters which are
so characteristic of rancid butter is due to Penicillium glaucum
and especially to Cladosporium butyri. The former, however,
only forms esters in symbiosis with Oidium lactis, which mould
also promotes ester formation by Cladosporium butyri. Certain
mycodermse which do not themselves hydrolyse fat may also
promote the formation of esters. According to the investigations
oiH.C. Jacobsen1, exactly the same microorganisms are responsible
for rancidity in margarine.

All the above-mentioned organisms grow better and quicker
in unsalted butter than in salted butter. The water bacteria
are particularly sensitive to salt, and by packing the butter
in large casks the moulds, which require air, will obtain the
smallest possible surface for attack ; Danish butter and similarly
prepared butters therefore contain, as a rule, only lactic acid
bacteria and yeast. Among the yeasts, however, there are
certain torulae which are by no means innocuous and which
may, especially in symbiosis with lactic acid bacteria, hydrolyse
butter-fat to varying extents. The reason why the lactic
acid bacteria promote hydrolysis in such cases is that the
yeasts in question thrive best in presence of a little lactic acid 2.
While the lactic acid bacteria, generally speaking, promote the
growth of yeasts and moulds, they inhibit the action of the fat
hydrolysing bacteria to a considerable extent, so that it is difficult
on the whole to say whether the souring of cream will as a general
rule contribute towards the preservation of butter. According to
the observations of Rogers and Gray 3, butter from pasteurised
sweet cream keeps better than that from pasteurised sour cream,
and the addition of a little lactic acid to the pasteurised cream
has the same undesirable effect as souring with a starter. In all
circumstances it is essential that the starter used should contain
only pure cultures of good lactic acid bacteria, and no yeasts or
moulds. The safest method of preventing yeasts and moulds as
well as other harmful organisms from developing to any extent in

1 " Folia Microbiologica," 1918, V. 2.

2 Orla Jensen, " Maelkeritidende," 1910, p. 965.

3 Experimental Station Kecord, 1909, No. 5.

124 DAIRY BACTERIOLOGY

the butter is to wash it so thoroughly that the organisms will
hardly find any nutrient matter therein1.

THE ABNORMAL FLORA

As in the case of milk (and, as we shall see later on, of cheese),
the normal changes have been dealt with before the abnormal
changes, though in the present instance one may note the im-
portant difference that while the souring of milk may occasionally
be a desired change, the turning rancid of butter is under all
circumstances a defect, and if the defects of butter are classified in
the same way as those of milk, i.e., into original and secondary
defects, most of the latter will be found to be closely connected
with the normal process "of turning rancid, although in their first
and indeterminate stages they may go under many different
names.

The original butter defects may often be divided into
defects in appearance and defects in taste ; they originate in the
milk, in the ripening process or in the processes connected with
churning, and they may be of a chemical or of a biological nature.

As the nature of the feed largely influences the melting point
of the butter -fat, it has an important bearing on the consistency
of the butter., The consistency is also affected by the water
percentage and the way in which the water is distributed. It is
the fine state of division of the water in the butter which renders
the product pliable and easily spread on bread. Storch's re-
searches2 have shown that the water globules in butter should
neither be too large nor too small. If they are too small, the
butter will be dull in appearance and " thick," and if too large,
they will tend to coalesce, giving a wet butter. Normal good butter
contains about 3J million water globules per cubic millimetre,
while " thick " butter contains about 12 J millions in the same
volume. Storch has found that during the souring of cream
bacteria may develop, which make the butter thick, but this
defect may also be due to causes, chiefly chemical and mechanical,
which give rise to a high water percentage. This is especially
brought about by churning and working at too high temperatures 3.
If the cream is oversoured the casein may collect in large lumps,
which may persist in the butter in cheesy lumps and spoil its
appearance and keeping properties.

1 In Weigmanns work, " Versuche zur Bereitung von Dauerbutter "
(" Milch wirtschaftliches Zentralblatt," 1915, p. 353), many valuable hints
will be found.

2 " Forsogslaboratoriets 36 Beretning," 1897.

3 Qrla Jensen, " Maelkeritidende," 1907, p. 943.

MICROFLORA OF BUTTER 125

The original defects in taste 'are due to original milk defects
(and therefore also possibly to the feed), to milk defects arising
at a later stage, i.e., secondary defects, faulty ripening and impure
salt. What has been said under the heading of milk regarding
stable, grass, turnipy and bitter tastes, applies also to butter.
When butter has a strong taste of grass it often contains numerous
small gas bubbles, which show the defect to be due to gas-producing
organisms, i.e., intestinal bacteria. Yeasts which ferment lactose
may produce gas bubbles and give to the butter a peculiar yeast-
like taste. Defects of this nature are generally to be avoided by
pasteurising the cream.. On the other hand, pasteurisation will
not prevent a metallic taste (which may arise in butter through
washing with ferruginous water), and secondary defects due to
faulty souring. A cooked taste is not so often due to heating the
cream to too high a temperature as to heating it for too long, as
may occur if the cooling after pasteurisation is too slow ; the
taste often arises through the milk having too high an acidity,
which causes the proteins to separate and burn on the pasteuriser.
A burnt taste may also arise in this way ; as already mentioned, a
burnt, oily or tallowy taste may also be due to faulty souring, while
the effect of sunlight or copper salts in producing tallowiness has
also been alluded to. If the salt used contains appreciable amounts
of magnesium salts, it may give a bitter taste.

Although for the sake of uniformity distinction has been made
between original and secondary defects, it must be admitted that
the line of demarcation between the two is by no means sharp.
Thus a defect such as unclean taste, which is produced by various
putrefactive bacteria (proteus, coli bacteria, etc.), may develop
sooner or later in the history of. the butter. The term original
defects will be applied to such defects as come out immediately or
during the first few days after the butter has been made, and then
disappear either partially or completely ; by secondary defects
will be understood those which develop gradually and become
worse as time goes on.

Secondary defects are of course counteracted by any condi-
tions tending to increase the keeping powers of the butter, i.e., good
raw material, efficient pasteurising, pure starter, proper churning,
thorough washing with good water, proper working and salting,
clean air, sterile and airtight packing, and the most thorough
cooling possible. Defects in appearance include mouldy spots
which are generally accompanied by a mouldy smell ; it must be
mentioned that many kinds of moulds may grow on butter besides
those which generally cause rancidity, and although most of these
organisms may be able to hydrolyse fat, moulds are known which
do not do so. Oidium lactis cannot as a rule be seen in butter

126 DAIRY BACTERIOLOGY

with the naked eye, but most of the other moulds form green,
brown or black spots. As has already been mentioned, butter from
sweet cream may be coloured red by certain torulse, which in
symbiosis with lactic acid bacteria hydrolyse fat and also give an
oily taste in butter. Actinomyces chromogena turns butter brown
and gives it an unpleasant earthy smell. Among the secondary
defects of taste, the sour cheesy taste deserves special considera-
tion. The acid is chiefly due to lactic acid rod bacteria, so that
the defect is particularly likely to arise when the butter has not
been properly freed from buttermilk or when it contains lumps of
casein in which these cheese bacteria may initiate a cheese-
ripening process. The defect, however, only attains its worst
form when a symbiosis with yeast gives rise to fat hydrolysis 1.
Certain yeasts may produce a fishy or train-oil taste. In marshy
districts or where the land is occasionally inundated by sea or
brackish water, this defect may appear in fresh butter, and is
then due to the grass or small crabs which- are found in great
numbers in the grass. Certain bacteria are also said to be able to
produce a fishy taste by forming trimethylamine from lecithin 2.
Some of the defects which may appear in butter after keeping for
any length of time are of a purely chemical nature, like the oxida-
tion process discussed above, and their appearance may be
accelerated by iron and possibly other salts. It is therefore of
importance that the salt used in butter should be chemically pure 3.
It has been said that dairy salt may contain fat hydrolysing
bacteria 4. Fresh pure salt is of course sterile, but when kept in
the dairy, numerous organisms (thousands per gram) may collect
on its surface, and Weigmann has therefore proposed to dry the
salt in an air oven at 100° C. before use.

As butter defects which are apparently of the same nature may
arise in different ways, and conversely, as is often seen to be the
case in butter grading, the same defect may pass under different
names, it is hardly possible in the present state of our knowledge
to go into further detail as regards the secondary defects. As,
moreover, most of the defects sooner or later pass into the stage

1 The author was the first to show that the ability of certain yeasts to
hydrolyse fat is promoted in the presence of lactic acid bacteria : " Bak-
teriologisehe Studien iiber die danische Butter" (" Centralblatt f. Bakt.,"
2 Abt., 1911, Bd. XXIX., p. 610). This was later confirmed by Sandelin :
"Die Hefen der Butter," Helsingfors, 1919.

2 Thus, Cusick (" Journal of Dairy Science," 1920, Vol. III., p. 194) is of
the opinion that this defect can be produced by Bad. ichthyosmius, a motile
Gram-negative rod showing dirty white surface growth, peptonising milk
while producing slight amounts of acid, and producing gas from cane sugar,
but not from lactose.

3 Rogers, Berg and Potteiger, U.S. Dept. of Agric., Bureau of Anim. Ind.,
1913, Bull. 162.

4 Wolff, " Milchzeitung," 1914, p. 545.

MICROFLORA OF BUTTER

127

known as rancidity, it is probable that they are only forerunners
of this principal defect.

The dairies can easily control the keeping qualities of the
butter which they produce, and thus of any possible defects, by
making a practice of keeping samples of the freshly worked
product in small jars at about 8° to 14° C., and tasting these after
a week and a fortnight.

Chapter VI

The Ripening Processes of the Different

Cheeses

THE methods of preservation, involving the use of preservatives
or bactericidal substances, include the use of harmless acids such
as acetic or, better, lactic acid. The latter is not added but
produced by allowing the material to be preserved to set up a
lactic acid fermentation. This method is applied to the preserva-
tion of beet slices, white cabbage (sauerkraut) and other sugar
containing fodders and foodstuffs which contain too much water
to be dried without the aid of artificial heat. In such cases it is
only necessary to exclude air as completely as possible so that the
acid-consuming moulds are kept down. The same process may
also be applied to milk. If the yeasts and moulds are destroyed
by pasteurising in bottles, and the milk is inoculated with a
vigorous culture of lactic acid bacteria, it will remain unchanged
after the souring process has been completed ; a similar principle
is applied in the making of cheese, which is always based on a
vigorous development of lactic acid bacteria. Under normal
conditions the acid which is produced will always inhibit the
growth of other bacteria, and in the closely pressed mass of cheese
no moulds will develop, owing to the absence of air. If, as in the
case of the hard cheeses, an additional protection is afforded in
the shape of a firm cheese rind, all risk of infection is excluded, and
a permanent product is obtained.

The origin of cheese making was without doubt a desire to preserve
the valuable constituents of milk in a permanent and easily market-
able form. The primitive process, therefore, only involved the
drying and salting of the curd, a process which is still employed
in several places in the East. Subsequently it was discovered
that the curd would also keep without drying, and that with
suitable treatment it would acquire other valuable properties into
the bargain. The art of cheese making thus became not only one
of mere conservation, but also the production of a palatable food,
and in the case of the soft cheeses it may well be said that attention
has been concentrated on the latter point.

The curd is separated by the action of either rennet or lactic
acid ; we may commence with an examination of the mechanism
of the two processes.

RIPENING PROCESSES OF CHEESES 129

The active principle in rennet is a proteolytic enzyme Chymosin,
the action of which does not cease with the coagulation of the milk
and the contraction of the casein, but continues in the cheese
with the formation of soluble proteins. The author's researches
have established that cheese rennet exerts a powerful solvent
action on the proteins of milk, and that this action is promoted by
the addition of small amounts of acid l. American researches 2
have shown that the ripening of cheese is accelerated by increasing
the amount of rennet used. As pepsin is also present in rennet,
it was formerly supposed that the solvent action on the proteins
was exclusively due to this enzyme. Careful investigation,
however, has conclusively shown that chymosin itself is a proteo-
lytic enzyme which can act in the presence of smaller amounts of
acid than pepsin 3. The addition of pepsin to the milk used for
cheese making appears to have no influence on the ripening
process. Trypsin, on the other hand, has a decided influence on
cheese, but may easily cause a bitter taste 4. In this connection,
mention may be made of galactase, a proteolytic enzyme which,
according to researches by Babcock and Russel5, is a normal
constituent of milk, and which just after its discovery in
1897 was assumed to play a most important part in the ripen-
ing of cheese. Subsequently it transpired that galactase does
not play any notable part in the ripening of soft cheeses,
in which, however, it is particularly plentiful 6 ; and the fact
that this enzyme is not indispensable to the ripening of hard
cheeses is amply demonstrated by the fact that good cheese
has been produced on a large scale from milk which has" been
heated to over 80° C., at which temperature the enzyme is
destroyed.

As regards the action of lactic acid, this is of interest not only in
the making of sour milk cheeses, but also in the making of rennet
cheeses. As is well-known, casein occurs in milk as a calcium
salt, a dicalcium casemate (the calcium compounds of the proteins
usually form milky solutions in water), and when casein is precipi-
tated by acid it is not due to transformation into paracasein as
occurs with rennet, but simply to the abstraction of lime by the
acid. At the same time a little lactoglobulin is precipitated.
The greater the percentage of casein in the milk, the more acid

1 " Landwirtschaftliches Jahrbuch der Schweiz," 1904, p. 404, and
1907, p. 97.

2 Seventeenth Eeport of Wisconsin Agricultural Experiment Station,
Madison, 1900.

3 Petry, " Wiener Klinisclie Wochenschrift," 1906, p. 143.

4 Orla Jensen, " Nyt Tidskrift for Fysik og Kemi," 1897, p. 92, and
" Landwirtschaftliches Jahrbuch der Schweiz," 1901, p. 197.

5 Wisconsin Agric. Expt. Station, Bull. 14, 15 and 19.

6 Orla Jensen, " Centralblatt f. Bakt.," 2 Abt., 1900, Bd. VI., p. 793.

D.B. 9

130 DAIRY BACTERIOLOGY

will be required for complete coagulation. The higher the tem-
perature, the easier the coagulation. Thus, at 18°, 30°, 40° and
100° C., 0-6, 0-5, 0-25 and 0-1 per cent, respectively of lactic acid
will generally be required, corresponding to 80, 72-5, 53-5 and
27-5 c.c. of decinormal caustic soda per 100 c.c. respectively.
For this reason, milk is often warmed when the casein is to be
precipitated. If a temperature of 70° C. is exceeded, appreciable
amounts of albumin are also precipitated ; larger yields of cheese
are therefore obtained by using milk pasteurised at high tem-
peratures. On heating to over 90° C., all the albumin and globulin
are precipitated ; the more completely this takes place the more
easily will the milk coagulate, for all solid particles, including fat
globules, serve to stiffen the coagulum. The same milk will
therefore coagulate at a lower acidity .when pasteurised at a high
temperature than when pasteurised at a low temperature. Acid
coagulum differs from rennet coagalum in not contracting so much,
so that it does not separate such large quantities of whey when
allowed to stand undisturbed. This affords a means of distin-
guishing between the two types of coagulation. ,In the case of
a bacterium which coagulates milk, a titration will at once decide
whether a quantity of acid sufficient to be entirely responsible for
the coagulation has been produced ; if not, then the bacterium
must also have produced a coagulating enzyme like rennet. The
coagulating power of rennet is increased by the addition of acid ;
the reason for this is not only that the acid promotes the action of
the enzyme, but also that it forms soluble calcium salts which
facilitate the precipitation of the paracasein. The coagulation of
pasteurised milk by rennet may thus be promoted by the addition
of a fair amount of acid, e.g., in the form of buttermilk 1, and also
by the addition of calcium chloride (100 c.c. of a 40 to 50 per cent,
solution to 100 litres of milk). The contraction of the rennet
coagulum is also promoted to a certain degree by the addition of
acid, the whey separating most readily at a concentration of
about J- per cent, of lactic acid, when all the casein is converted
into the mono-calcium salt ; for this reason, the cheese dries
better if prepared from slightly acid milk than if prepared from
fresh milk. If the milk is so sour that there is a suspicion that
the resulting cheese will be too dry, all that is necessary is to
dilute it with water before adding the rennet ; in this way, the
concentration of both the free, acid and the soluble calcium salts
will be decreased. Conversely if a higher acidity is desired, the

1 In the making of Cheddar cheese from pasteurised milk, Sammis and
Bruhn (Bureau of Anirn. Ind., 1913, Bull. 165) recommend besides the
addition of starter milk containing f per cent, of acid the use of 1 part of
normal hydrochloric acid per 100 parts of milk.

RIPENING PROCESSES OF CHEESES 131

milk may be allowed to ripen at a temperature favou»able to the
development of good lactic acid bacteria (15° to 20° C.), or an appro-
priate amount of buttermilk, starter or other lactic acid cultures
may be added. The longer' the time occupied in making the
cheese, the sourer will be the curd. The "temperature of the curd
when placed in the cheese press determines the species of lactic
acid bacteria which shall obtain predominance. If scalding is
omitted, and the curd is cooled by kneading after the whey has
been run off, the bacterial flora will be quite different from that
which results when heat is applied and the curd is taken direct
from the warm whey. By scalding and carefully cutting the curd,
the separation of the whey will be facilitated ; a means is thus
afforded of shortening the curd forming process, and of regulating
the degree of acidity of the cheese.

As nearly half the natural acidity of milk is due to the
casein, it follows that the whey must have a much lower acidity
than the milk from which it was made. In order to control
the process of souring during cheese making, the whey may be
titrated at different stages. Although the indications thus
obtained will be of some value, they are far from accurate,
and should be supplemented by a careful examination of the
consistency of the curd. On coagulation, with careful treat-
ment of the milk, most of the bacteria become enclosed in
the curd and consequently lactic acid fermentation will take
place far more rapidly within the curd than in the whey 1. In
the curd, however, most of the acid produced will be neutralised
by the lime of the casein and the phosphates ; accordingly it
will sometimes be found that in spite of a vigorous lactic acid
fermentation the acidity of the whey may remain unaltered
during the process, or even decrease as happens in the making
of Emmental cheese, for owing to the fact that the whey is
scalded at a fairly high temperature, the loss of carbonic acid
more than counterbalances the gain in lactic acid taken up by the
whey from the curd. If the bulk of the whey is removed at an
early stage, the remaining whey will be more acid than usual at
the end of the process as the lactic acid which has diffused out of
the curd will have been diluted to a less extent. When the casein
loses its lime it becomes, as van Slyke and Hart were the first to
show2, fairly readily soluble in a 5 per cent, solution of sodium
chloride, especially at 50° to 55° C., but this property is lost in the
presence of an excess of acid. The author has shown3 that this

1 Orla Jensen, 'w Tiber die im Emmenthalerkiise stattfindende Milch -
sauregarung." " Landwirt. Jahrbuch der Schweiz," 1906, p. 287.

2 New York Agric. Exp. Station, Bull. No. 261, 1905.

3 " Zeitschrift f. physiologische Chemie," 1914, Bd. XCIIL, p. 283.

9—2

132 DAIRY BACTERIOLOGY

is due to the easy solubility of monocalcium casemate and para-
caseinate, while free casein and paracasein are practically insoluble
in salt solution. As cheese is always salted, these observations
have an important bearing on cheese making ; it will now be
understood why the acidity of the curd and the way in which it is
salted come to nave such an important effect on the consistency of the
cheese. When a slightly acid curd is salted or placed in brine, it
will swell as if tending to dissolve, becoming elastic and semi-
transparent. On the other hand, on excessive acidification or
slow dry-salting, the curd will retain its original crispness for a
long time. Monocalcium caseinate, though soluble in dilute salt
solution, is insoluble in strong brine, for which reason the brine
used in the cheese making dairies should contain at least 25 per
cent, of salt ; as lactic acid is constantly diffusing into it from the
cheeses, it should be neutralised and filtered from time to time,
as is done in Holland. According to Rosengren, 1 to 1-2 per
cent, of salt in cheese can be considered normal ; 2 per cent, or
more makes the cheese dry, and inhibits the fermentation
processes.

The action of acid has been dealt with at some length not
only because the degree of acidity influences the consistency
of the cheese, but also because it has a determining in-
fluence on ihe course taken by the ripening process which,
owing to the empirical methods of treatment in vogue, deter-
mines more than anything else the nature of the resulting
cheese. As the maximum acidity attained by the cheese de-
pends first and foremost on the amount of whey which it con-
tains, the classification into more or less sour cheeses accords
fairly well with the division into hard and soft cheeses. In
the former the production of acid will practically be limited
to the amounts required to neutralise the lime, and the cheese
will therefore ripen uniformly throughout. On the other
hand, the relatively large amounts of acid present in the soft
cheeses will only be neutralised after a considerable lapse of time,
and as a rule the process is only completed by the aid of the
ammonia which is formed on the surface. It follows that in such
cheeses the ripening process will start at the surface and work
inwards by degrees ; for this reason the ripening may be accelerated
by giving the cheeses a large surface relative to their bulk. The
foregoing points are illustrated by the accompanying table which
gives the percentages of soluble proteins and their decomposition
products in different cheeses. Sol. N. stands for soluble nitrogen,
i.e,. the nitrogen of the soluble proteins, plus that of the protein
decomposition products present, Dec. N. for the nitrogen of the
protein decomposition products, or amino acids which are not

RIPENING PROCESSES OF CHEESES

133

precipitated by phosphotungstic acid 1. Am. N. stands for
ammonia nitrogen.

Per cent, of
total Nitrogen.

Per cent, of
soluble Nitrogen.

Sol. N.

Dec. N

Am N

Dec N.

Am N

HARD CHEESES.

1

Emmental cheese, five
months old.

Interior.

35-82

17-36

—

48-47

—

i Exterior.

29-22

12-57

—

43-02

—

9

Emmental cheese
twelve months old,
ripe.

Interior.

33-15

17-35

2-37

52-34

7-15

1

Edam cheese four
months old, ripe.

Interior.

26-90

3-00

0-60

11-15

2-23

1

Prize Danish dairy
cheese from pasteur-
ised milk, three
months old.

Interior.

35-5

9-7

0-22

27-6

0-6

2

Prize Danish dairy
cheese from pasteur-
ised milk, six months
old.

Interior.

34-9

9-0

0-17

25-8

0-5

1

Swiss skim milk cheese,
eight months old, ripe.

Interior.

41-51

7-90

6-40

19-03

15-40

Exterior.

35-90

7-40

5-20

20-61

14-50

o

Swiss skim-milk cheese,
sixteen months old,
over-ripe.

Interior.

43-54

6-66

6-85

15-29

15-73

Exterior.

53-59

9-11

5-37

17-00

10-02

1

Roquefort cheese, ripe.

The whole
bulk.

52-50

23-64

4-99

45-03

9-51

kSoFT CHEESES.

1

Brie cheese, not quite
ripe.

Interior.

47-10

7-58

5-14

16-10

10-91

Exterior.

53-50

21-33

12-37

39-87

23-12

1

Camembert cheese, ripe.

Interior.

95-52

8-71

8-71

9-12

9-12

1

Limburger cheese, six
weeks old.

Interior.

24-82

5-27

4-37

21-23

17-60

Exterior.

55-10

12-58

4-51

22-83

7-85

Limburger cheese, ripe.

Interior.

99-82

4-33

1L-97

4-52

11-99

§1

1

Schabzeiger.

The whole
bulk.

37-35

16-58

5-80

44-39

15-53

al

gs

£°

1

Norwegian Gammelost
(old cheese), ripe.

Interior.

51-35

31-99

4-23

62-33

8-23

Exterior.

69-47

38-57

7-42

55-52

10-68

The chief feature of the ripening process is the conversion of
insoluble proteins into soluble substances. The table shows that

1 For cheese analysis, see " Zeitschr. f. Nahrungs und Genussmittel "
1906, Bd. XII., p. 193.

134 DAIRY BACTERIOLOGY

in the hard cheeses usually only one-third of the casein becomes
converted into water soluble proteins, whereas in the soft cheeses
nearly all the casein undergoes conversion. This explains the
apparent richness in fat of the soft cheeses, for when anything
gives the sensation of melting in the mouth distinction is as a rule
not to be made between melting proper, as in the case of butter,
and a solvent action. Further examination of the soluble sub-
stances in hard cheeses shows that a larger proportion of these
have been converted into amino acids than in the soft cheeses ;
these conditions may be summed up by saying that in hard
cheeses the ripening is less extensive but more thorough., while in the
soft cheeses it is extensive but not so thorough. This definition,
however, cannot be regarded as a rigid one, for when the hard
cheeses are not much older than the soft ones (see, for example,
Edam cheese), the proportion of Dec. N. to Sol. N. is not large,
and if the soft cheeses do only contain small amounts of amino
acids, this is because these are quickly broken down into ammonia
on the surface l. Now as ammonia readily combines with the
lime free casein to form soluble salts, the high proportion of
Sol. N. in the soft cheeses is partly attributable to the thorough-
ness of the ripening process. The most important factor in the
ripening of the rennet cheeses, besides the various microorganisms,
is the rennet. It is the rennet which produces the perceptibly
soluble proteins and the microorganisms carry the degradation
further. Further, as the action of the most important of the
enzymes of the cheese ripening bacteria is inhibited, while that of
rennet is promoted by the presence of acid, it will easily be under-
stood why the practically neutral hard cheeses contain larger
proportions of amino acids than the soft cheeses.

In the ripening of cheese, not only is the casein converted into
easily digestible and palatable products, but the fat and the
lactose also undergo changes. As might be expected, the fat is
hydrolysed most rapidly in cheeses like Roquefort which are
permeated with moulds. Rapid fat hydrolysis also takes place
in cheeses made from separated milk, for here the fat globules are
very minute and thus expose, a large surface for attack. On the
other hand, fat hydrolysis proceeds with extreme slowness in the
common hard rich cheeses. As, however, butyric, caproic and
capric acids have a very persistent taste, they contribute very
largely to the aroma of the cheese even though they may only be
present in small amounts. The reason why the choicer cheeses
must pass through a long period of ripening in order to attain
their characteristic piquant taste to the full, is that the processes

1 Orla Jensen, " Centralblatt f. Bakt.," 2 Abt., 1900, Bd. VI., p. 773.

RIPENING PROCESSES OF CHEESES 135

of fat hydrolysis and ammonia formation to which the production
of the sharp taste (not to be confused with the salt taste) is due
proceed with extreme slowness in these cheeses. Both of these
processes seem also in the case of hard cheeses to start at the
surface $nd gradually penetrate inwards.

While the fat is usually slowly decomposed in cheese, the
lactose is the first constituent to undergo change. In hard
cheeses it usually disappears in a few days and in soft cheeses in
a week or two. Normally it is completely converted into lactic
acid, which is neutralised by the lime and other bases present.
Calcium lactate is not necessarily the end product, for it is quite
frequently more or less completely converted into propionic acid by
the fermentation process described on p. 41, whereby the normal
" eyes " or cavities are produced in cheese. The butyric acid
fermentation of calcium lactate however, must be looked on as
a disease of cheese, for the evolution of gas will be too vigorous,
while unpalatable or even poisonous substances may be produced.
The author's researches on the volatile acids of cheese, summarised
in the accompanying table, show that normally no more butyric
acid is found in the ordinary rennet cheeses than will originate
from fat hydrolysis.

After these general remarks, we may pass on to consider the
microorganisms which play the chief part in the ripening of the
different kinds of cheese. The flora of the hard rennet cheeses
will be dealt with first. Duclaux, who was the first to investigate
this field, found in cheese various sporing rod bacteria which he
named Tyrothrix, i.e., " cheese threads." He found both aerobic
and anaerobic forms, i.e., what now would be called hay and potato
bacilli and anaerobic putrefactive bacteria. As the aerobic forms
were able to decompose and dissolve casein, and the anaerobic forms
produced the odour characteristic of cheese, Duclaux and all other
contemporary investigators were agreed that the ripening of
cheese was due to the combined action of these bacteria 1. The
constant success of Duclaux in isolating Tyrothrix bacteria from
cheese was due to the fact that he employed an enrichment
method which particularly favoured the growth of these bacteria.
He introduced a small piece of cheese into broth in which the
lactic acid bacteria did not thrive, and it was therefore only
necessary to have a few Tyrothrix spores present from the outset
in order that a Tyrothrix film should form on the surface of the
liquid after some time, after which the anaerobic sporing bacteria,
protected from the air, could grow rapidly.

The careful researches carried out by Freudenreich over a period

1 Duclaux, "Le Lait," Paris, 1894, pp. 213 — 258.

136

DAIRY BACTERIOLOGY

of many years 1 have established that the Tyrothrix bacteria are
comparatively rare in cheese, that normally they cannot develop
in cheese even if introduced in large numbers, owing to their
sensitiveness to acid, and finally that if conditions favourable
to their growth are secured by using pasteurised milk and emitting
the addition of lactic acid bacteria, they produce a disgusting
taste of putrefaction in the cheese. The bacteria living in the

Kind of Cheese.

Found in 1,000 grams Cheese.

As cc. normal.

In grams.

Total volatile acids.

Total ammonia.

Produced
by fat
hydrolysis.

Produced by the splitting up of
the Casein (paracasein) and Lac-
tose (lactic acid).

Total volatile acids.

|

i

H

Caproic acid.

Butyric acid.

Valerianic acid.

| 1 Butyric acid.

TJ

•a

c«
O

1

1

4-218

_o

1
•<

Formic acid.

Emmental
cheese.

Interior.

88-0

75-0

0.116

0-176

—

1-680

—

6-190

1-275

Exterior.

75-0

55-0

0-928

1-232

—

—

2-812

0-900

—

5-872

0-935

Edam cheese.

Interior.

15-6

15-0

—

—

—

0-224

0-678

0-057

0-959

0-255

Swiss skim-
milk cheese.

Interior.

81-6

267-5

0-986

1-496

—

—

2-405

1-200

0-138

6-225

4-548

Exterior.

100-0

207-5

1-682

2-552

—

—

2-775

1-080

0-046

8-135

3-528

Roquefort
cheese.

Whole
bulk.

38-0

115-0

0-928

1-672

—

—

—

0-540

0-092

3-232

1-955

Camembert
cheese.

Interior.

6-6

175-0

0-081

0-246

—

—

—

0-069

0-082

0-478

2-975

Brie cheese.

Interior.

11-3

95-0

0-139

0-572

—

—

—

0-204

0-008

0-923

1-615

Exterior.

8-7

217-5

0-128

0-466

—

—

—

0-120

0-013

0-727

3-698

Limburg
cheese.

Interior.

111-0

200-5

0-058

0-440

1-581

— 5-180

1-140

0-046 8-445

3-409

Exterior.

104-5

220-0

0-232

1-003

1-550

j 4-529

0-822

0-046 8-182

3-740

Glarner
Schabzeiger.

Whole
bulk.

258-2

215-0

1-195

1-848

—

4-452

9-102

3-198

—

19-795

3-655

interior portions of the cheese are nearly exclusively of the
lactic acid group, and it must consequently be these which are
principally responsible for the protein hydrolysis which takes
place. Freudenreich, who devoted special attention to Emmental
cheese, found chiefly the rod-shaped species, and succeeded in
showing that these were able to attack casein if only the lactic
acid which they produced was neutralised by chalk as completely
as it is neutralised in the hard cheese. This forms the basis of a

1 The first of Freudenreich' s papers on this subject appeared in " Annales
de Micrographie," 1889, p. 257, while the subsequent ones are nearly all
published in " Landwirtschaftliches Jahrbuch der Schweiz," 1891 to 1906.

RIPENING PROCESSES OF CHEESES 137

correct understanding of the ripening processes which take place
in these cheeses.

According to the author's researches, the proteolytic enzyme of
the lactic acid rod bacteria must be regarded as an endoenzyme, as
it is not liberated by the living cell, and its action recalls that of
erepsin, for it produces amino acids from the casein without
forming albumoses and peptones as intermediate products l.
The large amounts of amino acids which are found in many hard
cheeses are principally due to the action of this endoerepsin2.
The bacteria themselves take no part in the process, for most of
them will be dead before the bulk of the amino acids have been
produced 3 ; they do not thrive without sugar, and as all the
lactose in hard cheeses will have been fermented in the course of
a day or two, the ripening bacteria will already have reached their
maximum (over 100,000,000 per gram) by then, after which they
gradually fall off. Dead cells which are not exposed to desiccation
generally digest themselves more or less completely owing to the
action of the intracellular enzymes which thus set themselves free,
so that they become able to exert their digestive action on the
surrounding medium ; this is what happens in cheese where the
endoenzymes act under favourable conditions, not involving too
great dilution. The author has demonstrated the presence of
such enzymes in both hard and soft cheeses 4. The fact that the
ripening of hard cheeses depends on enzyme action pure and
simple, and is not directly dependent on the action of living
bacteria, is also demonstrated by the ripening of these cheeses at
temperatures far below the minimum for cheese bacteria 5.

Of the various rod-shaped lactic acid bacteria isolated by
Freudenreich from Emmental cheese, one particular species,
Thermobacterium helveticum, seems to be indispensable for the
production of the typical sweetish taste, and it is extremely
interesting to see how the methods of manufacture, arrived at by
practical experience, favour the development of this bacterium
throughout the process. As the organism occurs in the fourth
stomach of the calf, it will develop freely when the stomach is
extracted in a warm place with " Schotte " (see p. 50). This is
the usual practice in the Swiss dairies, the ripening bacteria thus

1 " Centralblatt f. Bakt.," 2 Abt., 1900. Bd. VI., p. 840, and 1904, Bd.
XIII., p. 521.

2 Freudenreich and Orla Jensen, "Landwirt. Jahrbuch der Schweiz,"
1899, p. 167.

3 Orla Jensen, " Landwirt. Jahrbuch der Schweiz," 1906, p. 303.

4 " Studien iiber die Enzymen im Ease," " Centralblatt f. Bakt.,"
2 Abt., 1900, Bd. VI., p. 734.

5 Babcock and Russel, Eighteenth Annual Report of the Wisconsin
Agricultural Experiment Station, 1901, p. 136.

138 DAIRY BACTERIOLOGY

being introduced with the "natural rennet"*. As the stomachs,
especially those of bad quality, also contain many harmful bacteria,
it is advantageous to inoculate the rennet at once with a mixed
culture of Thermobacterium helveticum and the mycoderma men-
tioned on p. 50 2, or to secure favourable conditions for the lactic
acid bacteria simply by adding lactic acid (e.g., some boiled acid
Schotte) to the rennet. It is safer to use the pure factory made
rennet, together with pure cultures of Thermobacterium helveticum,
in sterile milk 3. In order to secure the development of this
organism to the exclusion of other bacteria, the cheese must be
made warm and kept warm in the presses. This is achieved by
scalding at a comparatively high temperature, taking the curd
out of the warm whey in a lump, and above all by making the
cheeses so large that they will retain their heat for a long time.
Cheeses made in this way maintain a temperature of 50° to 35° C.
during the first twenty-four hours, and under these conditions
Thermobacterium helveticum is certain to obtain predominance.

At the same time a lactic acid streptococcus (Sc. thermophilus)
having a high optimum temperature grows luxuriantly 4. As this
organism does not attack casein, it hardly plays any important
part in the ripening of the cheese beyond the souring process, and
possibly assisting in the production of favourable conditions for
the growth of the lactic acid rod bacteria, in much the same way
as occurs in Yoghurt and similar products.

While the drastic scalding weakens the gas-producing bacteria,
the high temperature of the cheese while in the press favours the
growth of the pseudo lactic acid bacteria, and if the milk was
dirty there will already be a very vigorous development of blow
holes. (Such a cheese (Presslis) develops a large number of pin-
holes.) Coli and aerogenes bacteria, as well as butyric acid
bacteria, may also develop at a later stage, and it is therefore
advisable to keep the cheese as cold as possible after it has left
the press, and until all the lactose has been fermented. Any
evolution of gas will do far more harm in the compact mass of
Emmental cheese than in most other cheeses from which the gas
can partially escape through the narrow fissures which mark off
the original particles of curd. For this very reason the normal
cavities or " eyes " are able to attain to a much larger size in

1 Freudenreich and Orla Jensen, " Centralblatt f. Bakt.," Abt. 2, 1897,
Bd. III., p. 545.

2 J. TJidni, " Bacteriologische Studien iiber Labmagen und Lab."
("Landwirt. Jahrbuch der Schweiz," 1906, p. 181.)

3 Rosengreen and Haglund, " Meddelande No. 101 fran Centraianstalten
for forsoksvasandet pa jordbmksomradandet," Stockholm, 1914.

4 Orla Jensen, " Uber die in Emmenthalerkase stattfindende Milch-
siiuregarung." " Landwirt. Jahrbuch der Schweiz," 1906, p. 437.

RIPENING PROCESSES OF CHEESES 139

Emmental cheese than in other makes. The eyes should only
begin to form when the cheese has ripened to such a degree
that it is sufficiently plastic to allow of their rounding off ;
if they form too soon, they become irregular in shape, and if the
mass is made too dry, it will never become sufficiently plastic ;
in this state the cheese is known as " Gldsler," having elongated
cavities or clefts instead of proper eyes. In order to promote the
formation of normal eyes, the cheese is brought into a room at
18° to 22° C. when it is two weeks old. Here the ripening process is
accelerated and the propionic acid bacteria will gradually develop
so that in the course of four to six weeks the eyes will have been
fully formed. The cheeses are then brought into a cold place
again. As the propionic acid bacteria are very sensitive to sodium
chloride, it is possible to regulate the formation of eyes by the
addition of more or less salt * ; the chief reason for the adoption of
the somewhat troublesome method of dry salting in the case of
Emmental cheese, instead of methods by which the cheese
receives its full amount of salt at an earlier stage, is doubtless that
the cheeses would, in the latter case, only develop small eyes or
none at all, as often occurs with small cheeses, which naturally
tend to become salted too quickly. If the cheese has not been
made sufficiently dry, or contains too many propionic acid bacteria
at the outset, the development of the normal eyes will be excessive,
and this is a defect which may be just as objectionable as the
blowing at an earlier stage, which has been described above. On
the other hand, if neither the ripening process nor the development
of the propionic acid bacteria have progressed sufficiently in the
" warm cellar," it may happen that the eyes will suddenly begin
to form at a still later stage ; an after fermentation of this nature
will always tend to give a variable product. The pre'sence of
unusually large numbers of butyric acid bacteria will also give
rise to an abnormal eye formation which, it may be noted, will be
of the worst possible type2.' In such cases a very energetic
butyric fermentation sets in when the cheeses are from ten to
fourteen days old, and under these circumstances they will not
stand exposure to the temperature of the warm cellar.

Closely connected with the ripening process is the formation of
drops of liquid or " Tears " in the eyes, a process which often first
starts when the Emmental cheese is eight months old. The
conditions determining the collection of this liquid are, first, that
it shall not be too viscous, and second, that the pores of the cheese
shall not be too fine. In the ripening of the cheese an appreciable
proportion of the dissolved proteins are converted into amino

1 Orla Jensen, " Landwirt. Jahrbuch der Schweiz," 1906, p. 437.

2 J. Thoni, " Landwirt. Jahrbuch der Schweiz," 1906, p. 157.

140 DAIRY BACTERIOLOGY

acids, whereby the viscosity of the liquid is decreased, and as
increasing amounts of the curd become soluble the pores in the
cheese will become larger. The salt which at an earlier stage
tended to cause the cheese to swell now has the opposite effect,
but as the rind prevents contraction of the mass as a whole, the
pores must become still larger. The salting and drying of the
cheese finally bring about the precipitation of the sparingly
soluble amino acids as white crusts, and these so-called " salt
stones " may be distributed throughout the whole mass.

The reason why the ripening of Emmental cheese has been
discussed so thoroughly is that this cheese has been studied more
thoroughly than any other. In this respect, the large-holed
Swedish manor farm cheese, which resembles the Danish-made
Swiss cheese, comes next. According to Gerda Troili-Petersoris
researches J, the ripening of this cheese depends on peptonising
tetracocci in addition to the lactic acid rod bacteria, and the
formation of eyes on certain glycerine-fermenting aerogenes
bacteria as well as the propionic acid bacteria. According to
Gorini 2, peptonising tetracocci also play an important part in the
ripening of Parmesan cheese, in the making of which the curd is
scalded at a high temperature ; this cheese differs from Emmen-
tal in being prepared without the addition of strongly acidifying
lactic acid rod bacteria, a dough made from chopped calves'
stomachs being used instead of the acid " natural rennet " de-
scribed above. The peptonising cocci, just mentioned, generally
develop freely in the fresh curd, for, unlike the tyrothrix bacteria,
they can grow in the presence of acid. They are, however,
quickly suppressed if too much acid is present, and will therefore
be most prominent in cheeses which sour slowly or in cheeses made
from milk which has been ripened at a temperature below 15° C.,
as they grow better than the true lactic acid bacteria under
this temperature. The proteolytic enzyme secreted by the
peptonising tetracocci is less sensitive to acid than the erepsin of
the lactic acid rod bacteria, and its mode of action is intermediate
between that of the latter and that of rennet. It would therefore
appear to be best defined as an " exotrypsin" The best investi-
gated of these cocci is Tetracoccus liquefaciens, which produces
the characteristic taste of Tilsit cheese or Russian Steppe cheese,
and there is no doubt that it plays an important part in the

1 "Centralblatt f. Bakt.," 2 Abt., 1909, Bd. XXIV., p. 343. »

2 Among the numerous papers by this investigator, special mention may
be made of " Kecherches sur les coccus producteurs d'acide et de pressures
du fromage " (" Eevue generate du lait," 1910, vol. 8, p. 337). The true
significance of the action of the peptonising micrococci in the ripening of
cheese was first pointed out by Weigmann, 1896 (" Centralblatt f. Bakt.."
2 Abt., Bd. II., p. 151).

RIPENING PROCESSES OF CHEESES 141

ripening of these cheeses. Similarly, it is probable that it assists
in the ripening of Gouda cheese, which is made on the smaller
Dutch farms from perfectly fresh milk without the addition of
lactic acid bacteria ; under such conditions these organisms will
have ample opportunity for development. Boekhout and de
Vries have shown that the peptonising bacteria do not assist to
any appreciable extent in the ripening of Edam Dutch cheese,
which is nowadays generally made with a lactic acid starter 1, and
they have not succeeded in demonstrating other specific ripening
bacteria 2. As Edam cheese is on the whole poor in typical pro-
ducts of bacterial action (amino acids and volatile acids), it would
appear that the bacteria here play only a subordinate part, and
that the principal changes are due to the rennet, as is indicated by
van Dam's researches3. "Salt stone" is occasionally found in
Edam cheese, but in this case it consists of calcium lactate and
phosphate. Again, in Cheddar cheese, the peptonising bacteria
do not bring about any changes of importance ; here also a great
deal of the action is due to the rennet, and the main feature in the
making of this cheese is the production of a curd having a fairly
high acidity at the outset, so that when the curd is pressed against
a hot iron it can be drawn out into long silky threads (the hot-
iron test). American investigators have found the flora in
Cheddar cheese to consist almost exclusively of the true lactic acid
bacteria, chiefly streptococci. The bacterial count reaches its
maximum immediately after the cheese has been made and then
decreases steadily, the rate of decrease being quicker the higher
the temperature at which the cheese is kept4. By allowing the
curd to become strongly acid through the agency of different
streptococci, the author has succeeded in producing cheeses resem-
bling Cheddar in texture and taste, so that there can be no doubt
that in this case such bacteria actually accomplish more than the
mere production of acid 5. . The streptococci cannot, however, be
responsible for the large amounts of amino acids found in ripe
Cheddar, so that here also we have evidence of the action of lactic
acid rod bacteria. As a matter of fact, the author has never met
with any kind of cheese in which lactic acid rod bacteria (strepto-

1 " Kevue generate du lait," 1910, vol. 6, p. 1.

2 " Centralblatt f. Bakt.," 2 Abt., 1905, Bd. XV., p. 323, and 1906, Bd.
XVII . , p. 1 49. These authors have described a diplococcus-like rod bacterium
which converts lactic acid into acetic acid, carbon dioxide and hydrogen,
and which is said to have some significance in the formation of eyes in
Edam cheese. It will stand 4^ per cent, of salt, and its optimum tempera-
ture is 21° C. (ibid., 2 Abt., 1918, p. 130).

3 "Centralblatt f. Bakt.," 2 Abt,, 1910, Bd. XXVI.. p. 189. '

* Harrison, " Centralblatt f. Bakt.," 2 Abt., 1904, Bd. XI., p. 637. Sub-
sequently Harding and Prucha made a thorough study of the flora of
Cheddar cheese (New York Agric. Exp. Station, Bull. No. 8, 1908).

5 " Centralanstalten's 97 Meddeise."

142 DAIRY BACTERIOLOGY

bacteria) could not be demonstrated in large numbers, a natural
result of the ability of these organisms to overgrow all other
bacteria under circumstances favourable to themselves. Thus,
they also obtain predominance by slow degrees in the Danish
dairy cheese, although this cheese, like Cheddar, is freely inocu-
lated with lactic acid-producing streptococci by the addition of
buttermilk.

Barthel ascribes to the streptococci a more important part in
the ripening of cheese than has hitherto been done. He has
shown tfrat certain strains form, at ordinary temperatures, quite
appreciable amounts of Sol. N. (see table, p. 143). The author
has also found, with fair regularity, streptococci, especially strains
of Sc. cremoris, which are conspicuously able to split casein into
soluble products ; as these bacteria gradually lose this power
when cultivated on artificial media, it may be surmised that
streptococci which do not hydrolyse casein to any appreciable
extent may acquire this property when cultivated in milk and
cheese. It must also be pointed out that many lactic acid bacteria
grow better in milk to which rennet has been added than in
ordinary milk ; that is, their action is promoted by the rennet
just as, conversely, the action of the rennet is promoted by the
lactic acid.

In addition to the action of certain lactic acid streptococci
which hitherto have not been further investigated, the chief
factors in the ripening of the hard rennet curd cheeses are
seen to be the rennet, the exotrypsin of the peptonising tetracocci
and the endoerepsin of the lactic acid rod bacteria. The relative
importance of each of these factors varies considerably in different
cheeses, and determines the characteristic properties of each variety.
Further differences arise owing to the fact that the above-men-
tioned groups of lactic acid bacteria include many different
species, an illustration of the production of special characteristics
owing to the action of a particular species of organism being seen
in the case of Emmental cheese. The table on page 143 has
been drawn up to illustrate the action of the several factors in
ripening ; the amounts of soluble nitrogen (Sol. N.), nitrogen of
decomposition products (Dec. N.), and ammonia nitrogen (Am. N.)
formed in milk after two months are given (cf. table, p. 133). All
the cultures were treated with chalk and shaken regularly so that
the lactic acid produced was neutralised.

Taking a wider view of the ripening process, as including not
only the decomposition of -the casein, we must also consider the
propionic acid bacteria as a ripening factor. Weigmann is of the
opinion that the aromas of the various kinds of cheese are princi-
pally due to the action of bacteria other than the lactic acid

RIPENING PROCESSES OF CHEESES

143

organisms1, but as yet he has not been able to adduce any evidence
as to the correctness of his theory. Rodella's 2 contention that
the specific-taste-producing bacteria are obligate anaerobic sparing
organisms can hardly be supported, for it has never been proved
that these bacteria reproduce themselves in rennet curd cheese 3,
and when on rare occasion they have been found in large numbers,
a harmful action has also been observed. The interior of a hard
cheese should have a clean taste and smell, but putrefactive pro-
cesses in the rind cannot be avoided unless special precautions are
taken, and if the cheese becomes very old the interior may also

Milk with Chalk.

Percentage of Total Nitrogen.

Sol. N.

Dec. N.

Am. N.

Streptococcus lactis ....
Rennet ......

2-51
11-75

2-02
0-00

0-23
0-00

Streptococcus lactis and rennet
Tetracoccus liquefaciens
Thermobacterium helveticum

60-56
75-70
36-12

5-32
12-12
34-60

0-36
1-89
3-91

become affected. An ordinary dairy cheese, not too poor in fat,
may even be made to acquire quite a piquant flavour in the course
of three to four months, if only certain changes are promoted in
the outer layer, as is done in the case of the " smeared " soft
cheeses. This method was at one time successfully applied by
Fru Hanne Nielsen. Certain hard cheeses, e.g., Tilsit cheese, have
a tendency to undergo similar changes spontaneously, and may
therefore be regarded as intermediate between the hard and the
smeared soft cheeses. On the other hand, those hard cheeses
which become permeated with moulds form a link between the
hard cheeses and the soft mouldy cheeses ; they will therefore be
dealt with before passing on to the soft cheeses.

Cheeses like Stilton, Gorgonzola and Roquefort, which are per-
meated with moulds, resemble the other hard cheeses in their
mode of ripening, i.e., the process does not originate on the surface
and work inwards. The surface is kept as clean as possible, and
the cheeses are not made flat ; on the contrary, they are shaped
so as to expose a relatively small surface. In Stilton and Gorgon-
zola the moulds only develop slowly, as they owe their presence to

1 " Mykologie der Milch," Leipzig, 1911, p. 223.

2 " Centralblatt f. Bakt.," 2 Abt., 1903, Bd. X., pp. 499, 753; also
1906, Bd. XVI., p. 52.

3 Burri and Kursteiner. " Landwirt. Jahrbuch der Schweiz," 1909,
p. 442.

144 DAIRY BACTERIOLOGY

casual infection and not to artificial inoculation (at any rate, up
till recently they were not purposely introduced) ; the initial
stages of the ripening processes in these cheeses, therefore, follow
much the same course as in the typical hard cheeses. On the
other hand, Roquefort is inoculated with pure cultures of the
required mould, Penicillium roqueforti, which means that a new
factor comes into play at the outset, with the result that the
ripening of this cheese, although accomplished at a low tem-
perature (6° to 8° C.), only requires as many weeks for its com-
pletion as the ripening of the above-mentioned cheeses require
months. ThejDiquant taste and smell of this type of cheese are
principally due to fat hydrolysis, and it is therefore possible to
produce a similar aroma by inoculating the corresponding moulds
into butter. These organisms also affect the casein to a consider-
able extent, as is shown in the table on p. 133. Roquefort, in
common with the Norwegian Gammelost (the latter an acid curd
cheese which is permeated with mould), contains more amino acids
than any other cheese. The development of the moulds is pro-
moted by making the fresh cheese strongly acid, for which reason
the curd is not submitted to a prolonged working, the treatment
resembling that applied to the soft cheeses, while during the early
stages of the ripening the cheese is kept at 18° to 20° C. in order
to promote lactic acid fermentation. As air is necessary for the
growth of the moulds, , the cheeses must be stabbed immediately
after salting ; before doing this the surface of the cheese must be
cleaned carefully to avoid the transference of the organisms
growing thereon to the interior by means of the needle ; some of
these organisms may colour the cheese red or turn it bitter. This
is of particular importance if the cheeses are stabbed at later stages.
They are stored in a comparatively dry room and placed on edge
so that the holes shall not be closed up again. If it is desired to
produce a typical Gorgonzola cheese in which certain bacterial
fermentations will have taken place before the mould has com-
menced its action, the stabbing should not be commenced before
the cheese is two months old. The bluish-green veins of Gorgon-
zola owe their origin to the practice of interlaying the fresh curd
with acid curd which is twelve hours old and which has become
strongly infected with Oidium lactis and various species of Peni-
cillium on the surface. According to the results of the author's
investigations with butter l, the above-mentioned mixture of
moulds seems to be particularly well adapted to produce the
desired taste and appearance of Gorgonzola. In the case of
Stilton, the moulds simply penetrate through fissures in the

1 " Centralblatt f, Bakt.," 2 Abt., 1902, Bd. VIII., p. 369.

RIPENING PROCESSES OF CHEESES 145

surface. The species of Penicillium found in Stilton and Gorgon-
zola appear to be mainly Penicillium roqueforti. This circum-
stance is explained by Thorn and Currie x by the fact that P.
roqueforti can thrive with less oxygen than the other Penicillia,
and therefore obtains predominance in the interior of the cheese,
even though the cheese has not been inoculated with it. In the
making of Roquefort cheese the veins are produced where desired
by dusting with mouldy bread, of which about 0-1 per cent, of the
weight of the cheese is used ; the mould grows not only in these
places, but also in the holes made by the needle ; it is able to
overgrow the cheese bacteria owing to the low temperature (8° C.)
at which the cheese is ripened. The bread used for the cultivation
of the mould is best made from a dough containing equal parts of
rye, barley and wheat flour, and a little lactic or acetic acid, in which
acid fermentation is induced by the addition of some sour dough.
The bread is very thoroughly baked, which, in conjunction with the
action of the acid which has been added, destroys bacterial spores
which otherwise would develop when the bread is set aside for the
development of the moulds. It is then cut into slices and dipped
into sterile water containing J per cent, of acetic acid, into which
a culture of the mould has been stirred. The bread slices are
placed close together on sterile shelves in a damp room and
covered with sterile filter paper. As the change in temperature
caused by the growth of the mould, when this sets in, is consider-
able, the initial temperature should be kept down to 9° to 10° C.
When in the course of three to four weeks the bread has become
thoroughly mouldy, the crusts are removed and the slices are
dried for about ten days at 30° to 32° C., after which they are ready
to be ground and sifted. The yield is 45 per cent, of the original
weight of the bread.

Even if the work is carried out in rooms which are as sterile as
possible, it is difficult to avoid infection with foreign moulds.
The author has accordingly worked out a method 2 in which the
strain of P. roqueforti which is used is little by little accustomed
to stand comparatively large amounts of formaldehyde. Formalin
can be added to the water in which the slices are dipped, whereby
the development of the foreign moulds which cannot stand
formalin is inhibited. A mould powder made in this way is sold
by Messrs. Blauenfeldt and Tvede, of Copenhagen.

As already mentioned, the ripening process of the soft rennet curd
cheeses starts at the surface and works inwards. At the same time,
a ripening action due to the rennet takes place throughout the whole
mass, even though it may not be very obvious. While the amounts

1 " Journal of Biol. Chem.," 1913, vol. 15, pp. 249 and 259.

2 " Maelkeritidende," 1919, p. 277.

P.B. 10

146

DAIRY BACTERIOLOGY

of acid present in the interior of soft cheeses inhibit the action of
the ripening factors hitherto considered, they promote the proteo-
lytic action of the rennet and conserve its enzyme in an active
state for a much longer time than is the case in the hard cheeses *.
As will have been gathered from the preceding remarks, the soft
cheeses may be divided into two groups : (a) The " smeared "
cheeses ; in these the moulds are suppressed and prevented from
forming aerial hyphse by daily smearing the cheeses either with
a wet cloth or by hand, keeping them moist and protecting them
as much as possible from the air by packing them closely together.
(b) The mouldy cheeses ; in these the growth of moulds is favoured
by touching the surfaces of the cheeses as little as possible, keeping
them dry and providing for free access of air.

The copious production of ammonia on the surface of the best-
known of the smeared cheeses, Limburger and Romadour, has been
found by the author2 to be due to a symbiosis of peptonising
tetracocci and Bacterium casei limburgensis. The latter is a
non-motile, very irregular short rod which does not ferment
lactose ; it forms a film on media containing calcium lactate, and
oxidises the lactic acid to acetic and carbonic acids ; it does not
attack casein, but forms traces of ammonia in milk, from the
amino acids, which has a slight solvent action on the casein.
This organism is therefore to be classed with the bacteria which
produce a soapy taste in milk. Its most characteristic property
is its ability to carry further the decomposition of the products of
protein hydrolysis which have been produced by other micro-
organisms ; this is most strikingly illustrated by inoculating it
into milk together with Tetracoccus liquefaciens.

Percentage of Total N.

Milk without Chalk.

,

Sol. N.

Dec. N.

Am. N.

1

Bacterium casei limburgensis

0-60

2-36

1-30

Tetracoccus liquefaciens

61-61

8-73

1-43

Bact. casei limburgensis and Tetracoccus

liquefaciens .....

67-59

32-13

14-99 .

The smeared crust of Limburger cheese consists, according to
the unpublished researches of Freudenreich, chiefly of Bacterium
casei limburgensis together with smaller numbers of peptonising
organisms (chiefly Tetracoccus liquefaciens, a small spore-forming

1 Orla Jensen, " Centralblatt f. Bakt.," 2 Abt., 1900, Bd. VI., p. 795.
* " Studier over de flygtige Syrer i Ost, etc.", Doctoral thesis, 1904. p. 74.

RIPENING PROCESSES OF CHEESES 147

rod bacterium, and yeast) ; in view also of the results set out in
the table on p. 146, there can hardly be any doubt as to the
cause of the thorough-going decomposition processes which take
place in this cheese. A similar flora will also develop on the rind
of hard cheeses if they are kept damp, and it will be understood
how the ripening of these cheeses may be modified so as to conform
with the type generally associated with Limburger cheese. The
ammonia formed on the surface gradually diffuses into the interior,
and not only does it neutralise the lactic acid, thereby rendering
possible the activity of the bacterial enzymes, but it also converts
the casein into the readily soluble ammonium caseinate. The
ripening of the soft smeared cheeses involves further complexities.
In the ripe state these cheeses contain appreciable amounts of
valerianic acid, an acid which is only produced by the above-
mentioned bacteria in very small amounts, while on the strongly -
smelling surface there are formed various typical products of
putrefaction, such as hydrogen sulphide and indole. It is obvious
that under normal conditions organisms other than those mentioned
above participate in the process — Weigmann mentions Plectridium
fcetidum — but whether their activity is to be regarded as at all
desirable may be an open question ; it is quite possible that Lim-
burger cheese might have a wider market if it contained no
products of putrefaction. According to Laxa1, Oidium lactis
participates in the ripening of certain Bohemian cheeses, which
resemble Limburger cheese (Harrach and Knoppist) ; these
cheeses therefore form a link with the mouldy cheeses.

The moulds play a part in the mouldy cheeses similar to that
of the peptonising bacteria in the smeared cheeses. As a type we
may take Camembert, which, thanks to the researches of Roger,
Maze 2, and Thorn 3, is one of the most thoroughly investigated
cheeses. The moulds which develop in this cheese are Oidium
lactis (according to Maze, Oidium camemberti, other oidium species,
and a mycoderma), Penicillium camemberti, and P. candidum.
The moulds of the Oidium group grow the quickest, while the
Penicillium group first begins to develop after five to six days,
when the surface has become a little drier. The chief difficulty
lies in maintaining the proper humidity. If the air is too damp,
the surface of the cheese becomes too slimy, as in the case of the
smeared cheeses, and bacteria, yeasts, oidium, and even certain
mucors, gain the upper hand ; if too dry, on the other hand, the
Penicillia grow too freely, causing the rind to shrivel ; all the green
Penicillia which develop (not excepting P. roqueforti) produce

1 " Centralblatt f. Bakt.," 2 Abt., 1899, Bd. V., p. 755.

2 " Annales de 1'Institut Pasteur/' 1910.

3 U.S. Dept. of Agric., Bureau of Animal Industry, Bull. 115, 1909.

10—2

148 DAIRY BACTERIOLOGY

an undesirable taste. The moulds should not completely over-
grow the surface, neither should they form too many conidia.
The drying is regulated by providing for a suitable draught and
by laying the cheeses on straw mats ; after three weeks, when
the cheeses have softened at the corners, they are transferred to
another room and placed direct on the shelves, or sometimes on
mats infected with bacteria. The moulds are now displaced by
rod bacteria resembling Bacterium casei limburgensis, which
appear as red and brown spots between the mouldy ridges, and
the ammoniacal fermentation which starts under these spots
gradually penetrates to the centre, resulting in the production
of ammonium caseinate and other soluble proteins. The quicker
the cheese ripens, the quicker will it become perfectly liquid and
putrefying. A good saleable cheese should, therefore, be made
firm and ripened at a low temperature ; during the first few days,
when the cheeses are salted and develop acid fermentation, they
should be kept at 18° to 20° C. ; the drying room should be kept
at 13° to 15°C., and the ripening room, to which the cheeses are
finally transferred for the chief fermentation to take place, at
10° to 12° C., or even lower. Frequently the process is completed
in the boxes in which the cheeses are packed for transport.
Cheeses never get the proper taste if they are put in the boxes
too early. As soon as the interior of the cheese has become
neutral in reaction the lactic acid bacteria will be able to act as
in hard cheeses, but they act slowly and have no time to produce
any noteworthy change before consumption. While Oidium
lactis and the necessary bacteria will establish themselves of their
own accord, there is some difficulty in establishing the desired
species of Penicillium in places where the manufacture of Camem-
bert is to be started. Thorn recommends their cultivation on
dry, sterilised rusks, wetted with a suspension of the conidia in
water. After keeping for ten days at 20° C., the rusks will be
overgrown by the mould ; these mouldy rusks are then shaken
vigorously with water, in which the cheeses are dipped immediately
before salting. The French mode of procedure is more scientific ;
the Institut Pasteur (Service des vaccins. 35, Rue Dutot, Paris)
supplies three different cultures for the purpose, consisting of
an ordinary lactic acid starter, a culture of moulds, and a culture
of the ammonia-producing bacteria. The two first-mentioned of
these, or sometimes all three, are added to the milk before the
rennet. The mould culture need only be used for the first ten
days, after which it will have established itself in the mats used
in the drying room, from which subsequent batches of cheese
will become inoculated. The third culture is applied directly to
the mats on which the cheeses rest during the last stage of

RIPENING PROCESSES OF CHEESES 149

ripening. The cultures may also be obtained in powder form
for sprinkling on the cheeses 1.

Only the acid curd cheeses remain to be discussed. These include
both hard and soft varieties, and the latter include both smeared
and mouldy cheeses — even cheeses which resemble Roquefort in
being permeated with moulds, e.g., the Norwegian Gammelost.
The making of the hard acid curd cheeses2, the Danish cheeses,
Appetitost, Knapost, the Norwegian Pultost, and the Swiss Green
Alpine Cheese (Schabzeiger) presents several points of interest; they
are scalded at a high temperature or for a long time and ripened
before they are shaped (the Norwegian Pultost is not shaped at
all). The scalding produces the same results as in the pasteurisa-
tion of milk : the development of the sporing bacteria and the
suppression of the lactic acid bacteria ; as the author has shown
in the case of Schabzeiger3, and v. Klecki in the case of another
acid curd cheese4, the curd develops a vigorous butyric acid
fermentation ; in both cases the butyric acid bacteria were motile ;
the butyric acid gives these cheeses a sharp taste. The ripening
proper is due to lactic acid rod bacteria, which possibly act in
conjunction with other microorganisms. As regards the decompo-
sition undergone by the casein, the hard acid curd cheeses do not
differ greatly from the hard rennet curd cheeses (see table,
p. 133). According to Olav Johan-Olsen5, the ripening of Pultost,
in which the author has found a fair amount of valerianic acid, is
accomplished by yeast, Oidium lactis, and especially by a species
of Mucor. Johan-Olsen has also carried out a detailed investi-
gation of the ripening of the Norwegian Gammelost. Here the
active moulds are especially species of Penicillium and Mucor,
which turn the cheese green and brown respectively. As the curd
or the cheese itself is strongly heated, it is improbable that the
moulds are derived from the sour milk ; they must find their
way into the cheese at a later stage, and gradually penetrate from
the surface throughout the whole mass. This already takes
place during the first six weeks, as the curd is fairly acid and
porous.

If it is desired to utilise separated milk for cheese, this is
best done by turning it into one of the above-mentioned acid

1 According to our experience, it is sufficient to infect the mats with
P. Candidum. The bacteria come of their own accord while the Oidia
only do harm,

2 Several of these cheeses contain, in addition to casein, smaller or larger
amounts of albumin (Zeiger), which ajso has an influence on the ripening
process.

3 " Centralblatt f. Bakt.," 2 Abt,, 1904, Bd. XIII., p. 755, and 1907.
Bd. XVII., p. 225.

* "Centralblatt f. Bakt.," 2 Abt., 1896, Bd. II., p. 169.

5 " Undersogelser over Ost og Ostegaering," Kristiania, 1905.

150 DAIRY BACTERIOLOGY

curd cheeses, for, in spite of the lack of fat, a piquant product
will result, owing to the formation of the sharp -tasting substances.
In the ripening of the soft acid curd cheeses, e.g., Harz Cheese,
the following organisms appear to play the principal part :—
Oidium lactis, and, possibly, also certain yellow cocci1, a
mycoderma 2, yeast, and lactic acid bacteria. In the ripe state
these cheeses generally contain more amino -acids and less
ammonia than the soft rennet curd cheeses. Oidium lactis has
a more pronounced action on casein in presence of large amounts
of lactic acid ; it forms only small amounts of ammonia.

1 Eckks, "Landwirt. Jahrbuch der Schweiz," 1905, p. 503.

s EaJin, " Centralblatt f. Bakt.," 2 Abt.f 1906, Bd. XV., p. 786.

Chapter VII
Defects of Cheese

FROM the scientific point of view, the defects of cheese must be
classified according to their origin. They may be due to milk of
abnormal composition, faulty treatment in the manufacture (including
production and treatment of the curd, pressing, salting, and treat-
merit during ripening) or to bacterial action. Only the last-
mentioned cause comes within the scope of this work. In practice,
however, the various causes are so interdependent that it is
difficult to make any hard and fast distinction. By judicious
treatment of the milk and the cheese, it will generally be possible
to avoid the development of harmful organisms, whereas with
careless treatment, even when starting with good milk, their
development may easily be encouraged.

To take an instance, if the milk has curdled badly (a fault
which might have been corrected by raising its temperature or
adding a little calcium chloride), owing to the presence of raw
milk or milk from cows which are getting towards the end of their
lactation period, the curd will not dry readily and the cheese will
tend to become spongy, even though the milk used could be described
as good from the bacteriological point of view. Sponginess may
thus be due, not merely to the presence of large numbers of gas-
producing organisms, but to an excess of lactose in the curd.
As has been mentioned, gas-producing organisms may come from
the udder (Aerogenes mastitis), though in the great majority of
cases they owe their presence to diarrhoea among the cows and
unclean milking ; the acuter the digestive trouble, the richer
will the manure be in gas-producing bacteria and the greater
will be the difficulty in keeping the milk free from infection through
the manure. The gases which are formed consist chiefly of
hydrogen and carbon dioxide, the former being the more objection-
able, for the water in the cheese will absorb large amounts of
carbon dioxide before the slightest tendency towards sponginess
becomes apparent. At 15° C. and at atmospheric pressure water
will absorb its own volume of carbon dioxide, at two atmospheres
pressure twice, and at three atmospheres three times its volume,

152 DAIRY BACTERIOLOGY

etc. ; not only are the hard cheeses submitted to great pressures
in the press, but any tendency towards expansion from within
will meet with considerable resistance, owing to the close texture
of the curd, and particularly the rind. By storing at a low
temperature, a tendency towards sponginess may be kept in check
in two ways, for not only is the absorption of carbon dioxide by
water increased by lowering the temperature, but the development
of the gas-forming bacteria is checked. It will thus be understood
why sometimes bacterial growth may appear to commence afresh
when the cheese is taken out of a cold cellar on a warm day, even
though no new development of gas-forming bacteria actually
takes place. It will also be understood why eyes are more readily
formed in Emmental cheese when the curd has been saturated
with carbon dioxide during the fermentation of the lactose ;
under these conditions the distinction between normal and
abnormal eyes may not become so sharp as might be expected,
even though the two processes are due to totally different bacteria.
Unlike carbon dioxide, hydrogen is only very sparingly absorbed
by water ; it follows that those organisms which produce the most
hydrogen are capable of doing the most harm. Thus the butyric
acid organisms may transform the cheese into a large-holed^
spongy mass in the course of two days ; the non-motile butyric
acid bacteria are especially dangerous, owing to their ability to
ferment calcium lactate, and thus to cause damage after all the
lactose has been fermented. The aerogenes bacteria, especially
the colon bacteria, also produce hydrogen. On the other
hand, these organisms are able not only to effect respiration by
means of atmospheric oxygen, but they will transfer loosely-bound
oxygen from oxidising agents, like saltpetre, to sugar, and thus
consume the sugar completely, so that no hydrogen is liberated
or lactic acid formed. Saltpetre is thus an excellent preventative
of the harmful effects of these bacteria ; as a rule it will be
sufficient to add 30 to 50 grams of potassium nitrate per 100 litres
of milk 1. As the propionic acid bacteria and the lactose fer-
menting yeasts do not form other gases than carbon dioxide,
they are less dangerous to the fresh curd. The yeasts can,
however, blow the soft cheeses in which there are particularly
large amounts of sugar to be fermented ; their growth is pro-
moted by small amounts of free lactic acid. The most natural
means of preventing sponginess, as well as most other cheese
defects, is the use of a vigorous lactic acid starter consisting, as

1 According to Eosengren, it is dangerous to use saltpetre in Emmental
cheese, even in small amounts ; 10 grams per 100 litres may produce an
unclean taste and turn the cheese red. Feebly acid cheeses, like Gouda,
are best able to stand the addition of larger amounts of saltpetre.

DEFECTS OF CHEESE 153

far as possible, of a culture of the specific organisms of the cheese
in question ; in this way the ripening of the cheese will also be
accelerated. Saltpetre in small amounts is without effect on the
ripening process, while a low temperature and plentiful salting
delay it. The last-mentioned expedient is very effective ,in
regulating the formation of normal eyes, but it is too slow in
operation to prevent the defect of sponginess unless the curd is
salted direct, before moulding, though this method is not appli-
cable to the choicer varieties of cheese. (See the " Ripening of
Em mental Cheese.")

As was mentioned above, the degree of plasticity of a cheese
is determined by its content of acid and salt. If the curd is too
acid or if it has been made too dry, e.g., by over-scalding, it will
become crisp and, therefore, easily crack or crumble to pieces on
rough treatment, and particularly if much gas is produced. A
hard coat or rind produced by injudicious salting or over-drying
will, of course, easily crack. The coat will also tend to crack
when large amounts of whey collect beneath it ; this may occur
when the cheese is pressed too hard to begin with, so that the
outer layer becomes too compact before a sufficient amount of
whey has been expelled, or, again, as mentioned above, the
trouble may be due to slimy whey. If the cheeses are too damp
without, however, being particularly acid, they will flow or be
liquefied, especially if the temperature is high. 8c. liquefaciens,
if abundant, will always cause this defect. It has a strong
peptonising action on the curd and produces, at the same time, a
bitter taste.

While defects in colour no longer play any important part as
far as milk is concerned, they are of great importance in the case
of cheese making, for in cheese they have ample time to develop.
Distinction may be made between cases in which the colour appears
evenly or in spots throughout the whole cheese, and those cases
in which it only appears on the surface.

Light spots in the interior are due to the reduction of the colour
which has been added to the cheese 1. The commonest colour
defect is the turning grey or blue of the curd, due to admixture of
salts of iron or copper. Iron may come from the water, rusty
pails or, if the milk is heated by direct steam, from the steam-
pipes. Copper may come from the cheese vat or, in the case of
Parmesan cheese, from the untinned copper vessels in which the

v

1 According to Campbell ("Trans, of the Highland Agric. Soc. of Scot-
land," 1898) this defect may be avoided by the use of a good lactic acid
starter. The reducing organisms may be colon bacteria (Harrison, " Eevue
Generale du Lait," 1902, vol. 1, p. 457) and torulse (Harding, Rogers and
Smith, New York Agric. Exper. Station, Geneva, 1900, Bull. No. 183).

154 DAIRY BACTERIOLOGY

evening milk is often kept for eighteen hours. In the case of iron
the colour is due to ferrous salts, for which reason the outer
portion of the cheese will not be coloured, while the colour
disappears from a slice of the cheese which is exposed to the air.
In the case of copper the conditions are reversed ; as the colour
is due to the green cupric salts, the outer portions will be the most
affected, and a slice cut out of the cheese will only develop the
full colour after exposure to air for some time. Metallic sulphides,
stable in air, will be produced where hydrogen sulphide is formed.
The minute coloured spots which are sometimes found distributed
throughout the whole mass are of greater interest from the
bacteriological point of view, as they are colonies of chromogenic
organisms, which develop in the same way as on other solid media,
such as nutrient gelatine or agar x. The conditions of bacterial
growth consequent on the cheese being a solid medium are not
so strikingly illustrated in the case of the lactic acid bacteria,
which grow rapidly throughout the fresh curd, and thus appear
to be evenly distributed ; but the slow-growing organisms which
appear at a later stage will appear in this characteristic manner.
Thus Bacillus cyaneofuscus (which, however, dies out before the
cheese has fully ripened) forms blue spots in Edam cheese, while
the chromogenic propionic acid bacteria form red and brown spots
in Emmental cheese. In all probability the eyes in Emmental
cheese are formed in those places where the colonies of propionic
acid bacteria are particularly abundant. According to Connel 2.
the so-called rusty spots in Cheddar are caused by Bacillus rudensis,
an acid-producing organism, which may possibly belong to the
propionic acid group, as it is generally found in or near the eyes
of Emmental cheese ; being particularly prevalent in spring,
it is supposed to originate from the fresh grass ; if once established
in the dairy, it will appear in successive batches of cheese unless
all the appliances are sterilised.

The form of discoloration resulting in the production of a red
colour just inside the rind, but not in the rind itself, may be
regarded as intermediate between interior and exterior discolora-
tion. It is said to be due to the diffusion of colouring matter from
the shelves into the cheese ; shelves of white pine, but not of red
pine or fir, are said to be objectionable in this respect. This
explanation hardly holds good in all cases, for the red zone may
spread after the cheeses have been removed to another place, and

1 The staining of cheese sections to show the natural position of the
bacteria was first accomplished by Miss Gerda Troili-Petersson, 1904
("Centralblatt f. Bakt.," 2 Abt., Bd. XI., p. 212).

2 "Discoloration of Cheese," Canadian Dept. of Agric. Bull., 1897, and
Harding and Smith, New York Agric. Exper. Station, Geneva, 1902,
Bull. No. 225,

DEFECTS OF CHEESE 155

the coloration is generally accompanied by an unpleasant taste.
In the latter event the trouble is most likely due to chromogenic
organisms which have penetrated the rind. Among the bacteria
which produce a red colour in cheese, both cocci and rod forms
are known 1. Several of them liquefy gelatine, and many of them
are chromogenic on cheese, but not on the common media, and
some of the organisms which produce a red colour on the surface
of the soft mouldy cheeses are only chromogenic in presence of
the decomposition products formed from the casein by the
moulds which are characteristic of the cheese in question. Bacteria
are known which colour the rind yellow 2 and brown ; in fact, the
rind always becomes brown when kept damp. Atmospheric
oxygen plays an important part in all these colour changes.
Many moulds also produce surface colorations, e.g., Oidium
aurantiacum, Penicillium casei, Cladosporium herbarum, and
Monilia nigra. Red and yellow Torulse are also believed to play
some part in the process. Many moulds penetrate the rind and
make the surface uneven. In this connection mention may be
made of cheese mites and maggots, though their detailed descrip-
tion does not come within the scope of a work on bacteriology.
In order to avoid the transference of harmful organisms from
cheese to cheese, the uninfected cheeses should always be washed
before those which are infected, and the cloth should be boiled for a
quarter of an hour daily. The only efficient means of preventing
the defects under consideration is the thorough disinfection of
the ripening room and the shelves (see p. 56).

Defects in taste and smell, originating from the fodder, will
generally disappear in time. Similarly, the bitter taste due to
Streptococcus liquefaciens and Torula amara, and sometimes also that
due to certain lactic acid rod bacteria 3, may disappear during a
later stage of the ripening. Many of the chromogenic organisms
produce unpleasant tastes. The tallowy taste sometimes found in
rich cheeses (e.g., Gouda cheese) is probably due to those organisms
which turn milk and butter tallowy. Bitter and tallowy tastes
are defects which chiefly occur when fresh cheeses are kept too

1 Thus Adametz has described two micrococci which do not liquefy
gelatine, or, at any rate, only do so very slowly ; they form red colonies on
gelatine and agar. Gratz has isolated a liquefying bacterium, Micrococcus
rubri casei, which forms pink colonies, and Weigmann has isolated two
liquefying organisms, Micrococcus chromoflavus and Bacterium casei fusci,
which form chrome yellow and cream-coloured colonies respectively on the
common media, but which turn the surface of the cheese red. The red
Bacillus firmitatis, isolated by Roger from Camembert cheese, grows only
in the decomposition products produced by moulds."

2 Thus Barthel has found that yellow spots may be produced by Micro -
coccus flavus, a liquefying organism commonly found in air.

3 Harding. Eoqers and Smith, New York Agric. Exper. Station, 1900,
Bull. No. 183,

156 DAIRY BACTERIOLOGY

damp and cold 1. In the French soft cheeses Penicillium brevicaule
produces a taste of cabbage.

The organisms which produce spongy cheese may also give rise
to unpleasant, mostly bitter-sweet tastes. Spongy cheeses, more-
over, dry too readily, so that they ripen too slowly.

As cheese poisoning, like meat poisoning, is chiefly due to
certain colon bacteria and sometimes to non-motile butyric acid
bacteria, spongy cheese must always be regarded with suspicion 2.

1 At a low temperature the peptonising bacteria develop more rapidly
than the good lactic acid bacteria, or the bacterial metabolism may be too
sluggish to cause the disappearance of all the atmospheric oxygen in the
cheese. Sunlight promotes the oxidation of the fat in cheese, as in butter.
As already mentioned, copper salts and carbonic acid which latter is
copiously produced in the ripening of cheese, may turn the fat tallowy.

2 H. Kuhl (" Zeitschrift f. Untersuchung der Nahrungs und Genuss-
inittel," 1913, Bd. 25, p. 193) reports on a case of poisoning by cheese
which was due to an aerogenes bacterium. A number of references to the
literature of this subject are given in this paper.

Chapter VIII
The Grading of Milk

THIS chapter deals with the more important methods of judging
of the cleanness and freshness of milk and its suitability for the
making of good dairy products.

EXAMINATION FOR TASTE AND SMELL

As the senses of taste and smell are our best aids in avoiding
putrid or harmful food, every careful examination of milk should
include tasting and smelling. Unfortunately the test is of a
decidedly subjective nature, as the senses in question are soon
dulled. Further, considering that the temperature of the milk
varies considerably on arrival, and that the smell and taste are
most pronounced when the milk is warm from the cow, it is
evident that milk can only be graded very roughly by this method,
and that the detailed classification in fifteen grades according to
the taste and smell alone, as was previously carried out by the
Danish milk grading associations, was futile.

ESTIMATION OF DIRT

Milk which contains visible amounts of pus, blood, manure, etc.,
must of course be regarded with suspicion from the outset. It
may justly be demanded that milk retailed in towns shall show
no sediment when 1 litre is allowed to stand at rest for two hours
in a vessel of colourless glass. It is advantageous to use vessels
tapering to the bottom, so that the sediment may readily be
collected for examination after the milk has carefully been decanted
off. As only the heaviest particles will separate, it is more usual
nowadays to filter a definite quantity through a^ disc of cotton
wool, which will become more or less dirty according to the state
of the milk. The discs may be dried and kept, so that a " dirt
scale" may be prepared for future reference and comparison1,
and the dirtiest discs may perhaps produce some moral effect if
sent to the suppliers responsible for them. As it is impossible to
take an average sample of dirt from a large quantity of milk, it

1 See, for example, Hoyberg's Scale in " Maanedsskriftet for Sund-
hedspleje," 1910, p. 49.

158 DAIRY BACTERIOLOGY

will be necessary in dairies to filter the contents of each can or
churn through a separate cotton wool disc, and also to examine
the empty cans carefully, for most of the dirt usually remains in
them. As has been previously pointed out, the chief impurity of
milk is cow manure, which contains 80 per cent, of water, and
soluble matter, both of which are completely incorporated with the
milk owing to the shaking up which occurs in transit ; at the same
time, the bacteria which constitute an appreciable proportion of
the solid matter, are distributed throughout the milk. The above
estimate applies to normal dung ; if the cows are suffering from
diarrhoea, the dung will contain a still larger proportion of soluble
matter, and consequently the filtration test will show less dirt.
Further, the more liquid the dung the greater will be the propor-
tion of dangerous bacteria introduced into the milk. The estima-
tion of dirt therefore furnishes no measure of the bacterial contents
of the milk, and it must also be remembered that very dirty
milk which has been well cooled may contain fewer bacteria than
less dirty milk which has been inadequately cooled. In reality,
therefore, the only sure indication afforded by the dirt test is
whether the milk has been properly cleaned or not, either by
straining, filtering or centrifuging.

TROMMSDORFF'S LEUCOCYTE TEST

The particles of dirt and foreign bodies which are found in
suspension in milk are removed much more completely by centri-
fuging than by sedimentation or filtration. Thus, of the white
blood corpuscles or leucocytes which normally do not separate out
when the milk is allowed to stand for a relatively short time,
3 to 50 per cent, are separated by centrifuging, or even more if the
milk is warm 1. The heating should, however, not exceed 70° C.,
for otherwise precipitation of the proteins may occur. As was
first shown by Barthel, the centrifuge slime therefore consists very
largely of leucocytes2. By direct microscopical counts, normal
milk has been shown to contain J to 1 J million leucocytes per cubic
centimetre 3. The number of leucocytes increases as the yield of
milk decreases, being particularly high at the beginning and
towards the end of the lactation period. It is still higher in cases
of udder disease, and on this fact Trommsdorff has based his test,
the object of which is to ascertain whether or not the milk has
been derived from healthy cows. The test is carried out as

1 Campbell, U.S. Dept. Anim. Industry, Bull. 117, p. 19.

" Revue generate du lait," 1901, vol. 1, p. 193.

3 Prescott and Breed, " Journal of Infectious Diseases," 1910, vol. 7,
p. 632 ; Breed and Stiger, ibid., 1911, vol. 8, p. 361.

THE GRADING OF MILK

159

follows : 10 c.c. of the milk are introduced into a glass tube, the
end of which is drawn out into a capillary graduated in thousandths
of a cubic centimetre. The tube is closed by a rubber stopper and
whirled at a speed of 1200 revolutions per minute. Normal milk
will generally yield a sediment measuring 0-002 to 0-004 c.c.,
while milk drawn from diseased udders will yield a sediment
measuring 0-01 c.c. or even more. The amount of the sediment
is, however, not a decisive criterion by itself ; the test must be
supplemented by a microscopic examination, and only in cases
where large numbers of bacteria characteristic of udder disease,

FIG. (34. — Leucocyte Sediment from the Milk of a Cow suffering from
streptococcic mastitis. (After Ernst.} X 1000.

e.g., streptococci, are found can definite conclusions be drawn.
As these bacteria cannot be distinguished from the harmless milk
bacteria by direct inspection1, the test is only of value when
applied to milk fresh from the cow, and can therefore only be
applied as an aid to veterinary control at the farm. In mixed
milk the characteristic features are completely lost on account
of the dilution alone.

THE CATALASE TEST

This test is supplementary to the leucocyte test. The various
constituents of blood, especially the corpuscles, are rich in catalase ;

1 Capsule formation and disc-like cells, which by some authors are
regarded as characteristic of Sc. mastilidis, can be observed in all strepto-
cocci. On the other hand, as already mentioned under Sc. mastitidis, the
red colour in casein starch stab cultures is very characteristic.

160

DAIRY BACTERIOLOGY

£L

n

for this reason, milk drawn from cows with diseased udders or from
cows which are approaching the end of the lactation period, or
colostrum, will liberate large amounts of oxygen from hydrogen
peroxide. As the fat globules carry with them large numbers of
leucocytes, unpasxeurised cream, obtained either
by spontaneous separation or by the use of the
separator, will be richer in leucocytes than the
corresponding skim milk. Many ingenious and
sometimes complicated forms of apparatus have
been devised for the carrying out of the test.
In the author's laboratory Lind's apparatus is
used ; this consists simply of a graduated tube
of 20 c.c. capacity, into which 15 c.c. of milk are
introduced ; sufficient hydrogen peroxide (1 to 3
per cent.) is added to fill the tube, the rubber
stopper carrying the bent tube is inserted, and the
apparatus is inverted as shown in the illustration.
The maximum amount of oxygen is obtained at
20° to 25° C., so that no water bath or thermostat
is necessary. The number of cubic centimetres
of oxygen evolved in six hours is taken as the
catalase number. Fresh milk from healthy cows
will not yield more than 2-5 c.c. In mixed milk
the blood corpuscles will be too sparsely distributed to produce
any recognisable effect. The milk must be perfectly fresh
when tested, as many bacteria decompose hydrogen peroxide.
The common sarcinae and micrococci (see p. 38), e.g., Micrococcus
candicans, and most of the putrefactive bacteria are particularly
active in this respect, while the true lactic acid bacteria and the
butyric acid bacteria l do not produce catalase. Milk which has
stood for any length of time at a low temperature, or old pas-
teurised milk, will accordingly show high catalase values. Hesse 2
has proposed to apply the catalase test to butter as follows :
100 grams of butter are warmed to 45° C. and shaken with 40 c.c.
of water at this temperature ; the aqueous liquid is separated and
tested in the same way as milk. It is obvious that butter which
has been made from pasteurised cream ripened with a pure starter
of lactic acid bacteria will show a low catalase figure ; if the
butter has been washed with bad water it may show a high catalase
figure, for Bacterium fluorescens liquefaciens is particularly active
in decomposing hydrogen peroxide.

1 Orla Jensen, " Det. kgl. danske Videnskabers Selskabs Oversigter "
(Danish Academy of Sciences), 1906, No. 5, p. 306.

2 " Molkereizeitung," Hildesheim, 1912, No. 6.

FIG. 65. — Catalase
Test Apparatus.

THE GRADING OF MILK 161

THE RENNET TEST

As milk drawn from diseased udders generally coagulates badly
with rennet, this test affords the same indications as the two tests
just described ; it has also a practical significance, for milk which
coagulates badly can hardly be made into good cheese, even if it
is satisfactory from the bacteriological point of view. The test
may be applied with advantage when difficulty is experienced in
making the curd sufficiently dry, so that the particular consign-
ments of milk which are at fault may be detected. In the labora-
tory the test is generally carried out by means of Schaffer's
apparatus, which consists simply of a shallow water bath with a
false bottom on which a number of beakers can be placed. One
cubic centimetre of one of Hansen's rennet tablets, No. 2, in
500 c.c. of water is added to 100 c.c. of milk which is kept at 35° C. ;
normal milk will coagulate in nine to nineteen minutes. MarschalUs
apparatus is better suited for use in the dairy ; this consists of a
graduated enamelled iron cup having a fine opening in the bottom.
The hole is closed by a finger, the cup is filled, and milk is allowed
to run out until the level comes to the top graduation mark in the
cup. One cubic centimetre of the rennet solution is rapidly and
thoroughly mixed with the milk, and the finger is removed from
the opening. The milk will cease to run out the moment coagula-
tion sets in, so that the capacity for coagulation will be inversely
proportional to the amount of milk which has run out. Con-
versely, this apparatus may be used for the estimation of the
coagulating capacity of rennet, which is a test of some importance
in the Swiss dairies which make their own rennet. The means
available for correcting a deficient capacity for coagulation are
discussed on p. 130.

DETERMINATION OF ACIDITY

The degree of acidity is generally understood to be the number
of cubic centimetres of standard sodium hydroxide solution
required to neutralise a given volume of milk, with phenol phthalein
as indicator. The Soxhlet-Henkel degrees, which are largely used
on the Continent, express the number of cubic centimetres of
quarter normal alkali per 100 c.c. of milk. In British and American
works, degrees of acidity are understood to represent the number
of cubic centimetres of normal sodium hydroxide per litre of milk ;
the titration is often carried out with decinormal alkali, using
smaller quantities of milk, the results being calculated to the
above standard ; as little as 10 c.c. of milk is sometimes used, but
more exact results can be got by using larger quantities. No
water should be added to the milk or cream before titration, as

D.B. 11

102 DAIRY BACTERIOLOGY

this will lower the acidity. EllbrecMs titration paper " Exact "
has been designed as a colour standard in order to ensure that the
same end-point is reached each time. Richmond gives the follow-
ing method in his " Dairy Chemistry " : " 10 c.c. of milk are
titrated with decinormal baryta, or 11 c.c. with eleventh normal
strontia, using 1 c.c. of J per cent, phenol phthalein ; as a standard,
an equal volume of milk is coloured with one drop of 0-01 per cent,
rosaniline acetate in 96 per cent, alcohol." The degree of acidity of
normal milk generally lies between 16 and 19 (6-4 to 7-6 Soxhlet-
Henkel). If under 15 (6 S.-H.), the milk is probably derived from
cows which are sick or approaching the end of the lactation
period, or it may have lost a portion of its natural carbonic acid
by having been kept in shallow vessels, shaken or warmed. If the
degree of acidity is over 21 (8-4 S.-H.) the milk may be derived
from cows suffering from streptococcic mastitis or it may contain
colostrum the acidity of which may be as high as 55. As a rule,
however, high acidity will be due to incipient lactic acid fermenta-
tion. Mixed milk having an acidity of over 21 will usually
coagulate on mixing with an equal volume of 68 per cent, alcohol ;
this is the basis of the so-called alcohol test. The boiling test is
based on the fact that milk having an acidity of over 27-5 coagu-
lates on boiling. It is, however, impossible to be certain that the
milk will stand pasteurisation if the acidity exceeds 22-5 *. Fresh
milk shows an amphoteric reaction towards litmus, i.e., it turns
red litmus blue and blue litmus red ; if the degree of acidity is
under 12-5, the reaction towards litmus will be alkaline. Accord-
ing to Hoyberg2, the rosolic acid solution proposed by Hilger for
the detection of added soda may be used with advantage in
testing milk from the individual quarters with a view to detecting
udder disease. If 5 c.c. of 96 per cent, alcohol and 0-5 c.c. of a
1 per cent, rosolic acid solution are added to 5 c.c. of milk, an
orange colour will be obtained with normal milk, and a red colour
with alkaline milk. Eugling 3 has shown that a saturated alcoholic
solution of alizarin may be used for the same purpose ; if 5 to
10 drops of the solution are added to 50 c.c. of milk, a red-violet
colour will be produced with normal milk, a violet-blue colour
with alkaline milk, and a yellowish colour with sour milk. Morres 4
combines this test with the alcohol test in the Alizarol test, 0-05 per
cent, of alizarin being dissolved in the 68 per cent, alcohol, so that
an indication may be obtained as to whether the coagulation is
due to acid- or rennet-producing bacteria. If narrow test tubes

1 Henlcel, " Milchwirtschaftliches Zentralblatt," 1907, Bd. III., p. 378.

"Skandinavsk Veterinaertidsskrift," 1911, p. 23.

" Handbuch f. die praktische Kaseriei." Leipzig, 1901, p. 20.
4 " Oesterreichsche Molkerei-Zeitung," 1912. A colour scale for this
test is supplied by Dr. N. Gerbers Co., Zurich.

THE GRADING OF MILK 163

%

are used, 2 c.c. of milk and 2 c.c. of alcohol will suffice for this test.
It is recommended that the dairies should apply it to each can of
milk and reject all milk which shows a precipitate or an abnormal
colour.

THE FERMENTATION TEST

This test shows if the milk has become infected with an undue
proportion of gas-producing organisms, which first and foremost
include the pseudo lactic acid bacteria. We have already seen
that these are among the most objectionable organisms that can
be met with in dairy practice, as they cause trouble in various ways,
including the spoiling of milk for the purpose of cheese making.
As the pseudo lactic acid bacteria are usually brought into the
milk with the cow dung, and are particularly plentiful when the
cows are suffering from diarrhoea, the fermentation test will give
evidence of undue contamination and thus give warning that the
milk may possibly be dangerous for human consumption. This
test is of value not only in judging of the suitability of milk for
cheese making, but also in the laboratory control of retail milk,
particularly that which is to be used for infant feeding. While
most of the milk bacteria (excepting the thermobacteria, which,
however, are rare in fresh milk) grow best at about 30° C., many of
them ceasing to develop at temperatures above 38° C. , the pseudo
lactic acid bacteria, being typical intestinal organisms, have their
optimum at blood heat, and will gain predominance most readily
at a slightly higher temperature, for which reason the fermentation
test is carried out at 38° to 40° C. It is important that the tem-
perature should not be allowed to vary beyond these limits, as at
a higher temperature good milk may appear to be bad, while at a
Lower temperature bad milk may appear to be good. If fine
distinction is made between the different types and degrees of
fermentation, these temperature limits are too wide, and the
temperature should be kept constant at 38° C. In order that the
results of the test may be strictly comparable, it is advisable
always to use tubes of a certain diameter (about 2 cm.), into which
40 c.c. of milk are introduced. The shape of the tubes is shown
in the accompanying illustration ; they should be strongly made
and provided with a graduation mark at 40 c.c., and with a frosted
square so that they can be marked in pencil. An average sample
of each supplier's milk should be taken, preferably from the
weighing or measuring vessel, by means of a small measure
furnished with a pointed spout. The measure should be rinsed
several times with the milk which is being sampled before taking
the actual sample. The tubes are marked, covered with a small
cap of aluminium or zinc, placed in stands and brought into the

164

DAIRY BACTERIOLOGY

water bath. Failing a proper thermostat, the temperature may
be kept fairly constant by means of a spirit lamp or gas burner,
provided that a large well-insulated water bath is used, and that
the temperature of the room does not vary too much. The
samples are examined after twelve and twenty to twenty-four
hours. If quite fresh, the milk will remain liquid for twelve hours.
The sooner visible alteration occurs, the greater the importance
to be attached to the results of the test. On the other hand, milk

01 02 03

FIG. 66. — Gelatinous Types.

&1 &2 &a

FIG. 67. — Blown Types.

containing relatively few bacteria gives unreliable indications,
duplicate tests often showing different types of fermentation.

According to Peter's classification, we may distinguish between
the following types in the fermentation test : — Fluid (/), gela-
tinous (g), blown (b), spongy (s) and cheesy (c) (see his illustrations).
In the gelatinous type the true lactic acid bacteria predominate ;
a perfectly homogeneous coagulum is denoted by gv a coagulum
with very few streaks and bubbles by g2, and one with few streaks
and bubbles by g3. In the blown (gassy) milk the pseudo lactic

THE GRADING OF MILK

165

acid bacteria predominate ; bv b2 and 63 indicate progressive degrees
of intensity of gas evolution. While there is only a difference of
a degree between g3 and 6l5 all the casein will have been driven
to the surface in b3. The spongy types, sv s2 and «3, differ from
the blown types in being finer in curd texture. In si the coagulum
forms so fine a network that it may easily be mistaken for gr
Milk which is poor in true lactic acid bacteria often becomes
spongy in the fermentation test, owing to the gas production

FIG. 68. — Spongy Types.

FIG. 69. — Cheesy Types.

being in full swing before coagulation occurs. In this case the
gas-producing organisms may sometimes be lactose fermenting
saccharomycetes. The cheesy type is distinguished by a well-
marked separation of clear whey, due to organisms which secrete
rennet-like enzymes, especially peptonising lactic acid strepto-
cocci ; cl5 c2 and c3 indicate progressive degrees in the contraction
of the curd. If at the same time much gas has been formed, this
type cannot well be distinguished from the spongy or the blown
type. In actual practice distinction is only made between the

166 DAIRY BACTERIOLOGY

highly objectionable blown types 62, ^3 and s3 on the one hand,
and all the remaining types on the other. In examining the
fermentation types the following should also be watched for :
sediment (and possibly pus), sliminess, and evil-smelling milk.
Alkaline milk keeps fluid for a long time, and often putrefies before
going sour.

The combined rennet and fermentation test suggested by
Fr. Jos. Herz is a special form of the fermentation test, 2 c.c. of
the rennet solution mentioned above being added to each test
tube. Milk which coagulates badly will have separated compara-
tively little whey and formed a soft and non-coherent coagulum
after twelve hours. On examination after twenty to twenty -four
hours the coagulum should have assumed the form of a smooth
cjdinder, which only shows small holes in longitudinal section.
If the coagulum contains large holes, and especially if it forms a
screw-shaped sponge floating on the surface of the whey, the milk
must be considered unsuitable for cheese making.

In the Emmental dairies, where a home-made rennet is used
which, at the same time, is also a culture of the more important
ripening bacteria, but which not infrequently contains colon and
aerogenes bacteria, it is of the greatest importance to test the milk,
both with and without the addition of rennet. If a bad result
is obtained without rennet and a good result with rennet, the milk
is certainly bad, though the lactic acid bacteria in the home-made
rennet will be able to counteract the defect. The milk is only
unsuitable beyond doubt if the test with rennet turns out badly.
If the last-mentioned result is obtained in spite of the fact of the
milk being good, then the rennet will be known to be unfit for use.

The tubes and caps used in the fermentation test must be
cleaned immediately after use by rinsing carefully with hot soda
solution, after which they should be placed in the stands and
covered completely with water into which steam is then passed
for about fifteen minutes. Finally, they are to be dried in a warm
place.

THE REDUCTASE TEST

All living cells, including microorganisms, have reducing
properties, which are well illustrated by their behaviour towards
methylene blue. The rate at which milk decolorises methylene
blue will, therefore, depend on the number of microorganisms in
it. The reductase test devised by Barthel x and the author 2 is
based on this fact.

1 *' Kungl. Landtbruks-Akademiens Handlingar och Tidskrift," No. 6,
1907.

2 " Maelkeritidende," 1909, p. 359.

THE GRADING OF MILK 1(57

It must not be assumed that the time taken to decolorise
methylene blue (reduction time) under standard conditions is
an accurate measure of the number of microorganisms in the
milk ; for, in the first place, all microorganisms do not reduce with
equal rapidity, and, second, the milk itself, as obtained from the
cow, contains reducing substances. Of the milk bacteria examined
by the author Streptococcus liquefaciens appears to have particu-
larly marked reducing powers, while the true lactic acid bacteria
are among the organisms which reduce slowly. The obligate
anaerobic bacteria reduce rapidly, a fact easily explained on
considering that they derive their energy from reduction processes.
The reduction time is shortened by the addition of a little alkali,
and lengthened by the addition of a little acid. As regards the
reducing substances natural to milk, the most important is
aldehyde reductase, an enzyme which appears to be associated with
the fat globules 1, but which has no significance in the present
connection, as it only decolorises methylene blue in presence of
formaldehyde. Greater interest attaches to the leucocytes, which,
like other living cells, are able to reduce methylene blue2. They
will, however, only exert an appreciable effect on the reduction
time if present in large numbers, while it can hardly be considered
a drawback to the test that milk rich in leucocytes should appear
to be worse than its bacteriological condition would warrant,
inasmuch as such milk should always be regarded with suspicion.
Of still greater importance is the fact that milk contains substances
other than enzymes which exert a reducing action in the absence
of oxygen. Thus Burri and Kursteiner have shown 3 that newly-
sterilised milk which has been prevented from absorbing oxygen
decolorises methylene blue rapidly, and Barthel 4 has shown that
raw milk behaves similarly when the dissolved oxygen is expelled
by a current of hydrogen or carbon dioxide, from which he
concludes that the decolorisation of methylene blue in milk is
due to the action of the milk itself, and that the microorganisms
only act indirectly by consuming the dissolved oxygen. This

1 Orla Jensen, " Uber den Ursprung der Oxydasen und Reduktasen der
Kuhmilch," Det Kgl. Danske Videnskabernes Selskabs Oversigter (Danish
Academy of Sciences), No. 5, 1906, and Centralblatt f. Bakt. II. Abt.
1907, XV1IL, p. 211.

2 Olav Skar, " Skandinavsk Veterinaertidsskrift," 1913, p. 51.
" Milchwirtschaftliches Zentralblatt," 1912, p. 269.

4 " Skandinavisk Veterinaertidsskrift," 1916, p. 155. The author has
been able to confirm BartheVs results, and has found that milk with a low
bacterial count which is kept free from atmospheric oxygen by passing a
current of hydrogen through it reduces methylene blue in forty-five
minutes at 40° C., no matter whether raw or sterilised. (On the addition
of a little formaldehyde decolorisation was complete in ten minutes.)
As pure lactose solutions were not found to decolorise methylene blue
under similar experimental conditions, the reducing action of milk itself
cannot be due to the lactose.

168 DAIRY BACTERIOLOGY

explanation, however, hardly covers the case of the obligate
anaerobic bacteria, but, as far as the aerobic organisms are
concerned, the reducing action of the milk itself is, no doubt, a
contributing factor, so that in the reductase test the degree of
aeration of the milk is a condition which must not be overlooked.
Milk will already begin to absorb oxygen freely as it comes in
a fine stream from the udder, and its oxygen content will naturally
be increased on pouring from vessel to vessel, and especially during
any process of aeration to which it may be submitted. The
dissolved oxygen will gradually be consumed as the micro-
organisms increase in number ; the less the milk is shaken, and
the deeper the vessels in which it is kept, the quicker will the
oxygen content fall off. The temperature at which the milk
is kept is also an important factor in this connection, for not only

FIG. 70. — Apparatus for Reductase and Fermenting Test to take 200 Samples.

will the bacteria develop more rapidly at higher temperatures,
but the individual cells will consume more oxygen at higher than
at lower temperatures 1. In order to equalise these conditions,
it will be advisable always to shake the milk well before subjecting
it to the reductase test.

It will be seen that the theory of the reductase test is not
so simple as was formerly supposed ; nevertheless, numerous
experiments with mixed milk of commerce have shown that the
reduction time, as determined by the reductase test, furnishes
just as satisfactory a measure of the bacterial contents as the
troublesome method of plate counts, which, as a matter of fact,
is by no means less subject to error than the reductase test.
Moreover, the differences between the reducing powers of the
different species of bacteria are not greater than the differences
displayed in this respect by the members of the same species

1 This is clearly shown by C. Lind's work, " Reduktaseproven sammen-
io-net med Bakterietaellingsmetoden " (" Ma-elkeritidende," 1915, p. 921),

THE GRADING OF MILK 169

when subjected to different conditions. The more favourable
the conditions for bacterial development, the shorter will
be the reduction time. It follows that not only does the
reductase test give an estimate of the number of organisms
present, but the result will be influenced to some extent
according to the vitality of the organisms, which is a factor
as important as any other in determining the keeping power
of the milk. As the reductase test will reveal any appreciable
bacterial increase before this becomes apparent through the
presence of lactic acid, it is not only the most convenient, but also
the most sensitive, method for the grading of milk, whether for
retail direct or for the making of dairy products. On the other
hand, it must be admitted that milk from individual cows or milk
that has been subjected to unusual treatment may give divergent
results. The translator l has thus found that milk which has
been kept for a long time at low temperatures appears to be better
than it really is, in the reductase test carried out at 38° C., because
the majority of the bacteria in it are greatly weakened at the
temperature in question. Such milk is decolorised quicker at
28° C. A comparison between the reduction times of the same
milk at 28° and 38° C. can thus furnish information as to how this
milk has been treated. According to the author's proposal,
however, the test should be carried out at 38° to 39° C., as this will,
in the majority of cases, give the shortest reduction time. On
the other hand, it is a wrong principle to employ a still higher
temperature, as was oTiginally done, for then the development of
all the most common milk bacteria will be hindered.

The methods of sampling and the apparatus required for the
reductase test are the same as have already been described under
the fermentation test. Particular care must be taken to measure
out exactly 40 c.c. of milk, either by means of a graduation mark
on the tube or by taking the sample in a measure holding exactly
40 c.c. when filled to the rim. As the preparations of methylene
blue on the market are very different in their properties, and the
solutions made from them are not permanent, it is necessary to
use a fresh solution of definite strength, made from a standard pre-
paration. The tabloids prepared for the purpose 2 are readily
soluble in warm water ; each tabloid makes 200 c.c. of solution,
1 c.c. of which is required for each test with 40 c.c. of milk. The
colour is mixed with the milk by rolling the tube in the hands,
then pressing the mouth against a clean portion of the palm of the

1 "Analyst," 1918, 43, p. 1.

2 By Messrs. Blauenfeldt and Tvede, of Copenhagen, who also supply all
requisites for the reductase test, including complete outfits suitable for
dairies dealing with milk from 100 to 400 suppliers.

170 DAIRY BACTERIOLOGY

hand and shaking vigorously. For each tube a different portion
of the palm should be used, and, when completely wetted, the
hand should be carefully washed before proceeding any further.
In actual practice, the colour is added when all the samples have
been taken and the tubes are in position in the stand. After
placing in the water bath, the samples should be examined at
frequent intervals during the first twenty minutes, after which
they need only be examined every quarter or half hour. As
mentioned above, the author has found it best to employ the same
temperature as in the fermentation test ; in this way there is the
additional advantage that the two tests may be combined. The
combined reductase and fermenting test l gives information regard-
ing both the number and the nature of the organisms in the milk.

The joint investigations of Barthel and the author 2 have shown
that it is possible by means of the reductase test to grade milk
and cream into four classes, as follows :—

Class 1. — Good milk, not decolorised in five and a half hours,
containing, as a rule, less than | million bacteria per cubic centi-
metre 3.

Class 2. — Milk of fair average quality, decolorised in less than
five and a half hours but not less than two hours, containing, as
a rule, | to 4 million bacteria per cubic centimetre.

Class 3. — Bad milk, decolorised in less than two hours, but not
less than twenty minutes, containing, as a rule, 4 to 20 million
bacteria per cubic centimetre.

Class 4. — Very bad milk, decolorised in twenty minutes or less,
containing, as a rule, over 20 million bacteria per cubic centimetre.

If samples of retail milk from different dairies are to be com-
pared, they must, of course, be examined simultaneously ; thus
it would be unfair to sample one dairy on a cold morning and
another on a warm afternoon. Unpasteurised milk, as sold in
large towns, should retain its colour in the test for at least two
hours, and the pasteurised rmlk for at least five and a half
hours. Most of the milk retailed by the large dairies in Copen-
hagen fulfil these requirements 4. The pasteurised milk supplied

1 " Maelkeritidende," 1909, p. 359.

2 " Milch wirtschaftliches Zentralblatt," 1912, No. 14.

3 As the counts were found by the plating method, they were really
under-estimated ; milk bacteria usually occur in pairs and not infrequently
in long chains, or large clumps, which are not broken up on shaking, and in
such cases only one colony is obtained. The counts should at least be
doubled, and in some cases they should be trebled or quadrupled.

4 Orla Jensen, " Maanedsskrift for Sundhedspleje," 1909, p. 239. The
translator has found the reductase and fermenting test to be of great use
in the control of milk as it arrives from the farms, and that the conditions
of treatment of the milk on various farms were found to correspond with
conclusions which had previously been drawn from the behaviour of the
samples in this test.

THE GRADING OF MILK 171

by these dairies is, generally speaking, not so good as might be
expected ; this matter deserves attention, for pasteurised milk
which is rich in bacteria can only be regarded as a highly objection-
able product. It should, therefore, be forbidden to sell pasteurised
milk as raw milk.

In the dairies it is impossible to lay down a definite line of
demarcation between first and second grade milk without unduly
lowering the standard, for obviously milk cannot be expected to
conform to the same standard on a close summer's day as on a
frosty day. The author has therefore proposed the adoption of
the average reduction time of all the samples tested at one time as
the standard for that particular batch of samples. Only milk
which is better than the average will be placed in class 1. Milk
showing a reduction time equal to or less than the average, but
not less than two hours, will be placed in class 2, and other cases
can be dealt with as detailed above. As it may be inconvenient to
have the samples examined for longer than twelve hours, the
reduction time of any samples not finished in this time may be set
down as twelve hours in calculating the average ; such cases will
only occur on cold winter days. All reduction times may be
estimated to the nearest quarter of an hour, the reduction time
of milk in class 4 being set down as a quarter of an hour.

If it is desired to combine the grading according to taste and
smell and according to the results of the fermenting test, with the
grading according to the reductase test, the following system may
be adopted : — Samples placed in classes 1, 2 or 3 according to the
reductase test are degraded by one class if the taste and smell are
decidedly bad, and samples thus placed in classes 2 or 3 are further
degraded by one class if the results of the fermentation test show
62, 63 or s3. If the average reduction time should be under five
and a half hours, class 1 milk which is decolorised before this
time, and which is bad according to the fermenting test, should also
be placed in class 2. It will be seen that, excepting in the case
just mentioned, the results of the fermenting test are not taken
into account in the cases of samples placed in classes 1 and 4, the
reasons for this being as follows : — Milk which contains relatively
few bacteria will generally show a bad result in the fermenting
test as it will be particularly poor in true lactic acid bacteria ;
moreover, the results observed in the combined reductase and
fermenting test in such cases may often be worse than would be
shown in the fermenting test alone, as methylene blue exerts a
certain toxic effect on bacteria, especially the true lactic acid
bacteria, and the fewer the bacteria the more of the poison will
each cell have to reduce. On the other hand, the fermenting
results are not affected by the small amount of methylene blue

172 DAIRY BACTERIOLOGY

used in the reductase test, in the case of milk containing a greater
number of bacteria. Conversely, milk which is very rich in
bacteria generally shows good results in the fermenting test, for
even if it may contain millions of gas-producing bacteria, it will
generally contain still greater numbers of true lactic acid bacteria,
and therefore become sour so quickly that the former type will
not be able to gain predominance. No error will be committed in
ignoring the results of the fermenting test in the cases mentioned,
for the small numbers of bacteria in the best milk, whatever their
nature, will not be able to exert any influence on the mixed milk
of the dairy, while the worst milk will already have been placed in
the lowest class, and cannot therefore be degraded any further. In
the combined test, the fermentation need not be observed until
after the lapse of twenty to twenty-four hours as the reductase
test gives a far more accurate estimate of the number of bacteria
present than the fermenting test after twelve hours. It should
be clearly understood that while the indications afforded by the fer-
menting test are purely qualitative, those afforded by the reductase
test are purely quantitative, and it is only by combining the two tests
that any real insight will be obtained into the nature of the bacterial
contents of the milk.

In large dairies it will only be possible to take an average sample
of each supplier's milk. If the number of suppliers is not large,
there is no reason why the morning and evening milk should not
be tested separately provided that the cans are properly marked
according to the respective meals as they always should be.
When the milk only comes to the dairy in the morning, it will
generally be found that the morning milk will be better than the
evening milk, though if the milk comes from a long distance, the
reverse may be the case during warm weather if the morning milk
has not been cooled. In such cases it will also be necessary to
cool the morning milk. The more frequent the test the juster will
be the impression formed as to the relative goodness of the milk
from different suppliers. Large dairies which of course should
keep a well-equipped laboratory under the guidance of a chemist,
who has been trained in bacteriology, will do best to test each
supplier's milk daily. In the co-operative dairies, samples from
all the suppliers should be tested once a week, and one of the
members of the association should be present in turn, as an
impartial witness. The co-operation of expert milk tasters is no
longer absolutely necessary in view of the more objective methods
which are now at our disposal ; the examination of samples in the
reductase test requires no particular scientific knowledge, and may
be performed by any reliable boy or girl. Samples which need
not be subjected to the fermenting test should be removed from

THE GRADING OF MILK

173

the water bath in the evening ; next morning the blown or very
spongy samples may quickly be picked out. The following table
has been drawn up in order to illustrate the system : —

1

Time for decolorisation.

Classification according to

p

a

Remarks on

Fermenta-

Also taste

: .1
1

Taste and Smell.

Exact.

In round
numbers.

tion test.

Reductase
test alone.

Also taste
and smell.

and smell
and fer-
mentation

test.

03

1

19 min.

ihr.

ft

IV.

IV.

IV.

2

3 hr. 31 min.

3| hr. *

*a

II.

II.

III.

3

24 min.

fhr

q

III.

III. III.

4

5 hr. 30 min.

5| hr.

X

I.

I. I.

5

Acid

5 min.

i hr.

g

IV.

IV. IV.

6

over 12 hr.

12 hr!

g

1.

J

I.

7

1 hr. 57 min.

2 hr.

63

II.

II.

III.

8

7 hr. 5 min.

7 hr.

g

I.

I.

I.

! 9

52 min.

fkr

02^i*

III.

III.

III.

10

11

Flat or slight
turnip taste
Bitter

5 hr. 1 7 min.
1 hr. 40 min.

If hr.

Ml

I.
III.

II.

IV.

II.
IV.

1

The average reduction time was 3| hours. The samples were placed : — 3 in

Class L, 1 in Class II., 4 in Class III., 3 in Class IV.

* In practice round figures only need be noted. To facilitate the calculation of
the average, i may be written as |. When the fermentation result is denoted by two
letters, this denotes an intermediate form.

The average of the four weekly tests places the milk in the class
which has to be reckoned with during the corresponding month.
The figure is rounded off to the nearest unit, fractions of | or over
being counted as 1. The grading of milk has been dealt with at
length because it furnishes just basis for payment according to
quality, the only really effective means available for improving
the quality of milk. If such a system were established there
would be some hope that the farmers would exert themselves to
ensure the clean treatment and proper cooling of their milk, just
as the more progressive of the Danish farmers have succeeded in
increasing the fat content of their milk since the system of pay-
ment according to " fat units " was adopted by most of the co-
operative dairies in Denmark. The two systems may easily be
combined by allowing a small increase in the price per fat unit for
first-class milk, and making a corresponding deduction in the case
of third-class milk. Double the amount should be deducted for
fourth-class milk. There would be no surer means of guarding
against butter and cheese defects, and there could be no better re-
commendation for the dairy products than the fact that they had
been made from milk sufficiently clean and fresh to be palatable

174 DAIRY BACTERIOLOGY

to the most critical. This reform would also have far-reaching
results contributing indirectly to the welfare of coming generations ;
for once the principles of hygiene have gained a footing in the cow-
shed they will gain admission everywhere else. Healthy and clean
cows, good milk, healthy children.

INDEX

ACID, acetic, 1, 16, 30, 128

butyric, 42, 135, 139

lactic, 30, 31, 32, 49, 128, 129
optical isomers, 32

propionic, 41, 139
Acidity of cream, determination of, 107

of milk, determination of, 161

of starter, determination of, 117

of whey, determination of, 131
Acids, fatty, 122, 134, 144
Actinomyces bovis, 29

chromogena, 126
Actinomycetes, 29
Actinomycosis, 70
Aeration of cream, 112

of milk, 78, 168

of starter milk, 116
Aerobic bacteria, 10
Aerogenes bacteria, 39, 153

mastitis, 151

After-fermentation of cheese, 139
Albumin, 81, 96, 97, 113, 130, 149
Alcohol test, 162
Alcoholic fermentation, 50, 102
Alizarol test, 162

Alkali -forming bacteria, 63, 65, 75
Amino acids, 4, 140
Ammonia formation in cheese, 134, 135,

136, 146, 147

Ammonium casemate, 119, 134, 135
Amylase, 13

Anaerobic bacteria, 10, 19, 42
Anthrax, bacterium of, 10, 71
Anti-toxins, 14, 97
Appetitost, 149
Aroma bacteria in butter, 108, 110, 116

in cheese, 142
Aspergillus repens, 87
Autoclave, 17
Average reduction time, 171

BACILLUS, 27, 29
Bacillus, abortus, 71

acidi lactici, 32

acidophilus, 38

anthracis, 10, 47

bifidm, 33, 101

botalinus, 48

bulgaricus. See " Thermobacterium
bulgaricum."

butyricus immobilis, 43

mobilis, 43

Chauvoei, 43

coli, 40, 45, 65

cyaneofuscus, 45, 154

Delbrucki. See " Thermobacterium
cereale"

firmitatis, 155

mesentericus, 47

Bacillus, mycoides, 47, 67, 71, 101
putrificas, 47
rudensis, 154
subtilis, 46, 47
tetani, 48
typhosus. See " Bacterium typho-

.w/w."
Bacteria, 1, 27, 29

classification of, 26, 29

in air, 61

in butter, 20, 120

in cheese, 20, 137

in earth, 47, 61

in manure, 61, 63, 70, 158, 163

in milk, 19, 20, 62, 64, 71, 69, 163,

166, 170
in salt, 126
in straw, 61
in udder, 61, 63

in water, 21, 40, 44, 68, 74, 75, 123
lactic acid, classification of, 32
pathogenic in milk, 69, 71, 97
staining of, 24, 25
Bactericidal substances, 55, 89, 113

of milk, 64, 97
Bacteriological methods, 16
Bacterium abortus, 71
acidi propionici, 41
bifidum, 33, 101
carotce, 74
casei a. See " Streptobacterium

casei"

casei y. See " Betabacterium breve"
casei 8. See " Betabacterium

longum."
casei e. See " Thermobacterium

helveticum."
casei fusci, 155

limburgensis, 146, 148
caucasicum, 35
cloacce, 40, 104
coli, 40; 45
cyaneofuscus, 45
enteriditis, 40
erythrogenes, 46

fluorescens liquefaciens, 44, 46, 74, 160
lactis, 32

acidi. See " Streptococcus

lactis.''''
aerogenes, 40
innocuum, 63, 75, 107
longi, 36
acidi, 32

minimum mammce, 69
pneumonice, 40
prodigiosum, 46
pyocyaneum, 26, 44
pyogenes, 69
sapolacticum, 75

176

INDEX

Bacterium syncyaneum, 26, 45

synxantum, 46

typhosum, 27, 40, 71

vulgare, 27, 46

Zopfii, 45, 75
Betabacterium, 33, 67

breve, 34

caucasicum, 34

longum, 34

Betacoccus, 33, 37, 38, 65
Biological characteristics, 15
Biorisation, 83
Bitter taste in butter, 125

in cheese, 129, 155

in milk, 75

Blood in milk, 69, 159
Blown cheese, 139, 151

milk, 164

Blue milk bacteria, 73
Boiling test, 162
Bottling of milk, 91

stoppers for. 91
Brie cheese, 133
Buchner, 12
Budde's process, 89
Burnt taste, 79, 118, 125
Burri's Indian ink method, 25

tubes, 19, 65
Butter, aroma of, 108, 110, 116

defects, original, 124
secondary, 125

flora, abnormal, 124
normal, 120

making of, 106

milk, 98, 118

rancidity of, 51, 52; 121, 127
Butyric acid bacteria, 42, 139, 149, 152, 160

CABBAGE taste in cheese, 156

Calcium lactate, fermentation of, 41, 42,

135

Camembert cheese, 133, 136, 147
Capsule formation by bacteria, 7
Carbol fuchsine, 25
Carbon, sources of, 9
Carbonic acid, 122, 151, 156
Casein, 13, 14, 97, 99, 129, 131
Catalase, 14, 33, 90

test, 159
Cell walls, 5
Cells, 4

Cellulose, fermentation of, 43
Certified milk, 60, 62
Cheddar cheese, 43, 130, 141, 154
Cheese curd, consistency of, 131, 132

making of, 129

defects, 73, 129, 151

flora, abnormal, 151
normal, 135

in milk, 72, 164

poisoning by, 156

rennet, 129, 166

ripening of, 128, 132
course of, 132
depth of, 134
extent of, 134

salt in, 131, 132, 139
Cheesy milk, 72, 164

sour butter, 126

Chlamydospores, 4
Cholera bacteria, 28, 71, 82
Chymosin, 14, 81, 129, 138, 143, 165
Cilia, 26, 46
Cladosporium, 51
butyri, 51, 123
herbarum, 51, 56, 155
Classification of bacteria, 28

of lactic acid bacteria, 32
Cleaning, 54, 58, 59

separator, 90
Clostridium, 3, 4
Coagulating milk, 72
Coagulation of milk proteins, 81, 129
Cocci, peptonising, 45
Coccus, 27
Coccus liquefaciens. See " Streptococcus

liquefaciens."

Coli bacteria, 39, 45, 65, 125, 152, 153
Colonies, 20, 23
Colostrum, 69, 71, 151, 162
Colour defects in butter, 126
in cheese, 45, 153
in milk, 73

producing organisms, 44, 45, 46, 73
Condensing of milk, 86
Conidia, 4

Cooked taste in butter, 125
in cream, 81
in milk, 8l
Cooling of cream, 81, 111

of milk, 65, 77, 91, 95
Copper in cheese, 153

in milk, 75

Counts, bacterial, of air, 61
of butter, 20
of cheese, 20
of earth, 61
of manure, 61
of milk, 19, 20, 62, 64, 170
of straw, 61
of water, 21

Cow-sheds, floor covering of, 58, 59
Cows, diarrhoea of, 57, 72, 151, 158, 163

diseases of, 57, 69, 71
Cracks in cheese, 73, 153
Cream, cooling of, 81, 111
fat in, 110

homogenisation of, 80
pasteurisation of, 109, 111, 120
pathogenic germs in, 109
souring of, 36, 106
whipped, 67

Cultivation methods, 16, 17, 22
Cultures, lactic acid, 108, 117

dry, 109
mould, 145, 148
pure, 21, 24, 145, 148
single cell, 21, 24
Curd, acidity of, 131

consistency of, 131, 132
making of, 131

DANISH dairy cheese, 133, 136, 142

Swiss cheese, 140
Defects of butter, 122, 124

of cheese, 151

of milk, 68, 71
Diarrhoea of cows, 57, 72, 151, 158, 163

INDEX

177

Diastase, 13, 47
Diet of cows, 57, 68
Dietetic preparations, 100
Diffusion slices, 58
Diphtheria bacteria, 29, 71, 82
Dirt in milk, 60, 61, 90, 157
Disease germs in milk, 29, 71, 82
Diseases of cows, 69, 71
Disinfecting, 55
Dry cultures, 109
Dutch cheese, 36, 73
Dysentery bacteria, 82

EARTHY smell in butter, 126

Edam cheese, 36, 73, 133, 136, 141, 154

Emmental cheese, 34, 41, 42, 131, 133,

136, 142, 152

Endoenzymes, 12, 137, 142
Enrichment method, 22
Enzymes, 12, 15, 97, 129, 137, 140, 142, 167
Erepsin, 14, 137, 140, 142
Exoenzymes, 12, 140, 142
Eyes in cheese, abnormal, 139
normal, 41, 42, 138, 152

FAT butter, hydrolysis of, 121, 134

oxidation of, 121

Fat hydrolysis by microorganisms, 44
in butter, J.22
in cheese, 134, 144

in cream, 110

in cheese, 134

in milk, 96

Feeding of cows, 57, 68
Fermentation processes, 11, 30, 41, 42,

44, 106, 119, 128, 139
Fermenting milk, 72

test, 163, 170

combined with reductase test.

166

Ferments, 12
Fishy taste in butter, 126
Flagellae, 26, 46
Flash pasteurisation, 82
Flies, 60

Floor of cow-sheds, 58, 59
Fluorescent bacteria, 44, 65
Fodder, influence of, on milk, 68
Foot and mouth disease, 82
Freudenreich flask, 17
Fuchsine, 25

GALACTASE, 90, 129
Gassy milk, 72
Gentian violet, 25
Gioddu, 100
Gldsler, 139
Globulin, 130
Gorgonzola cheese, 143
Gouda cheese, 141, 155
Grading of milk, 157, 170
Gram's staining, 25
Grass taste in butter, 125

in milk, 74

Green Alpine cheese, 133, 136, 146, 149
Growth and reproduction, 2
Gruyere cheese, 34, 136

HANSEN'S distinction of yeasts, 49

single cell culture, 24
D.B.

Harrach cheese, 147

Hay bacteria, 46, 83

Hard cheeses, 132, 149

Harz cheese, 150

Heat action on bacteria, 19, 80

on enzymes, 12
Holder pasteurisation, 82, 93
Holes in cheese, abnormal, 139

normal, 41, 42, 138, 152
Homogenisation, 79, 80, 85
Household pasteurisation apparatus, 83,

97
Hydrogen peroxide, 14, 89, 90, 93, 159

ICE milk, 60, 78, 94

Immunity, 15

Incubation period, 11

Incubator, 17, 18

Infants, nutrition of, 88, 96, 97

Infection, 56

by milk, 69, 71, 82
Invertase, 13
Involution forms, 16
Iron in cheese, 153

in milk, 98
Isigny butter, 110

JORGE NSEN'S moist chamber, 23

KAELDER milk, 102
Kefir, 51, 100, 103
Knapost cheese, 149
Knopist cheese, 147
Koch, 17
Kumys, 100, 102

LACTASE, 13

Lactic acid, 30, 31, 32, 49, 128, 129

optical isomers, 32

bacteria, pseudo, 30, 38, 39, 118
true, 32, 114, 160, 171

fermentation in cheese, 129, 153
psuedo, 30, 163
true, 30, 106, 114, 128

micrococci, 38

rod bacteria, 33

streptococci, 35
Lactobacillus, 33
Lactose, 96, 135
Leben, 100, 102
Lecithin, 98, 126
Leucocyte test, Tromsdorff's, 158
Leucocytes in milk, 93, 158, 167
Leuconostoc, 7, 38
Limburg cheese, 133, 136, 146
Lipase, 13
Liquefying of cheese, 153

organisms, 14, 153
Lofotrich bacteria, 26
Long milk, 7, 8, 36, 72, 100

whey, 72
Low temperature pasteurisation, 82, 93

MALT taste in butter, 118

Maltase, 13

Manure, bacteria in, 61, 63, 70, 158, 163

preservation of, 119
Mastitis, 69, 72, 151, 159
Mazun, 100
Meat poisoning, 46

12

178

INDEX

Medicines passing into milk, 68
MetaUic taste in butter, 125

in milk, 75
Methyl violet, 25
Methylene blue, 25, 166
Microbacterium, 33, 38
Micrococcus, 27, 65
Micrococcus candicans, 160

casei amari. See " Streptococcus
liquefaciens"

casei liquefaciens. See " Telracoccus
liquefaciens."

chromoflavus, 155

flavus, 155

Freudenreichii, 72

rubri casei, 155
Microorganisms, 1

cultivation of, 16, 17, 22

mode of growth, 3

multicellular, 1

size of, 2

spore formation by, 3

unicellular, 1
Midday milk, 79
Milk, acidity of, 161

aeration of, 78, 168

bacterial counts of, 62, 170

bottling, 90

certified, 60, 62

coloured, 73

coagulation of, 72, 87

condensing of, 86

cooling of, 65, 77, 81, 91, 94, 95

defects, primary, 68
secondary, 71

dirt in, 60, 61, 90, 157

dried, 88

fermented, 100

fermenting, 72

flora, abnormal, 68
normal, 63

from cows long in milk, 69, 151, 162

gassy, 72, 164

grading of, 157, 170

homogenisation of, 79, 80, 85

keeping of, 65, 77

pails, 60

pasteurisation of, 70, 81, 93, 94, 162

pathogenic germs in, 69, 71, 80, 97

poisoning of, by bacteria, 83

powder, 88

preparations, 100

preservation of, 77

ropy, 100

separated. See " Separated milk.''

skim. See " Separated milk."

slimy, 100

souring of, 66, 67

stale, 67

sterilised, 79

strainers, 60

town supplies, 90, 95, 170

sugar, 96, 135

unclean, sour- tasting, 73
Milking, 59

machines, 59
Moist chamber, 23
Monilia nigra, 51, 155

species of, 51

Monotrich bacteria, 26

Morphology, 4

Mould spots on butter, 125

on cheese, 155

on walls, 56
Moulds, 2, 49, 51, 118, 121, 122, 125, 134,

143, 145, 148, 155
Mouldy cheeses, 143, 146
Mucin, 5

Mucor, species of, 4, 56
Mucorinse, 29
Mycelium, 2

Mycoderma, species of, 50, 104
Mycoderma cerevisice, 50
Mycomycetse, 29

NIELSEN'S steriliser, 80
Nitrates, reduction of, 33
Nitrogen, sources of, 9
Norwegian Gammelost, 133, 149

Kaeldermaelk, 102
Nucleus, 5

Nutrient media, 7, 17, 19
Nutrition of animals, 96

of bacteria, 7, 15, 16, 17

of infants, 88, 96, 97

OIDIUM, 4, 51

Oidium aurantiacum, 155

camemberti, 147

lactis, 6, 51, 67, 107, 122, 123, 125,

144, 147, 148, 149, 150
Oily taste in butter, 118, 125
Optimum temperature, 12
Orla-Jensen's household pasteuriser, 97
Oxidases, 14, 167
Oxygen, sources of, 9

PARACASEIN, 13, 129

Parmesan cheese, 140, 153

Pasteur, 32

Pasteurising of cream, 81, 109, 111, 120

of milk, 81, 93, 162

of milk for starter, 113

of separated milk, 70, 84

of water, 1 22

Pathogenic germs in butter and cream,
109

in milk, 69, 71, 80, 97
Payment according to quality, 61, 173
Penicillium, 52, 144, 145, 148

brevicaule, 53, 74, 156

camemberti, 53, 147

candidum, 53, 147

casei, 53, 155

(jlaucum, 2, 6, 52, 122, 123

roqueforti, 52, 145, 147
Pectins, 43, 57
Pepsin, 14, 129
Peptones, 13

Peptonisation, 13, 129, 133, 137, 140
Perhydrase milk, 90
Periirich bacteria, 26
Petri dishes, 19
Petruschky flasks, 17, 20
Plectridium, 3, 4

foetidum, 47, 147

INDEX

179

Poisoning by cheese, 156

by meat, 46
Poisons in milk, 68, 83
Potato bacteria, 46, 83
Preservation of butter, 120

of milk, 77, 89, 128

of stable manure, 119
Preservatives, 89, 120
Pressler, 138
Propionic acid bacteria, 41

fermentation, 41, 42
Proteins, hydrolysis of, 44, 48, 81, 129,

133, 136, 143, 146
Proteolytic enzymes, 13, 129, 133, 137,

140, 143

Proteus bacteria, 40, 45, 65, 104, 125
Protoplasm, 4
Pultost, 149
Pure cultures, 21
Pus in milk, 69
Putrefaction, 11, 40, 143, 147
Putrefactive bacteria, 44, 125

QUALITY, payment according to. 61. 173

RANCIDITY, 51, 52, 121

Ray fungi, 29

Red milk bacteria, 73

Reductase and fermenting test, 95, 170

test, 166, 170
Reductases, 14

in milk, 167
Rennet, 14, 81, 129, 138, 143, 165

and fermentation test, 166

curd cheeses, hard, 135
soft, 145

test, Marshall's, 161

Schaffer's, 81, 161
Reproduction, 3
Retting, 43

Ripening of cheese, 128, 132, 134
Romadour cheese, 146
Ropy milk, 7, 8, 36, 72

whey, 36, 73

Roquefort cheese, 134, 136, 143
Rosolic acid test, 162
Rotting, 11, 57
Russian Steppe cheese, 140
Rust spots in cheese, 155

SACCHAROMYCETES, 49, 65, 87
Salt, bacteria on, 126

milk, 68

stones, 140
Salting of butter, 120

of cheese, 131, 132, 139
Saltpetre, 152
Salt stones in cheese, 140
Sarcina, 27, 38, 160
Scarlatina, 71
Scarlet fever, 71

Schabzeiger cheese, 133, 136, 146, 149
Schotte, 86, 137
Sediment in milk, 157, 159
Self-souring of cream, 106

of milk, 64, 67, 72
Separated milk, 84, 89, 118, 149
Separator, cleaning, 91
Single cell cultures, 24

Skim milk. See Separated Milk.
Skuta, 106

Slime formation, 38, 42
Slimy milk, 72

whey, 36, 73
Smeared cheeses, 146
Smell and taste of milk, 72, 172
Soapy taste, 75
Soft cheeses, 132, 145, 150
Soil bacteria, 11, 38, 43, 44, 47

formation, 11, 43
Sour cheesy smell in butter, 126
Sour milk cheeses, hard, 1 49

soft, 150
Souring

defects, 118

of cream, 36, 106

of milk, 66, 67, 112

of separated milk, 112, 118

of starter milk, 112
Soxhlet's nursery pasteuriser, 83
Species, 15
Spirillum, 28, 29
Spongy cheese, 139, 151

milk, 165
Sporangia, 4
Spores, 3, 4, 49, 80, 83

endogenous, 3

exogenous, 3

germination of, 3
Stab cultures, 10, 21, 38
Stabbing of cheese, 144
Stable floors, 58, 59

smell in butter, 125

in milk, 74

Staining of bacteria, 24, 25
Staphylococcus, 27
Starter for cream, 108, 112, 113
Sterilised milk, 79, 167
Sterilising, 19, 79, 167
Stilton cheese, 143
Storch's reaction, 89, 93
Streak cultures, 21
Streptobacterium, 33, 34, 35, 67

casei, 34

plantarum, 34
Streptococci, 27, 33, 35, 63, 65, 107, 120,

141, 159
Streptococcus acidi lactici, 32

agalactice. See " Streptococcus mas-
titidis."

brassicce, 38

casei amari. See " Streptococcus
liquefaciens. ' '

cremoris, 35, 36, 73, 102, 112, 114,
115, 120

fcecium, 36, 65, 85, 102

glycerinaceus, 85, 102

hollandicus, 36

lacticus. See ''Streptococcus cre-
moris"

lactis, 27, 36, 37, 65, 66, 102, 114, 118,
143

liquefaciens, 37, 72, 75, 140, 153, 155

mastitidis, 36, 69

pyogenes, 36

thermophilus, 35, 36, 65, 85, 114, 138
Streptococci, heat-resisting, 85

pathogenic, 159

180

INDEX

Spirillum, 28

Storch's reaction, 14, 89, 93
Stribolt's anaerobic cultivation, 22
Sugar beet slices, 58

in condensed milk, 86
Sugars, fermentation of, 23
Sulphur bacteria, 29, 75
Swedish manor farm cheese, 140
Swine erysipelas bacterium, 10
Swiss cheese making, 86
Swiss skim milk cheese, 133, 136
Symbiosis, 102, 103, 109, 146

TALLOWY taste in butter, 118, 122, 125
in cheese, 155
in milk, 75

Taste and smell of milk, 157
Tears in cheese, 139
Tetanus bacillus, 10
Tetracocci, 33, 38, 72, 140, 142
Tetracoccus liquefaciens, 39, 140, 143, 146
Thermobacterium, 33, 67, 85

bulgaricum, 30, 31, 34, 101 116

cereale, 33

helveticum, 34, 50, 116, 143

jugurt, 34

Thermophilic bacteria, 47, 83, 85
Thermostat, 18
Thick butter, 124
Thread bacteria, 29
Tilsit cheese, 140
Torula amara, 50, 75, 155
Torulce, 49, 50, 67, 123, 126, 153
Town's milk supplies, 00, 95, 170
Toxins, 14

Train oil taste in butter, 126
Trommsdorf's leucocyte test, 158
True lactic acid bacteria, 30, 72
Trypsin, 14, 129, 140, 142
Tubercle bacteria, 70, 82, 97, 109
Tuberculosis, 69, 71, 82

Turnip taste in butter, 125

in milk, 74

tops, 57

Typhoid bacteria, 27, 40, 71, 82
Tyrothrix bacteria, 135, 140

UDDER, bacteria in, 61, 63

inflammation of, 69

tuberculosis, 69
Ultra-violet rays, 90
Unclean taste in butter, 125

in cheese, 155

in milk, 73
Urda, 106

VACUOLES, 5
Variability, 15
Variants, 15
Variety, 15
Vibrio, 28, 29
Vitamines, 98

Volatile acids in butter, 121
in cheese, 136

WASHING of butter, 111, 118, 124
Water, bacteria of, 21, 40, 44, 74, 76, 123
Wet butter, 124
Whey, 86, 119, 131

champagne, 106 .

long, ropy or slimy, 36, 73

sparkling, 106

YEAST, 2, 3, 49, 67, 105, 118, 121, 123

taste in butter, 125
Yeasts, fermentation of sugars by, 50
Yellow milk, bacteria of, 46
Yoghurt, 100, 138

ZOOGLO3A, 7

Zygosaccharomycetes, 87
Zymase, 12

THE WHITEFRIARS PRESS, LTD., LONDON AND TONBRIDGE.

: -

ISJJOOK IS DUE ON THE LAST DATE
'ED BELOW

T.C 8861