Skip to main content

Full text of "Reports of the Committee upon anaerobic bacteria and infections. Report on the anaerobic infections of wounds and the bacteriological and serological problems arising therefrom"

See other formats


Cornell University Library 
RD 131.G78 



Reports of the Committee upon anaerobic 




3 1924 000 323 687 



Special Report Series, No. 39. 



^-31 NATIONAL HEALTH INSURANCE 



MEDICAL RESEARCH 
COMMITTEE 



Reports of the Committee upon Anaerobic 
Bacteria and Infections 



Report on the Anaerobic Infections ofWounds and 
the Bacteriological and Serological Problems 
arisii J therefrom 




LONDON : 

PUBLISHED BY HIS MAJESTY'S STATIONERY OFFICE. 

To be purchased through any Bookseller or directly from 
H.M. STATIONERY OFFICE at the following addresses : 

Imperial House, Kingsway, London, W.C. 2, and 
28 Abingdon Street, London, S.W. 1 ; 
37 Peter Street, Manchester ; 
1 St. Andrew's Crescent, Cardiff ; 
23 Forth Street, Edinburgh ; 
or from E. PONSONBY, Ltd., 116 Grafton Street, Dublin. 

1919. 
Price 6s. ')iet. 



ptibj? OTouncil 
MEDICAL RESEARCH COUNCIL 

[Formerly Uedkal Besearch Committee, National Health Insurance). 

LIST OF PUBLICATIONS 

November 1920. 

The following publications relating to the work of the Medical Kesearch Council 
can be purchased through any bookseller, or directly from H.M. Stationery Office 
at the following addresses: Imperial House, Kingsway, London, W.O. ^, and 
28 Abingdon Street, London, S.W. 1 ; 37 Peter Street, Manchester ; 1 St. An- 
drew's Crescent, Cardiff; 23 Forth Street, Edinburgh ; or from E. Ponsonby, Ltd., 
116 Grafton Street, Dublin. 

Annual Ebpoets op the Medical EeSeaech Committee: — 

No. 1. 1914-1915. [Cd. 8101.J Price 3d., post free 4^d. 
No. 2. 1915-1916. [Cd. 8399.] Price S^d., post free 5i. 
No. 3. 1916-1917. [Cd. 8825.] Price Qd., post free Id. 
No. 4. 1917-1918. [Cd. 8981.] Price U., post free Qd. 
No. 5. 1918-1919. [Cmd. 412] Price 6d, post free 8d. 



Special Report Series 

Tuberculosis : 

No. 1. First Report of the Special Investigation Committee upon the Incidence 

of Phthisis in relation to Occupations. — The Boot and Shoe Trade. Price 

3d., post free 4d. 
No. 18. An Investigation into the Epidemiology of Phthisis Pulmonalis in 

Great Britain and Ireland. Parts I and II. Price Is. 3d., post free Is. bd. 
No. 22. An Inquiry into the Prevalence and Aetiology of Tuberculosis among 

Industrial Workers, with special reference to Female Munition Workers. 

Price Is. 6d, post free Is. 7\d. 
No. 33. Pulmonary Tuberculosis : Mortality after Sanatorium Treatment. 

Price 2s., post free 2s. '2i^d. 
No. 46. An Investigation into the Epidemiology of Phthisis in Great Britain 

and Ireland, Part III. Price 2s. 6d., post free 2s. 8d. 

Cerebro-spinal Fever : 

No. 2. Eeport of the Special Advisory Committee upon Bacteriological Studies 

of Cerebro-spinal Fever during the Epidemic of 1915. Out of print. 

No. 3. Bacteriological Studies in the Pathology and Preventive Control of 

Cerebro-spinal Fever among the Forces during 1915-1916. Price Is. &A, 

post free Is. S\d. 

No. 17. I. A Eeport upon the Seasonal Outbreak. of Cerebro-spinal Fever in the 

Navy at Portsmouth, 1916-1917. 

II. The Treatment of Cerebro-spinal Meningitis by Antimeningococcus 

Serum at the Royal Naval Hospital, Haslar, 1915-16-17. Price 2s. 6d., 

post free 2s. 9d. 

No. 50. Cerebro-spinal Fever. Studies in the Bacteriology, Preventive Control, 

and Specific Treatment of Cerebro-spinal Fever among the Military Forces' 

1915-19. Price 4s., post free 4s. 3|d. ' 

Dysentery : 

Reports upon Investigations in the United Kingdom of Dysentery Cases received 

froin the Eastern Mediterranean. 
No. 4. I. Amoebic Dysentery and the Protozoological Investigation of Cases and 

Carriers. Price Is., post free Is. 2d. 
No. 5. II. Report upon 878 Cases of Bacillary Enteritis. Price 4s 6d post 

free 4s. 8d. ' ^ 



3944 



B 



NATIONAL HEALTH INSURANCE 



MEDICAL RESEARCH COMMITTEE 



REPORTS 



OF THE 



COMMITTEE UPON ANAEKOBIO BACTEEIA 
AND INFECTIONS 



REPORT ON THE ANAEROBIC INFECTIONS 
OF WOUNDS AND THE BACTERIOLOGICAL 
AND SEROLOGICAL PROBLEMS ARISING 
THEREFROM 



'Nb, l&IX^ 
Medical Research Committee 

{National Health Insurance.) 

The Hon. Waldorf Astob, M.P, {Chairman). 

The Eight Hon. Christopher Addison, M.D., M.P. 

The Viscount Gosohen, C.B.E. {Treasurer). 

0. J. Bond, O.M.G., P.R.C.S. (Hon. Colonel). 

Professor William Bulloch, M.D., P.E.S. 

Henry Head, M.D., P.E.C.P., F.E.S. 

Professor P. G. Hopkins, D.Sc, P.E.O.P., F.E.S. 

Major-General Sir William Leishman, K.C.M.G., C.B., F.E.S. 

Professor Noel Paton, M.D., F.B.S. 

Sir Walter M. Fletcher, K.B.E., M.D., P.E.S. {Secretary). 

15, Buckingham Street, 

Strand, W.C. 2. 




The Conim.ittee upon Anaerobic Bacteria and 
Infections. 

With a view to the better co-ordination of inquiries into the 
characters of anaerobic organisms, with special reference to the 
bacteriology of anaerobic wound infections, the Medical Eesearch 
Committee invited the following to serve as a special investigation 
Committee for this purpose : 

Professor William Bulloch, M.D., P.E.S. {Chairman). 

W. B. Bullock, M.D. 

S. E. Douglas, M.E.C.S., L.E.C.P. (Captain I.M.S. retd.). 

Herbert Henry, M.D. 

James McIntosh, M.D. 

E. A. O'Brien, M.D. 

Miss Muriel Eobertson, M.A. {Secretary). 

C. a L. Wolf, M.D. 

{with corresponding members). 



.INTRODUCTION 

Among the wound complications which confronted all the belli- 
gerents in 1914, none was more serious than ' gas gangrene ', whether 
considered from the point of view of incidence or of mortality. 
Although this remarkable condition was known and described in 
pre-antiseptic days, it had not been a disease of any consequence 
in any of the wars of the twentieth century. The outbreak of the 
Great War, however, brought it early into prominence, and it was 
at once recognized as a toxic-infective condition. The unusually 
high incidence on the Western Front, where fighting took place upon 
a long cultivated soil contaminated with human and animal excre- 
ment, led to the belief that the infective agents were soil bacteria, 
and it was early estabhshed in all the countries interested that these 
rnicro-organisms were of the anaerobic type. In spite of the accumula- 
tion of much knowledge on the subject of anaerobes, a study of 
bacteriological literature before the war shows clearly that' the 
descriptions published by different writers were for the most part 
widely divergent, and agreed in only a few instances. Prom the 
frequency with which Bacillus welchii [B. perfringens) was found, 
apparently in pure culture, it was considered that this was the chief 
aetiological agent, and it was actually named ' the bacillus of gas 
gangrene '. 

A careful scrutiny of the discharges from wounds by more refined 

methods during 1916 and 1917 revealed the fact, however, that gas 

gangrene could not be regarded as an aetiological entity, but could 

be caused by a series of pathogenic anaerobes acting singly or in 

combination with each other or in association with certain well 

defined non-pathogenic anaerobes. By degrees Pasteur's Vibrion 

sepUqv£ was disentangled from the mass of anaerobes, and a number 

of other anaerobes, new to science, were discovered, among which 

may be specially mentioned the highly toxic B. oedematiens. The 

isolation of these bacteria was chiefly the work of MM. Weinberg 

and Seguin of the Pasteuf Institute in Paris, whose conclusions were 

confirmed and extended by English bacteriologists, notably Majo^ 

McNee, Captain Adrian Stokes, Captain Herbert Henry, Captain 

W. E. Bullock (now Gye), with the Forces Overseas, Dr. James 

Mcintosh, working at the London Hospital in the service of the 

Medical Eesearch Committee, and Miss Muriel Eobertson of the 

Lister Institute of Preventive Medicine. Major W. J. TuUoch in 

the laboratories of the Eoyal Army Medical College and the Lister 

Institute made at the same time unusually successful researches into 

the bacteriology of tetanus. Captain C. G. L. Wolf, working for the 

Medical Eesearch Committee, by the application of exact methods 

added important contributions to the biochemistry of anaerobes, first 

at Boulogne and later at Cambridge. In America, Major C. G. Bull 

and Dr. J. L. Stoddard added considerably to our knowledge of the 

bacteriology of anaerobes. 

(5012.) Wt, 47167. 226. 1250. 10/19. O.U.P. 

(3944.) Ps. 2208. Wt. 21317. 309. 1000. 11/20. O.U.P. 

A2 



IV 

The advances which were made were the result of applying the 
most refined and most difficult forms of bacteriological technique, 
and they depended also upon the early recognition of the fact that 
many, if not most, anaerobes have an inveterate tendency to live 
in close association with other anaerobes. This is an association 
which can only be recognized by the most careful and prolonged 
scrutiny of what appear to be pure cultures, under diversified and 
often complicated conditions. To the failure to reaUze this must be 
ascribed the fact that in the bacteriological literature of the Central 
Powers the knowledge of gas gangrene was and still remains (1919) 
in a state of hopeless confusion, and it cannot be doubted that 
it has been the researches of French and EngUsli bacteriologists 
that have cleared away the mystery of the aetiological agents pro- 
ducing gas gangrene. The outcome of this advance in knowledge 
led the French to test at an early date the prophylactic and curative 
effects of the serum of animals immunized with pure cultures of 
specific pathogenic anaerobes. Among the most successful workers 
in this field must again be mentioned Weinberg and Seguin, and 
Major C. G. Bull. They were the first to apply the serum treatment 
in man. In England the earliest attempts to produce sera were 
those of Mcintosh and O'Brien, the latter of whom, in association 
with Captain H. Henry and Captain W. E. Bullock, ultimately 
undertook, with the assistance of the Committee, the preparation of 
sera in the Wellcome Eesearch Laboratories on a large scale for the 
British Army. Throughout the course of these inquiries the British 
workers were greatly aided by French bacteriologists and surgeons, 
among whom may be specially mentioned Weinberg, Vaucher, 
Duval, and Chutro. It is a pleasant duty for the Medical Eesearch 
Committee to record their thanks for the help received from those 
French workers who with unusual generosity placed their unpub- 
hshed records in the hands of the British, and kept us in close touch 
with all the developments in their experience of gas gangrene. 

This wide co-operative effort towards gaining new knowledge and 
applying it to the production of curative sera was chiefly focussed 
m this country m the work of the Committee upon Anaerobes which 
the Medical Eesearch Committee had appointed in March 1917 in 
the hopes of bringing more closely together the various workers 
engaged The constitution of this special committee is given at 
the head of this Eeport. In the autumn of 1918 it seemed certain 
ti.at if the war continued there would be available large supplies of 

^w^l P^*f* ''''''^' ^°*^ ^^'^ PJ^eventive and curative use, in 
which the defence agamst the three chief organisms already men- 
tioned was to be combmed. Happily the armistice intervened to 
remove the urgent need for this, and the practical demonstration 
of the fruits of these long mvestigations must be expected now only 
in the infinitely rarer occasions of peace. ' 

It appeared to the Medical Eesearch Committee that every effort 

£7ii f T^' ^l^^^^f "^ ""^ ^l^"fy *^« ««ie^tific results^ whS 
had so far been obtamed The prolonged study of gas gangrene in 
all the chief countries had led, moreover, to the production of many 
writings both good and bad, and this gave further reason fo?S 
preparation of an authoritative and criti^l report uponThe wious 



parts of the subject by workers with first-hand knowledge. The 
special Committee upon Anaerobes were invited accordingly to 
complete their labours by preparing the Eeport which is now pre- 
sented. 

At first it was intended to deal completely with the aetiological, 
experimental, and sero-therapeutic aspects of the gas gangrene 
problem, and work on these lines had proceeded to a considerable 
extent, when in 1918 appeared the admirable, exhaustive monograph, 
entitled La Gangrene Gazeuse, by MM. Weinberg and Seguin. This 
necessitated a revision of the plans of the Committee, who deemed 
it inadvisable merely to traverse the same ground as the French 
writers. It was considered best to concentrate especially on a critical 
analysis of the experiences of the individual members of the Com- 
mittee, and to undertake, where necessary, fresh investigations to 
clear up the doubtful .points and to utilize the practical results 
gained by the use of the British-made sera in the treatment of gas 
gangrene. 

The Eeport as it stands is the conjoint work of the members of the 
Committee. For the sake of completeness, however, several sections 
were written by others who had special knowledge of certain aspects 
of the anaerobic and even aerobic microbic complications of wounds. 
Thus the Committee received a report on ' the chnical features and 
treatment of anaerobic infection of wounds and gas gangrene ' by 
Dr. John Fraser, of Edinburgh, who had an almost unique experience 
of the disease while serving in France. Dr. Alexander Fleming 
prepared a summary of his extensive researches on the ' aerobic 
infections of wounds ', while Major W. J. Tulloch gave the benefit 
of his wide experience on the bacteriology of tetanus. Dr. E. H. 
Kettle, of St. Mary's Hospital, undertook for the first time an exact 
inquiry into the finer pathological anatomical changes in human 
and experimental gas gangrene, which is published in the Appendix. 
Professor W. Bulloch compiled an extensive bibliography of the 
literature on anaerobic infections, and personally examined almost 
all the papers for the purposes of this report, as a help to future 
investigators who may not have been in touch with much of the 
medical war literature. 

The Medical Eesearch Committee believe that they are speaking 
for a wide circle of workers when they express their own cordial 
thanks to the members of the Anaerobe Committee, not only for 
their arduous individual work in the laboratory, but even more for 
the time and effort they have given so ungrudgingly to share their 
knowledge with others and to contribute in discussion to the common 
object in view. Thanks are due especially to Professor WilUam 
Bulloch, who from the beginning has acted as chairman of the Com- 
mittee, and to Miss Muriel Eobartson, who by kind permission of 
the governing body of the Lister Institute was able to undertake 
the difficult duties of secretary, which brought a heavy burden of 
correspondence with many workers at home and abroad. 

The work of the special Committee was necessarily done in close 
touch with the Army Medical Service, and here the Committee 
■would offer their grateful acknowledgements to the Director-General, 
A M S , and to many other officers, for the encouragement and direct 

a3 



VI 



assistance which made it possible to link effectively the work done in 
civiHan laboratories at home with that done in the armies and with 
the practical problems offered by Service needs and conditions. To 
the Army Medical Department the Committee are indebted for the 
supply of many data from Medical Case Sheets, which have been 
important for the study of the sero-therapy of gas gangrene ia the 
field. 

To the editors and pubhshers of the Journal of Pathology and 
Bacteriology the Committee are indebted for permission to reproduce 
a large number of the illustrations given in the plates. Other 
illustrations are taken from the report by Dr. James Mcintosh, 
previously published by the Committee (Special Report Series, 
No. 12). Figures 76, 77 and 78 are reproduced from the beautiful 
drawings made by Mr. Thornton Shiels from pieparationp by 
Dr. Mcintosh, and the Committee beheve that these are perhaps 
the most exact representations yet pubhshed of the infective con- 
ditions displayed. 

Medical Eesearch Committee 
15 Buckingham Street, 
Strand, W.C.2. 

September 1919. 



REPORT ON THE ANAEROBIC INFECTIONS OF 
WOUNDS AND THE BACTERIOLOGICAL AND 
SEROLOGICAL PROBLEMS ARISING THERE- 
FROM 



CONTENTS 

FAOE 

iNTEODtrCTIOIf ........... 6 

I. Thb Clinical Ebatubbs and Treatment of Anaerobic Infection oe 

Wounds and Gas Gangbenb . - . . . . 6 

1. General Clinical Considerations ....... 6 

(i) The Type of Wound . 6 

(ii) The General Condition of the Wounded Man .... 7 

(iii) ^he Nature of the Infecting Agent ..... 7 

(iv) The Vascular Supply ....... 7 

(v) The Regions and Tissues Affected ..... 7 

{vi) The Question of Interrupted Drainage fro-in the Wound . . 8 

(vii) The Clinical History of the Wound ..... 8 

2. Varieties of the Bisease ....... 9 

3. Symptoms and Signs of Gas Gangrene ...... 9 

(i) Symptoms ......••■ 9 

(ii) Physical Signs . . . . . ■ ■ . . 10 

(iii) The Group Type ........ 10 

(iv) The Massive Type . ■ . ■ ■ ■ ■ .10 

(v) The Fulminating Type . . ■ . ■ ■ . U 

4. Complications of Gas Gangrene . . . . . . .11 

5. The Diagnosis of Gas Gangrene . . . . . . .11 

6. The Treatment of Gas Gangrene 12 

(i) Preventive Treatment . , . . ■ • .12 

(ii) Treatment hy the Administration of Alkalies ... 12 

(iii) Surgical Treatment ........ 12 

(iv) Methods of Amputation in Oas Gangrene .... 14 

(v) General and Post-operative Treatment 15 

II. Incidence oe Gas Gangbenb 15 

III. BACTBEIOLOffY 1^ 

Introduction ......•■■• 1° 

1. Pathogenic Anaerobes liable to set up Conditions of Gas Gangrene 

with Symptoms of General Intoxication . . . .17 

(i) B. welohii . ^7 

(ii) Vibrion septique : with note on B. chauvoei ... 20 

(iii) B. oedematiens 24 

(iv) B. histolyticus 25 

(v) B. botulinus 26 

(vi) B. fallax 27 

2. B.tetani '. ^* 

(i) Bacteriological account ... ... 28 

(ii) Recent Experimental work on B. tetani . . . .29 

3. AnaerobestheActionofwhiohmaybeAnoillarytotheCondition 

of Gas Gangrene .....••• ^5 

(i) B. sporogenes ^" 

(ii) B. parasporogenes ^^ 

(iii) B. tertius *° 



(iv) B. ooohlearius 

(v) B. tetanomorphns 

(vi) B. aero-fetidua 

(vii) B bifermentans 

(viii) B. putrificus . 

(ix) B. sphenoides . 

(x) B. butyricuB . 
(xi) B. multifermentans tenalbus 

4. Classification of the Anaerobic Bacilli found in Wounds 

(i) Morphology 
(ii) Gvltural Characteristics 
(iii) Agglutination Beactions 
(iv) Toxin-antitoxin Beactions . 

5. Biochemistry .... 

(i) The Mechanism of Anaerobiosis 
' (ii) The Biochemistry of certain micro-organisms found in Wounds 

6. Experimental Gas Gangrene .... 

(i) Infection ...... 

(ii) The established disease .... 

7. The Aerobic Infections of War Wounds 

(i) B. mesenterious, B. subtilis, B. myooides groups 

(ii) B. aero-tetanoides 

(iii) B. aero-tertius 

(iv) Staphylococci . 

(v) Streptococci 

(vi) Diphtheroid Bacilli 
(vii) Goliform Bacilli 
(viii) B. pyocyaneuB 

(ix) B. proteus 

(x) Gram-negative cocci 

(xi) Miorococous tetragenus 

8. Influence of the Aerobic on the Anaerobic Infection of Wounds 

(i) Influence of Staphylococci and Streptococci on the growth oj 

B. welchii in Milk 
(ii) Influence of Aerobic Organisms on the growth of B. welchii in 

Serum neutralized vnth Acid, in Serum the antitryplic pouier 

of which has been neutralized with Trypsin . 
(iii) Influence of Aerobes on the growth of Anaerobes other than 

B. welchii 
(iv) Influence of B. welchii on the growth of Streptococcus and 

Staphylocooons 
(v) Effect of a Diphtheroid Bacillus on the growth of Streptococcus 

pyogenes 

9. B6sum6 of the Literature on the Bacteriology of Gas Gangrene, with 

an Account of the Incidence of the various Types of Patho 
genie Anaerobes ....... 

10. The Isolation of Anaerobic Organisms of Wounds in Pure Culture 
(i) Material suitable for examination .... 

(ii) MethodiS of Isolation ....... 

A. The separation by mechanical methods of individua 

organisms in a mixture ..... 

B. The selection by appropriate heating of spores coniained 

in a mixture . 
0. The use of selective media 

D. The separation of a Pathogenic Anaerobe by Animtd 

Experiment 

E. Summary 
^iii) The Preservation of Cultures 



PAGE 

40 
41 
42 
43 
43 
43 
44 
45 
45 
45 
46 
52 
55 
55 
56 
59 
65 
65 
67 
70 
72 
72 
72 
73 
73 
77 
80 
81 
82 
83 
83 
84 

85 



86 
86 
87 
87 



88 
95 
95 
95 

96 

98 



101 
103 



:5 



B. 



by B. 



welchii. 



vibrion 



Toxins 



IV. Serology . . ... 

1. Toxins and Antitoxins 

(i) Historical Intrcduction 

A. B. welchii 

B. Vibrion septique 

C. B. oedematiens 
(ii) The Preparation of Toxin . 

A. Toxin of B. welchii . 

! Vibrion septique . 
B. oedematiens 

2. The Properties of the Toxins elaborated 
septique, and B. oedematiens . 

(i) Physical features 
(ii) Biological characteristics . 

A. The toxin of B. welchii 

B. The toxin of Vibrion septique 

C. The toxin of B. oedematiens 

3. The Symptoms and Pathological Changes produced by these 
and the Estimation of their Minimum Lethal Dose in Animals 

(i) The Toxin of B. welchii 
' A. Mice 

B. Rabbits . 

C. Guinea Pigs 

D. Pigeons . 
(ii) The Toxin of Vibrion septique 
(iii) The Toxin of B. oedematiens 

4. The standardization of Antisera . 
(i) B. welchii antitoxic sera 

(ii) Vibrion septique antitoxic sera 
(iii) B. oedematiens o»<jif(ia;i6sero 

5. Preparation and Standardization of Anti-gas 
(i) Introduction .... 

(ii) Preparation of Antitoxins 

A. B. welchii 

B. Vibrion septique 

C. B. oedematiens 
Methods of estimating th^ value of Ai 

A. Antitoxin of B. welchii 

B. Antitoxin of Vibrion septique 

C. Antitoxin of B. oedematiens 
The Serum Therapy of Gas Gangrene . 

(i) The Collection of Data 
(ii) The Serum employed 
(iii) Analysis of Records . 

A. Deaths 

B. Recoveries 
(iv) Discussion 

(v) Summary 
(vi) TMe of Cases . 



(iii) 



gas-gangrene Sera 



6. 



BiBLlOGBAPHY 

Addendum 



Appendix ..... 

The Histopathology of Ga-, Gangrene 

Dbscbiption of Pla.i'ES .... 



gangrene Sera 



PAGE 

104 
104 
104 
104 
105 
105 
105 
105 
108 
108 

109 
109 
109 
109 
111 
lU 

III 

111 

111 

112 

113 

113 

113 

114 

115 

115 

116 

116 

117 

117 

118 

118 

119 

119 

120 

120 

120 

120 

121 

121 

121 

122 

123 

126 

130 

131 

131 

146 
172 

173 
173 

180 



INTEODUCTION 

The serious consequences which may result from the multiplication 
of anaerobic bacteria in wounds have furnished a medical problem 
in the present war which, both because of its gravity and of its 
novelty, has been studied by cHnicians and bacteriologists of most 
of the belligerent nations. 

The following pages summarize the present state of knowledge of 
wound anaerobes, of the clinical conditions which they may cause, 
and of experience in the preparation and the use of antisera. 

The advances chronicled are due to many investigators, who, by 
means of improvements in technique which have made it easier to 
obtain pure cultures, have studied the large amount of material 
which the war has provided. 



I. THE CLINICAL PBATUEBS AND TEEATMBNT OF 
ANAEKOBIC INFECTION OP WOUNDS AND GAS 
GANGEENE 

By John Peasbb, M.C, M.D., P.E.C.S. Ed. 

Anabeobio infection with its sequel of gas gangrene has been 
a common and an extremely serious complication of infected war 
wounds. 

When the present war began our knowledge of the subject was 
limited. The condition had been noted as occurring in previous 
campaigns, and on rare pccasions it had been met with in civil 
practice, but we possessed httle trustworthy information regarding 
the pathology and the treatment of the condition. It has been 
through the medium of expensive and painful experience that the 
knowledge which we now possess has been gained. 

In the subsequent remarks the anaerobic infection of wounds is 
dealt with from the point of view of clinical features and treatment. 

1. General Clinical Considerations. 
(i) The Type of Wound. 

There are certain types of wound which are especially Uable to 
anaerobic infection and these are wounds which illustrate one or 
both of two conditions, imperfect drainage and extensive devitaliza- 
tion and death of tissue with extravasation of blood. 

The first mentioned condition is typically illustrated in the deeply 
penetrating wounds with a small point of entrance, widespread 
destruction of the more friable muscles and possibly retention of the 
infected projectile at the base of the wound. Such wounds are 
usually produced by fragments of bombs and grenades, by shrapnel 
bullets, and by small fragments of H.E. shell. The projectile is 
generally traveUing at a comparatively low velocity and hence the 
tendency for the missile to be retained. 



The second type of wound associated with tissue death in the 
part affected is usually caused by irregular fragments of H.E.. shell. 
These fragments are generally travelling at such high velocity that 
there is widespread tissue destruction not only in the course of the 
missile but also in a considerable zone of tissue outside the 
actual track. 

(ii) The General Condition of the Wounded Man. 

It is certain that the general condition of the wounded man has 
an important bearing on the development of the infection. It has 
been observed that gas gangrene is more prevalent among fatigued 
troops, who have undergone long spells in the trenches, than it is 
among fresh soldiers who have gone direct into action. In a similar 
way, men who have suffered from severe haemorrhage and shock 
subsequent to the wound are specially liable to the development of 
this compUcation. 

(iii) The Nature of the Infecting Agent. 

It is essentially true that the more irregular the projectile fragment, 
■the more likely is anaerobic infection to follow. There are certain 
definite reasons for this statement. (1) An irregular fragment of 
shell produces widespread devitahzation and tissue destruction. 
This is well seen when such a wound is studied in comparison with 
that produced by a rifle bullet. A short range rifle bullet on the 
other hand may produce as much destruction as a large irregular 
shell fragment. (2) The more irregular the projectile fragment 
the more likely is it to carry into the wound pieces of infected 
clothing. (3) As the majority of H.E. shells burst on percussion 
with the ground, irregular fragments carry with them into the 
body of the wounded man portions of soil which in highly fertilized 
districts may be heavily charged with anaerobes. 

(iv) The Vascular Supply. 

A factor of extreme importance is the question of the blood 
supply to the part. A dirninished supply of blood to the wounded 
area predisposes the part to the development of an anaerobic infec- 
tion. A variety of causes may lead to the interference with the 
vascularity : the severe haemorrhage which is so often the con- 
comitant of war wounds results in a general anaemia : the nature 
of the wound is often such that a tourniquet has been applied and 
the distal blood supply has thus been arrested : in a more local 
sense the contusion and tissue destruction which has resulted from 
the wound leads to a thrombosis in the blood vessels for some 
extent around the wound. At any rate, in whatever way it may 
l^ave been caused, an impoverished blood supply to the part is 
a potent factor in favouring the development of an anaerobic 
infection. 

(v) The Regions and Tissues Affected. 

Anaerobic infection spreads with great rapidity in muscle and 
more slowly in areolar tissues. It is interesting to observe that all 



muscles are not equally liable to the infection ; for example, it is 
comparatively rarely that one finds any widespread infection of the 
muscles of the back. The larger and coarser-fibred muscles, e.g. 
the glutaei appear to be the most liable to the infection. It is 
possible that this distinction really depends upon the blood supply. 
In scalp wounds uncomplicated by compound fracture^ of the skull 
and in wounds of the face, gas gangrene is a rare complication. 

The serous cavities of the body would appear to offer considerable 
resistance to anaerobic infection. The disease in these situations 
progresses as a rule more slowly and is naturally hmited in extent 
to the region of the infected membrane. Although the infection 
of a serous membrane is thus for anatomical reasons a localized 
infection, yet the absorption of toxine from such a focus may yield 
a progressive intoxication which either by itself or in conjunction 
with an ensuing anaerobic septicaemia is fatal to the individual. 
This is particularly true of infection of the pleura. 

(vi) The Question of Interrupted Drainage from the Wound. 

Any factor which interferes with efficient drainage from the 
wound predisposes to the anaerobic infection. This fact may be 
illustrated by the type of wound. The deep narrow punctured 
wound is obviously liable to the infection. A slit-like skin wound 
may contract, effectively blocking the escape of discharge and at 
a deeper level there may be considerable retention of infected 
material. In the same way a foreign body may prevent efficient 
drainage, and in the early days of the war there were numerous 
instances in which packing of the wound or insertion of antiseptic 
pastes acted in a similar way with disastrous results. 

(vii) The Clinical History of the Wound. 

The period of time which elapses before the wound shows signs 
of infection varies over a wide range. Changes which indicate an 
anaerobic infection may be apparent within an hour of the injury 
being sustained ; on the other hand, several days may elapse before 
distinctive features appear. The changes in the wound which are 
associated with the infection may be summarized as follows. 

There is a scanty f oul-smeUing discharge composed of broken-down 
blood clot ; it is of a brownish colour, and mixed with it there are 
hubbies of gas. The odour of the discharge appears to depend on 
the type of infecting organism, and on the state of the infection : 
in the early stages a characteristic smell is absent, later it becomes 
foul and acrid. The skin around the wound assumes a faint 
purphsh colour, and there is a local swelling due to the accumulation 
of brown serous fluid in the subcutaneous tissues. There is a similar 
infiltration throughout the intermuscular septa and connective tissue 
planes. Throughout the precincts of the wound there is an infiltra- 
tion of gas and it extends to further hmits along certain lines— 
the subcutaneous tissues, the perivascular tissues, and the planes 
of intermuscular connective tissue. 

Up to this point in the clinical history of the wound, the muscle 
IS not mfected. When it does become infected, the nomenclature 



9 

must be alteted ; it is no longer an anaerobic infection of the wound 
but the condition of gas gangrene. The muscle is infected in one 
of two possible ways : 

1. The muscle (usually an injured one) is gradually invaded by 
the organisms, which starting from the wound make their way 
towards the extremities of -the muscle until it is totally destroyed. 
The blood supply of the muscle is intact. In such cases, the line 
of invasion may be seen and distinct zones of infection recognized. 

. At the actual wound, the muscle is black, friable, and diffluent. 
Next comes a zone in which the predominant colour is red. Separ- 
ating the red zone from the healthy muscle is a yellow band, irregular 
in outline, somewhat raised above the surface and hard to the 
palpating finger. In the majority of cases, however, no such 
differentiation into zones is apparent. In these a loss of contractihty 
is the first thing to be observed, and when this has been established 
the muscle rapidly changes its colour through varying stages of 
dirty red and greenish yellow to a dark diffluent mass. 

2. The muscle may undergo what is really a post-mortem infection, 
for it has already died in consequence of the blood supply having 
been cut off. In such a case the appearance of the muscle varies 
from a purplish red to a greenish black diffluent mass. 

Soon after the muscle has been infected, gas can be demonstrated 
in it, at first as bubbles between the muscle fibres, afterwards in the 
surrounding areolar tissues. The gas has a deleterious effect in so 
far as it exerts pressure within the fascial sheath of the limb, con- 
stricting the blood-vessels and producing that amount of interference 
with the blood supply which is so favourable to an extension of the 
infection. 

2. Varieties of the Disease. 

Clinically, several types of the disease may.be recognized. 

1. Localized anaerobic infection in the wound. 

2. Slowly spreading anaerobic infection in the wound. 

3. Gas gangrene of the ' group ' type when a single muscle or 
a group of muscles is attacked. 

4. Gas gangrene of the ' massive ' type where a whole segment 
of a limb is involved. 

5. The fulminating type. 

3. Symptoms and Signs of Gas Gangrene. 
(i) Symptoms. 

The features depend on whether it is early or late. In the early 
stages, distinctive symptoms are absent but pain is generally marked. 
The unusual amount of pain probably depends upon increasing 
pressure within the wound. 

When the infection becomes estabhshed and begins to spread, 
distinctive signs arise. The pain increases and there is a feehng 
of numbness in the superficial parts of the limb. Constitutional 
symptoms begin to make their appearance ; the patient looks 
distressed and ill ; the lips acquire a cyanotic colour ; the pulse 
rate is rapid and the temperature rises by three or four degrees. 



10 

Vomiting is frequently present. If tKe disease continues to extend, 
the constitutional symptoms become more marked. The pulse 
becomes more rapid and is finally running and uncountable ; vomiting 
becomes more frequent, the extremities are cold and blue and the 
temperature falls. It is a remarkable. fact that the mind generally 
remains acute even to the end. In the terminal stages some degree 
of general icterus may be present. 

Death frequently occurs with such dramatic suddenness as to 
suggest the occurrence of pulmonary embohsm, although it is to be 
noted that this has never been found at autopsy. 

(ii) Physical Signs. 

The physical signs differ according to the type of disease which is 
present. If there is a localized anaerobic infection of the wound, 
the only distinctive sign may be a foul-smelUng discharge mixed 
with bubbles of gas. As the disease extends, swelhng of the part is 
noted and percussion of the swollen area yields a tympanitic note. 

(iii) The Group Type. 

In the ' Group ' type the limb may show no departure from the 
normal or it may be swollen to a greater or less extent. The skin 
may be normal in appearance or it may be tense or blanched from 
the underlying swelling. As a rule, the area around the wound is 
tympanitic and crepitation may be detected. In a further stage the 
swelling of the limb increases, the skin acquires a dusky hue and 
tympanitis and crepitation are more marked. At a still later 
period, the skin shows mottling with purple patches and finally 
becomes a greenish yellow. It must be borne in mind that the 
gangrene of the muscles may be far advanced though covered by 
a skin which is normal in appearance. In certain cases the skin 
may show the development of large irregular buUae filled with 
a blood-stained serous fluid. 

If the muscle is exposed, it presents at first a dry brown appear- 
ance changing to a dirty red and, finally, it becomes pultaceous, 
and black in colour with a slimy surface. 

(iv) The Massive Type. 

This type occurs generally in a segment of a hmb from which the 
blood supply has been cut off by the interruption of the main vessel. 
There are two groups of features which are met with in this variety 
of the disease. 

1. The Hmb which is already in a condition of dry gangrene from 
the interruption of its blood supply suddenly becomes swollen and 
tympanitic and the patient exhibits the constitutional symptoms, 
pain, vomiting, rise of temperature, and rapid pulse. 

2. The gangrene, the tympanitis, and the constitutional features 
manifest themselves at the same time. 

In both varieties, during the early stages the appearance of the skin 
is that seen m ordinary arterial gangrene, but infection by bacteria 
causes a more rapid appearance of the signs of decomposition. 



• 11 

(v) The Fulminating Type. 

The fulminating type stands out in contrast to all the other 
varieties. Here it would appear that either the individual powers 
of resistance are low or the infection is very virulent. The type 
may appear as a primary manifestation, more commonly it occurs 
as a secondary feature of the ' Group ' or ' Massive ' type. It is 
associated .with severe pain, extensive swelling of the affected part, 
rapid spread of the disease, and severe constitutional features. 

4. CoMPLICATlbNS OF GaS GaNGKBNB 

Apart- from the local spread of the infection and, the gas gangrene 
toxaemia which is invariably present, there are two complications 
which may occur : 

1. Gas gangrene septicaemia. 

2. Gas gangrene pyaemia (metastatic gas gangrene). 

Gas Gangrene Septicaemia. Under certain conditions -invasion of 
the blood-stream with the organisms of gas gangrene occurs ; such 
an invasion produces the features of gas gangrene septicaemia. 
It is interesting that the general signs of the condition are by no 
means so acute as those of gas gangrene toxaemia, but in contrast 
to the latter, the general appearance is good, there is a swinging 
temperature, normal in the morning and 102 or 108 in the evening ; 
there is progressive wasting and this in spite of the fact that the 
appetite may remain good. In the majority of cases pyogenic 
organisms eventually gain entrance to the blood-stream and to this 
super-added infection the patient rapidly succumbs. 

Gas Gangrene Pyaemia. This complication is rare, but a number 
of instances of its occurrence have been recorded. It generally 
follows a gas gangrene infection associated with a compound 
fracture. There is an invasion of the blood-stream and secondary 
deposits of gas gangrene organisms occur throughout the body ; 
crepitant swellings containing gas appear in the muscles and sub- 
cutaneous tissues. These secondary foci of gas gangrene often 
develop in various parts of the body where the tissues have suffered 
some slight damage as, from the introduction of a hypodermic 
injection or an infusion of saline, or they may develop in tissues 
subjected to prolonged pressure as in the buttock when the patient 
lies in bed tilted on one hip. The prognosis is extremely bad. 

5. The Diagnosis of Gas Gangrene. 

There are certain conditions which may be mistaken for gas 
gangrene. Tiiese are : 

(a) Extensive haemorrhage into the tissues, generally from a 
wound of a large blood-vessel. 

(b) Post-traumatic oedema of the tissues. 

In the case of haemorrhage, the limb while swollen is firm to the 
touch, dull to percussion and there is an entire absence of the 
constitutional features which are so typical of gas gangrene infection. 
The situation of* the wound may give an indication of the source 
of the haemorrhage ; blood clot is generally escaping from the 



12 

wound and X-ray examination demonstrates the absence of gas in 
the tissues. Pain while present is generally much less severe than 
in gas gangrene. If the swelling is due to tissue oedema, there is 
acute pain on palpation, a dull note on percussion, herniation of the 
underlying tissues through the wound, and absence of constitutional 
symptoms. 

The diagnostic features of a gas gangrene infection are : pain, . 
crepitation, resonance on percussion, and, above all, severe constitu- 
tional symptoms. 

6. The Treatment of Gas Gangrene. 

(i) Preventive Treatment. 

There are certain conditions which undoubtedly favour the 
development of gas gangrene. They may be summarized as follows : 

(a) Eetention of extravasated blood. 

(b) Interference with the local circulation. 

(c) The presence of masses of devitalized tissues. 

(d) Extensive fractures and comminution of long bones. 

(e) Eetention of wound secretions by dressings, pastes, or packing. 
(/) Deky in the mechanical cleansing of the wound. 

(g) Eetention of foreign bodies. 

It is, therefore, evident that steps should be taken to avoid any 
of these predisposing factors. It is of the utmost importance that 
any constricting influence on the limb should be avoided. 

Cases in which there is any suspicion of gas gangrene occurring 
should be evacuated to some centre where eflBcient surgical treatment 
■can be carried out as rapidly as possible. 

(ii) Treatment by the Administration of Alkalies. 

All patients suffering from this disease should be given alkali by 
the mouth. If there is vomiting and the general symptoms are 
inarked the alkah (4 per cent, sodium bicarbonate in normal sahne) 
should be given intravenously— fifteen to twenty ounces are generallv 
■sufttcient- ° -^ 

(iii) Surgical Treatment. 

The surgical treatment will necessarily be guided by the general 
condition of the patient and the extent of the gangrene Thus 
a patient who is in good condition can be subjected to an operation 
for extirpation of the diseased muscles while another patient with 

'^l!rto7:^ifj'^r::^^^^^^ ^--^ --^^*- - ^ad, wiu 

tolttTeSrdSrrs :^ ^'°P*^' '-' '''' '^^"^^^-^ --^-g 

■ (1) A localized infection. 

(2) Group gangrene. 

(3) Massive gangrene. 

.(4) Fulminating gangrene. 



13 

(1) Localized Wound Infection. The most efficient treatment of 
this condition is complete removal en bloc of the affected part. 
After primary excision of the wound with skin and fascia, and 
after prolonging the incision to expose the deeper structures, an 
attempt is made to remove the whole of the injured and infected 
tissue en masse together with any foreign material. This leaves a 
large fresh wound surface which though almost invariably infected is 
at least easier to sterilize than the original wound. Excision, en Uoc, 
is often impossible on account of the involvement of such structures 
as bone, large arteries, or nerves. In such instances dead muscle, 
and other soft structures which are obviously a source of danger are 
cut away, after all injured and ragged skin has been cleanly excised ; 
infected tissues which cannot be removed are carefully cleansed. 
The wound is then immediately subjected to a process of continuous 
sterilization by the Carrel-Dakin method. .It is important that the 
wound should not be closed in any way immediately after operation, 
even though complete excision has apparently been carried out. 
The question of secondary suture can be considered when repeated 
bacteriological examinations of the part show that the proper degree 
of sterility has been secured. The operative treatment outlined 
must be carried out with all possible speed — a long operation is 
most harmful. 

(2) Group Gangrene. In this type of the disease the treatment 
will vary according to whether the general condition of the patient 
is good or bad. 

(a) Patient in good condition. The mechanical cleansing of the 
wound is carried out by the method which has already been 
described. The affected tissues are exposed by adequate longi- 
tudinal incisions. The condition of the muscles is carefully 
observed. All muscular tissue is removed which does not contract 
or bleed when cut into, or which shows any departure from the 
normal colour. 

Certain muscles may require to be removed in toto, that is from 
origin to insertion ; in this way a limb may be preserved which, 
iudging from the degree of crepitation and tympanitis present, would 
otherwise have been sacrificed. After excision has been performed, 
a Carrel-Dakin dressing is applied. 

(h) Patient in had condition. This is most likely to occur in 
fractures of the long bones. In these cases, amputation at the site 
of fracture is generally indicated. If on account of the crepitation 
and tympanitis of the limb, it is thought that a higher amputation 
is necessary, it is advisable to investigate the condition of the muscles 
through skin incisions which are made at the level of the injury : 
in this way it can be ascertained whether the muscles are dead at 
the level of the proposed removal. 

When gas gangrene affects the tissues of the leg it is seldom 
justifiable to amputate through the thigh ; in these cases, a guillotine 
amputation is performed through the head of the tibia or through 
the knee-joint, the heads of the gastrocnemius muscles being then 
removed from their attachment to the femur. By this procedure 
the operative shock is lessened, a painful stump is avoided and an 
increase of length of limb is gained. 



14 

(3) Massive Gangrene. As this type is usually the sequel to injury 
of the main vessel of the limb, every attempt should be made_ in 
cases of wounds of the larger blood-vessels to maintain the circulation 
until at least such a time as the collateral circulation has been 
estabUshed. This may be done by arteriorrhaphy or by the use of 
Tuffier's tubes. 

When gangrene has occurred and the limb is obviously dead, 
amputation is the only possible method of treatment. As regards 
the level at which the amputation should be done, the condition of 
the muscles may be accepted as a guide and the limb should be 
removed at the lowest level of the living muscle. 

In cases rn which there is grave injury such as a shattered joint 
or a fracture of a long bone, it may be necessary to disregard the 
extent of the gangrene and amputate at a higher level as dictated 
by the injury. In massive as in group 'gangrene, amputation 
through the knee-joint or head of the tibia is to be preferred to 
amputation through the thigh. 

(4) Fulminating gas gangrene. Here the rate of spread of infection 
is so rapid that within a few hours not only may the whole Hmb be 
gangrenous but the process may have extended for some distance 
on to the trunk. Immediate amputation by the most rapid means 
is the only treatment which offers any possibility of success. The 
disease may have extended to such a level that the amputation must 
be performed through tissues which are already infected. This need 
not be taken as a contraindication to operation, for if the flaps 
are stitched back and continuous sterilization by the Carrel-Dakin 
method is carried out, the progress of the disease becomes arrested 
in a certain proportion of cases. 

This type of the disease often appears after amputation through 
the thigh ; in these cases re-amputation at or near the hip is almost 
uniformly fatal and should only be undertaken with caution. 

In cases in which the disease has spread from the limb to the 
trunk, high amputation should be carried out, and the infected area 
above the amputation should be treated by free incision. 

(iv) Methods of Amputation in Gas Gangrene. 

Whenever possible, short flaps of som^ description should be made ; 
the wound is kept open for several days and is irrigated by the 
Carrel-Dakin method. Below the knee, where amputation will 
almost always be of a provisional nature, the guillotine method is 
often the best. It may be performed through the head of the tibia 
or through the knee-joint and the cut surface of the amputation 
stump should be carefully examined. It should be borne in mind 
that the lesion is not a horizontal one, but owing to certain muscles 
being infected and others not, the upper level of the disease is very 
uneven, mounting much higher at one point than another. For 
this reason the appearance on section of an individual muscle may 
afford evidence that it is infected, the disease having extended 
locally above the general level. If such be the case, the individual 
muscle should be dissected out. In this way the disease is eradicated 
and length of hmb saved. 



15 

(v) General and Post-operative Treatment. 

The usual remedies for the treatment of shock are carried out. 
Stimulants are given freely with advantage. As all cases are 
suffering from some degree of acidosis, it is important to administer 
an alkah. It may be given by mouth (sodium bicarbonate 8 drachms 
to one pint of water) or if the general condition is poor and there is 
persistent vomiting, it may be given intravenously (4 per cent. 
solution of sodium bicarbonate). The serum treatment, both pro- 
phylactic and therapeutic, is under trial and as yet no definite 
statement can be made as to its efficiency. 



II. INCIDENCE OP GAS 6ANGEENE. 

During the different phases of the war the incidence of gas gangrene 
among the wounded varied greatly. The highest figures were 
always observed during active operations, when the wounded were 
not collected with the usual rapidity and when treatment was, on 
that account, delayed. Again, the incidence varied according to the 
site of the battle area, being most frequent in highly cultivated soils. 
The figures given below must be looked upon as approximate 
rather than accurate on account of the difficulty in collecting 
the necessary information. 

The highest incidence of gas gangrene undoubtedly occurred in 
the early period of the war about the time of the first battle of the 
Marne and the heavy fighting in the Ypres salient. No actual figures 
have as yet appeared, but surgeons in charge of casualty clearing 
stations in these areas were of opinion that among the wounded the 
incidence was well over 12 per cent. From that date the condition 
of affairs improved very markedly in the British army. In the spring 
and summer of 1918, during heavy fighting, the figure was only about 
1 per cent. In the first and second armies, among 13,303 wounded, 
there were 158 cases (1-1 per cent.) of gas gangrene in the forward 
stations, while for the same period in the base hospitals, out of 
23,792 wounded, there were 92 cases of gas gangrene (0-39 per cent.). 

The figures given by Bowlby (1919) are very similar in that out 
of a total of 25,060 patients at certain base hospitals during 1917 
and 1918 there were only 84 patients (0-83 per cent.) with severe 
or ' massive ' gas gangrene, while at one base during the great retreat 
of March 1918 the incidence of gas gangrene among 20,000 wounded 
was 1-0 per cent. 

Many factors contributed to this reduction, chief amongst which 
may be mentioned, more efiicient surgical cleansing, earlier evacua- 
tion of wounded, and prophylactic measures. 

In the French armies the incidence appears to have been rather higher. 
Ivens (1916) had 107 cases of gas gangrene among 1,694 wounded 
(6-3 per cent.), while Chalier and Chalier (1) give the figure as 5-4 per 
cent., and Ombredanne as 13 per cent, (cited by Guermonprez). 

For the German army on the Western front, Wederhake puts the 
incidence, at 2-2 per cent, and Franz at 2 per cent. Eumpel's figures 
are more detailed : in one war hospital, from May to September 
1916, among 8,036 wounded there were 114 cases of gas gangrene, 



16 

an incidence of 3-7 per cent. He also gives statistics for one of the 
armies from the middle of December 1916 until the end of March 
1917 ; there were 5,921 wounded with 170 cases of gas gangrene, 
a total incidence of 2-8 per cent., but during an offensive the mcidence 
rose to 7 per cent. By January 1917, however, following upon more 
energetic surgical treatment and the employment of anti-gas gangrene 
serum, the incidence was reduced to 0-6 per cent. 

On the Eastern front, on the other hand, the general incidence was 
lower, as might have been expected. Out of 5,000 wounded, Wieting 
observed gas gangrene in 1-43 per cent, of the cases. 

Although the morbidity from gas gangrene was considerably 
reduced amongst all the belligerents, nevertheless even an incidence 
of 1 per cent, constituted a serious loss on account of the high 
mortahty (20-50 per cent.). 

III. BACTEEIOLOGY. 
Introduction. 

A perusal of the pre-war literature of anaerobic bacteria reveals 
the state of disorder which prevailed in this branch of bacteriology. 
First, in the matter of technique there was no method of choice 
by which surface growths could be easily obtained ; many 
methods were detailed but none was entirely satisfactory. Further, 
it cannot be doubted that most of the cultures of the anaerobes 
described were impure, and consequently the descriptions were 
inaccurate. This is seen in the case of such a well-known and impor- 
tant organism as B. oedematis maligni, which is described by von 
Hibler and later by Fehx von Werdt (1912) as proteolytic. It is highly 
probable that B. enteriditis sporogenes, Klein, represented a mixture 
of B. welchii and B. sporogenes. Such errors were numerous, were 
copied from book to book, and led to sterile controversy which ended 
in general confusion. 

No attempt will be made in this Eeport to clear up the confusion 
except where absolutely necessary, nor to put names to the numerous 
organisms insufficiently described in the older books and journals. 

The errors were due for the most part to the fact that the 
problems of anaerobic infections were of academic interest and were 
not thoroughly studied ; and in part to the very great inherent 
difficulties of the subject. 

The war transformed these problems into questions of high 
importance and urgency; the sufferings of our men and the 
grievous losses of our armies compelled professional bacteriologists 
to attack the subject with renewed vigour and enlisted to their 
aid colleagues from kindred branches of biological science. The 
united efforts of all these workers, with their different disciplined 
experiences, have advanced knowledge, enlarged our comprehension 
of wound infections, and aided in abating suffering. 

The advances in knowledge may be thus summarized : 

It may be said that it is now not much more difiacijt to obtain 
surface cultures of such anaerobes as B. tetani or vii)rion sepique 
than of common aerobes. This fundamental advance is due to 
improved methods of anaerobic cultivation, among which should 



17 

be especially noted the beautiful, adaptation by Mcintosh and 
Pildes of Laidlaw's method of obtaining anaerobiosis by the use of 
finely divided platinum or palladium as a catalyst. 

The pathogenic and probably most of the non-pathogenic anaerobes 
which may infect wounds contaminated with soil, have been isolated 
and described. The list of these microbes is long, but the majority 
are apparently non-pathogenic. The principal pathogenic species are 
B. welchii, vibrion septique, and B. oedematiens. 

These three organisms, acting alone or in combination, are respon- 
sible for almost all of the acute rapidly-evolving cases of gas gangrene 
where the whole course of the infection may be run in the space of 
8 to 48 hours. They may be helped in their pathogenic activity by 
the so-called non-pathogenic anaerobes and also by aerobes, but our 
knowledge on this point is not precise. It is certain, however, that 
they all assist one another. 

B. welchii is responsible for a larger percentage of cases of gas 
gangrene than either of the other two, though it is now known that 
the percentage is not so high as it was formerly thought to be. 
Vibrion septique is commoner than B. oedematiens. 

The role played by the more numerous but less dangerous non- 
pathogenic anaerobes has not been clearly defined ; they may be 
responsible for some of the features of gangrenous wounds — such as 
putrefactive odour, blackening and digestion of tissue — both in the 
acute cases and in local anaerobic infections. Certain strains of 
B. histolyticus undoubtedly play a large part in some cases of gas 
gangrene (Weinberg). 

It should be noted that any one, or possibly any combination, of 
these organisms may be found in a wound which is not, and does not 
become, gangrenous. This point will be discussed in the section 
devoted to experimental gangrene. 

The part played by B. tetani will be dealt with separately. Finally, 
in some cases, specific names are used in this Eeport for species which 
are believed to comprise several varieties. 

The anaerobes infecting wounds may, therefore, for practical 
purposes, be considered under the following headings. 

1. Pathogenic anaerobes liable to set up conditions of gas gangrene 
with symptoms of generahzed intoxication. 

2. B. tetani. 

3." Wound infecting anaerobes whose action may be ancillary to 
the causal agents of gangrene and tetanus. 

1. Pathogenic Anaerobes liable to set up Conditions of 
Gas Gangrene with Symptoms of Gbnbea"l Intoxication. 

(i) B. welchii. Migula, 1900. (Synonyms : Bacillus of Achalme, 
1891 ; B. aerogenes capsulatus, Welch and Nuttall, 1892 ; B. phleg- 
monis emphysematosae, Eug. Fraenkel, 1893 ; B. perfringens, Veillon 
and Zuber, 1898 ; B. enteritidis sporogenes, Klein, 1895). 

B. welchii is a non-motile organism, uniformly Grani-positive in 
young cultures; in old cultures Gram-negative individuals are 
frequently met with. 

B 



18 

Morphology. The length of the bacilli varies from 4 to 8 /u, and 
the breadth from 1 to 1-5 /x. In general they are straight and the 
ends are usually rather square. 

Very short, almost coccal forms, and long filaments may be found 
under certain conditions of growth. Some strains show curved rods. 

Involution forms showing considerable ' pleomorphism ' (club 
shapes, filaments, tadpole forms, granular types, &c.) are found, 
particularly in old cultures upon coagulated serum. 

A capsule may be demonstrated ; it is most conspicuous in rods 
from the exudate of infected tissues or from media containing serum, 
but it does not appear to be absent at any time (M. Eobertson). 

The organism has no flagella. 

Spores. Sporing forms of B. welchii are sometimes to be found in ' 
infected wounds, but are most readily formed in such media as casein 
broth, alkaline egg broth, and' coagulated serum, all of which are rich 
in protein and free from fermentable carbohydrates. They are rarely 
seen in media containing s*ubstances which the bacillus is able to 
ferment. , Individual strains of the organism vary a good deal in the 
readiness with which they form spores even in the most favourable 
media. The spores are relatively large, oval in" shape and with 
slightly flattened ends. They are subterminal or central in position ; 
the body of the rod in which they develop disintegrates very rapidly. 

Anaerohiosis. B. welchii does not require very strict anaerobic 
conditions for its growth. 

Colony. Surface colonies on nutrient agar, serum agar, or glucose 
agar are circular in contour. In 24-48 hours they may attain a size 
of 1 to 2 mm. in diameter ; older cultures may become much larger, 
the edges of the colonies sometimes being crenated. Young 
colonies (12 hours) are translucent ; they soon become opaque and 
granular. There is a variety of the colonies of B. welchii -which shows 
a very thick opaque centre and a granular slightly crenated margin. 
The latter are of a viscid consistency and tend to adhere to the 
medium In deep agar shake cultures the colonies are lenticular. 

Cultural reactions. 

Meat medium : Very rapid growth, with the development of a pinkish 

colour ; gas ; and an acid reaction. The meat is not 

blackened, and there are no macroscopic signs of 

digestion. The odour is sour but not putrefactive. 
Milk medium : Vigorous reaction ; rapid acid clot with evolution of 

much gas, i.e. so-called 'stormy fermentation'. 

This reaction takes place as a rule within 12 to 

48 hours ; spores are not formed. 
Coagulated serum : No change; no liquefaction. Spores are usually 

formed in this medium ; filaments, granular forms, 

and many involution types may be seen, especially 

in old cultures. 
Alkaline egg broth : The medium becomes evenly opaque but is not 

precipitated in large flocculi. 
Gelatine : ^ Liquefied. 

hJ J'^w^'"'"!*''^^ Report the gelatine reaction is recorded upon the result obtained 
by incubating the inoculated tube at 37°C. for 48 hours, and thereafter coolin" Tby 
immersion in a beaker of cold water. i.""uu„ is uy 



19 

Cultural reactions (continued). 

Broth : Active growth ; medium becomes turbid and opaque ; 

there is evolution of gas; after a period of- time 
varying between about 2\ to 48 hours the organisms 
sink to the bottom of the tube. The reaction 
becomes strongly acid. 

, Substances fermented. Subsrances not fermented. 

Glycerine (variable) Mannite 

Glucose Dulcite 

Laevulose Salicin. 

Galactose 
Maltose 
Saccharose 
Lactose 

Inulin (variable) 
Starch. 

Animal reactions. — The pathogenicity of the various strains differs 
considerably, many being non-pathogenic, but under constant condi- 
tions the virulence of a given strain does not, as a rule, show much 
alteration. Other strains, however, which have been propagated 
upon artificial media for long periods of time may show considerable 
loss of virulence. 

The lethal dose of different strains of B. welchii in broth culture 
is variable within fairly wide limits for pigeons, guinea-pigs, and 
mice ; rabbits show a considerable resistance, but they can be 
infected if massive doses are used. 

Washed bacilli suspended in saline are non-pathogenic. Spores 
of B. welchii, when washed off serum slopes with saline, fail as a rule 
to produce any effect whatsoever in the animal, though slight local 
induration of the limb injected may occur. 

If a lethal dose (0-1 to 1 c.c. according to the virulence and condi- 
tion of the culture) of a broth culture is injected into the muscles of 
the thigh of a guinea-pig, an extensive oedema is produced which 
involves the limb injected and which may spread over the whole 
abdomen and into the axillary region. The animal usually dies 
within 24 to 48 hours, but spontaneous recoveries do occur even 
with doses of a pathogenic broth culture which are lethal for the 
majority of animals of the same weight. Injection in a guinea-pig 
may produce a perforating necrosis of the skin and subcutaneous 
tissue, in which case recovery not infrequently takes place. Sub- 
lethal doses produce lesions of greater or less severity from which 
the animal recovers. 

A post-mortem examination reveals an extensive oedema, usually 
slightly blood-stained, but never to the same extent as with infection 
by vibrion septique. There may be some gas in the tissues ; the 
muscles of the injected limb are of a pale pink colour. They are very 
friable and soft and give the appearance of being digested. The 
odour is sour but not putrefactive. 

The internal organs in guinea-pigs do not show any conspicuous 
macroscopic alterations, except the suprarenal glands, which may 
be of a deep red colour, especially where the death of the animal 

B2 



20 

has taken place rapidly. In other cases they may he mottled^or 
hardly altered at all. If the animal survives until the third day, the 
suprarenals do not as a rule show discoloration. 

In the case of mice the liver is small and pale in colour, the kidneys 
and suprarenals are congested, and the bladder is sometimes full of 
blood-stained fluid. The duodenum is constantly dilated and often 
pink in colour from congestion of its vessels. 

In all cases the blood-stream is invaded at a relatively early'stage 
of the infection. 

Microscopic examination. Very large numbers of bacilli are to be 
found in the muscles at the site of inoculation, and these may also 
be numerous in the exudate. Spores are not formed in the body_ of 
animals experimentally infected with cultures of B. welchii. Chains 
of 3 to 5 elements may be seen on the peritoneal surfaces. Long 
filaments are never observed. 

Toxin. . A soluble toxin can be demonstrated in young (24 to 48-hour) 
broth cultures ; the nature and action of this toxin is discussed in 
another section. 

Agglutination. Agglutinins for B. welchii cannot be demonstrated 
in the blood of rabbits which have been repeatedly inoculated intra- 
venously with the organisms. Weinberg claims that horses which 
receive massive intravenous injections of B. welchii emulsions over 
a long period of time do sometimes develop agglutinins for the 
horiiologous strain, but the reaction is usually restricted to this and 
cannot be applied for diagnosis. . 

As a general rule a good emulsion can be made in saline, although 
certain strains show a tendency to spontaneous agglutination. 

Distribution in nature. B. ivelchii is a very frequent and widely 
spread organism. It is present in the intestine of men and animals. 
It is found in many samples of milk, in earth and dust, and can be 
cultivated from clothing and practically from any object exposed to 
dust. 

(ii) Vihrion septique. Pasteur et Joubert (1) (1877). Synonym 
B. oedematis maligni. Koch. 

This microbe has been the centre of much controversy ; it has, 
however, become clear that an organism agreeing in, characters with 
Pasteur's vibrion septique is of frequent occurrence in wounds. 
Numerous strains have been isolated and the characters of the bacillus 
are now perfectly well known. Vibrion septique and B. oedematis 
maligni of Koch are probably identical. Many writers subsequent 
to Koch, such as von Hibler, G. 0. Jensen, von Werdt, and others, 
have, however, repeatedly described B. oedematis maligni as liquefying 
serum and digesting meat with the production of a putrid odour. 
These reactions do not obtain in pure cultures of undoubted vibrion 
septique, the inference being that these writers and even certain quite 
recent workers, such as Conradi and BieUng, were dealing with impure 
cultures (see p. 93). 

Confusion has also arisen in the tendency to consider vibrion septique 
as identical with B. chauvoei (bacillus of Eauschbrand). This is 
undoubtedly an error. B. chauvoei is quite distinct from vibriovr 



21 

sepUque, but strains of vibrion septique have been isolated from ■ 
cases which appeared chnically to be symptomatic anthrax and 
also from accidental wounds in animals. The most recent discus- 
sion of this sjabject will be found in the paper of K. P. Meyer (1915). 

Morphology. Vibrion septique is a Gram-positive organism ; it is 
motile in young cultures and in the exudate from infected animals. 
It presents a rather wide range of different forms according to the 
conditions of culture. In broth or in meat medium the organisms 
appear as rods of varying length somewhat more slender than 
B. welchii. Spores are readily formed and are usually situated 
towards one extremity ; central spores are, however, not uncommon. 
Deeply stained bulb-like types may be present, especially in young 
cultures. In fluid media containing fresh tissue and on coagulated 
serum very varied appearances may be seen, such as * navicular ' or 
' citron ' types, i. e. pale, citron or boat-shaped bodies with deeper 
staining points at one or both extremities, deeply staining club- 
shaped forms, filaments, and bulb-like types often growing in short 
chains. The navicular forms may be observed in films made directly 
from infected tissues and blister fluid, &c. 

Cultural Beactiofis. Colony. Vibrion septique is a strict anaerobe. 
Surface colonies can be obtained on plates under good conditions 
of anaerobiosis or upon the surface of serum agar slopes. Serum 
agar containing 1 per cent, of salicin produces a more robust colony 
than other media. There is always a noticeable tendency for growth 
to occur in a continuous film instead of in discrete colonies, especially 
if the agar surface is too moist. The colonies are transparent and 
faintly opalescent, the contours are smooth or indented, and the 
edges may be more or less crenated. 

Colonies in deep agar shakes are delicate and branching, the centre 
of the colony is never very compact or well defined. 

Meat medium : Gas ; colour varying from bright red to pink ; no 

blackening ; odour is rancid but not putrid. 

Milk : Acid and clot ; some gas may be formed : the change 

in the milk is slow ; the clot does not as a rule 
appear before 3 to 6 days. 

Coagulated serum : No liquefaction. The morphology of the organism 
is very variable. 

Alkaline egg broth : The medium is rendered more opaque ; there is no clot. 

Gelatine : Liquefied. 

Broth : The medium becomes turbid : after a period of growth 

the organisms settle down to the bottom of the 
tube, leaving a perfectly clear supernatant fluid. 

Substances fermented. Substances not fermented. 

Glucose Glycerine 

Laevulose Saccharose 

Galactose Inulin 

Maltose Mannite 

Lactose Dulcite. 
Salicin. 

Animal reactions. Pigeons, guinea-pigs, mice, rabbits, and dogs are 
all susceptible' to infection with vibrion septique. Individual strains 



22 

vary a little in pathogenicity, but the pathogenic quality is retained 
unaltered over years of subculture. The lethal dose of a given strain 
remains fairly constant, provided the same growth conditions are 
repeated.! 

The most infective inoculum for Experimental animals is a 24 to 48- 
hour culture in glucose, broth with or without a piece of fresh tissue. 
Ordinary nutrient broth cultures are slightly less infective. 

If a lethal dose (0-01 to 0-5 c.c, according to virulence and condition 
of culture) of a living culture of vibrion se^tique is injected into the 
muscles of the thigh of a guinea-pig, the animal dies in 12 to 24 hours 
with oedema and the development of gas in the tissues. Spontaneous 
recovery in guinea-pigs which have become infected with vibrion 
septique does not occur. Sub-lethal doses produce no symptoms 
whatsoever, and •with.vibrion septique it may be said that for guinea- 
pigs an infecting dose is a lethal dose. 

On post-mortem examination an extensive blood-stained oedema 
is observed and a considerable amount of gas is found to have been 
developed in the areolar tissue around the muscles involved. Pockets 
of gas are almost invariably seen in the groins and axiUae. The 
muscles affected have a characteristic very intense deep red colour 
and are softened, but there is no putrid odour. 

There may be a collection of fluid in the peritoneal cavity and in 
the pericardium ; this is not as a rule very evident in guinea-pigs 
which have died in 12 to 24 hours, but is very well-marked 
in rabbits which have succumbed to the infection after 24 to 
48 hours. The suprarenal glands may show a variable amount of 
redness, but this is not so regular or so marked a feature as in guinea- 
pigs which have died from infection of B. welchii. In mice the local 
lesion is of the same character as that described for guinea-pigs. 
The organs do not, however, show any great changes. The adrenal 
bodies are most frequently affected, and they may show congestion 
varying in degree from a light pink tp a deep red colour. 

Microscopic examination of the exudates of animals infected with 
vibrion septique shows very numerous motile rods and usually the 
characteristic ' navicular ' or ' citron ' types. Swollen club-ended 
rods and short chains of intensely staining bulb-shaped individuals 
may be seen. Spore-bearing rods are, however, not as a rule to be 
found until some hours after death. 

The peritoneal surface of the liver shows long snake-hke filaments 
which are of diagnostic importance. It was this appearance which 
led to the erroneous idea that the organism was a vibrio. It may 
be noted, however, that when dealing with mixed cultures of vibrion 
septique and other spore-bearing anaerobes, ' citron ' types in the 
muscles and filaments on the guinea-pig's hver are not alwavs 
present. -^ 

Toxin. A soluble toxin can be demonstrated in 24 to 48 hour 
broth cultures. (See later, section IV). 

Tf\-5^i?n''°'''^''i'°'' °*- f 'i?*''''' <^,°'^' °* ^ 'i^i'ig "'^'"o i« notoriously inexact, 
rlfn'n r,?Tr^-'"'fW misleading With the vibrion septiqw than in most cases. The 
r/ .nn1iHnn=' ^^^ Pf "bably a iairly constant amount of toxin is required toproduce 
the tissues 'nfection, i.e. for the proliferation of^the bacteria in 



23 

Agglutination. Agglutinins are produced in the blood of rabbits 
inoculated intravenously with heated and washed cultures of vibrion 
sejptique. In using monovalent agglutinating sera, it is found that 
the strains fall into several well-defined serological groups. 

Distribution. Vibrion septique is said to be present in the intestine 
of men and animals and in the soil of cultivated land ; it is, however, 
less frequently found in nature than B. welchii. 

B. chauvoei. 

B. chauvoei. This organism is the cause of the disease known as ' Black- 
leg ', ' Quarter ill,' or ' Blackquarter ', as ' Charbon symptomatique ' by 
the French and as ' Rauschbrand ' in the German literature. There has 
been a good deal of confusion in regard to this bacillus, although it has been 
known and studied for a long period. It was probably with impure 
cultures of this organism that Grassberger and Schattenfroh (1, 2 and 3) 
carried out their extensive studies which led to conclusions now recognized 
as being erroneous. 

The characters of B. chauvoei are very briefly as follows : 

It is a very strict anaerobe. Although it has not so far been recorded as 
occurring in man (see p. 92), it is very pathogenic for mice and guinea- 
pigs. Rabbits are relatively insusceptible. 

Morphology. In broth and meat media, clostridial forms and rods with 
subterminal spores are to be seen ; the spores are of a more elongated oval 
shape than those of vibrion septique. In serum media or in the presence of 
fresh tissue, navicular, forms and club-shaped types are produced. The 
riaviculaf individuals are, however, longer and more slender than those 
usually seen in vibrion septique. cultures. In infected guinea-pigs the 
navicular and club-shaped elements are also present, but no long 
filaments are found on the peritoneal surfaces of the liver. This is quite 
an important feature in the difEerential diagnosis of B. cjiauvoei and vibrion 
septique. 

It is frequently stated that B. chauvoei is a Gram-negative organism. 
In fresh smears from infected tissues and in young cultures the bacillus is 
definitely Gram-positive . In older cultures Gram-negal ive individuals may, 
however, frequently be met with. : 

Cultural reactions. 

Meat Medium : Gas ; pinkish colour which may fade ; no putrid 

odour ; no blackening of the medium. 
Milk : Acid ; clot in 3 to 6 days ; some gas may be evolved. 

Coagulated serum : No liquefaction. 
Gelatin : Liquefied. 

Substances fermented. Substances not fermented. 

Glucose . ' Glycerine 

Galactose Mannite 

Laevulose ' Dulcite 

Maltose Inulin 

Saccharose Salicin, 
Lactose. 

It should be noted that B. chauvoei ferments saccharose and does not 
ferment salicin, whereas the contrary is the case with vibrion septique. 
Toxin. A specific soluble toxin is produced by B. chauvoei. 



24 

(iii) B. oedematiens. Weinberg and Seguin (1), 

This organism, which was discovered by Weinberg and Seguin in 
1915 and subsequently identified by Legros (3), Vaucher (1)^ Dalyell 
and others, is closely allied to, if not identical with, B. oedematis 
maligni II described by Novy in 1894. It is motile under strictly 
anaerobic conditions, for instance when examined in a sealed capillary 
tube ; under a cover-slip, however, it is non-motile. In shape 
B. oedematiens is a stout rod as thick as B. welchii (0-8 to 1 ix) but 
usually longer. The rods are frequently curved and the- spores which 
are readily formed in all media are large and oval with slightly 
flattened ends. They are subterminal or central in position. The 
morphology of the organism should be studied in young cultures as 
autolysis sets in very early. There is relatively little variation in 
the appearances presented by B. oedematiens but short chains and 
filaments are occasionally formed. 

Cultural reactions. B. oedematiens requires strict anaerobic con- 
ditions to ensure surface growths. Surface colonies are flattened and 
tend to be confluent forming a translucent film. In agar shakes the 
growth is dehcate, resembling snowflakes, but more solid colonies like 
conventional bursting grenades may also be seen. 

Meat medium : Gas ; pinkish colour which fades quickly ; in some 

samples of meat medium the colour may not change 

very markedly with growth. 
Milk : After long incubation (10 to 30 days), there is an acid 

reaction and clot, the acid reaction may be seen in 

vigorous cultures after 4 or 5 days, but the clot is 

always delayed. 
Coagulated serum; no change in medium. 
Broth : Early flocculation and sinking to the bottom of the 

tube where the organisms rest as a semi-opaque 

cloud (18 to 36 hours). 
Gelatine : Medium is liquefied. 

Fermentation : Production of acid is feeble. 

Substances fermented. Substances not fermented. 

Glucose Glycerine Mannite 

Laevulose Galactose Dulcite 

Maltose Saccharose Inulin 

Lactose Salicin. 

Animal reactions. The pathogenicity of different strains varies 
somewhat ; gmnea-pigs, mice, rats, and rabbits are all susceptible ' 

The mjection of 0-25 to 1 c.c. of a 24 hour broth culture into the 
thigh muscles of a guinea-pig produces death within 24 to 48 hours 
At the post-mortem examination the muscles immediately at the 
site of inoculation are found to be red, softened but not diffluent 
As a rule very httle if any gas develops. There is a spreading 
gelatinous oedema which is usually quite colourless but if the 
infection has progressed very rapidly the oedema may be tinged 
with pmk. The muse es of the abdominal wall are Unaltered in 
appearance. The baciUi are to be found in numbers only lithe 
site of inoculation. In general the oedema fluid and the peritoneal 



25 

surfaees of the liver show only infrequent individuals. Cultures 
from heart blood are generally positive, except in cases where the 
animal succumbs to the effect of the toxin present in the inoculum 
and not to an actual infection with the bacilli. 

Toxin. B. oedematiens develops an active toxin in broth cultures. 
(See section IV).' 

Agglutination. Agglutinins can be produced in rabbits by the 
intravenous injection of washed bacilli. The sera thus obtained, 
however, agglutinate only the homologous strain. Agglutination 
"tests must be carefully controlled as many strains show auto- 
agglutination. 

Distribution. B. oedematiens has been isolated from samples of 
earth from cultivated areas by injecting small quantities into the 
muscles of guinea-pigs protected by means of antitoxic sera against 
B. welchii, B. tetani, and vibrion septique. 

(iv) B. histolyticus. Weinberg and Seguin (12). 

B. histolyticus holds a position that is intermediate between the 
acutely pathogenic, invading anaerobes and the secondary purely 
saprophytic forms. 

B. histolyticus is a motile rod very frequently arranged in pairs. 
It is about 0-5 to 0-8 microns broad and about 3 to 5 microns long. 
Very young cultures (14 to 16 hours growth) should be examined to 
obtain information concerning the appearance of this organism. 
The bacilli degenerate very soon in cultures, and after 24 to 48 hours 
of growth many Gram-negative individuals may be seen together 
with fusiform and skittle-shaped involution types. The spores, 
which are readily formed in all media, are oval and usually sub- 
terminal ; they are considerably wider than the rod in which they 
arise. 

Cultural reactions. Surface colonies are delicate and flat ; they 
have crenated or irregular edges. Colonies in deep agar are arbor- 
escent or coral like with fine woolly ends to the branches. 

Meat medium : Digested with production of a white deposit of tyrosin. 

Coagulated serum : Liquefied. 
Gelatin : Liquefied. 

Substances fermented. 
Glucose 
Laevulose 
Maltose. 

Distribution. This organism is difficult to isolate ; it is found in 
wounds and has been obtained from earth. 

The strains of B, histolyticus vary very greatly in pathogenicity 
as tested by injection of young broth cultures (16 to 18 hours) into 
guinea-pigs. Some strains produce very serious lesions culminating 
in the death of the animal, while others are only very slightly if at 
all pathogenic. The bacillus is actively proteolytic and digests 
living tissue (Weinberg and Seguin (15) ). This can be demonstrated 
by injecting 1 c.c. of a young broth culture of a virulent strain into the 
thigh of a guinea-pig ; in the course of 12 hours the skin and muscles 



26 

become digested and a haemorrhagic liquefaction of the soft, partj 
of the Umb takes place. This digestion of the, tissue may spread 
over the abdomen and the animal may die in, the. course of the next 
12 to 24 hours or it may recover with a more. or less complete necrosis 

of the limb. , ,, i, e-iA 4-^ 

Toxin. A toxin can be demonstrated m broth pultureS .ot 14 to 
16 hours' growth, but is destroyed if growth is allowed to proceed 
beyond 18 hours. It is difficult to filter the toxin successfully as 
it is largely retamed in the candle. The action of the toxin 
resembles that of vibrion septique in that when injected intravenously 
(in rabbits) there is practically no incubation period, death occurring 
within five to fifteen minutes. An antitoxin neutrahzmg this 
substance has been produced in horses (Weinberg and Seguin). 

(v) B. hotulinus. van Ermengem. 

This bacillus was first described by van Ermengem (1) in 1896 in 
connexion with an epidemic of meat poisoning in Flanders. It has 
not been found in wounds, and so far as is known is in no way involved 
in any aspect of the gas gangrene question. A description of its 
chief characters is appended here for the sake- of completeness as it 
ranks among the highly toxigenic anaerobes. 

Although it has not been frequently isolated the recent in- 
vestigations of E. C. Dickson would indicate that its distribution 
in nature is wider than was formerly believed. 

Morphology. B. hotulinus is a motile Gram-positive rod, some- 
what larger than vibrion septique but resembling it in general 
appearance. The bacilli are usually single but short filaments are 
not infrequently formed. Spores are not readily produced, but 
when present are small and oval, usually sub-terminal in position 
and do not distend the rod to any great extent. 

Cultural reactions. B. hotulinus is difficult to cultivate upon 
artificial media and requires suitable conditions. The organism 
will not grow in media the reaction of which is acid. Even the 
presence of COg inhibits the growth. The best results are obtained 
by the use of media containing fresh animal tissue or glucose, and 
growth occurs between 18° C. and 35° C. Good conditions of 
anaerobiosis are essential. Surface colonies can be obtained upon 
glucose media. They may attain to 1 mm. in diameter in 24 to 
48 hours. They are flat, irregular in shape, and of a greyish 
colour. If they are examined with a hand lens they are found to be 
irregularly mottled. Madsen described colonies on glucose gelatine 
as being circular, transparent, and yellowish in colour ; they were 
coniposed of coarse granules which are seen to be in continual move- 
ment. The gelatine was Uquefied round the colony. Upon further 
incubation the colonies increased in size and sometimes reached 
several millimetres in diameter. They do not assume an opaque 
feathery appearance. In deep shakes of glucose agar, semi-opaque 
biconvex or kidney shaped colonies are produced of about 1 mm. 
in diameter after 24 to 48 hours' incubation. There is usually 
a central nucleus and older colonies may send out irregular pro- 
jections. A considerable amount of gas may be produced. 



27. 

Meat medium : Poor growth. This medium is not suitable for the 

organism. 
Milk medium : Growth is scanty- and frequently fails altogether. No 

change in the medium. 
Coagulated serum : Not liquefied. ' 
Gelatine : Liquefied. 

Substances fermented. Substances not fermented. 

Glycerine Galactose 

Glucose - Saccharose 

Maltose Inulin 

Lactose Mannite 

Starch. Dulcite 

Salicin. 

Animal reaction. B. botulinus grown in broth, or in broth to which 
a piece of Hving tissue has been added, produces a characteristic 
toxin which causes the death of guinea-pigs within 36 to 48 hours, 
A tenth of a c.c. of such culture injected subcutaneously into a guinea- 
pig of 250 grm. causes typical signs of botulismus in 36 hours, there 
is a complete muscular paralysis, dilatation of the pupils, shallow 
breathing, intense salivation, and the death of the animal supervenes 
after a short period. The death is mainly due to the action of the 
toxin. The organism can, however, be cultivated post mortem from 
the tissues and the blood of the injected animal (Mcintosh). 

Animals very susceptible to botulinus toxin are rabbits, guinea- 
pigs, mice, cats, and monkeys. Madsen using a filtered toxin found 
that 0-0015 c.c. killed a guinea-pig in 1 to 2 days and 0-0010 c.c. killed 
in 4 to 5 days. The symptoms he describes are marked muscular 
relaxation, greenish discharge from the mouth, aphonia, aphagia, 
constipation, dilatation of pupils, and great loss of weight. 

(vi) B. Jallax. Weinberg and Seguin (5). 

This is a motile bacillus which occurs in infected wounds and 
may be found in cases of gas gangrene. It is occasionally present 
in blood cultures from the patient. When recently isolated certain 
strains are pathogenic for guinea-pigs (Weinberg and Seguin (9)), 
this character is, however, rapidly lost upon cultivation. A strain, 
which had lost its pathogenic properties for mice, killed however 
when injected along with a small dose of calcium chloride. 

B. fallax is a somewhat slender rod, of about 3-6 microns in length, 
with rounded ends. It is often slightly curved. Gram-negative 
elements are frequent and there is altogether rather a feeble capacity 
to retain the stain. Spores are not readily formed on any media 
but do occur in small numbers in meat and coagulated serum. They 
are oval and usually subterminal. 

Growth reactions. On surface plates the colonies are round or 
crenated- and slightly granular. Deep colonies are lenticular, 
irregular or bean shaped. 

Meat medium : Gas ; pinkish colour ; no digestion. 

Milk : Acid clot after some days (3-7). 

Coagulated serum : No liquefaction. 

Gelatin : No liquefaction. 



28 

B. fallax ferments glucose, laevulose, and maltose. Some strains 
also attack galactose. Accounts vary somewhat in regard to the 
fermentations of B. fallax ; Henry considering that in addition to the 
above sugars, galactose, saccharose, starch, inulin, and salicin are 
also fermented. 

Toxin. A soluble toxin is produced. 



2. B. Tbtani. Nicolaier. 
(i) Bacteriological account. 

Bacillus tetani was described by Nicolaier in 1884 and cultivated 
by Kitasato in 1889. 

It is extremely difficult to isolate in a pure state ; many of 
the cultures obtained from serum institutes and from laboratory 
collections being found to contain some other anaerobe in addition 
to B. tetani. The contaminating organisms usually belonged to 
the type of B. sporogenes, but oval and round end-sporing bacilli 
such as B. cochlearius and B. tetanoides, were also found. It is to 
the impurity of many of the cultures of B. tetani that the great 
variety in the cultural characters ascribed to this organism is due. 
(Mcintosh, Eobertson). 

It may be noted that agglutination tests showed that all these 
laboratory cultures belonged to one serological type (TuUoch (1) ). 

Morphology. B. tetani is a motile Gram-positive rod. In young 
cultures it is rather a stout bacillus about 4 to 8 microns in length 
and 0-4 to 0-6 microns in breadth. 

The spores are spherical and always strictly terminal. The 
bacillary rod becomes thinner after the spore is formed but remains 
attached to it for some time ; this condition gives the familiar 
drumstick appearance. In pure cultures in the ordinary culture 
media spores are not formed as a rule until the 3rd or 4th day. 

Cultural reactions. B. tetani is a very strict anaerobe but surface 
growths can be obtained on serum agar with or without the addition 
of glucose under good conditions of anaerobiosis. 

Surface colonies are flat and delicate, sometimes with finger-hke 
projections. After 48 to 72 hours the centres become raised a Uttle 
above the surface of the medium and a ground glass appearance 
may be observed. They are generally small, being not more than 
1 mm. in diameter. 

In agar shakes (preferably containing glucose) the colonies appear 
as dehcate filamentous outgrowths spreading from a small central 
nucleus. They may attain a diameter of 2 mm. after 48 hour's 
growth. 

Meat medium : Pink colour or no change in medium according to the 

samples of meat ; softening of the consistency of 
the meat. The odour is characteristic but not 
putrefactive. 

^ilk : Poor growth ; no change in medium. 

Coagulated serum : Little or no liquefaction. 

Gelatine : Liquefied. 



29 
None of the following substances are fermented 



Glycerine 


Maltose 


Starch 


Glucose 


Saccharose 


Mannite 


Laevulose 


Lactose 


Dulcite 


Galactose 


Inulin 


Salicin. 



Animal reactions. B. tetani when injected in the form of a broth 
culture of 3 to 4 days' growth produces death with the characteristic 
symptoms of tetanic intoxication. This condition is due solely to 
Ihe absorption of toxin ; the bacilli do not invade the tissues. 
Washed and heated emulsions of bacilli and spores, that is to say, 
cultures deprived of their toxin, are not pathogenic. 

Toxin. A characteristic toxin is produced by the growth of 
B. tetani in broth. 

Agglutination. Washed and heated organisms injected intra- 
venously into rabbits provoke the production of agglutinins in the 
serum of the animals injected. 

By means of the agglutination test the strains of B. tetani can be 
divided into 4 serological groups. This matter along with its 
bearing on serum prophylaxis is discussed in detail in the following 
section by Major Tulloch. 



(ii) Recent Experimental Work on B. tetani. 
By W. J. Tulloch, M.D., Bt.-Majoe E.A.M.C. 

Lecturer in Bacteriology, University of St. Andrews. 
Member of the War Office Committee for the Study of Tetanus. 
Prom the Laboratories of the K.A.M. College and the Lister Institute of Preventive 

Medicine. 

The prevalence of tetanus in the war has been studied statistically 
by many bacteriologists, Lumiere, Chavasse, H. Pribram. In this 
country the papers published by Bruce (1,2, 3, 4, 5, and 6), byLeishman 
and Smallman, and by Cummias and Graeme Gibson, show that, 
roughly speaking, the disease was comparatively common in the 
ea;rly months of the war, but that its incidence has steadily fallen since 
the autumn of 1914 when prophylactic inoculation with antitetanic 
serum was introduced. 

The tetanus bacillus can generally be cultivated from the wounds 
of actual cases of the disease and is by no means uncommon in the 
wounds of patients showing no clinical evidence of tetanus. Thus, 
from 100 wounds of such cases examined, true tetanus bacilli were 
recovered from 19 at one time or another. The bacilli may persist 
in wounds for long periods, as is shown by the fact that in one case 
B. tetani was obtained 882 days after infliction of the injury. 

A brief description of B. tetani has been given above and will 
therefore not be repeated here. Nor will the generally known and 
well established facts of tetanus be touched upon, attention being 
devoted to the new experience gained since 1914 and particularly 
to the researches which have been conducted on behalf of the tetanus 
committee of the War Office. 



30 

A new technique which has been found of great use in dealing 
with large numbers of cases of tetanus, or supposed tetanus, is 
described in the Appendix. 

One of the most interesting facts estabhshed by the study of 
a considerable number of strains of tetanus bacilli is that there exist 
at least 4 distinct serological types. Type I is the standard U.S.A. 
bacillus and is the organism which appears to have been usually 
employed in laboratories in Europe prior to 1914 for the preparation 
of antitetanic serum. Types II, III, and IV have been differentiated 
since. 

The relative, frequency of occurrence of these types, their value 
as toxin producers, the similarity or dissimilarity of their toxins and 
•consequently of antitoxins and other evident problems have been 
studied. 

With regard to the first of these points it should be emphasized 
that, in all probability, most of the anti-tetanus serum manufactured 
in England and in America in the first years of the war was obtained 
by the employment of toxin from Type I of the bacillus. If such 
monovalent serum be more efficient against infection with the 
■corresponding type of bacillus, one would expect that a census of 
tetanus bacilli found in wounds, from actual cases of the disease 
among inoculated men, would show a relative preponderance of 
infections due to Types II, III, and IV. 

Of 100 strains of B. tetani obtained from cases of tetanus among 
inoculated men, Type I bacillus was found in 41 instances. Type II 
in 22, Type III in 33, and Type IV in 4. But of 25 strains of true 
tetanus bacilli isolated from wounds of men who showed no signs 
of the disease 76 per cent, were of Type I, 12 per cent, of Type II, 
8 per cent, of Type III, and 4 per cent, of Type IV. 

This very large discrepancy between the percentage of Type I 
in the two series is possibly due to the greater efficiency of a mono- 
valent serum against the corresponding type of bacillus. The 
methods which have been employed in the investigation of this have 
been both statistical and experimental. On the statistical side an 
.analysis of the series of 100 cases of tetanus has shown that of 
91 cases known to have received tetanus antitoxin prophylactically 
the case mortahty in the men infected with Type I bacillus was 
lowest. This, together with the figures for the other types of bacillus, 
is shown in the accompanying diagram (Fig. I), which deals only 
with the 91 cases definitely known to have received serum prophy- 
laxis ; in the remaining 9 cases the records are incomplete. 

Figure II shows the mortahty in cases of tetanus occurring" in 
men moculated prophylactically with antitoxin. The onset of 
symptoms occurred within 14 days of the infliction of the wound. 

The influence of the antitoxin employed— presumably monovalent 
Type I— on the percentage occurrence of the different types of 
B tetani m ' mdifferent wounds' and in wounds of actual eases 
•of tetanus, is shown graphically in Figure III. 

This statistical analysis suggests that the serological type of the 
infectmg bacillus may be of importance in relation to the patho- 
genesis of tetanus in inoculated men. To prove this by such methods 
-would require an investigation of a very large number of cases of 



31 



/O 20 


50 


40 


10 




lO 


30 ac 










' 






1 I 










^■^r~ 






11 










E 




















1 BT 








NIL.-Bt 









The Actual mj/nher ofc/eafhs c/ue fo each fi/pe I 



Fia I. 



10 20 30 4-0 



SO 



^ 'yyyyyyyy. y/yy/z^^y -tt ttttt^ 



^ZZ'ZZZZZZZZ'ZZZZZZZZITZZZZZI-S. 



r^'TT? ^yy//y7:7. 



''/////A 



m 



IncA/ence. \WW^/A Death- r<xte 

Fm. II. 



lO 20 30 iO 



60 /O 











■ I 








y777^77777. 


/'/'////'///, 


\////////^ 




































■■117 

^777S nr 



From casM X////^/A From "fnc/iWere/tt" wounef'S 

Fig. III. 



32 

the disease. The matter was, therefore, submitted to experimenti 
inquiry. 

• It may be stated at once that the results obtained by exper 
mental methods, though they are of considerable general interes 
are inconclusive as to the special point under consideration. Takin 
first the antitoxins, it has been shown that a mono-typical antitoxi 
is equally active in neutralizing toxin of any or of all the types c 
B. tetani when the test animals are rats, mice, or guinea-pigs. Bu 
there is some evidence, not conclusive, that monotypical antitoxi 
sera may under certain circumstances exhibit specific anti-infectiv 
properties in relation to the types. 

The difi^erences may be bound up in the different types of bacill 
but it may be said that anti-bacterial sera containing no antitoxi: 
or negligible quantities of it, do not prevent infection in the experi 
mental conditions under which one is forced to conduct the investiga 
tion, nor is there so far, any evidence that anti-bacterial sera o 
a given antitoxin content afford more adequate protection agains 
infection, than do pure antitoxic sera of the same antitoxin content 

This statement is made with reserve as anti-bacterial sera weri 
not available in sufficient quantity to permit of crucial experimenti 
being carried out. 

Infection in Tetanus. Tetanus is rarely produced by the inocula 
tion of bacilli or spores deprived of their toxin by washing or, in th( 
cases of spores, by washing and heating combined. SusceptibL 
animals withstand very large numbers of the organisms, 2,000 millioi 
in the case of guinea-pigs and 200 miUion in the case of mice. Th( 
toxin of the tetanus bacillus as ordinarily prepared is feebly aggres 
sive, differing in this respect from the toxins of the organisms o: 
gas gangrene. When the toxin of B. welchii, in sub-lethal doses 
is inoculated along with washed and heated spores of B. tetani 
the spores germinate and the animal succumbs to tetanus. Th( 
toxin of vibrion septique acts in a similar manner though not sc 
regukrly. Chemical irritants have been used to initiate tetanus 
infection in animals. Lactic acid is very uncertain in action: 
trimethylamine in certain concentrations is able to permit the 
development of tetanus in guinea-pigs but in mice rarely ; saponine 
will invariably provoke tetanus in guinea-pigs, seldom in mice; 
calcium chloride, on the contrary, almost invariably induces the 
development of infection in the latter. 

The degree of tissue destruction with lactic acid, trimethylamine 
and saponine is considerable ; with soluble calcium salts it is 
small. 

The quahty of tissue ' debility ' at the nidus of infection is, there- 
fore, of great importance, and experiment has shown that the 
degree and quahty of such debihtation is of greater importance tlian 
the actual number of organisms injected. 

When an insufficient degree of local disturbance is produced in 
gumea-pigs, typical tetanus, due to infection, occurs, but some of 
the animals recover and in ^uch circumstance local tetanus has 
frequently been noted. Cases of local tetanus in experimental 
animals was also noted, when the degree of tissue debilitation was 
relatively great, provided, in this instance, that the animals had been 



33 

passively imnmnized with anti-tetanus serum before, or soon after, 
the injection of the ' spore debiUtant ' mixture. 

The importance of the symbiotic factor, i.e. the influence of the 
products of B. welchin,, &c., in the initiation of tetanus infection is 
evident. On the one hand it has been shown that the toxin of 
B. welchii may provide the conditions required by the tetanus 
bacillus in initiating the disease ; on the other hand it has been- 
shown that certain mixed cultures from wounds are non-toxic 
although they contain true tetanus bacilli. From these cultures the 
tetanus bacillus may be recovered in a sufficient state of purity to 
permit of its being agglutinated in the presence of type sera. Further, 
the organisms are not deficient in toxigenic capacity as is shown 
by injecting into animals the washed spores, derived from such 
cultures, together with tissue debilitant, when tetanus develops. 
Moreover, by growing one of the stock cultures of B. tetani in the 
presence of these mixed growths, the toxigenic capacity of the 
tetanus culture was greatly reduced. The particular combination 
of organisms which leads to this depression of pathogenicity is not 
yet fuUy determined, nor can it be stated whether it is the infectivity 
or toxigenicity which is depressed, though presumably the latter. 

All these facts show that the conditions, under which tetanus may 
occur naturally, are very complex, and that when the infection has 
begun factors may come into operation which increase or diminish 
the violence of the disease. 

An inquiry into surgical procedures in relation to degree and persistence 
of Anaerobe Infection of Wounds. 

This investigation, which was carried out by Miss Cayley, indicated 
that 

(a) None of the antiseptics investigated could be specially recom- 
mended as valuable for the elimination of anaerobic infection. 

(b) Anaerobic bacilli, even those of pathogenic significance, may 
persist in wounds until the completion of the process of repair. 

(c) The degree of anaerobic infection of wounds that have been 
excised is, on the whole, less than in those that have not been so 
treated. While excision therefore does not eliminate infection it 
does alter the condition of the wound in such a fashion that the 
harmful capacity of any anaerobes present is much reduced. 

(d) In examining the influence of brilliant green and other ani- 
line dyes on growth of anaerobes in vitro, it was found that their 
activity was very much reduced when the cultures were made by 
Tarozzi's method. This reduction of activity is more marked than 
that caused by the presence of serum. This observation has a direct 
bearing on the application of aniline dyes as antiseptics in surgical 
therapeutics. 

Technique. 

In making the investigations summarized above, the following 
technique was used. 

Swabs taken from wounds — whenever possible these were taken 
from the deeper parts — were emulsified in about 3 c.c. of saline. 

'c 



34 

Of this, 1 c.c. was inoculated into ordinary meat water medium 
and incubated anaerobically— Culture A. This gave an index of the 
non-sporing anaerobic bacilU present. 

The remainder— 2 c.c— was heated to 80° C for 10-15 minutes 
or to 60° C for 30^0 minutes. 

Of this 1 cc. was inoculated into a tube of meat water medium 
and incubated anaerobically— Culture B. This gave a growth of 
the sporing anaerobes present. 

The 1 cc. which remained was inoculated into a tube of ' selective ' 
medium designed to give overwhelming growth of endsporing bacilli 
—Culture C. 

The selective medium was prepared thus : 

Take 1 lb. of chopped meat, add 1 litre of tap water, and boil for 
30 minutes. Cool to 45° C, make shghtly alkaline to litmus and 
add trypsin as for the preparation of Douglas' broth ; then incubate 
in an open vessel for 4-5 days at 37° C. allowing the material to 
undergo natural putrefaction. Pass the putrescent material so 
obtained through paper pulp to clear, and make neutral to phenol- 
phthalein at room temperature. The material is then sterilized by 
filtration through a Berkfeld and a Doulton candle in series. The 
medium is stored in sterile flasks under a layer of paraffin to which 
sodium formate has been added to the extent of 1 per cent, of 
the total volume of medium. The flasks should be provided with 
a hooded delivery tube, so that the material may be distributed 
as required. It keeps fairly well — about three weeks — but should 
never be sterilized by heat. 

Before use the sterility of the medium should be tested by inocu- 
lating quantities of from 5 c.c. to 0-1 c.c. into tubes of meat water 
medium and incubating anaerobically for at least 7 days. When the 
medium is to be employed for cultures that may ultimately be used 
for animal inoculation, it must be shown that the medium itself is 
non-toxic. Before use, fresh sterile rabbit kidney is added to the 
medium, J^th P^^^t of a kidney being sufficient for 5 c.c. The tubes 
to which kidney has been added should be used within 3 days. 

Prepared thus, the medium inhibits the growth of B. sporogenes 
but allows of the growth of B. tetani, atoxic round endsporing 
bacilli, and certain oval endsporing bacilli. 

Latterly, in place of using meat water tubes for culture- A and B, 
the following medium has been employed : 

The flesh of one rabbit is chopped, 1 litre of water containing 
5 grams sodium carbonate is added, and the mixture is allowed to 
decompose for 16-24 hours at 37° C. This mixture is then examined 
and the reaction again adjusted to be slightly alkahne to litmus, 
and 2 per cent, of trypsin added. The flask is now returned to the 
mcu^ator for a further period of 16-24 hours. The material is then 
filtered through paper, made slightly acid, and boiled to coagulate 
protems, filtered agam and neutraUzed. In neutrahzing, the mean 
of two titrations is taken at room temperature (the specimens for 
tiie tests having been previously boiled and rapidly cooled) (1) with 
phenolphthalem and (2) for a-naphtholphthalein, and the requisite 
amount, of sodmm hydrate is added to the bulk of the medium. 
Boil again and finally filter. The medium may then be tubed and 



35 

autoolaved or autoclaved and stored in bulk. Before use, j-^h 
part of fresh sterile rabbit kidney is added to tubes of 5 c.c. Tubes 
which have been stored should be boiled for 30 minutes and rapidly 
cooled before the kidney is added. The medium should be used as 
soon as possible after the addition of the kidney. This medium 
gives very heavy growths of B. tetani and appears to have some, 
though not marked, selective properties. 

Tubes A, B, and C were examined at two days' intervals and when 
C showed a growth consisting mainly of endsporing bacilli the 
culture was centrifuged, washed, and tested by the agglutination 
technique. Not infrequently culture failed to show growth, 
although round endsporing bacilli might ultimately develop in 
tube A or tube B. In such cases A or B or both A and B were 
heated to 60° C. for 45 minutes, and subcultures made from them 
into the selective medium. In this way a number of cultures of 
B. tetani could be obtained, although they failed when directly 
inoculated into the selective medium. 

The technique of the agglutination test and that of the preparation 
of agglutinating sera are described fully in papers dealing with the 
serological differentiation of B. tetani. 

3. Anaerobes the Action op which may be Ancillary to the 
Condition of Gas Gangrene. 

The less directly pathogenic anaerobes which proliferate in wounds 
comprise a very large number of diverse types. It is proposed to 
describe only the more frequent and important species. It is not 
claimed that even this list is complete ; the forms included are, 
however, those which have appeared with sufficient regularity to be 
considered of some definite importance in wound infection. It is ad- 
mittedly difficult to estimate the role that these ancillary organisms 
play in gas gangrene and in the very much more frequent local 
infections caused by anaerobic bacteria. 

Blood culture from the patient occasionally reveals the presence 
of some of these types, notably B. sporogenes, but the fact that 
organisms can be cultivated from the patient's blood is not neces- 
sarily significant when wounds of large area are present. In pure 
culture they are practically all to be classed as non-pathogenic for 
laboratory animals. With few exceptions soluble toxins have not 
been found in cultures of any of these organisms but in no case 
so far have they been exhaustively examined for this property. 

In mixed cultures they are for the most part, though not 
invariably, favourable to each other's growth, and it is pre- 
sumed that they act with an equally successful mutual assistance 
in wounds. 

It is well-known that it is not until two or three days after the 
infliction of the wound that the anaerobic flora of this mixed, less 
pathogenic type, arises. Some of the characters of these infections, 
such as putrefactive odour, digestion, and gas formation can be 
explained from the known reactions of organisms isolated from 
these cases. These sub-acute conditions may show one or several 
of the pathogenic anaerobes already dealt with and this mixed type 

02 



36 

of infection is always a state that may proceed to acute gas 
eanerene or to acute tetanus. . i,„-4„j 

In wounds in which these sub-acute local infections have subsided 
and also in late septic wounds without obvious local symptoms ot 
anaerobic action, many of these bacterial types Imger on especiaUy 
when sequestra or adherent sloughs are present, multiplymg suffi- 
ciently to be very readily cultivated from qmte_ small samples of 
wound exudate. It is not precisely known what, if any, importance 
these organisms have in weakening the tissue resistance and m delaymg 
the healing process, but it is probable that their local action is 
unfavourable to the patient. ^, i, i „„ 

Pathogenic anaerobes may prohferate as apparently harmless 
saprophytes, but their presence, even as spores m sequestra and 
in the scars of healed wounds, are a source of great potential danger 
especially when disturbed by injury or surgical interference. 

(i) B. sporogenes. Metchnikoff. 

This organism was first described by Metchnikoff, 1908, who 
distinguished two varieties or races ; ' A ' derived from the faeces 
of healthy persons, and ' B ' derived from persons suffering from 
chronic cohtis. He distinguished them chiefly by their morpho- 
logical characters, A being more slendej than B. Both races 
have filamentous woolly colonies and agree with each other in 
cultural characteristics. 

B. sporogenes as found in wounds appears to belong to Metchnikoff 's 
race A. 

It is present in a very large proportion of wounds, and appears 
in acute cases of gas gangrene as well as in conditions in which 
the wound is progressing in a satisfactory manner. It has been 
cultivated from the blood of the patient during hfe, sometimes 
alone, but more often in association with other bacteria. 

B. sporogenes is widely distributed in nature ; it is frequently 
found in earth and in practically all materials exposed to dust. 
It is a common inhabitant of the alimentary canal of man and 
animals and can be cultivated from a great variety of sources includ- 
ing milk. The organism is extremely tenacious of life as is shown 
by its capacity to survive 8 days in a 5 per cent, phenol solution 
• (Schiitze), or an exposure in a capillary tube to 100° C. for 45 
minutes. 

The name B. sporogenes is used in a wide sense and the strains 
designated by this title are capable of further sub-division upon 
minute characteristics. 

Morphology. B. sporogenes is an actively motile rod of about 
3 to 7 f^i in length ; it closely resembles vibrion septique in appear- 
ance and is more slender than B. welehii. Spores are very readily 
formed in all media, are oval in shape and central or subterminal 
in position. The range of bacillary forms is more restricted than 
in the organisms already described ; long filaments being the only 
variation. The baciUi are Gram-positive, but in old cultures Gram^ 
negative individuals are found. 
Cultural reactions. B. sporogenes is an anaerobic organism, but it 



37 

does not exact a perfect condition of anaerobiosis ; it grows readily 
on agar plates in an exhausted cylinder. 

Colonies. Surface colonies of B. sporogenes are characterized by 
the production of woolly tangled filaments which grow out all round 
the more solid centre and down into the medium. Young colonies 
are very transparent and irregular in shape ; later, they become 
opaque and woolly with a definite raised centre. Old colonies may 
be yellowish in colour. They have a tendency to grow large and 
are solid in type. Isolated colonies on thinly sown plates may 
attain a diameter of several millimetres. Deep colonies are woolly. 

Meat medium : Vigorous growth ; gas ; alkaline reaction ; digestion 

of the medium and blackening ; the extent of the 
change of colour varies with different samples of the 
meat medium. Putrid odour. 

Milk : Digestion occurs without the production of a firm 

clot, leaving a somewhat turbid supernatant fluid : 
the reaction becomes alkaline. 

Coagulated serum : Liquefied and becomes darker in colour. 

Alkaline egg broth : There is a flocculent precipitate or even a soft clot 
which sinks to the bottom of the tube leaving 
a clear supernatant fluid ; the clot is digested to 
greater or less extent. 

Gelatine : Liquefied. 

Sugars : Glucose, laevulose, and maltose are alone fermented. 

Micrdbic association. One feature of the» behaviour of B. sporo- 
genes calls for particular notice, namely, its extraordinary capacity 
for persisting in the presence of other organisms. Only bitter 
experience can convince workers, unaccustomed to the clinging and 
pervasive character of this bacillus, of the extreme difficulty of 
dislodging it from the cultures of any other organism with which it 
has become associated. B. sporogenes will linger unsuspected for 
weeks or years in a culture of jB. welchii or of vibrion septique, only 
to appear when some particularly favourable set of circumstances 
(such as an alkaline protein medium) permits it to obtain the upper 
hand in the association. Moreover B. sporogenes will fall into 
a commensal adjustment with some other organism such as B. tetani, 
B. terti/us, B. cochlearvus or vibrion septique, lending its putrefactive 
and proteolytic characters to these types and thus leading to great 
confusion. These mixed cultures will keep a consistent appearance 
in regard to type of colony and cultural reactions over long periods, 
and it is. only by ringing the changes over a wide range of media and 
conditions that the observer has reason to suspect the composite 
character of the strain which he beheved to be pure. 

This difficulty of freeing anaerobes from one another is not confined 
to B. sporogenes, but there is no doubt that it is the most frequent 
and most persistent intruder. The tendency to grow in mixed 
colonies, the power of cryptic proliferation within the characters of 
another species without being detectable, and the capacity of mutual 
adjustment possessed by anaerobes and by B. sporogenes in particular 
are at the root of nearly all the more serious confusions and dis- 
crepancies in the literature of this subject. 



38 

Animal reactions. In general B. sporogenes cultures are not lethal • 
for laboratory animals in doses up to 4 c.c. Some strains are, how- 
ever, capable of producing a putrid perforating gangrene of the limb 
injected and a small number of strains are described as being defi- 
nitely lethal in doses of about 1 c.c. and upwards. _ 

The presence of B. sporogenes in combination with B. welchii 
quite definitely enhances the pathogenicity of the latter and pro- 
duces a mixed putrefying type of gangrene closely resembling the 
putrefying gangrene of wounds which was of frequent incidence 
in the earhest days of the war. B. sporogenes added to a sub-lethal 
dose of B. welchii produces a virulent and rapid type of gangrene in 
experimental animals. In this condition an extensive oedematous 
swelling of the limb is observed a few hours after inoculation. 
The skin takes on a shiny bhstered appearance, and a curious 
green colour. The hair is generally shed over the bhstered parts. 
Death may supervene at this stage. If the animal survive for 
twenty-four hours the skin over the oedema becomes sodden and 
perforates, with the production of a foetid sloughing gangrene of the 
limb. The animal may die after the gangrene has set in or it may 
gradually recover even in cases where a considerable amount of 
tissue has become necrosed. The post-mortem examination of 
a guinea-pig, dead from a mixed infection of B. welchii and B. 
sporogenes shows the following features. The skin of the abdomen 
is soft and greenish, and the hair is easily detached if it has not 
already come out before death ; there is an extensive fluid oedema, 
blood-stained and oily in appearance, which extends as a rule 
over the whole abdomen up to the axillary region. Haemorrhagic 
patches are sometimes to be observed in the subcutaneous tissues 
remote from the lesion. The site of the inoculation is oedematous ; 
the tissues are friable and are dark in colour. There is some gas 
formation ; the odour is foetid and distinct from that produced by 
B. welchii alone. The parietal peritoneum is purple and the blood- 
vessels are injected. Lungs, liver, and spleen are unchanged in 
appearance. 

The injection of B. sporogenes with vibrion septique gives less 
certain results, but the general effect is the same. 

B. sporogenes therefore cannot be neglected in the gangrene 
syndrome though it may be difficult or impossible to gauge its effect 
or importance in any given case. 

Non-specific toxic products of B. sporogenes. The growth of 
several of the non-pathogenic anaerobic bacilli, in particular that of 
B. sporogenes, is accompanied by the production of toxic substances. 
These are for the most part of a non-specific nature and appear 
to be formed as the result of the breaking down of the food 
material. _ Without doubt, similar substances are produced by the 
pathogenic anaerobes in addition to the formation of specific toxins. 
The exact relationship of these two kinds of products is not very 
clear, but the non-specific products lead to much of the confusion 
concerning the presence of the specific bodies. In certain instances 
their presence together with the great lability of the specific toxins 
masks more or less completely the presence of the latter. 

The demonstration of the non-specific toxic substances is best 



39 

made by growing the bacilli in a meat medium, as suggested by 
Besson. This is prepared by placing minced meat in a flask 
with enough water to cover it. Several cubic centimetres of soda 
(1 per cent.) are added and the medium is then sterihzed at 115° C. 
for 15 minutes. If this is inoculated before it is completely cold the 
anaerobes will grow quite well if the cotton-wool plug be replaced 
by a rubber stopper. 

The maximum toxicity seems to be developed about the sixth 
day, after which it gets rather less. The fluid is decanted off the 
meat, the juice expressed, and the two fluids mixed and filtered. 

Toxicity. The toxicity of such a fluid varies considerably, but as 
a rule from 1 c.c. upwards injected into the peritoneum of a young 
guinea-pig will produce almost immediate signs of intoxication 
which are not infrequently followed by a fatal issue. If death does 
not occur in the course of a few minutes the animal usually recovers 
completely. After the. injection of a lethal dose the animal imme- 
diately becomes restless, jumps about, jerking its head upwards in 
a pecuhar manner, then becomes convulsed and falls over on its side. 
The symptoms observed are therefore quite distinct from those 
which follow the intraperitoneal injection of fatal doses of the true 
anaerobic toxins. 

Nature of the poison. Dale and Barger stated that the poisonous 
substance present passed over when the media was distilled. An 
analysis of the distillate convinced them that they had to deal 
with a volatile poison probably of the nature of an ammonium base. 

A series of experiments by Mcintosh and Pildes suggested that 
some sulphide — probably ammonium sulphide — played an im- 
portant part ; the presence of sulphides in the cultures can readily 
be demonstrated, and the injection of weak solutions (0-5 c.c. of 
a J per cent, sol.) of ammonium sulphide produces symptoms of 
immediate intoxication and even death. 

The influence of the growth products of B. sporogenes upon the 
action of the toxin of B. welchii has been studied in mice (W. E. 
Bullock and Cramer '(1 and 2)). When a mouse has been injected 
with a sub-lethal dose of sterile toxin of B. welchii together with sterile 
filtrate from a culture of B. sporogenes, in an amount of either 
substance insufficient of itself to produce symptoms, the animal dies. 

Specific toxin. A true soluble toxin has not as yet been demon- 
strated for B. sporogenes. 

Agglutination. Agglutinins can be produced in the blood of 
rabbits. So far as is at present known strains of B. sporogenes can 
be divided into at least two serological groups by means of this test. 

(ii) B. parasporogenes. Mcintosh. 

This organism was originally isolated and described as Type XII 
by- J. Mcintosh (1917). It agrees with B. sporogenes in all its 
characters with the exception of two. In the opinion of the com- 
mittee these differences, which are quite constant, are considered to 
be of sufficient importance to make it advisable to place the organism 
in a separate species and the name parasporogenes is adopted. 

B. parasporogenes differs from B. sporogenes in the nature of its 



40 

colony ; in agar shakes it is lenticular or slightly irregular in shape 
but not woolly. 

The second feature which distinguishes B. parasporogenes is the 
production of specific agglutinins. In morphology and in cultural 
characters and in all other respects the two organisms resemble 
each other closely. 

(iii) B. tertms. Henry, 1917 ^ (Eodella's Bacillus III) (von Hibler's 

Bacillus IX). 

B. tertms is a saccharolytic oval endsporing anaerobe. It is found 
both in early and in late wounds and may be present in severe cases 
of gas gangrene, or in ordinary wound infections. 

Morphology. B. tertius is sluggishly motile. The rods are slender 
and about 3 to 5 microns in length, they are Gram-positive when young 
(12-18 hours), but soon tend to become Gram-negative especially 
in fluid media. Spores are readily formed, and as stated, are oval 
and strictly terminal. The morphology of the organism is best 
studied in cultures 24 to 36 hours old. 

Cultural reactions. B. tertius requires only a moderate degree of 
anaerobiosis for growth. 

Surface colonies are small (1 mm.) and are rounded or crenated ; 
when young (24 hours) they are delicate and almost transparent ; 
examined under a hand lens by transmitted light they have a char- 
acteristic greenish-blue iridescent appearance not at aU unlike 
a small opal. On certain media they become more opaque and 
slightly granular when incubated for two or three days. 

Colonies in agar shakes are small, lenticular or irregular in shape 
and do not show branching filaments. 

Meat medium : Pink colour ; gas ; no blackening ; no digestion. 

MUk : Acid clot in 3 to 6 days ; some gas. 

Coagulated serum : Not liquefied. 

Gelatine : Not liquefied. 



Substances fermented. 


Substances not fermented, 


Glucose 


Glycerine 


Laevulose 


Inulin 


Galactose 


Dulcite. 


Maltose 




Saccharose 




Lactose 




Mannite 




Salicin. 





(iv) B. cochleariMs. Douglas, Fleming, and Colebrook (3). 

This organism was first described as type III C. by Mcintosh 1917 
_ It IS an endsporing bacillus very frequently found in wounds'. 
Like ii. tertms it may be present both in acute early infections and 
m late septic cases. It is frequently associated with B. tetani. 

mi^rltg'armS J:Ss'"'" '^ ^^"^ '^"^"^^ '* ^^« ^'^^ '"^^ --* — °" 



41 

Morphology. B. cocMearius is an actively motile slender rod ; 
the length is variable, and there is a tendency to give up the stain 
in Gram's method. The spores are strictly terminal and are 
oval when fully developed, giving the organism a spoon-shaped 
appearance, hence the name cochlearms. There are stages, how- 
ever, during their growth in which the young spores may be 
almost spherical. In perfectly pure cultures on ordinary media 
spores are not readily formed. The morphological features can be 
well observed in an alkaline digest broth containing fresh tissue, 
as this medium is particularly favourable to the formation of 
spores. The examination should be made after 24-36 hours' in- 
cubation. An impure culture containing a slight admixture of B. 
sporogenes gives a vigorous growth of B. cochlearius with active 
sporulation — an instance of stable microbie association between two 
organisms. Such a culture resembles B. putrificus (Bienstock) 
and has frequently been designated by this name. It is proteolytic, 
digests meat and inspissated serum and gives off a putrid odour. 
It is often extremely difficult to dissociate B. cochleariiis from 
B. sporogenes in such a mixed culture. 

Cultural reactions. In pure culture B. cochlearius is one of the less 
vigorous anaerobes. Surface colonies can be obtained upon serum 
agar plates or serum agar slopes under good conditions of anaero- 
biosis. They are delicate, glass-clear droplets, sometimes with 
faintly crenated edges. When colonies show minute opaque spots 
or opaque centres or have rootlets penetrating into the medium 
it is a sign of admixture with some other anaerobe, probably B. 
sporogenes. Colonies in agar shakes are lenticular in shape. 

Meat medium : Very little change in the colour. There is no digestion 

and very little gas. There is a characteristic odour 
which is not putrid. 

jyClk : Scanty growth ; no change in medium. 

Coagulated serum : Not liquefied. 

Gelatine : Not liquefied. 

None of the following substances are fermented : 

Glycerine Maltose Mannite 

Glucose Saccharose Dulcite 

Laevulose Lactose Salicin. 

Galactose Inulin 

Agglutinins can be produced in rabbits by the inoculation of 
washed baciUi. 

The inoculation of living cultures of B. cochlearius intramuscularly 
into guinea-pigs in doses up to 3 c.c. does not produce pathogenic 
results; 

B. cochlearius is distinguished from B. tertims by its failure to 
ferment any of the sugars, by its motiHty, characteristic odour, and 
the production of agglutinins in the blood of rabbits. 

(v) B. tetanomorphus. 

This is the organism described by Mcintosh and Fildes as ' Type IX, 
B. pseudotetanus '. It closely resenibles B. tetard in its morphological 



42 

appearance ; but differs in cultural characters, its power to ferment 
maltose and glucose, the absence of liquefaction in gelatine and in 
its, inability to produce the specific toxin. It is frequently found 
in wounds, especially in late infections, and is often present along 
with B. tetani. 

Morphology. B. tetanomorphus is highly motile ; the rods are 
Gram-positive. The spores are spherical, terminal in position, and 
are formed in all media in the course of 24-36 hours. 

Growfhreactions. B. tetanomorphus grows best in an alkaline medium. 
When a sample of wound exudate containing B. tetanomorphus is in- 
cubated in meat medium the organism makes its appearancein numbers 
about the 3rd to the 5th day, that is to say considerably later than 
B. sporogenes. The conditions produced in the medium by the 
growth of B. sporogenes are favourable to B. tetanomorpJius. 

Surface colonies can be obtained upon serum agar slopes and 
plates. They are deHcate, flat, and have a slightly crenated edge. 
There is a tendency to grow in a continuous surface film. Colonies 
in deep agar are small in size and irregular in shape but are not 
woolly or branched. 

Meat medium : Pink colour ; gas ; no digestion or blackening. (This is 
the best medium for the growth of B. tetanomorphus.) 
Milk : No change ; scanty growth. 

Grelatine : No liquefaction. 

Substances fermented. Substances not fermented. 

Glucose Laevulose Lactose 

Maltose. Galactose Inulin 

Saccharose Glycerine 

Salicin Dulcite. 

Ma unite 

Agglutinins for B. tetanomorphus can be demonstrated in the serum 
of rabbits inoculated intravenously with the bacillus. B. tetand' 
morphus is not pathogenic for guinea-pigs. 

(vi) B. aerofetidus. Weinberg and Seguin (8). 

Morphology. This is a small slender baciUus about 3-5 fi in length ; 
it has a tendency to be Gram-negative. Motility is shght. 

Growth reactions. Spores are not readily formed ; they are 
subterminal. Surface colonies on serum are round and trans- 
parent and may attain to 1 or 2 mm. in diameter after 24 hours' 
incubation. 

Deep colonies are small and irregular. 

J^eat : Putrid odour ; change of colour, first reddening and 

then blackening. 
Milk medium : Clot and gas in 24-48 hours, later, a certain amount of 

digestion. 
Coagulated serum : Putrid odour ; liquefaction. 
Gelatine : Liquefied. 

Alkaline egg broth : Soft clot which may be partially digested. 

This organism is not pathogenic in pure culture for guinea-pigS, - 



43 

Fermentation. There is some divergence of opinion as to the 
sugars fermented by this organism. Glucose, maltose, and lactose 
were, according to Mcintosh, the only sugars fermented, whereas 
Henry (2) found that laevulose and salicin were also affected. 

(vii) B. bifermentans. Tissier and Martelly. 

This bacillus was first described by Tissier and Martelly in 1902 
under the name of B. bifermentans sporogenes. It received the name 
of bifermentans owing to its being the first anaerobe in which a 
capacity to split both proteins and sugars was definitely recognized. 

B. bifermentans is occasionally found in acute cases of gas gangrene, 
as well as in late septic cases. 

Morphology. In the non-sporing state it is a stout, non-motile 
rod closely resembling B. welchii. Individual elements may be very 
short. Chains are frequently formed. Spores are readily formed 
in all media ; they are central or subterminal. 

Cultural reactions. Surface colonies are- round or crenated. Deep 
colonies are lenticular or irregular but without filamentous out- 
growths. 

Meat medium : Blackened and digested ; putrid odour ; gas. 

Milk medium : Casein is precipitated and later digested. 

Coagulated serum : Liquefied. 

Gelatine : Liquefied. 

Alkaline egg brotli : Soft clot wMcli is later digested to a variable extent. 

Fermentation. Glucose, laevulose, and maltose are alone fer- 
mented. B. bifermentans is non-pathogenic in pure culture for 
guinea-pigs. Agglutinins can be produced in rabbits inoculated 
with the organism. 

(viii) B. putrificus. Bienstock (2). 

This organism was first described by Bienstock in 1899 under the 
name" of B. puirijicus. He and many subsequent writers consider 
that it is strongly proteolytic. Experience during the war has shown 
that many of the cultures of B. putrificus so-called are really mixtures 
of B. cochleariMS or B. tertius with B. sporogenes. 

The following is a brief synopsis based upon the incomplete 
descriptions of Bienstock, Salus, Tissier and Martelly, Metchnikoff, 
Wiircker, Weinberg and Seguin, and others. 

B. putrificus is a slender Gram-positive rod, the spores are oval 
and strictly terminal giving the drumstick appearance described by 
Bienstock. 

The cultures are strongly proteolytic; gelatine and serum are 
liquefied and milk digested with or without the formation of a clot. 

The organism is described as being non-pathogenic (Bienstock, 
Tissier and Martelly). 

(ix) B. sphenoides. Douglas, Fleming, and Colebrook (3). 

This is a small motile bacillus. Gram-positive when young but 
rapidly becoming Gram-negative. In the non-sporing state it is 



44 

fuisiform in shape and often arranged in pairs placed end to end not 
unlike a Hoffmann's pseudodiphtheria bacillus. _ _ 

The spore when first formed is subterminal, but as it mcreases m 
size its position becomes terminal ; it is perfectly round and is 
of large size, being considerably broader than the body of the 
bacillus. As the spore reaches its maximum development the 
body of the bacillus assumes a wedge shape the point being situated 
at the end opposite to the spore. It was the appearance of the 
bacillus at this stage which suggested the name sphenoides. As 
the body of the bacillus degenerates in older cultures, the orgamsm 
becomes shaped Hke a drumstick, but the end opposite the spore 
always remains pointed. 

Cultural reactions. Surface colonies on agar attain to a size of 
about 1 mm. in diameter. They are round and usually have smooth 
edges ; occasionally, however, slight irregularities are seen. 

Meat medium ; A little gas is formed ; no change is produced in the 

colour. 
Milk medium : Acid is formed and occasionally a soft clot. 

Coagulated serum : No change. 
Gelatine : Not liquefied ; good growth. 

Broth : The growth is profuse, an even turbidity being 

produced. 

The fermentation reactions of this organism appear to be some- 
what variable. Glucose, maltose, galactose, lactose, and saHcin are, 
however, fermented by the three strains isolated. Two of the 
strains ferment mannite, saccharose, dextrine, and starch in addition. 

Incidence. It was isolated in pure culture from three cases and 
was observed in two others in a total of sixty-one (Douglas, Flencdng, 
and Colebrook (3)). The majority of the patients had had gas 
gangrene. 

(x) B. hutyricus. 

B. hutyricus is a non-pathogenic bacillus rarely met with in 
wounds. In 1861 Pasteur (1) described, under the name of ' Vibrion 
butyrique ', an anaerobe which was capable of producing butyric 
fermentation. This organism of Pasteur is probably identical with 
the Clostridium butyricum of Prazmowski, the Bacillus, amyldbacter 
of Gruber and with the mobile ' Buttersaurebacillus ' of Grassberger 
and Schattenfroh (1 and 2). 

The literature deahng with the organism is full of confusion and 
it is almost impossible at the present date to decide exactly with 
what organisms the above authors worked. The characters of what 
the committee consider to be B. hutyricus are detailed below from 
the study of a single strain isolated by Fleming from a wound. 

Morphology. It is a small Gram-positive motile bacillus. The 
spores are small and oval in shape, usually, central though occa- 
sionally subterminal in position. The most frequent appearance 
is that of an elongated Clostridium. 

Cultural reactions. Surface colonies on serum-agar are small, 
flat, irregular in shape, greyish in colour and semi-transparent. 
Shake colonies are small irregular lenticular masses. 



45 

Meat medium : No digestion ; some gas. 

Milk medium : A firm acid clot appears in 24 tours. 

Coagulated serum : No change in the medium. 

Gelatine : Not liquefied. 

Fermentations. Glucose, maltose, lactose, saccharose, and starch 
are fermented. 

(xi) B. multifermentans tenalbus. Stoddard (3), 1919. 

This is a non-pathogenic organism recovered from a case of gas 
gangrene. It is an infrequent organism but is of interest as it shows 
a morphological resemblance to vibrion septique. 

Morphology. It is a Gram-positive motile bacillus. The spores are 
central or subterminal. When grown upon egg or serum media, 
navicular types, filaments, and swollen club-shaped individuals are 
formed. 

Cultural reactions. Surface colonies, cultivated under anaerobic 
conditions, may attain a large size (3 to 5 mm.) ; they are round, with 
slightly irregular edges ; after incubation for several days they 
become white and opaque and rise up considerably above the surface 
of the agar. In glucose-agar shakes the colonies are white and 
opaque, irregular or lenticular in shape with projecting outgrowths. 

Meat medium : G-as ; no obvious digestion ; no blackening ; the odour 

is not putrid. 
Milk medium : Acid and clot. 

Coagulated serum : No change. 
Gelatine : Not liquefied. 

Substances fermented. Substances not fermented. 

Glycerine Glucose Mannite 

Maltose Saccharose Dulcite. 

Lactose Inulin 

Eaffinose Salicin. 

4. Classification of the Anaeeobic Bacilli found in Wounds. 

The characters available as a basis for the classification of the 
anaerobic bacilli can be grouped under the following headings : 

Morphology. 

Cultural characteristics. 

Serological reactions. 

Production of characteristic toxin. 
Each of these categories deals with a different aspect of the organism, 
and owing to the difficulties inherent in the study of bacteria no one 
of these aspects taken by itself furnishes the data for a satisfactory 
classification. Since, therefore, a combination of these several 
different series of characters has to be- relied upon, the order and 
relative value of the different criteria must be considered. 

(i) Morphology. 

The first subdivision is based upon morphology and the next upon 
cultural reactions. The production of characteristic toxins is a quality 



which is valuable in classification where available, and serological 
reactions form a final series of distinctions which may or may not 
be of specific value. 

The anaerobes may be divided into three groups upon their 
morphology : 

(a) Organisms with oval spores which are central or sub-terminal 

in position. 
(h) Organisms with oval spores which are strictly terminal in 

position, 
(c) Organisms with spherical spores which are strictly terminal 
in position. 

(ii) Cultural Characteristics. 

Further divisions can be made upon cultural reactions such as 
the capacity for decomposing protein, carbohydrates and alcohols. 

All organisms have a certain capacity for attacking proteins and 
splitting sugars, but these characters have to be sufficiently marked 
to be readily appreciable under the test-tube conditions of artificial 
cultivation and have to be somewhat arbitrarily defined for the 
purposes of classification. 

In the case under consideration proteolytic characters are judged 
by the capacity of liquefying coagulated serum and of gelatine. 
In working with the anaerobes the gelatine test has been used in 
the special manner first employed by Eiineberg and differs from 
that generally employed in bacteriology. Owing to the unsatis- 
factory amount of growth of the anaerobes obtained at 25° C, the 
inoculated gelatine stab is incubated at 37° C. for forty-eight hours 
under anaerobic conditions, thereafter it is cooled by being placed in 
a beaker of cold water and the reading taken. In the case of the 
saccharolytic organisms the fermentation is recognized by a definite 
capacity for producing acid or acid and gas in sugar-containing media. 

Subsections based upon these characters can be arranged as follows : 

A. Organisms showing proteolytic and saccharolytic properties. 

B. Organisms showing proteolytic properties only. 

C. Organisms showing saccharolytic but no proteolytic pro- 

perties. 

D. Organisms showing no obvious proteolytic or saccharolytic 

properties. 

Subsection A would be subdivided into : 

A 1. Organisms predominatingly proteolytic with, however, a 
restricted but definite capacity of fermenting certain 
carbohydrates. 

A 2. Organisms predominatingly saccharolytic possessing, how- 
ever m addition, slight proteolytic properties as shown by 
the liquefaction of gelatine. 

The criterion in accordance with which an anaerobe is here classified 
as a predominatmgly proteolytic organism is the capacity to produce 
liquefaction of coagulated serum. r j r 

The accompanying table shows the arrangement of the organisms 
classmed m these terms. 



47 





■j§^ 








1.1 






§ 


:^§ 


% 


'^j 


o 





a 

03 

I 






f^ 



ll-i 
III 

W o O 

« S t 



^ 

-« 

^ 



^ s i i 



«t 



T3 
ID . 

ccCS 



Si, eo 

O ^3 



cqoj 



fl^ 



^ ^ © ca 




n^uaqcqa; 



s 




-g 


r 


te 


05 


g 


^f 


s^ 


&< 




h 






eg 


1 


is 


.fi 








y 


p^ g 


ttf) 


PM 







§ ? g i 5 



a 



I 



s 



£ 






eg 


K, 


1 


m 






















1 
o 


CD 



•^ a 1* h 

? p-ap . 

^ o 53 a S 



■s 




o 




o 






TJ 


p< 


O 


OQ 


S 



=«-« 2 



60 



© d 2 0) .H 
gi3 a CI 3 o 



CLi fc; qP+3 ^ is 
K^ 00 -? «M f4j ^tt 

a t^-S S ° o - 

S -S" S 2 rt 'o 



H in 



ClO 

o 

I 

a, 
o 






d 
s o 



1== 



.id ■« "S 

g •• rt 3 



•fed 



^ 



03 ' 



© . 



DQ 









:S 



0.2 
.2 c3 « 

■>S -a. ^ SP 

p - 
;h DO 






ai 



bo 

a 



13 -a 

•fed 



g. 



O 'o 

-2^2 i 



d M fl ^ 

^ ^ 2-§ 



;>! t 



•S fe 



^S?- 



P^ O 
O' ' 



^ ^^ 



O eg 

-2 a 

°-|8 






-5 



I 



|s-,'d 2 
^g S' " « 1=^ 



"O A 2 S A 



o -Es a ^ a 



ee o c 



s 



i|2 















+ 



+ 



-S) 



1 






02 




s. , 


.&> 


• 


M O 


•» «0 








II 






1 ^i 






«;«» 


«j "~ 







ta-^ ea o n ca n "*? 






BJ'S 

flag 



•Be o tta S -a 

.a a s "^ s 2 3 

tH ■** t-< GQ H ta 









■E.S «2 .a 



"ia: 



is 



2<fla'SS'3§-fc 



c3 a 

s 

cc bo 

■u 




M 



( ■ CD +3 CD rt 

•3 lis I 

op c5 'C O +3 

° l>— g 



9 S3 " ^ a 
d a 



a^ 



■H=H. 



o at 

- o 



COP 

^^!z;al 



3T3 









313 



S 



§£|.aa 
=^-al'^^ 

p o 



Opt^ S fl S 3 






a 






^ga--§ 



^1 



o .S o *w 



11 



2 3 

3 O a 
— . cS 



1 sg, 



O.S 

•I :1s 

.^ P so 

o .S = 
S S S3 



•S-^ 3 



B -e if 

^ 3 a 



s & 






^ 






•s.S1§ 



2 a B,§ Ml 
o o g-s a 



g §3.3.2., 

a c!_ 
S"J3 as 






••gg 



c3 






P5a 



'H ■■ ■ 

es a 

3 53 -M 



C3 



oSlo 



T3 
C3 



is a- 









b'^'f^ 



IIP 

o ^ 

!i'§^ a 



3 « 
° 60 
T3t3 

.a § 



oqS a o g o 3 



^ -^ if 



I « o o o a 3 

„. dg'•ag||=^^• 
is Sg"a*3 

3ro"S3SS60t3 






CQ 



ilil 






^t 



_^ a 
3 s?-^ 



CS' 



.3 



1^ 




CO 


&> 




o 


-d n 


01 > 


■s« 


a ■" 

n <D 


«^ 


m 


S J2 



& " 3 

go g 

" g<aj a 



t3^t3 



ft 



CO 

O .3 OJ 



"g£- 
S,a l>^ "3 

'"^H.s a 



+ 



05 ^ 






J 



o 
o 

A 
ft 

S 

O 

M 
OQ 
1^ 

m 
o 
« 



IS! 

o 



8 
w 

o 

pq 

EH 




Ota • 9 iS ■£ 

" ^ CJ to „ ^H te 

M-S boo >trtC5 



EjU 



a p So 



as i6 



CO 



o S« 

.sag ^s 

.S o 43 53 o 



S ^ o 
I'll 




•" a 
•ai 

sob 



a" n p 03 







.si ca 



.1 s.« 



11 
1== 



3* 



3 



I 



e 



3 2 

o so 



j3 g 
.2 c 

■a §> 

■• e 

O o4 



\ 1^ 









^ ^ § 






.2 I" 

ri' IT- 
'S ° 



s « 
o S. 



o S ■» » 
hE^ CJ O -tJ © -to 

DQ fi !S O 



. O '^ O s O 



,53 

1^ s 



s sps 

.^ '^ 3 3 dj 






s 

cs bo 

— ( a 



a 2 5-3 



§ - ~ 

BO 5) 



fSi "S ;§''§' g "3 



I -a^gs 

^ J? ^ a 5 
^ d e >i ® © 






© pQ 



-a .a 



© r£ 

es GO 






ts S o 3 «3 " g " 

(h © ^ ^-1 © ^ I- 

ag^-sa-a^^ 



^ ^^ © ^3 .-, 
i? ^ 1^ 'S .2 S 
r^ ^ P d C ™ 



oq 






+ 



+ 



I 






51 



a § 
a " 



a a 

■I 2 J 






3 k. .S"< 



■e o oJ 

is a 

fid u 



e3 T3 d bO 

m oQ S <D O 

"H •-! ,J=^ O &, 

§3 M d £? 

£ 2 » S 

S3 a bog 



^ 






^ 

^ 



lid 


So 


o 


o 


■St 


jd 






ce 


13 


Oi 


& 



fe; 



8= 



o 



-a 



^.S'i 



br Ed <» 









3T3 

cri© 









I. 



•§ 



-s 



&. 



& 



Iz; 



^ 



!l 



13 

3 



00 M 

C5 3 



o 

O CI 



o a 



Q s 

iS o 
^1 



s 






=g.a 
o a 



O ® e' 



"2 S 
2 3 

eg o 

.is e<i 



I 






So 


bo 




ff^ 




g 


■5 


J3 



^ 



5-2 
o " 

aj -*3 
'3° 



3'a 
5P3 



°.2 § 
§1 

s 3 S 

H -S S 



-3 



&04i a 



S'3 



43 ^ 

MOO 

g § o 

'o .d 






^ r 



"3^ 






1^"^ 






^■^^ 



o M ® bo o 



^ 03 ce 



a >?a 

o »^ § 

P^ tH 2 'O 




-a a 



« - So 




03 eg ^ 

o -e o . 

O 2 B 3 

3 >u (c 



c3 3 



5-S 



T3 >>Ti 
u © 3 



DO >^ ^ "^ 



!s g £^ a a 

43 ^ (D 
43 43 

to 3 ^^ 

a, S'o "r-J 

.ir S 5 s ^s 3 



-0 



DQ-! 



t3 



M 3 

■s a 
CO s 



T3 3-rt 

^ O 

.S-3 ^ 



+ 



+ 






+ 



+ 



+ 



+ 



OQ 



i 1 § -S 

03 *■" 05 



o ■ 

S 



:^ 



.D 2 



cq 



cq 



i| 



52 

(iii) Agglutination Beactions. 
At a very early period in the history of anaerobic bacteria, it 
was found that the injection of the bacilli into animals led to the 
production of agglutinins. Thus Leclainche and Morel prepared 
a serum which agglutinated vibrion septique in dilutions of 1 in 500 
and upwards, while Leclainche and Vallee (1) prepared a serum which 
agglutinated B. chauvoei in dilutions up to 1 in 3,000. These writers 
claimed that the sera produced were rigorously specific. About the 
same time the production of specific agglutinins against B. tetani 
was demonstrated by Courmont. Markoff (1911) showed that aggluti- 
nation was of considerable value for the differentiation of certain 
anaerobes, in particular of ' Eauschbrand ', mahgnant oedema and 
' Geburtsrauschbrand ' and almost identical results were obtained by 
K. P. Meyer. 

The more recent researches deaUng with the identification of the 
various anaerobic bacteria of wounds, has again directed attention 
to the question of agglutinins. Amongst those who found that the 
agglutination reactions were sufficiently specific to be of diagnostic 
value may be mentioned, Gaehtgens, Landau, McLitosh, H. C. Plant, 
Pfeiffer and Bessau, Robertson and Weinberg and Seguin. Piirth, 
on the other hand, was much less successful. The consensus of 
opinion is, therefore, that the agglutination reactions of the majority 
of anaerobes with the exception of B. welcMi and B. tertius are 
of great value. According to the above writers the test may be 
used not only for the differentiation of species but also for the 
differentiation of sub-species. ' 

Agglutinating sera are best prepared by intravenous injection 
into rabbits, surface cultures or the centrifugalized deposit of broth 
cultures suspended in saline should be used for the injection. Eabbits 
are relatively insusceptible to these emulsions and a whole serum 
agar slope may be injected. Usually three or four injections made at 
three or four days interval are necessary to produce a high titre serum. 
It is of course essential that the cultures used to prepare the 
agglutinating sera should be absolutely pure. Failure to observe 
this will lead to endless confusion and worry. 

The agglutination test is best carried out in small test-tubes or 
m Pasteur pipettes according to Wright's method, immersed in 
a water bath at 56° C. The .bacterial suspensions should be obtained 
from young cultures, well washed in saline and diluted to give 
shghtly opalescent emulsions. As a rule living bacilli are more 

Xr twotr *^°'' ^^^'^ ^^ ^''*- ^'^^^^' ^^^ *^i^«^ 

arenas foHowsf '"^ *''*' '"^ ''^''^ *^ *^' ^"^' ''^^^^^ ^^^^^obes 
1. B. welchii. 
Werner (1905) prepared agglutinating sera against four ' Gasbrand ' 

1 in 1,000. No agglutination was obtained Jn ?ross ailutin^tin^ 



was agglutinated to 1 in 40, ten strains to 1 in 2Q and ten strains 
not at all. Weinberg (1918) obtained serum from two horses which 
agglutinated the homologous strains in 1 in 500 and 1 in 2,000 
respectively, but he does not state the action of these sera on other 
examples of the organism. 

Bull (1917) mentions the agglutination and lysis of B. welchii by 
normal rabbit and guinea-pig sera and also agglutination in immune 
sera but has given no exact details of his results. Gaehtgens (1917) 
found he could prepare agglutinins to all his anaerobes with the 
exception of B. welchii. Pfeiffer and Bessau claimed to have pro- 
duced a serum against B. welchii which agglutinated the homologous 
strain in 1 in 40. This serum, however, had no effect on fourteen 
other strains of B. welchii. Mcintosh (1917) found it impossible to pro- 
duce definite agglutinating sera against any of his B. welchii strains. 

It may be assumed from the above evidence that the agglutination 
test is of no value in the diagnosis of B. welchii. 

2. Vibrion sejptique. 

As previously mentioned, this was one of the first anaerobes 
for which specific agglutinins were demonstrated. Weinberg and 
Seguin (18) in titrating a monovalent agglutinating serum against 
nine strains of vibrion septique isolated from human cases found that 
four of these (including the homologous culture) were agglutinated in 
a dilution of 1 in 1,000, eight in 1 in 600 and one in 1 in 10 only. 
The same serum agglutinated a type strain of vibrion seftique from the 
Pasteur Institute collection 1 in 500 and a stock strain of so-called 
B. chauvoei from the same source in 1 in 10 only. 

More recent research upon the agglutination reactions of vibrion 
seftique strains has shown that there are at least three serological 
types, which while producing different agglutinins are found to be a 
single group as regards their toxin antitoxin reactions. (Eobertson.) 

3. B. oedematiens. 

Weinberg (18) has prepared rabbit sera which agglutinate the 
homologous organism 1 in 100, but have no action on other strains 
of the organism. The organism is a difficult one to work with, 
because most strains tend to agglutinate spontaneously. 

B. tertius. Attempts to obtain an agglutinating serum from 
rabbits for this organism have failed. 

4. B.fallax. 

The following table of agglutination reactions is given by 
Wemberg (18). 

Strains. Serum 1 Serum 3 Serum 4 

rabbit. rabbit. human. 

1 1/500 1/100, 1/50 

2 1/50 1/50 

3 1/500 

4 1/100 

5 

None of the strains of B. Jallax dealt with were agglutinated by 
sera prepared against B. welchii, V. septique, B. oedematiens, B. 
sporogenes and B. aerofetidus, nor did the B. fallax sera agglutinate 
any of the above mentioned anaerobes. 



54 



5. B. sjporogenes. 

Metchnikoff in his original article on B. sporogenes described two 
types but did not mention agglutinins. Mcintosh (1917) by means 
of agglutination tests found at least three different types. 

The following table illustrates a series of experiments undertaken 
with four strains of B. sporogenes which were indistinguishable 
morphologically or culturally (Henry). 



Sera. 

Normal rabbit (5) 
Normal sheep (1) 
Normal horse (3) 
Mcintosh (rabbit) 
Weinberg (rabbit) 
Rabbit (against 3) 
Rabbit (against 4) 
Rabbit (against 1) 
Rabbit (against 1) 
Sheep (against 1) 
Goat (against 1) 
F8. 
T8. 
M8. 
8. 



Strains. 



SO 



50 









50 


50 








500 


500 








200 


200 


. 





4,000 


2,000 


— 


— 


4,000 


4,000 


4,000 


4,000 


— 


— 


8,000 


8,000 


— 


— 


64,000 


32,000 


— 


— 


250,000 


64,000 


— 


— 


2,000 


2,000 


50 


— 


500 


500 


— 


50 


1,000 


500 


— 


50 


1,000 


1,000 


— 


— 



F 8, T 8, M 8 and 8 were German gas gangrene sera prepared 
apparently by the inoculation into horses of mixed whole cultures 
of anaerobes. 

A glance at the table shows conclusively that the four strains of 
B. sporogenes, which were taken at random for the experiment, may 
be divided into two serological groups, each of which is distinctly 
demarcated from the other in the possession of the capacity to 
produce a specific agglutinin. 

An additional experiment with eleven other strains of B. sporogenes 
showed that of these only two came into the 1-2 group and only 
one into the 3-4 group. In the case of the remaining eight strains 
there was no agglutination. One of the latter, a strain isolated from 
a case of haemothorax was agglutinated by the patient's own serum 
and by his serous pleuritic exudate, both collected during convales- 
cence, in a titre of 1 in 2,000. 

The agglutination reaction would appear, therefore, to divide the 
B. sporogenes group of organisms into a number of sub-groups. 

6. B. tetani. 

TuUoch (1) has shownthat the strains of B. tetani he investi- 
. gated are divisible on their serological reactions into three groups. 

Summary. 

The application of agglutination as a diagnostic procedure in the 
case of the anaerobes results in a differentiation which ma^ be said 
to reveal evidence of tribal rather than of national characteristics. 

Where an anaerobe can be shown to be pathogenic, its capacity 
to produce a toxin which presents well-defined characters provides 
an unassailable basis for classification. Hence, in dealing with 
a group or collection of strains, the individual members of which 
have m common the property of producing one and the same toxin. 



65 

any further division into sub-groupa that results from the apphcation 
of an agglutination test leads to an unwarranted multipUcation of 
separate bacteriological types. In this respect the agglutination 
test as applied to the anaerobes becomes ultraspecific. 

(iv) Toxin-antitoxin Beactions. 

B. welchii, vibrion septique and B. oedematiens produce specific 
toxins, each of which has its own particular characters, and each of 
which, when inoculated into animals, gives rise to specific antitoxins. 

A well-washed emulsion of B. welchii may be introduced in 
large doses into experimental animals without producing infection. 
The combination, however, of a few washed bacilli with a sub-lethal 
dose of toxin results in the production of a rapidly fatal gas gan- 
grene. The toxin, therefore, is endowed with powerful aggressive 
properties. This aggressive action of B. welchii toxin can be 
successfully neutralized by B. welchii antitoxin, but not by other 
antitoxins. One is thus enabled to say quite definitely that if 
the toxin produced by an unknown anaerobe is completely de- 
prived of its lethal effect in animals when it is mixed with, for 
example, B. welchii antitoxin, then the organism in question is 
B. welchii. A simpler method of demonstrating the same fact 
consists in mixing a liquid culture of the organism with varying 
doses of antitoxin before inoculation into animals. If the infection 
be inhibited with B. welchii antitoxin then the organism is B. welchii. 
Or again, if it can' be shown that the inoculation of an animal with 
a B. welchii antitoxin renders it passively immune to infection with 
a particular organism, then one can be certain that the organism is 
B. welchii. 

The same is true for vibrion septique and for B. oedematiens ; the 
toxin of each gives rise to a specific antitoxin which is capable not 
only of neutralizing the corresponding toxin, but also of inhibiting 
the infection , of experimental animals by whole cultures of the 
organisms in question. 

This method of arriving at an exact diagnosis of the nature of an 
unknown pathogenic anaerobe is applicable both to pure cultures 
and to mixed cultures. 

The question of toxin-antitoxin reactions is dealt with in detail 
in the serological section of this Report. 

5. Biochemistry. 

It is to Spallanzani in 1776 that we owe one of the first, if not the 
very first, real inquiries into the behaviour of hving organisms in 
the absence of oxygen. 

In a set of experiments which are a model of experimental technique, 
he exposed to a vacuum animalculae in small tubes, sealed at one 
end. Similar control tubes were left in the air. The experiments 
were continued for 24 days, at the end of which time he observed that 
the animalculae in the anaerobic tubes were dead, while those 
exposed to air survived. 

Other organisms were not so susceptible to privation of oxygen, 
for he recounts that certain animalculae lived 35 days in vacuo, 



56 

and others died in 14, 11, and 8 days. Some were only able to with- 
stand oxygen want for 48 hours. 

He was able to observe that motility and the power of reproduction 
was not totally inhibited by a vacuum, but he believed that depriva- 
tion of air eventually stopped both these processes. 

He also showed, in the case of some organisms with which ho 
worked that a partial vacuum was favourable to their development. 
He thus anticipated by more than a hundred years the work_ of 
subsequent investigators on oxygen minima. This astonishing 
investigation appears to have been entirely overlooked by aU the 
later workers in this field. 

(i) The Mechanism of Anaerobiosis. 

Pasteur in 1861 recognized that certain organisms were able to 
exist and multiply in the absence of free oxygen. This belief in 
the possibility of protoplasm continuing to function under strictly 
anaerobic conditions has always encountered a certain amount of 
scepticism, but it may be said that the arguments advanced against 
the original Pasteur conception do not carry great weight with later 
workers. 

The matter of oxygen tolerance is one of a large number of grada- 
tions in the animal kingdom, and is therefore analogous to many 
other processes connected with life and growth. At the upper end 
of the scale one encounters animals which exist normally in an 
atmosphere containing twenty volumes per cent, of oxygen, but 
which can tolerate for long periods pressures of oxygen up to three 
atmospheres, or concentrations of oxygen as low as seven per cent. 
Above and below these pressures the animal rapidly succumbs. 
Other animals, such as Spirostomum, one of the ciliates, are killed 
by the partial pressure of the oxygen in the atmosphere, but exist 
normally at one-third of this pressure. 

With bacteria all gradations of oxygen tolerance are foimd. 
Some species cannot live without a certain partial pressure of oxygen, 
while others, the facultative anaerobes, are capable of hving in or 
without the presence of this element. The obligate aerobes are 
distinguished by the fact that continued absence of oxygen leads 
eventually to impairment of function and cessation of growth. 

Prom these forms, the transition to one which does' not require 
free oxygen for its existence is a natural one. In this case oxygen 
acts as a poison. For this toxic action we have an analogy, higher 
up in the scale, where a certaiu concentration of oxygen is necessary 
for hfe but a greater one is fatal. 

Much careful work has been done in establishing the fact that 
anaerobic bacteria hve in the complete absence of oxygen. To this 
may be added the consideration that no amount of reserve oxygen 
present m the cell would suffice to carry on such an intense metabolism 
as IS often found during the growth of these organisms. 

Whether the optimum growth of anaerobic bacteria takes place in 
complete absence of oxygen is still a matter for discussion. The 
elaborate series of investigations which were undertaken by Fermi 
and Bassu would lead one to believe that a certain small partial 



57 

pressure is necessary for intensive growth. These authors made 
many experiments to determine the conditions under which the 
strictest anaerobiosis was obtainable. It seems probable, judging 
from their experiments, that. the ordinary anaerobic technique never 
gives a medium completely free from oxygen. Another important 
point is disclosed by their research. While it is possible to obtain 
growth in the complete absence of oxygen, this is not so abundant 
as when a very small amount of the gas is present. Whether the 
small amount has a stimulating action, such as is seen with other 
poisons, is not known. 

In the case of some anaerobes, the oxygen tension at which growth 
takes place is very low. Chudiakow has shown that in the case 
of Bactridium huiyricum the limits are between and 15 mm. 
At 10 mm. of oxygen there was already a retarded growth. By 
way of comparison he gives an interesting account of the behaviour 
of a strict aerobe, B. suUilis. This bacterium can grow at 10 mm. 
of oxygen, but ceases to do so at 5 mm. 

The question of ' symbiosis ' in the growth of anaerobes is un- 
doubtedly an important one in soil infected wounds. A certain 
amount of work has been done, which is not for the present purpose 
very satisfactory. The subject is a difficult one but ought to yield 
most useful results if further investigated. 

One of the earlier ideas regarding ' symbiosis ' was that aerobes 
abstracted oxygen from the medium, and so made conditions 
suitable for anaerobic growth (Pasteur (2 and 4)). 

Kedrowsky claimed that aerobes elaborate certain substances 
favouring the growth of anaerobes, and that the absorption of 
oxygen by the former is not an important process. Scholtz and 
von Oeitingen criticized Kedrowsky's results and showed that no 
special substance was necessary. The idea of Pasteur was modified 
in such a way that von Oettingen came to the conclusion that the 
aerobe acted as does the presence of reducing substances in a medium. 
The anaerob'e has an incomplete capacity for oxidation. When it is 
in company with a micro-organism which is eager for oxygen, the 
latter acts synergetically and promotes growth. The action of the 
aerobe is therefore a secondary one. Von Oettingen's technique was 
ingenious but not beyond criticism. 

The question of the smallest quantity of oxygen necessary for 
growth was taken up by Beijerinck, who concluded that a very 
small amount of free oxygen always took part in anaerobic growth. 
Using his experiments as a basis, he proposed to replace the terms 
anaerobic and aerobic by micro-aerophilic and aerophilic. 

The experiments of Kiirsteiner, who used B. phosphoreum (Cohn) 
Molisch, as an indicator for free oxygen seem to point to the con- 
clusion that even the smallest quantity of oxygen is not at all essential 
for the vigorous growth of strict anaerobes, and in this respect his 
results confirm the earlier experiments of Chudiakow. An excellent 
critical review of this much debated question is given by Omelianski, 
who does not believe that free oxygen is necessary. Omelianski 
also points out the important influence of the composition of the 
mediuin on the degree of anaerobiosis necessary for the growth of 
individual organisms. 



58 

A full discussion on the mechanism of anaerobiosis will be found 
in the papers of Fermi and Bassu, von Oettingen, Beijerinck, OmeTian- 
ski, and Kiirsteiner (cf. Bibliog.). 

The chemical reactions which take place durmg the growth of 
anaerobes are manifold, and may roughly be divided into attacks 
on carbohydrates and the decomposition of proteins. In the 
fermentation of carbohydrates certain compounds are formed of 
great theoretical and technical importance, acetone, alcohols, 
especially butyl alcohol and acids of the fatty series. The acid 
encountered in dealing with anaerobes found in wounds is chiefly 
butyric acid. This is formed according to the reaction 

0,11.^^0,= C3H7COOH + 2H2 + 2CO2. 

This reaction may, however, have many by-products, and quite 
appreciable quantities of other fatty acids are formed at the same 
time. Non-volatile fatty acids such as lactic acid are formed in 
quantity. 

All the micro-organisms so far examined conform in general to the 
above scheme. The attack on albuminous compounds is a much 
more complicated affair, and individual differences between the 
anaerobes are more manifest. 

Eoughly speaking they all exert a certain amount of tryptic 
action and break down proteins to amino acids and ammonia. 
The so-called saccharolytic organisms B. welchii and vibrion septique 
have not this power to any great extent, and therefore one finds 
the particles in a culture such as cooked meat retaining their original 
appearance for long periods of time. An analysis shows, however, 
that a certain amount of digestion has taken place, and it is obvious 
that such large quantities of gas as are found in cooked meat cultures 
must have their origin from some type of protein cleavage. 

Apart from purely hydrolytic action whereby albumoses, peptones, 
and amino acids are formed from the original protein molecule, 
other processes take place, and it is noteworthy that both reductions 
and oxidations may be occurring at the same time. It is due to 
these simultaneous chemical reactions that the whole chemical 
mechanism of anaerobic metabolism is rendered so difficult to 
disentangle. 

The chemical processes may be divided into three groups (Ellinger). 

1. Eeduction. Eeplacement of the amino group by hydrogen. 
In this way phenyl, oxy-phenyl, and indol propionic acids are formed 
from phenyl alanine, tyrosine, and tryptophane. The general type of 
equation is the following : 

E.NH^COOH-i-H^ = E.H.COOH + NH3. 

2. The elimination of carbon dioxide with the formation of 
amines : 

E.NH2C00H= CO2 + E.H.NH2. 

3. The oxidative degradation of amino acids whereby the chain 
IS shortened. At the same time the amino group is removed. For 
example, phenyl acetic acid may be formed from phenyl alanine 

E.CH,.CHNH,.COOH + 0, = E.CH,COOH-t-CO, + NH,. 



59 

Apparently in these three groups of chemical reactions the reductions 
take place -with compounds having an aromatic nucleus. The 
oxidations are principally confined to the aliphatic amino acids. 

Very little indeed is known regarding the properties of individual 
bacteria for producing definite types of reactions. Experiments to 
determine this must be done by bacteria of undoubted purity of 
strain on definite chemical compounds. In the case of anaerobes, 
few of the experiments which have been hitherto performed will 
bear scrutiny. 

The chemical mechanism by which the necessary energy is pro- 
duced for anaerobic metabolism is not an economical one. E. Buchner 
showed that for B. subtilis it was necessary to use up 4-7 times as 
much material to produce a given amount of heat when grown 
under low tension as when the organisms grew in the presence of air. 

In ordinary aerobic metabolism, carbon dioxide is usually the 
result of the combustion of the carbon atom, the most efficient heat 
producing reaction into which carbon enters. In anaerobic meta- 
bolism, associated with the formation of this gas, the carbon dioxide 
is derived from the breaking down of a carbon chain with the simple 
elimination of carbon dioxide from a carboxyl group, or the formation 
of an acid of lower molecular weight, or the production of an amine 
or ah alcohol. These are reactions associated with comparatively 
little heat set free. 

(ii) The Biochemistry of certain micro-organisms found in Wounds. 

A certain amount of chemical work has been done during the war 
on both the obligate and facultative anaerobes found in wounds. 
Greater attention has been paid to the former class. The main results 
are to be found in the papers of Wolf and his co-workers (cf. Bibliog.). 

The directions in which biochemical investigations have taken 
place are two-fold, viz. those on the nature of the toxic products 
which are formed and those concerned with the growth metabolism 
of the organisms. 

Previous to the war, most of the chemical work on anaerobes dealt 
with those concerned in putrefaction. A certain amount of informa- 
tion was gathered in the studies of the souring of milk and the 
ripening of cheese. 

Eecently the bacterial chemistry of soils, of silos, and of manures, 
has contributed not a little to our knowledge of the metabolism of 
anaerobic bacteria, and it is obvious that this branch of agricultural 
bacteriology is closely alhed to the biochemistry of wound infection, 
for it was from the heavily manured fields of France and Flanders 
that most of the pathogenic anaerobes were derived. The later 
chemical investigators used pure strains of anaerobes for their 
researches and they were, therefore, at a definite advantage over 
earlier workers who were unable to do so. The work of Nencki, 
Tissier, and others on putrefaction gives a composite picture of the 
chemical changes taking place in an albuminoid medium under the 
influence of putrefactive and other anaerobes, while later work 
assigns to each orgariism a definite function in the process. Hence 
it is that the foundations of our knowledge of the metaboHsm of the 



60 

individual bacteria are based on the methods which have been 
elaborated for the isolation of pure cultures. . 

Surgeons have come to recognize three important processes taking 
place in gas gangrene infections which are capable of bemg studied 

biochemically. . , . ,, , • 

The first of these is gas formation, for with it there are obvious 
physical signs of the type of infection. The production of gas m 
the tissues is moreover important, for it leads to disturbances ot 
nutrition, whereby the progress of the infection is materially 
accelerated. It may be said at once that aU anaerobes are gas 
formers under suitable conditions ; furthermore that all anaerobes 
are capable of producing gas in a protein medium like cooked meat. 
The capacity for gas formation varies with the individual organism, 
and is, in some cases, largely dependent on the presence of available 
carbohydrates. 

The second sign of gas gangrene infection is that of proteolysis, 
as denoted by the breaking down of the tissue and the odour, which 
is sometimes one of the first diagnostic signs encountered in wounds. 

All anaerobes are proteolytic, in that they are capable of trans- 
forming a certain amount of protein into ammonia and substances 
liberating nitrogen when treated with nitrous acid. 

The differences between the individual species are, however, very- 
great. On the one hand, there are organisms of the type of B. welchii 
and vibrion septique in which the proteolysis can only be followed by 
refined chemical methods ; on the other, B. sforogenes and B. histo- 
lyticus produce as profound a disintegration of protein as that 
produced by trypsin. Indeed the reaction is a more far reaching 
one, for with the breaking up of highly complex nitrogenous groups 
to amino acids, deamination and other types of reactions occur 
whereby large quantities of ammonia are formed. B. oedematiens 
is rather more proteolytic than vibrion septique especially if incubated 
in protein media at 37° C. for a long time. The. proteolytic action 
has not the practical significance of that of B. sporogenes or B. histo- 
lyticus, which, under favourable conditions, will transform a damaged 
muscle into a stinking pulp within 48 hours. B. oedematiens has the 
power, as its name implies, of producing an alteration in the tissues 
leading to the exudation of a clear gelatinous oedema. The chemical 
investigations which have been made on the growth of this micro- 
organism have, however, not thrown any light on this peculiar 
change. 

It is worthy of note that the truly proteolytic organisms are not 
those which produce potent toxins. The attack on proteins, or the 
synthetic process which gives rise to toxin formation must be of 
a very special kind, and is not associated with any great protein 
destruction. Experiments which have been made indicate that 
toxin formation takes place under certain limited conditions especially 
with regard to the reaction of the medium. The ' fragility ' of some 
of the toxins produced is directly due to the action of the hydrogen 
ions upon them, when the hydrogen ion concentration of the nutrient 
medium is higher than that which exists at absolute neutrality. 

The third sign is the general toxaemia seen in gas gangrene. 
Three classes of toxic effect due to the products of bacterial meta- 



61 

bolism, have been distinguished. The first is the true toxic effect, 
and this varies widely both in degree and in nature. 

Some of the anaerobes, e. g. B. sporogenes, produce no toxin. 
With vibrion sepUque a toxin can. be prepared of which 0-25 c.c. 
injected intravenously will kill a large rabbit in 5 minutes. B. 
oedematiens produces a toxin of which 0-003 c.c. subcutaneously 
will kill a mouse (of 15 gr.) within 24 hours. The details of the 
toxicity of these products are given in the serological section. 

The second type of product which has been thought to cause toxic 
symptoms is the ammonia. This is certainly formed in large 
quantity by organisms such as B. sporogenes and B. histolyticus 
but it seems improbable that a toxic alkalosis develops as a result. 

The third toxic effect has been assumed to be due to the acids 
formed during the metabolism of these organisms. This view has 
occupied a prominent place in the discussion of the toxaemia of 
gas gangrene. Large quantities of acids, especially volatile acids, 
are formed in cultures of some of the organisms. Whether in large 
infections of muscle tissue the amount produced is sufficient by 
itself seriously to deplete the reserve alkali is a matter which has not 
as yet been fully demonstrated. A. E. Wright and Fleming have 
shown that a fall in alkaline reserve takes place and have looked upon 
this as the result of acid production in the wound itself. There are, 
however, so many factors concerned in the diminution of reserve 
alkali that it seems unjustifiable for the present to assume that acid 
formation in wounds occupies a principal place in the production 
of the toxaemia. 

Acid production in culture. All anaerobes, so far examined, produce 
volatile and fixed acids in the course of metabolism. It is highly 
probable that this is a property of all micro-organisms, in which 
behaviour they correspond to other forms of cell life. 

As might be anticipated where large amounts of organic acids are 
formed from amino acid complexes, ammonia is simultaneously 
produced in the process of deamination and reduction. The am- 
monia is utilized to preserve the neutrality of the medium, and this 
is one of the reasons why cultures remain viable in media like cooked 
meat or alkaline egg, when in fluids containing carbohydrate they 
are destroyed owing to the formation of acids without the corre- 
sponding quantity of ammonia. This is quite apart from any 
question of spore formation. 

In some instances, e.g. the cultures of B. ^sporogenes, the equivalent 
formation of ammonia and organic acids results in the practical 
stabilizing of the reaction of the medium. Absolute cessation of 
growth does not take place for weeks after inoculation. With other 
organisms, where the acids prevail over the ammonia, and the 
reaction of the medium progresses in an acid direction, growth stops 
by reason of the ultimate acidity of the medium. The point at 
which growth ceases is of theoretical and of possible practical 
importance. 

Taking a medium of a given constitution and given initial reaction, 
it is always found that the final reaction at which growth stops is 
constant for each organism. There are slight differences in the 
final reaction if the initial reaction is varied, but the variation in 



62 



the final reaction, especially when available carbohydrate is present, 
is so small as only to be discovered when the hydrogen electrode is 
used. 

The other point which is of practical importance is that growth 
of these organisms is limited by an acidity which is approximately 
that of one thirty thousandth normal solution of hydrochloric or 
sulphuric acid. 

Attention has been paid almost exclusively to the volatile acid 
formed during growth ; this is so because of the technical difficulties 
of performing serial determinations of the fixed organic acids. 
Earlier work has shown that a number of these latter acids are 
present, but the amount has not been determined. Eecent work 
(Wolf and Telfer), revealed that in cultures of B. sporogenes and 
B. welchii on milk and on 2^er cent, glucose peptone broth, 30 per 
cent, to 50 per cent, of the total acid production consists of non- 
volatile acids. 

Of the fixed acids, lactic acid undoubtedly forms a considerable 
part, but other acids, such as succinic, phenylacetic, and p.oxy- 
phenyl propionic are undoubtedly present. 

All the earlier qualitative work was done with impure cultures 
and therefore cannot be accepted without reserve. 

The Biochemical Characteristics and their Bearing on the Deter- 
mination of Species. The work which has been done on the bio- 
chemistry of these organisms leads one to inquire whether the 
available data are sufiicient to determine the identity of bacteria 
which have not been resolved by the usual cultural and morpho- 
logical methods. 

Harris, in a paper which is now in the press, has made for this 
purpose a careful study of two organisms about the individuahty of 
which some uncertainty exists. 

A comparison was made between a typical strain of B. sporogenes 
and the ' Reading ' bacillus, isolated from a wound by Donaldson. 
The cultural and morphological features of the two organisms were 
identical. 

_ The results of the work show that in every chemical character- 
istic the two organisms were the same within the hmits of present 
experimental error. In only one respect did they differ. The 
oxygen tolerance of the ' Reading ' bacillus was low. It is a verv 
strict anaerobe. The strain of B. sporogenes, however, tolerated 
large partial pressures oi oxygen in liquid media, although both 
behaved as strict anaerobes in surface growths. Whether the 
difference is a real criterion of individuahty it is impossible to say, 
but It IS certam that a large field of investigation is open to anyone 
who cares to examine the characteristics of closely related strains 
by the chemical method. "^ hwdmb 

The Chemical Characteristics of Anaerobic Bacteria. So far 
a detailed study has been made of those bacteria which are presen 
m wounds and are considered clinically important These are 
B. sporogenes B. welchid, vibrion septique, B.hisMykcmZdB 
oedemaUens. To these has been added a study oTvarious straps of ' 
B.proteus as accessory organisms in wounds. 

Of the above mentioned, B. sporogenes and" B. histolyticus are 



63 

strongly proteolytic. B. welchii and vibrion septique are feebly 
proteolytic, but attack certain carbohydrates with great vigour. 

B. oedematiens appears to stand by itself. When the quantity 
of medium is large and the observations are prolonged, it produces 
' considerable quantities of gas in glucose and lactose media, and also 
in cooked meat when available free carbohydrate is not present. 
At the same time its proteolytic activities are not inconsiderable in 
media free from sugars. 

Bacillus sporogenes (Metchnikoff). This organism is characterized 
by the violent attack which it makes on proteins and on carbo- 
hydrates. Its proteolytic action, especially in a medium which is 
rich in protein is very great. In a cooked meat medium containing 
0-87 per cent, of nitrogen, 71-5 per cent, will be changed to ammonia 
and amino acid nitrogen in the course of 286 hours. 

It forms large quantities of gas from solutions containing available 
carbohydrates, over 1,000 c.c. per htre being formed from milk. 
The amount of gas formed from a 2 per cent, glucose peptone may 
be 1,020 c.c. in the course of 96 hours. 

One of the notable and important features of this organism is the 
large amount of acids which are formed during fermentation both 
of carbohydrates and proteins. At the same time the ammonia 
formed is sufficient to maintain the reaction at the initial level, 
or one even finds the fermented medium more alkaline than at the 
start. The reaction resulting from fermentation is never acid enough 
to destroy the organism. 

Two other features of the growth action of this bacillus are worthy 
of notice. One is the putrid smell, the cause of which is not known. 
Volatile sulphides are formed, but these by no means account for 
the characteristic odour. The second feature is the rapid blackening 
which takes place in cooked meat. This is probably due to the 
action of the sulphides on the iron compounds present in the medium. 
The action can be developed more easily as Henry (1) has shown by 
the addition of carbonate of iron. 

B. histolyticus. This organism, as the name implies, is supposed 
to be characterized by its vigorous proteolytic action. This is so, 
but it does not appear from the experiments which have been 
carried out that the proteolysis is more rapid or extensive than that 
obtained with B. sporogenes ; neither is the gas production so 
vigorous. 

A detailed investigation of the acid production has not been made, 
but it must be very great, for with a large production of ammonia, 
the reaction of the medium remains almost constant. The volatile 
acids formed in a concentrated cooked meat medium reach the high 
level of a 5 per cent, butyric acid solution. 

One of the characteristics of B. histolyticus is the appearance in 
cooked meat medium of fine silky crystals. These have not been 
carefully examined, but they have the naked eye appearance of an 
amino acid like tyrosin. 

B. oedematiens. This organism has a proteolytic action much 
less vigorous than that of the two preceding bacilli, but it attacks 
proteins more readily than either B. welchii or vibrion septique. 
Its gas forming power is not great, although in milk as much as 



64 

750 c.c. of gas may be formed per litre of medium. The volatile acid 
production is comparatively small. 

Bacillus welchii. This organism is distinguished by its extra- 
ordinary power for gas formation. This is possibly best shown in 
a carbohydrate containing medium like milk. In a litre of this fluid 
as much as 3,800 c.c. of gas, consisting of carbon dioxide and hydrogen 
have been evolved in 20 hours. In a cooked meat mixture in which 
carbohydrates were not present 1,072 c.c. have been given off in 
8 hours. This was an exceptionally vigorous fermentation, but 
300 c.c. per litre in 10 hours is a very usual amount. 

The carbohydrate consumption can be very large indeed and 
corresponds to the stormy fermentation in milk which is so charac- 
teristic of this organism. On one. occasion 13-8 grams of lactose 
per litre of milk disappeared while 2,714 c.c. of gas were evolved 
from 1,000 c.c. of the medium. Even in a liquid like alkaline casein 
344 c.c. of gas per litre have been given off. 

The proteolytic action of the B. welchii cannot be disregarded. 
In media in which there is a large proportion of amino acids 
additional amounts are formed. In no case has a definite pro- 
teolysis been found wanting. Both ammonia and amino acids are 
produced. 

The volatile acid production naturally varies with the composition 
of the medium, but in a nutrient mixture like cooked meat the 
acids formed are only ^ to J^ of what is found with B. sporogenes. 
This is undoubtedly due to the acids slowing down metaboHsm by 
reason of their hydrogen ion effect. Whether in a medium more 
strongly buffered an increased acid production can be effected is 
not known, but experiments are now being made in this direction. 
The practical point in connexion with the acid production of B. 
welchii is, that in cultures in a medium which has a resemblance to 
muscle, the acids formed are not of the order which one would 
expect to produce systemic effects. 

The effect of hydrogen ion concentration on the growth of B. 
welchii has been examined, and it is found that metabohsm ceases 
at an average Ph. =4-82 in glucose peptone medium. The variation 
which takes place in the inhibiting concentrations of hydrogen ions 
due to individual acids are small. The principal volatile acid pro- 
duced in metabolism is butyric acid. 

Vibrion septique. This organism in its capacity for gas formation 
in a milk medium is in the same class with B. welchii. On one 
occasion 2,453 c.c. of gas were evolved from a litre of this fluid. 
The method of gas production is, however, quite distinct. While 
with B. welchii this amount of gas may be formed in 24 hours 
vibrion septique will occupy 218 hours in doing so. The last 1,000 c c' 
were formed in the particular experiment under discussion between 
the 168th and the 218th hours. This is certainly a ' late ' fermenta- 
tion. The proteolytic activity is of the same order as that of 
B. welchii. It does not appear to form as much volatile acid as 
B. welchii under similar conditions. 

Bacillus proteus. The biochemistry of this facultative anaerobe 
was examined because of the possible symbiotic effect it might have 
on the growth of true anaerobes. All the strains used were isolated 



65 

from wounds. None of the strains showed the putrefactive charac- 
teristics so often attributed to the micro-organism. It is a gas 
former, especially in meat media. Very moderate proteolysis was 
exhibited. No indol was produced. The volatile acid production 
was small. No notable difference was observed in the growth of 
cultures in the presence of air from those which were grown under 
anaerobic conditions. 

• 

6. EXPBKIMBNTAL GaS GaNGRBNB. 

A general account of the results, obtained by inoculating animals 
with broth cultures of wound anaerobes, has been given in the pre- 
ceding pages. It is proposed in this section to deal with the more 
general problems in the pathology of gas gangrene and the light 
thrown upon them by experimental work ; to show how far the 
accepted facts can be explained and whether the conceptions of 
the disease which have been formed as a result of general clinical 
observation and post-mortem experience can be verified by animal 
experimentation. 

(i) Infection. 

When a culture of a non-pathogenic anaerobe, such as B. sporogenes, 
is injected under the skin, as a general rule the animal remains in 
good health and is apparently unaffected. Examination of the site 
of inoculation,, however, shows that a local effect, generally oedema 
and congestion, occurs. If films are made and examined, the 
organisms are seen undergoing extra-cellular lysis and phagocytosis. 
When the number of bacilli injected is very large the leucocytes 
emigrate in such quantities as to form what amounts to an abscess, 
which may lead to necrosis, but which is almost invariably followed 
by the healing process. 

The same general results follow the inoculation of a saline suspen- 
sion, free of toxin, of the pathogenic anaerobes. 

It is evident, therefore, that the means by which the body protects 
itself against these microbes consists in lysis and phagocytosis and 
that the pathogenic organisms are of themselves non-virulent. 

When a culture of one of the organisms which elaborate toxin is 
injected — that is to say, when the organisms and their metabolic 
products are injected together — the result is different. Lysis and 
phagocytosis do not occur ; on the contrary the organisms multiply 
and invade the tissues, producing at the same time fresh toxins 
which the animal is unable to neutralize, and this process continues 
until death ensues. 

A study of these facts has shown very clearly that the main, 
indeed all-important, difference between the pathogenic and non- 
pathogenic anaerobes of gas gangrene, from the point of view under 
consideration, is this capacity to form a toxin ; and further that the 
toxins act as aggressins, that is to say, annul the animal defences 
and enable the microbes to proliferate freely in the tissues. 

Toxin does not stand alone in this respect. Other substances 
have been discovered which are able to rupture the defensive system 
of an animal ; their mode of action is different from that of toxin 
but it is not yet clearly understood. The first and probably the 



66 

most important of these substances are the ionizable calcium salts 
(Bullock and Cramer (1)). When washed and detoxicated bacilli ot 
B welcUi, for example, are injected together with small doses ot 
calcium chloride— 21 mg. for a mouse, 5 mg. for a gmnea-pig— under 
the skin, the animal becomes ill in 8 to 10 hours and almost invariably 
succumbs within 24 hours to a violent gas gangrene. 

Sterile distilled water in large doses— 1 c.c. for a mouse of about 
15 grm.— leads to the same result as do calcium salts, though not 
so regularly, nor yet with the same violence. Certain colloids show 
a similar behaviour when injected with the spores of vibrwn septique. 
The most efficient colloid so far examined is siUcic acid ; the less 
efficient are gelatine, colloidal gold, palladium, and iron. 

These facts, established during the war, call to imnd the work 
of a number of investigators in which it is claimed that sand, lactic 
acid, cultures of staphylococci, and other materials may enable 
washed and heated spores of B. tetani or of vibrion septique to gerimn- 
ate in animal tissues and set up infection. It has not been possible 
to confirm these statements. 

Such is the broad outline of the modes of infection in experimental 
gas gangrene. How far do they explain the occurrence of gas 
gangrene as observed during the war ? 

"When a man is wounded and the wound becomes gangrenous 

it is clearly not sufficient to infer that this is because the wound 

is infected with a pathogenic anaerobe, since we may with safety 

conclude that toxin, beyond an infinitesiinal amount, is never found 

in mud or soil. It is difficult to see why any wound should go 

beyond the stage already described for non-pathogenic and washed 

pathogenic anaerobes, namely a local gas abscess. The difficulty 

has been explained by the supposition that the nature of the wound 

is such that the infecting organism is able to grow and produce 

toxin. Thus, the presence of large masses of dead muscle and an 

impeded circulation, the condition of shock and depression of 

vitality are all factors which are supposed to explain this paradox, 

that non-virulent organisms may become violently virulent. These 

factors do undoubtedly play an important part in the gangrene 

process. Dead muscle is a good culture medium and in shock the 

circulation is defective and consequently resistance to bacteria is 

weak. Experiment has shown that cold may be an important 

factor in the evolution of the disease. But after giving due weight 

to all these facts and possibilities it cannot be said that they are 

sufficient to explain the occurrence of gas gangrene in the majority 

of cases. Experimentally it has not been possible to elicit with 

regularity a fatal gas gangrene in animals by imitating the conditions 

enumerated. Further, it is a matter of common observation that 

cases of fulminating gas gangrene sometimes occur in men in whom 

the amount of tissue destruction in the wound is small and the 

circulation is not noticeably impeded. 

The minor and comparatively unimportant type of anaerobe 
infection known as a gas abscess, in which simple incision and free 
drainage are generally sufficient treatment, corresponds no doubt 
to the result obtained in animals when whole cultures of the non- 
pathogenic or suspensions of washed pathogenic anaerobes are 



67 

injected into the subcutaneous tissues, except that in the wounded 
man there is generally a sufficient quantity of necrotic muscle to 
provide a fermentable pabulum. 

The fulminant cases of gas gangrene which occur in 6 to 72 hours 
after the infliction of the wound are similar in character to the 
experimental cases in which either a broth culture of the organism 
or calcium chloride and the organism have been injected. It is in 
these cases that cold, fatigue, and shock may play some part in the 
genesis of the disease ; but the influence of the soluble calcium 
compounds in the soil contaminating the wound is probably of 
paramount importance. 

It is a well-known fact that cultivated soil almost invariably 
contains soluble calcium compounds and that the quantity of these 
varies from time to time. They are produced by a complicated 
series of chemical reactions from insoluble calcium salts, notably the 
sulphate and phosphate, under the influence of bacterial activity, 
the sun and rain. It is, therefore, probable that the chemical com- 
position of soil plays a large part in the production of this most tragic 
disease of war. For a long time it has been known that tetanus is 
liable to follow wounds contaminated with heavily manured soil. 
Here also it is now clear that it is not merely the presence of 
the tetanus bacillus, in no matter what numbers, which determines 
infection, but that other substances, of which calcium is possibly 
the chief one, enable the bacillus to live and multiply in animal 
tissues. 

Whether calcium salts, silicic acid, and water, all constituents of 
soil, can reinforce each other in enabling organisms to break through 
the animal defences, has not yet been determined. 

There is at present no reasonable explanation of the mechanism 
of infection in the rare cases of gas gangrene which occur, generally 
Brfter secondary operation, weeks or months after the infliction of 
a wound. 

A further point in connexion with the subject of infection is the 
steady fall in the proportion of cases of the disease as the war con- 
tinued. It is a well-known fact and is discussed by Bowlby in the 
Hunterian Lecture for 1919. An altered condition of the soil, 
leading to a reduction in the numbers and a diminution of the 
virulence of the soil organisms, is one of the numerous causes brought 
forward to account for the decHne. The alteration of the soil which 
is of direct consequence in this matter is probably the disappearance 
of calcium salts and of other similar substances, under the conditions 
described by Bowlby. It is less likely that the organisms lose viru- 
lence than that they decrease in numbers. 

(ii) The Established Disease. 
The general features of gas gangrene in men are observed in 
animals inoculated with broth cultures of the causative organisms. 
The local condition of oedema and gas production, the rapid invasion 
of the tissues by the baciUi, the occurrence of septicaemia, and 
the final toxaemia are all reproduced with striking fidelity. Any 
differences which exist are easily referable to anatomical or biological 
pecuharities of the animal employed. Thus, in mice, which are 

B2 



very prone to septicaemia, B. welchii invades the blood stream an 
hour or two after a subcutaneous injection of a pure culture of this 
organism. 

The striking feature of the disease which distinguishes it from 
almost all other infections is the rapidity Avith which it may lead to 
a fatal issue. It is not uncommon for death to occur within 24 hours 
after the active infection has begun. 

It is indicated elsewhere in this Eeport that the toxins produced 
by B. welchii and vibrion septique are not very potent ; a killing 
dose of less than 0-1 c.c. for a guinea-pig being rarely obtained. 
When such a toxin is compared with that of B. tetani in which 
the killing dose for the cavy may be as smaU as 0-0003 c.c, it is 
evident that the potency of the toxins poured iiito the tissues cannot 
account for the rapidity with which death may follow infection.^ 
But in tetanus there is no general invasion of the tissues by the 
infecting organisms, whilst in gas gangrene this is a conspicuous 
phenomenon. It is probably safe to infer, therefore, that in the 
latter disease there is a greater multiphcation of the pathogenic 
organisms and consequently a much greater amount of toxin 
elaborated. But this does not explain completely the extraordinary 
fact, that in a disease like tetanus in which the toxin is about a 
thousand times more lethal than that of B. welchii, death does not 
often occur sooner than 2 to 3 days after the infection is estabhshed, 
whilst in gas gangrene due to B. welchii death may occur in 12 to 
18 hours. 

The basis of these differences has been at least partly revealed. 
It has been shown (Bullock and Cramer (2)) that exhaustion of the 
suprarenal glands is of constant occurrence in animals killed either 
with sterile toxin or with cultures of B. welchii and vibrion septique. 
This can be most easily demonstrated in mice by means of Cramer's 
osmic acid vapour method. The suprarenals show congestion of the 
cortex and a great diminution of the cortical hpoids, while in the 
medulla there is a complete disappearance of adrenahn. Similar 
changes have been observed in actual cases of gas gangrene in men 
It IS the change in the medulla which is of special significance : a 
diminution of the cortical hpoid occurs in a number of septic con- 
ditions of diverse origm. The disappearance of adrenahn is not a 
temporary effect such as may occur after excessive stimulation of 
Sf ^% I ^""^ *^ gland seems to be unable to re-form new adrenahn. 
The effect may be suitably described as an ' inhibition ' or ' paralysis ' 
of the organ. fai.a,±jaia 

inIhe%Ew?W^ °* *^' mode of death in gas gangrene consists 
hJ ot f f ^^'f'i^ ^"^^ ^^''^^^^ ^^ *l^eir action on the adrenal 

bodies by a variety of factors, some of which are inherent in the 

r^ t W^'^^t^.*^''' are adventitious to it and come into pay only 
tSL ri V^' T''^^ conditions obtaining in war woundk ^ 
Ihus It has been demonstrated experimentally that haemorrhaee 

the injection of acid, and exposure to cold lead to a disappe^ance^f 

r^hZ^t:rlt°:::r:'i:\'f^:t^^^ - -i-t^d intravenous^ « 

does not kill under approximXv 24 ifn,',.? same dose mjeotedsubcutaneously 



69 

adrenalin from the suprarenal medulla. It has also been shown 
that these three conditions transform doses of toxin which are 
non-lethal for a normal animal into lethal doses. 

The bacteria of gas gangrene are very active acid producers and 
are assisted in this respect by other common wound organisms, such 
as B. sporogenes, which by themselves are not pathogenic. The 
production of acid at the site of injection — or in a wound — proceeds 
therefore concomitantly with that of toxin. Both are continuously 
being formed in rapidly increasing quantities as the local lesion 
extends, are absorbed into the circulation and assist each other in 
exhausting the suprarenal gland, which has already been largely 
depleted of adrenalin by such extraneous factors as cold and haemor- 
rhage. The general conception of the. infection which emerges from 
experimental work is therefore as follows. 

The disease begins, not when a wound has become infected with 
the pathogenic anaerobes, but from the moment when a group of 
these bacteria have been enabled to surround themselves with 
a toxin sufficiently concentrated to abolish the local defences of 
the tissues. This condition may be brought about by one or more 
of the substances already enumerated which rupture the primary 
defences before toxin is produced. The organisms multiply rapidly 
and invade tissues which are often cut off from an active blood 
supply or are debilitated by the bacterial poisons. Acids are pro- 
duced, and these promote the pullulation of the microbes (Wright 
and Fleming) and assist the systemic toxic effects by acting on the 
suprarenal bodies, which are also affected by shock, fatigue, cold, and 
haemorrhage. The consequences which flow from this ' paralysed ' 
state of the adrenals favour the invading microbes. A vidious circle 
is established which, unless rapidly, broken, leads to that fulminant 
character of the disease which Wright has so aptly compared to the 
progress of an avalanche. 

A general theory of gas gangrene which has found wide acceptance 
is that it is essentially a disease of muscle, that is to say, that muscle 
is always affected and generally contains the primary focus of 
infection, and that unless muscle is involved the classical type of 
the infection does not follow. This belief finds no support in experi- 
mental work. It is almost as easy to produce a fatal gas gangrene 
in animals by subcutaneous as by intramuscular inoculation. 
Further, when a culture of vibrion septique for example is injected 
intravenously into a rabbit, if the animal does not die within an hour 
of the effects of the toxin, it may die in 24 to 48 hours of an infection 
of vibrion septique. Post-mortem examination shows the bacilli 
distributed throughout. the body but reveals no special predilection 
for muscle. It is possible also to produce the disease by intravenous 
injection of the organism deprived of its toxin. Thus if mice are 
injected intravenously with washed and heated spores of vibrion 
septique together with 0-7 mg. of siUcic acid, they die within 24 hours 
and the bacillary forms of the organism can be seen in large numbers 
in the heart blood of the animals. 



70 



7. The Aerobic Infections of Wab Wounds. 
Br Albxandkb FLEMnfa, F.R.C.S., Bnq. 

In the early stages of gunshot or shell wotrnds the association of 
anaerobic and aerobic bacteria is an almost constant one. The 
primary infection of these wounds, coming as it does from mud 
and material from the skin of the soldier, gives rise to a growth of 
anaerobic and of aerobic bacteria most of which, like the anaerobes, 
have a faecal origin. 

Stokes and Tytler have carried out a series of observations on 
the bacteria found in recently inflicted wounds on their arrival at 
a casualty clearing station. In most cases examinations were made 
within twelve hours of the infliction of the wound, and it was 
found that out of 365 cultures, 310 showed the presence of aerobic 
bacteria. In cases where the species of aerobe was identified they 
obtained the following results : 



Table I. 

Aerobic bacteria found in wounds on admission to a casualty 
clearing station. 

Organism, 
Haemolytic streptococcus . 
NoH-haemolytio streptococcus 
White staphylococcus . 
Yellow „ 

Tetragenus 
Diphtheroids 
Gram-negative cocci . 

„ „ bacilli 

Oram positive bacilli , 
Negative .... 

Total number of cultures examined = 165. 

These results should be contrasted with others, obtained in the 
later stages ot wounds, at a base hospital in France and at a 
hospital m England (Tables IT and III). 



Number of cultures 




obtained. 


Percentage 


30 


18-2 


64 


38-8 


89 


48-5 


23 


14 


22 


13-3 


12 


7-2 


15 


9 


38 


23 


59 


35-7 


23 


13-9 



Table II. 

Common types of bacteria, other than the spore-bearinq anaerobic 
bac4h, found m wounds at a base hospital in France {FleZnH^. 



Time since 

■ infliction of 

wound. 

1 to 7 days 
3 to 20 days 
Over 20 days 



No. of 

wounds 

examined. 

127 
66 

27 



Strepto- 


Staphylo- 


Conform 


cocci. 


cocci. 


bacilli. 


102 


40 


37 


51 


16 


18 


24 


19 


19 



Some of these bacilli were obligate anaerobes 



Diphtheroid 
bacilli.^ 

9 
21 
16 



On udmission to 


At any period during 


hospital. 


their stay in hospital. 


48 


53 


33 


45 


8 


23 


19 


29 


19 


29 


34 


43 



71 

Table III. 

Common tyj)es oj bacteria, other than the spore-bearing anaerobic 
bacilli, found, in wounds in a hospital in England^ (Douglas, 
Fleming, and Colebrook). 

Organism. 
Streptococcus 
Staphylococcus 
B. pyocyaneus 
B. proteus 
B. coli type . 
Diphtheroid bacilli 

Total number of wounds examined = 54. 

The faecal element of the infection tends to disappear. The large 
anaerobic spore-bearing bacilli become fewer in numbers, and the 
streptococci which in the primary infection are usually of the non- 
haemolytic faecal type, become replaced by others which usually 
conform to the type of S. pyogenes. In a wound, therefore, after 
about two weeks when the sloughs have largely disappeared and when 
the ' healthy ' suppuration has become established, the organisms 
usually found are streptococcus pyogenes, staphylococcus aureus, 
diphtheroid bacilH, and sometimes B. proteus and B. pyocyaneus. 
As these organisms are absent in most of the recently inflicted 
wounds, the question arises what is their source ? There can be 
little doubt that in most, cases they are hospital infections, and it 
is probable that they are spread from case to case in the dressing of 
the wounds, although sometimes,, doubtless, the secondary invaders 
gain access to the wound from the skin around. On several 
occasions I have taken cultures from every patient in a ward and 
have found that only one or two were infected with B. pyocyaneus 
or B. proteus, but when I took cultures again from the same patients 
in the same ward a week or ten days later I found that every patient 
was infected with one or other of these organisms. 

Again, as regards the haemolytic streptococcus (which in wound 
infections is practically always streptococcus pyogenes), this organism 
was found in only 18-2 per cent, of wounds on admission to a casualty 
clearing station (Table I). During the summer of 1918 Captain 
Porteous and myself examined a large number of wounds (com- 
pound fractures of the femur) on admission to No. 8 Stationary 
Hospital and found that only 20 per cent, of these were infected 
with haemolytic .streptococci where the wounded had been sent 
immediately to the base after the primary surgical cleansing. In 
similar cases, however, which had remained in the same base hospital 
for more than a week, being dressed every day, it was found that 
over 90 per cent, were infected with haemolytic streptococci. 

We see then that, whereas the primary infection of wounds is 
a mixed aerobic and anaerobic one, the secondary infection is almost 
wholly aerobic. This might be expected when it is remembered 
that the anaerobic bacilh of wounds will in general grow in a 
serous fluid only when the implantation is a large one, whereas 

=These men remained in hospital in London for an average of something like six 
weeks, during which time examinations were made of the bacteria about once a week. 



72 

the aerobes will grow with much greater readiness, and especially 
streptococcus pyogenes which is by far the most cominon micro- 
organism of wounds in their later stages; and the one which persists 
after all the others have disappeared. According to Sir Almroth 
Wright's nomenclature, the anaerobic baciUi are of the na,ture of 
sero-saprophytes, i. e. possessing the power of readily growing out 
in serum only when that serum has been altered from the normal 
especially in regard to a lowering of its anti-tryptic power, while 
organisms like streptococcus and staphylococcus are serophytes, 
i. e. possessing the power of growing freely in unaltered serum. Sir 
Almroth Wright's work in this connexion has done much to explain 
the change in the bacterial flora of wounds. 

Eeference to Table I will show that on examination at a casualty 
clearing station almost 36 per cent, of wounds contained aerobic 
Gram-positive baciUi other than diphtheroids, and most of these 
were spore-bearing baciUi. In the cases I have examined on their 
arrival at the base I found about one in three to contain these 
baciUi, which appear to have no pathogenic power, and their im- 
portance to the bacteriologist lies mainly in their morphological 
resemblance to some of the important anaerobic bacilli. In the 
wound it is possible that they have some importance owing to 
■their symbiotic action on the anaerobic bacteria. This question wiU 
■be dealt with later. 

(i) B. mesentericus, B. subtilis, B. mycoides groups. 

These are the most common types of aerobic spore-bearing bacilli 
found. They are large, Gram-po5itive, and in a film made from a 
wound may resemble B. welchii or one of the other large anaerobic 
bacilh. Frequently very long elements may be seen or the bacilli 
may be arranged in chains. Central, subterminal, and terminal 
spores are to be seen, but in most cases the spores are not much 
broader than the bacillary body. 

(ii) B. aero-tetanoides. 

This baciUus resembles B. tetani morphologically but grows 
aerobically. It is motUe and ciliated. It gives bluish colonies on 
agar with flowery prolongations. It is feebly proteolytic and does 
not ferment any sugar. 

(iii) B. aero-tertius. 

This baciUus I have recovered from wounds in cases admitted 
to a base hospital in Prance and also at a later stage in England. 
It bears a remarkable resemblance to B. tertius. It is Gram- 
positive, motile, and grows on ordinary media both aerobically 
and anaerobically. The colony on agar resembles that of B. tertiiis 
closely, being smaU and transparent with slightly irregular edges. 
It does not digest coagulated egg or serum, and its cultures have 
no putrid smeU. It does not hquefy gelatin. It is a powerful 
saccharolytic organism fermenting glucose, lactose, saccharose, 
clextrm, starch, salicin, mannite, and glycerin, with the formation 
ot acid and a smaU quantity of gas. 



73 

(iv) Staphylococci. 

. In the receiatly inflicted wound it is very common to find 
staphylococci of the albus variety while the aureus type is rare. 
These white staphylococci do not differ from the white staphylo- 
cocci found in open wounds in civil practice. 

Staphylococciis aureus is common in wounds in the later stages. 
It, is very seldom found in pure culture, being almost invariably 
associated with streptococci and other organisms. A severe staphy- 
lococcus infection is always associated with much tissue necrosis. 
Occasionally it gives rise to a generalized infection. The incidence 
of staphylodoccus aureus in war wounds seems to be very similar 
to its incidence in the open suppurating wounds which come into 
the casualty room of a London hospital. 

(v) Streptococci. 

This group is by far the most important of all the non-sporing 
bacteria found in wounds, and it is responsible for most of the 
deaths from sepsis in the later stages of wound infections. There 
are found in wounds two well-defined types of streptococci — the 
faecalis type, especially frequent in the recently inflicted wound, 
and the pyogenes type which is very common in the later stages. 
Other types occur but much more rarely. 

Streptococcus faecalis. This organism is referred to in French 
literature as the ' enterocoque '. It is part of the primary infection of 
the wound and it gradually - diminishes in numbers so that two 
weeks after the injury it cannot in most cases be discovered at all. 
It is present in the intestinal contents of man and animals and it 
is very commonly found in manured soil. Houston and McCloy 
recovered it on every occasion from mud scraped off the boots of 
wounded men on their admission to hospital. 

Streptococcus faecalis is a large oval coccus generally occurring in 
pairs, the individual cocci being often set at an angle to each other. 
In culture it tends to be ' pleomorphic ', large and small, round and 
oval cocci being found. It grows well on all the ordinary media. 
On peptone agar the colony is a little, but not much, larger than 
that of Streptococcus pyogenes, but when it is planted on Douglas ■ 
trypsin agar it grows with great luxuriance, the culture resembHng 
staphylococcus rather than streptococcus. It ferments glucose, 
lactose, saccharose, mannite, and salicin but not raffinose or inuHn. 
Usually it does not liquefy gelatin, but some strains have this power 
as also of digesting coagulated egg and serum. In broth it grows 
weU, giving an even turbidity, in sharp contrast to the Streptococcus 
pyogenes culture in this medium. 

This organism, although classed as an aerobe, prefers anaerobic 
conditions and sometimes when first isolated it will only grow in 
the absence of oxygen. In glucose broth under anaerobic conditions 
it grows very rapidly and luxuriantly and forms chains of twenty 
or more elements. 

One of its most striking characters is its remarkable resistance to 
certain adverse circumstances. It lives for a very long time in 
culture. It will resist heating for half an hour at 55° C, and 



74 



sometimes will survive after being heated for 10 minutes at 80° C. 
Its growth is not inhibited by ox bile and it grows well on Drigalski 
Conradi medium. Houston and McGloy have used heating to 
55° 0. for half an hour as a means of isolating the organism, and 
Weissenbach has employed glucose peptone water to which has 
been added one-tenth of its volume of ox bile to differentiate this 
organism from 8. pyogenes. In this medium Streptococcus pyogenes 
is completely inhibited while Streptococcus faecalis grows well. 

Animals inoculated with a vaccine of this organism develop 
agglutinins to the strain with which they were inoculated but they 
do not of necessity develop agglutinins to all the other strains. 
The subject, however, requires further investigation. 

Streptococcus pyogenes. In war wounds as in civil practice this 
is the most dangerous of all the pyogenic microbes met with, and 
it is the cause of nearly all the septicaemic conditions occurring in 
the later stages of woimd infections. During the eight months from 
April to November 1918, working with Captain Porteous at No. 8 
Stationary Hospital, we obtained forty-four positive streptococcal 
blood cultures and the streptococcus in every case belonged to this 
group. Of the other positive blood cultures which we obtained from 
woxmded men, two were of B. welchii, one of B. oedematiens, and 
two of Staphylococcus aureus. These cultures were aU made from 
cases of septic compound fractures of the femur and the results 
show the enormous preponderance of streptococcus pyogenes in 
generalized infections. ; 

Source of the streptococcal infection. As we have seen (Table I), 
haemolytic streptococci are present only in 18-2 per cent, of the 
wounds soon after infliction. In the later stages nearly aU the 
wounds contain this organism. It is usually a hospital infection. 
Levaditi and Delrez foimd that 54 per cent, of the Enghsh soldiers 
had streptococci habitually in the epithehal squames of the skin. 
Among the Belgian soldiers in rest, they foimd that only 12 per 
cent, had streptococcus in the skin although 62 per cent, showed 
the presence of this organism on the skin when they were coming 
out of the trenches. They state also that the wounds among 
the Belgians were less often contaminated with streptococci than 
•those among the EnglisTi. It may be then that the infection in 
some eases is by direct spread from the surrounding skin. It seems 
probable, however, that the infection is from case to case in the 
dressing of the wounds. It has long been known that B. pyoayaneus 
can easily spread from case to case in a ward unless the most rigid 
precautions are taken. The spread in this case is obvious to the 
surgeons owing to the colour imparted to the dressings, but unfortu- 
nately the streptococcal infection does not manifest itself to the 
naked eye until some disaster hke a spreading celluHtis or a general 
infection results. In view of Sir Almroth Wright's finding, that of 
all the microbes found in wounds streptococci will grow most easily 
m the unaltered serum, it is probable that it is much more easily 
carried from case to case than B. pyocyaneus. 

Characters of streptococcus pyogenes. In films of pus from wounds 
this orgamsm occurs either in pairs, or in long chains, which are 
usually seen to be made up of pairs of cocci. Very frequently the 



75 

cocci are intracellular and when so placed can be seen in all stages 
of disintegration. 

In young cultures the cocci are usually round and nearly uniform 
in size, but in older cultures they vary very much, and it is common 
to see one or two elements in a chain very much larger than the. 
others. The individual cocci may become elongated, pear-shaped,' 
or almost bacillary in old cultures. 

Streptococcus pyogenes grows on all the ordinary media, and on 
Douglas trypsin agar forms rounded, grey colonies about 1 mm. in 
diameter with a very sUghtly irregular margin and a definite dark 
area in the centre when seen by transmitted hght. In b'roth it 
grows in woolly masses which settle to the bottom or adhere to the 
sides of the tubes, leaving the supernatant fluid clear. 

Growth occurs slowly on gelatin at 20° C. It does not Hquefy 
gelatin or coagulated serum. 

In milk it usually produces acid and a firm clot which later contracts 
and expresses a clear whey. The clotting of milk by Streptococcus 
pyogenes is, however, variable and appears to depend very much 
on the calibre of the tube used in the test. The smaller the tube 
used the quicker does clotting occur, and sometimes there will be 
a definite clot in a small tube when even after long incubation no 
clot will be produced in a large tube. 

Sugar fermentations. The sugar reactions of this organism seem 
to be variable. Usually it produces acid in glucose, lactose, saccha- 
rose, and aalicin, but not in mannite, raffinose, and inulin. A certain 
number of the strains (about 12 per cent.) ferment, in addition, 
mannite. Apart from this difference, the mannite fermenters and 
the non-mannite fermenters appear to be the same morphologically, 
culturally, and serologically. 

Haemolytic power. All strains develop a haemolysin in culture. 
This haemolysin is very unstable, being rapidly destroyed by keeping 
or by heating to 60° C. McNee and Macleod showed that filtered 
broth cultures contained a powerful haemolysin. It was found 
impossible to produce an anti-haemolysin by the injection of the 
haemolysin into animals. 

Viability. It usually dies out in a month or less on agar or in 
broth, but it can be kept alive for a long time without subculture on 
Dorset's egg medium or Eobertson's meat medium. 

Methods of isolation. This organism, like streptococcus faecalis, 
in general prefers anaerobic conditions, and a certain number of 
strains will only grow anaerobically when first isolated, although 
after one or two subcultures they become accustomed to aerobic 
conditions. 

Usually an agar plate, preferably incubated anaerobically, suffices 
for the isolation of this streptococcus. It is often an advantage 
to use a blood-agar plate as the haemolytic action is then made 
manifest. The simplest method for this purpose is to pour a plate 
of ordinary agar and then after it has set to pour over the surface 
a ■'thin layer of blood agar. Thus the haemolytic action is not 
obscured hy the opacity of a thick layer of blood agar. 

If B. proteus is present, isolation by simple plating may be diffi- 
cult owing to the spreading growth of this microbe obscuring' the 



76 

streptococcus colonies. In such a case cultures can readily be ob- 
tained by Wright's pyo-sero culture method or by growing anaero- 
bically in glucose broth for six hours, diluting and then plating on 
agar or blood agar. 

Blood culture methods. Douglas and Colebrook have shoAvn that 
broth contaiping active trypsin is much better than ordinary or 
glucose broth. To 5 c.c. of broth 0-25 c.c. of trypsin (AUen and 
Hanbury's) is added and the tubes are incubated to ensure that 
they are sterile ; 1 c.c. of blood is then added to each tube. In such 
a medium it has been found that growth occurs with a smaller 
implantation and is noticeable earlier than when plain broth is used. 

Another method which has given very good results is to put 
2 or 3 c.c. of blood into sufficient sterile distilled water to produce 
complete laking. This diluted blood clots slowly and there is plenty 
of time to carry it from the bedside to the laboratory and deal with 
it before clotting occurs. The laked blood is mixed with 15 c.c. of 
melted agar at 45° C. and a plate is poured. This method, in 
addition to giving merely a positive or negative result, gives an 
indication of the number of bacteria in the circulating blood. In 
a case of generalized streptococcal infection following a septic wound 
the number of streptococcus colonies obtained from 1 c.c. of blood 
is usually under 100 and often only one or two colonies develop. In 
one case, however, which was nearing a fatal issue, we obtained as 
many as 1,000 colonies from 0-25 c.c. of blood. 

It has also been found that Eobertson's meat medium, such as 
is used for the growth of anaerobes, gives better results than does 
plain broth in the cultivation of streptococci from the blood, and 
this medium has the advantage that not only will it yield a growth 
of streptococci which are indifferent to oxygen, but also of- those 
which at first are strict anaerobes as well as the obligate anaerobic 
bacilli if they are present. 

Serum reactions of Streptococcus pyogenes. It has been found that 
a rabbit injected with a vaccine of Streptococcus pyogenes develops 
agglutinins to this organism, and that while the serum obtained in 
this way agglutinated all the strains of Streptococcus pyogenes up to 
the same dilution, it did not agglutinate other streptococci except 
ia a few cases, and then only in a very slight dilution. 

A rabbit received first a dose of 1,000 mUUon streptococci sub- 
cutaneously ; four or five days later 1,500 miUions were injected 
intravenously, and this was followed by three intravenous doses of 
8,000 millions at intervals of one week. The rabbit was bled nine 
days after the last injection. The serum thus obtained was found 
to agglutinate emulsions of the homologous streptococcus up to 
a dilution of 1 in 500. o f )^ 

t]^ P^^^*?™™n agglutination tests with Streptococcus pyogenes, 
diaculty has been experienced in obtaining a good emulsion, 
and IJouglas has devised the following method for growing fluid 
cultures of streptococci for this purpose. The medium consists of 
two parts of broth (Douglas') and one part of serum or hydrocoele 
Huid which has been heated to 60° C. for 30 minutes. The cul- 
tures were incubated in a slanting position. In this mixture the 
Streptococcus pyogenes was found to grow in such short chains 



77 

that on shaking up the culture an almost even turbidity was 
produced. The culture should then be diluted with 0-85 per cent, 
salt solution to a convenient strength. Twenty-four strains of 
Streptococcus pyogenes were found to be agglutinated by the serum 
of a, rabbit inoculated as above with one strain, to the same titre 
(1 in 500). It was found that the strains which fermented mannite 
but which otherwise had the characters of Streptococcus pyogenes, 
were agglutinated exactly as were the non-mannite fermenting 
strains. 

It appears from these experiments that the mannite fermenting 
and the non-mannite fermenting strains of Streptococcus pyogenes 
are, essentially the same. 

Anaerobic Streptococci. In 1915 I described an anaerobic strepto- 
coccus as being of frequent occurrence in wounds. It was found 
in 9 out of 12 wounds taken at random. Gottet found a true 
anaerobic streptococcus in 10 out of 33 wounds. 

This streptococcus was difficult to isolate, as it always occurred 
in association with Streptococcus pyogenes, and the colonies of the 
two were identical. It grew in long chains and in old cultures showed 
very marked involution forms. It did not clot milk, and gave no 
change of colour on neutral red egg medium {Streptococcus pyogenes 
always gives a bright red colour on this medium). In shake or stab 
cultures on glucose agar, growth only occurred in the depths. Only 
one of my strains was carried through many generations, and it main- 
tained its true anaerobic characters. The fact that these cocci did not 
give a red colour on neutral red egg medium indicates that they are 
a different type from Streptococcus pyogenes, which invariably gives 
a bright red colour. No serum reactions have been carried out in 
connexion with these cocci. 

It is unnecessary here to go into the characters of the various 
other streptococci which have occasionally been found in wounds, 
as the work which has been done on them during the war has added 
little to our knowledge of them. It is interesting, however, to note 
that Malone and Ehea found that in penetrating chest wounds it 
was common to find streptococci of the type normally to be found 
in the respiratory tract. 

(vi) Diphtheroid Bacilli. 

Bacilli of this type are more common in the later stages of wound 
infections (see Tables I, II, and III). They have generally been 
regarded as of little importance and as leading merely a saprophytic 
existence. 

True Diphtheria Bacilli. Fitzgerald and Eobertson reported that 
out of 67 cases arriving in Toronto between May 20 and June 7, 1917, 
true diphtheria bacilli were recovered in 40. These bacilli were 
identified both by cultural and inoculation, tests. In some cases, 
but not in all, the wounds showed the characteristic membrane of 
a diphtheritic infection. The extraordinary prevalence of this in- 
fection was partly explained by the fact that one of the nurses who 
was looking after these wounded men had a slight wound on her 
finger from which B. diphtheriae was isolated. 



78 

Fitzgerald and Robertson's report led to an investigation of the 
diphtheroid bacilli in wounds in some of the Canadian hospitals in 
England, and the results of this investigation have been published 
by Ad ami and others. Out of 306 oases investigated, bacilli were 
found in four which had the morphological and cultural charac- 
ters of B. difhtheriae, and of these four, two were found to be 
pathogenic to animals producing the lesions characteristic of the 
Klebs-Loeffler bacillus. 

Prior to this, in the latter part of 1916 and the early part of 1917, 
Douglas Fleming and Colebrook, in an investigation of wound 
infections at St. Mary's Hospital, found diphtheroid bacilli in 43 
out of 54 cases at some period during their stay in hospital, and out 
of these 43 diphtheroid bacilli isolated, five morphologically and 
culturally appeared to be true diphtheria bacilli. These five strains 
were inoculated into guinea-pigs, and four of them were found to be 
virulent. 

It seems clear, therefore, that a certain proportion of the wounds 
in England were infected with virulent diphtheria bacilH. In our 
series at St. Mary's Hospital none of the wounds showed any sign 
of a membrane, and they were all large flesh wounds healing up 
without any clinical sign of the presence of the bacillus. 

Anaerobic Diphtheroid Bacilli. Some of the diphtheroid bacilli 
seen in films from wounds, especially in the early stages, are obUgate 
anaerobes, and we isolated from wounds on a number of occasions 
a bacillus which we called the ' Wisp ' baciUus, which belonged to 
this group. It was Gram-positive, non-motile, long and slender, 
and it arranged itself more or less in the typical diphtheroid maimer. 
It grew anaerobically on agar or glucose agar in colonies shghtly 
smaller and more transparent than those of Sireptococmis fyo- 
^genes. We were unable to test its fermentative activities, and there 
seems to be no record of them elsewhere. Cultures on agar rapidly 
die out. 

In wounds in the later stages it is very common to see large 
numbers of diphtheroid bacilli in films of the pus, but when this is 
planted out there is only a very scanty growth. It is possible that 
some of these baciUi, and especially those which can be seen cram- 
ming the leucocytes, are really acne baciUi which have gained access 
to the wotmd from the surrounding skin. The acne bacillus is one 
of the most common inhabitants of the skin, and in pus or in culture 
it shows a very definite diphtheroid arrangement. In pus from an 
acne pustule, also, it is very frequently found in large numbers 
mside the leucocytes. This bacillus only grows freely under anaerobic 
conditions when first isolated, and even then it does not appear in 
culture for three or four days, so that in the ordinary routine of 
wound examination where plates are made and incubated for 24 
or 48 hours It would be missed altogether. The acne bacillus is to 
some extent a seropliyte, and it is not unreasonable to assume that 
as It IS so common on the skin it would sometimes gain access to 
a wound in the same way a^ staphylococci or B. pyocyaneus. 

Aerobic mpUhermd Bacdh. There is not complete agreement 
S?5:,r ^^\ff?^^^. *ph/ervers regarding the characters of the common 
diphtheroid bacilh found in wounds. At St. Mary's Hospital we 



79 



examined 34 strains, and found that they grouped themselves as 
follows : 

No. of 

Characters. 

Large opaque white or creamy colonies on agar. Short stout 
segtate bacilli. Only show a few Neisser's granules. 
Ferment glucose and saccharose with formation of acid. 
Do not ferment mannite, glycerin, or dextrin. 

As group 1 except that the culture medium becomes a rich 
tawny brown colour in the presence of oxygen. 

As group 1 except that they ferment mannite in addition. The 
mannite fermentation was always slow and there was no acid 
formation for two or three days. 

Bacilli longer than group 1 and growth on agar not so copious. 
Do not ferment any of the sugars. 

Bacilli have the morphological characters of B. diphtheriae with 
many Neisser's granules. 

Ferment glucose, glycerin, and dextrin with acid formation. 
Pathogenic for guirrca-pigs, killing them with oedema at the 
site of inoculation and with enlarged and hyperaemic adrenals. 

As group 4 except that they were not pathogenic for animals, 

In this series the sugar tests were made in Cole and Onslow's 
broth, using acid fuchsin as an indicator. None of these bacilli 
formed acid from lactose with the exception of one strain in Group 4. 

The Canadian observers, using Hiss's serum water medium, 
found that 16 strains out of 41 fermented lactose. They did not 
test the fermentative action on mannite. Their results are shown 
on Table IV. 

Table IV. 



Group. 


No. of 
Strains. 


1 


15 


la 


4 


2 


7 


3 


3 


4 


4 


4a 


1 



Tt/pe. No. 


Dextrose. 


Lactose. 


Saccharosi. 


Dextrin 


Wound diphtheroid 1 . .6 


A 




A 


A 


A 


„ 2 . . 6 


A 




A 


A 





B. diphtheriae (virulent) . . 2 


A 




A 





A 


„ (non-virulent) . 2 










Wound diphtheroid 3 (xerosis 












type) . . . .23 


A 







A 





B. Hoffmann .... 2 
















A = 


acid formation. 









The most common type of diphtheroid bacillus found in wounds, 
therefore, appears to be an organism of the xerosis type which grows 
Ijixuriantly on agar, and which ferments only glucose and saccharose. 

The dijjhtheroid organisms generally have been regarded as being 
of little importance in wounds, and it has been found that wounds 
could be closed by secondary suture when the number of diphtheroids 
present was relatively high. Some of the infections with true virulent 
diphtheria bacilh have apparently resulted in the formation of the 
characteristic membrane, but this has not been a constant featiire 
even when virulent diphtheria bacilli were isolated. 

We shall see later that the non-pathogenic diphtheroids have 
a powerful symbiotic influence on the anaerobic bacilli and on 
streptococci, and in the connexion they may have some importance 
in the wound. 



80 

(vii) Coliform Bacilli. 

In this group are included all the organisms which in a Gram- 
stained iilm resemble the B. coli type. Many varieties of these 
bacilli have been isolated from wounds in all stages, but usually it 
has been found that they are more common in the later periods 
than they are in the first few days after the wound is inflicted. Stokes 
and Tytler (Table I) found that 23 per cent, of the wounds on admis- 
sion to the casualty clearing station contained these bacilli. Their 
incidence in the later stages can be gauged from the following 
table compiled from figures given by Stewart working in Leeds 
and Fleming working in Boulogne. 





Table V. 




Time after 




PercenUtge incidence of 


infection. 


No. of .cases investigated. 


coliform bacilli. 




Stewart. Fleming. 


Stewart. Fleming. 


Under 7 davs 


17 127 


41 29 . 


8 to 20 days 


47 56 


42 32 


Over 20 days 


58 27 


74 70 



Prom this table it is obvious that infection with these bacilli takes 
place very largely in hospital. It is very noticeable that in a wound 
there may be on one occasion many coliform bacilli, but when 
cultures are taken a week later they have disappeared, and later 
they may be replaced by a coliform bacillus of a different type; 
With few exceptions they seem to be merely passing saprophytes 
which do not really infect the wounds but which, probably due to 
some temporary-favourable condition, gain access to the wound and 
proKferate in the discharges until such time as the condition which 
was favourable to their growth is changed. Some experiments by 
Wilson are very instructive in connexion with coliform infections. 
This observer planted a clean woimd copiously with a living culture 
of B. coli from the intestine of the patient and then observed the 
fate of the organisms. He found that they had nearly all disappeared 
m 24 hours, and at the end of 48 hours they could not be seen in 
films or recovered in culture. 

Two very definite types, B. pyocyaneus and B. proteus, occur 
frequently m wounds, but there are in addition a number of other 
types more or less definite which are met with. The classification 
of the cohform bacilli of wounds has been studied by Matthew J 
Stewart, who isolated 148 strams from 122 wounds and arranged 
them as shown in Table VT. • * 





Table VI. 


\ 


No. of 


Oroup. 


Strains. 


B.coli . 


49 


B. proteus 


29 


B. Morgan, No. 1 


7 


B. faecalis alkaligenes 


1 


Group X . . , ■- 


8 


Group Y . 


26 


B. pyocyaneus . 


24 


Unclassified 


4 



No. of Percentage 

Varieties. Case incidence. 

34 26 

4 24 

2 5-7 ' 
1 0-8 

3 5 

4 20 
1 20 
4 3 



81 

The chief characters which have determined the place of the 
organisms in the above table are given by Stewart. 

B. coll Fermentation of glucose and lactose with or without the 

formation of gas. 
B. Morgan, No. 1. Fermentation of glucose, laevulose, and galactose 
only with the formation of acid and gas. 

Formation of indol. 

Voges and Proskauer's reaction absent. 
Group X. Fermentation of glucose, saccharose, laevulose, galactose, 
and inosite with the formation of acid but no gas. Non- 
fermentation of lactose. 

Litmus milk rendered acid and then strongly alkaline. No clotting. 

Motility present. 

Formation of indol. 

Slow liquefaction of gelatin. 

Voges and Proskauer's reaction absent. 
Group Y. Fermentation of galactose without formation of gas. 

Non-fermentation of laevulose. 

Motility absent. 

Indol negative. 

Voges and Proskauer's reaction absent. 
B. proteus. Fermentation of glucose and saccharose with the forma- 
tion of acid and gas. 

Non-fermentation of lactose. 

Eapid liquefaction of gelatin. 

Clotting and bleaching of litmus milk and finally more or less 
digestion of the clot. 

No formation of indol. 

Voges and Proskauer's reaction absent. 
B. faecalis alhaligenes. Fermentation of none of the carbohydrates 
tested. 

Motility present. 

Litmus milk rendered strongly alkaline. 

Gelatin not liquefied. 

No formation of indol. 

Voges and Proskauer's reaction absent. 

It is unprofitable to discuss at length all these different varieties 
of coliform bacilli as they seem to play little part in wound infections. 
Stewart in his article deals with them fully. On account of their 
relative frequency, however, B. -pyocyaneus and B. proteus deserve 
some further mention. 

(viii) B, pyocyaneus. 

The war has not added much to our knowledge of this organism. 
The incidence of the bacillus in wounds varies much in different 
hospitals and in different wards of the same hospital, but in 
general it seems to be more frequent the longer the wounded man 
has remained in hospital. 

The colour produced by different strains on agar may vary frora 
almost jet-black to a very pale green. Some strains hardly produce 
aay colour. 

F 



82 

In somG cases it seems to exercise a definite pathogenic effect, and 
in some cases a vaccine of this organism has had a remarkable effect 
in reducing the temperature of a patient whose wound had become 
infected. A chart showing this is appended. 

(ix) B. profeus. 

This organism, Uke B. pyocyaneus, seems to be in most cases 
a hospital infection. Its importance to the bacteriologist is greatly 
enhanced by the fact that it usually spreads rapidly over the 
surface of the culture medium, thus rendering the isolation of 
other organisms difficult. It can usually be isolated from the other 
organisms ])y planting the material into the water of condensation 
of an apar slope and incubating aeiobically, when after 24 hours 




B. proteus will have spread as a sheet of growth up to the top of the 
tube, from which situation pure cultures can be obtained. Anaero- 
bically growth is much delayed, and if it is desired to isolate other 
organisms from B. proteus, anaerobic methods are of great assistance. 

This organism rapidly hquefles gelatin, and it wiU digest coagulated 
serum or egg, but its proteolytic activities are not nearly so marked 
as those of a bacillus like B. sporogenes. Cultures have an unpleasant 
but not a putrid smell. In trypsinized serum considerable quantities 
of gas are evolved in the growth of B. proteus. 

Sera of patients suffering from infections of this organism will 
frequently agglutinate this bacillus in some cases up to a dHutioh 
of 1 m 100 of the serum. 

An animal inoculated with a vaccine of B. proteus develops aselu- 
tinins, and It was found that the serum of a rabbit inoculated 
with a vaccme of one strain agglutinated not only this sti'ain but all 
the other strams to which it was tested (15 in aU) up to a dilution 



83 

of 1 in 20,000 of the serum. Another rabbit, however, inoculated 
with another strain of B. proteus agglutinated the homologous 
strain to a dilution of 1 in 5,000, but (3nly three of the other strains 
were agglutinated by this serum in a 1 in 1,000 dilution, and one 
out X)i the 15 strains was not agglutinated even in a 1 in 50 dilution. 
By means of absorption tests the existence of sub-groups of this 
organism can be demonstrated, but it appears that the differences 
between these are relatively slight. 

Pathogenicity. In many cases this organism appeared to cause 
little trouble in wounds, but sometimes it seemed to be the actual 
cause of much or all of the fever from which the patients were suffering. 
It has been noted above that the patient's serum in many cases 
agglutinates the organism, which one would not expect if it were 
leading only a saprophytic existence. It has been observed that 
better results have been obtained in the vaccine treatment of wounds 
with a mixed vaccine of B. proteus and streptococcus than were got 
by streptococcus alone. 

It is possible that the chief importance of B. proteus in a wound 
is due to its symbiotic action on the other bacteria. 

(x) Gram-negative Cocci. 

These are present in some of the wounds from the beginning (see 
Table I) and in the later stages they are occasionally to be found. 
Some of these Gram-negative cocci, and especially those found in the 
earlier stages of the wound, are strict anaerobes, and are similar to 
the Gram-negative cocci that can frequently be isolated from faeces. 

One variety of Gram-negative coccus found in wounds is an 
obligate aerobe. It is somewhat larger than a staphylococcus, 
non-motile, and apart from a certain number of diplococcal forms, 
shows no special arrangement. It grows readily on agar in large 
white round colonies with a dull surface. It ferments glucose with 
the formation of acid but it has no action on any of the other 
sugars. It is not pathogenic for animals, and apparently exists in 
a wound merely as a saprophyte. 

Wilson and Steer have made use of this organism for growing 
anaerobic bacilli in symbiosis without any other anaerobic pre- 
cautions. As this coccus fermented glucose only, they were able to 
test the sugar reactions of the anaerobes by growing them in open 
test-tubes containing the sugar broth, which they implanted with 
a mixture of the anaerobe and this Gram-negative coccus. 

(xi) Micrococcus tetragenus. 

' Cocci arranged in tetrads were very frequently seen in wounds, 
especially in the earlier stages. Some of these are obligate anaerobes 
which grow on agar or glucose agar in minute greyish colonies, 
i'hese is no evidence to show that these tetrads seen in wounds are 
the same as the Micrococcus tetragenus of Gaffky. 

Wound infections in the later period o/ the war. 

Wound infections as seen at base hospitals were to some extent 
different in the later period, of the war from what they were at the 

F 2 



84 

beginning. In all probability this wai3 due to the careful surgical 
cleansing which the wounds received at the casualty clearing station 
towards the end. In the early stages of the war the wounds came to 
the base containing large masses of slough in which all the faecal 
microbes which constituted the primary infection grew and flourished. 
In the later stages the infections seen at the base were much more 
limited to the pyogenic cocci and to the bacteria which gained access 
to the wormds at some period during the stay of the patients in 
hospital. In particular B. proteus, B. pyocyaneus, Streptococcus 
pyogenes, Staphylococcus aureus, and diphtheroid bacilU were fre- 
quently found. 

As a result of the primary surgical cleansing the original infection 
was apparently much diminished and the conditions were not so 
favourable for its development, so that it disappeared rapidly 
except in cases where the primary excision had been incomplete and 
where considerable sloughs or pieces of necrosed bone had been 
allowed to remain in the wound. 

8. Influence op the Aerobic on the Anaerobic Infection of 

Wounds. 

By Alexandbb Fleming, F.R.C.S., Estg. 

It is a function of the aerobic bacteria that they can, and prefer to 
make use of, the free oxygen dissolved in the medium and contained 
in the air above. This can easily be demonstrated by a very 
simple technique. 

A tube of broth is planted with an aerobic organism. The cotton-wool 
plug is pushed a short way down the tube and is saturated with melted 
vaseline, after which melted vaseline is pom-ed on the top of the plug to 
a depth of about one centimetre. The tube is incubated, and when the 
contents have become warmed to the temperature of the incubator the 
position of the plug is marked on the tube. It will be found that the 
vaseline is sucked down the tube owing to the absorption of oxygen during 
the growth of the bacteria. Thus, in a culture of a diphtheroid bacillus 
isolated from a wound, the plug was ultimately sucked down for such 
a distance that the volume remaining between the fluid and the vaseline 
plug was only four-fifths of the original volume. When pyiogallic acid 
and caustic soda were introduced into this tube there was no further 
reduction of volume and no blackening of the pyrogallic acid, showing 
that all the free oxygen had been taken up by the bacteria. 

It can easily be imagined, therefore, that the aerobic bacteria may 
have a powerful effect in making the conditions in a wound suitable 
for the growth of the obhgate anaerobes. Methods have long ago 
been devised for growing the anaerobic bacteria without special 
anaerobic apparatus in symbiosis with B. subtil-k and many other 
aerobes, and recently Wilson and Steer have made use of an aerobic 
bram-negative coccus for the same purpose. 

_ The influence which aerobic bacteria have on the production of 
mfection by spore-bearmg anaerobes excited much attention even in 
pre-war days Eoger showed that an inactive strain, of vibrim 
septique can be rendered pathogenic for a rabbit by injecting it in 



85 

association with B. prodigiosus. Vaillard and Eouget stated that 
tetanus spores -would grow out in the tissues when they were injected 
with aerobic bacteria which protected them against phagocytosis, 
and a great deal of work has been done along these lines. It has 
been common experience during the war that an organism like 
B. welchi'^ can produce a gaseous emphysema in animals in a much 
smaller dose if it is combined with some other organism, either 
aerobic or anaerobic. 

Tissier ((3)-(6)) attributes the spread of the primary anaerobic 
infection in wounds to the action of the associated aerobic bacteria. 
He maintains that when the aerobes are harmless saprophytes 
the anaerobic action is slow ; it is more rapid when the associated 
bacteria belong to the pyogenic group ; and it tends to become 
fulminating when the streptococcus is present. The aerobe not only 
favours the growth of the anaerobe, but it opens a way for it into 
the tissues. According to this author the commencement of the 
infection of the lymphatic channels is by streptococcus or staphylo- 
coccus. 

Douglas, Fleming, and Colebrook (1) have studied the question of 
bacterial symbiosis in vitro, and they have found that all the common 
aerobes found in wounds have the power of stimulating the common 
anaerobes. Not only does the anaerobe grow more rapidly in associa- 
tion with the aerobe, but in some cases a growth was obtained with 
an implantation one million times smaller than when the anaerobe 
was grown alone. They found also that this stimulation of growth 
was shown as well, when the cultures were made under strict anaerobic 
conditions, as when semi-aerobic conditions obtained. The various 
anaerobic bacilli grown in association with B. welchii also exercised 
a beneficial effect on the growth of this organism and a symbiotic 
effect was observed in some cases Avhen the different aerobic bacteria 
were grown together. 

A few typical experiments will indicate the extent of this symbiotic 
action. 

(i) Influence of Staphylococci and Streptococci on the Groivth of 
B. welchii in Milk. 

A broth culture of B. welchii was diluted by tenfold steps up to 
one in a million. 40 c.mm. of each dilution were planted into tubes 
containing (a) milk 1 c.c, (b) milk 1 c.c.+staphylococcus broth 
culture 20 c.mm. and (c) milk 1 c.c.+streptococcus broth culture 
20 c.mm. After 24 hours the tubes were examined and the growth 
of B. welchii was noted by the formation of the typical clot. The 
result was as follows : 



Micro-organisms planted. 

B. welchii only 

„ staphylococcus 
„ streptococcus 



Dilutions of B. welchii culture. 



1/10 


1/100 


1/1,000 


1/10,000 


1/100,000 


1/1,000,000 


GR 


GR 














GR 


GR 


GR 


GR 


GR 


GR 


GR 


GR 


GR 


GR 


GR 


GR 



GR = growth of B. welchii. = no growth of B. welchii. 



86 

(ii) Influence of Aerobic Organisms on the Growth of B. welchii in 
Serum neutralized with Acid, and in Serum the antitryptw 
power of which has been neutralized with Trypsin. 

In this experiment miniature test-tubes were used, and after the 
implantations had been made, melted vaseline was poured on the 
surface of the culture fluid to make a column about 1 centimetre 
in depth. The index of growth of B. welchii was gas formation 
evidenced by the vaseline plug being pushed up the tube. 

Dilutions of a broth culture of B. welchii were made as in experi- 
ment 1 and 10 c.mm. of each of these dilutions was added to 300 c.mm. 
of serum. The same procedure was repeated with serum infected 
with each of the aerobic organisms. The results obtained were as 
follows : 



organisms planted. 






Dilutions of B. 


welchii culture 




1/1 


1/10 


1/100 


1/1,000 


1/10,000 




■hii only 


GR 
















, diphtheroid 


GR 


GR 


GR 


GR 


GR 




, streptococcus 


GR 


GR 


GR 


GR 







coliform bac. 


GR 


GR 


GR 


GR 


GR 




, B. pyocyaneus 


GR 


GR 


GR 


GR 


GR 





1/100,000 


GR 


GR 





1/1,000,0( 




GR 




The same results were obtained when the alkaline reaction of the 
serum was neutralized with acid or when its antitryptic power was 
neutralized with trypsin. Sir Almroth Wright has shown that both 
of these procedures materially aid the growth of B. welchii in serum, 
but it appears from this experiment that the symbiotic action of the 
aerobic organism has nothing to do with its effect on the reaction 
or the antitryptic power of the serum. 



(iii) Influence of Aerobes on the Growth of Anaerobes other than 

B. welchii. 

(a) The technique employed was the same as in experiment 2. 
After 5 days' incubation the results were : 



■organisms -planted. 




• 


Dilutions of Broth Culture of B. 


sporogenef. 




rogenes only 

„ + streptococcus 


1/1 



GR 


1/10 



GR 


1/100 



GR 


1/1,000 



GR 


1/10,000 



GR 


1/100,000 



GR 


1/1,000,00( 

GR 



GR = growth of B. sporogeiies. 

(j)) Equal volumes of serum were implanted with B. sporogenes 
and B. tert%us. One tube of each was kept as a control, and to the 
others were added some diphtheroid baciUi staphylococci or strepto- 
cocci Melted vasehne was now run on to the surface of the fluid 
and the tubes were mcubated: The results after 3 and 5 days were 
as follows : •' 



Associated 
Micro-organisms. 

None 

Diphtheroid bacillus . 

Streptococcus faecaUs 

Staphylococcus 



B. sporogenes. 
3 days. 5 days. 



Growth Growth 



B. tertius. 
3 days. 5 days. 




Growth 



Growth 
Growth 



(iv) Influence of B. welchii on the Growth of Streptococcus and 
Staphylococcus. 

(a) Two tubes were put up, one containing 1 c.o. of serum planted 
with 10 c.mm. of streptococcus emulsion, and the other containing 
1 c.c. of serum planted, in addition with 40 c.mm. of a B. welchii 
broth culture. After eighteen hours' growth under anaerobic 
conditions the number of streptococci were estimated by Wright's 
method. The result was as follows : 

Tube containing streptococcus only =100,000,000 strepto- 

cocci per c.c. 

Tube containing streptococcus and B. welchii=B60,000,000 strepto- 
cocci per c.c. 

(6) Four tubes, each containing 1 c.c. of serum were taken. Two 
of them were planted with B. welchii and all four were incubated 
anaerobically for twenty-four hours. An equal quantity of staphylo- 
coccus was then planted into one of the previously unplanted tubes 
and into one of the B. welchii cultures. The other two tubes received 
an equal implantation of streptococcus. After a further twenty-four 
hours' incubation aerobically the number of cocci was estimated by 
plating on agar 10 c.mm. of a 1,000-fold dilution of the culture. 
The number of colonies of the cocci which resulted was as follows : 

Tube 1. B. welchii and staphylococcus gave about 600 colonies. 
Tube 2. Staphylococcus only gave 64 colonies. 
Tube 3. B. welchii and streptococcus gave about 1,000 colonies. 
Tube 4. Streptococcus only gave 1 colony. 

These two experiments show clearly that B. welchii stimulates 
the growth of streptococcus and staphylococcus. 

(v) Effect of a Diphtheroid Bacillus on the Growth of Streptococcus 

pyogenes. 

Into each of three small test-tubes 1 c.c. of serum was placed. 
To the first and second were added 10 c.mm. of a 1,000-fold diluted 
broth culture of streptococcus, while to the second and third were 
added 10 c.mm. of a 1,000-fold diluted broth culture of a diphtheroid 
bacillus. The second tube, therefore, had a mixed infection of these 
two organisms. The number of living microbes was estimated by 
plating out a known quantity on agar and counting the colonies. 
The tubes were then incubated for ten hours at 37 C. and the number 
of living bacteria was again estimated in the same way. The results 
obtained were as follows : 

Implantation. Content of living Streptococci Content of living Diphtheroid 

per c.c. bacilli per c.c. 

At time of After 10 hours' At time of After 10 Jwurs' 
incubation. planting. incubation. 



Streptococcus . . 13,000 2,800,000 — — 

^*^phtli°eroid t^ . 13,000 120,000,000 2,000 2,600,000 

DipMieroid . . - - 2,000 5,300,000 

This indicates that while the diphtheroid bacillus has a powerful 
stimulant effect on the .streptococcus the action is not reciprocal. 



88 

It seems clear, therefore, from the work of many observers .both 
in vitro and in vivo, that the aerobic organisms such as are found 
in septic war wounds have a powerful influence m stimulating the 
growth of the anaerobic spore-bearing bacilh, and this is manifested 
not only by an increased rate of growth of the anaerobe, but also 
by the fact that in association with the aerobic bacteria the anaerobes 
will flourish in a serous medium or will infect an animal in a very 
much smaller dose. In gas gangrene the anaerobe has been compared 
to the high explosive and the aerobe to the detonator. Bullock 
and Cramer have shown that other ' detonators ' exist in the shape 
of calcium and other salts, but it seems clear that the aerobic bacteria 
must not be forgotten in this respect. 

It has been held that this stimulating effect is due to the aerobe 
absorbing oxygen and so rendering the conditions more favourable 
for the anaerobe, but although this may in some cases be an important 
factor it cannot explain the whole of the symbiotic effect observed. 
The stimulant effect of the aerobe is manifest whether the cultiva- 
tions are made under perfect anaerobic conditions or whether the 
oxygen is not rigidly excluded. It has also been shown that the 
presence of a second anaerobe in a culture wiU stimulate the growth 
of B. welchii, and if mixtures of anaerobes are injected into animals, 
the lethal dose is very much diminished, and here there can be no 
question of the absorption of oxygen. 

9. Ebsume op the Literature on the Bacteriology of Gas 
Gangrene, with an Account of the Incidence of the 
Various Types of Pathogenic Anaerobes. 

When cases of gas gangrene began to develop with such remarkable 
frequency shortly after the outbreak of war, bacteriologists found it 
necessary to take stock of their knowledge of this group of infections 
and, it must be confessed, found it largely wanting. Many workers 
had neglected to apply the tedious methods of technique which was 
necessary for the study of the few cases which came in the ordinary 
routine of work and their knowledge was admittedly second-hand 
and based upon the experience of certain ' classical ' predecessors. 

It was understood that gas gangrene was not an aetiological entity 
but a syndrome capable of being produced by several bacterial 
species. Among these were the vibrion septique of Pasteur proved 
by Brieger and Ehrhchto be a cause of disease in man, B. aerogenes 
capsulatusot Welch and Nuttall and B. phlegmonis emphysematosae 
of Eugen Praenkel. Further, it was clear from a study of the litera- 
ture that a large number of bacteria had been described and named 
nnd thl^n/''^"'"' ^^''^ ^^^ ^° ^"^y ^^^'^^d points of difference, 
WP i".! V Ff'? ^^^' Tl^^'^'t *^^* '^^^^^1 of tb^se ne^ varieties 

There iLl.?m ""f S\V^' P°^^*^ «* ^^f^^^^^^e ^e^e illusory. 

uVon solid ^Z2 , ^^""""^^f .of plating out ' anaerobic bacteria 
seCdlv toThP tT *^^' obtaimng perfectly pure cultures, and 
rchnen bv in ?>, '1'^' of enriching' the anaerobic flora of a 
variernerLnTI -"^ ^t'*^' ^""'^ ^^^ia thus causing some 
variety, perhaps an unimportant one, to assume an undue prondnen.^- 



89 

After the outbreak of war, a greater familiarity with the disease 
and improvements in technique led to an increase of knowledge 
which in France and Great Britain, if not in Germany, tended to 
become more stabiHzed. The pronounced characters of B. welchii 
and the ease with which it can be isolated, naturally led it to a position 
of prominence although disguised under different names. Fleming, 
Weinberg, Eugen Fraenkel, Simonds, Henry, and many others 
found the bacillus in practically all wounds or articles of clothing 
examined. 

Much confusion has been associated with Pasteur's vibrion septique 
and Koch's B. oedematis maligni. Early in the war B. sporogenes 
Metchnikoff was found by many workers such as Eobertson, 
Dean, Goadby, and Lardennois and Baumel. This microbe, a moder- 
ately large, motile bacillus with a central or subterminal spore 
possesses the power of liquefying serum and digesting milk. Such 
an organism had been described by Jensen and Sand, Kitasato (1), 
Kerry, Silberschmidt, Theobald Smith, v. Hibler, v. Werdt, and others 
as B. oedematis maligni of Koch which itself was assumed to be 
identical with Pasteur's vibrion septique. Long ago, however, Sanfelice 
(1892) pointed out that an organism which produced the typical 
lesions of Pasteur's experimental disease was not putrefactive and 
proteolytic as was then being taught. During the war Weinberg 
and Seguin, Mcintosh, Sacquepee, and others began to report the 
presence of a motile, non-proteolytic bacillus in war wounds and from 
its pathogenic and morphological characters identified it with 
Pasteur's vibrion septique. An extended study of this microbe has 
confirmed this view and has shown that it differs entirely from the 
proteolytic jB. sporogenes, the assumption being that proteolytic 
cultures of B. oedematis maligni are not pure. In order to clear the 
confusion as to the real relationship of these non-proteolytic pathogenic 
organisms a short account of their history is given. 

Unfortunately, neither Pasteur nor Koch left ,any complete bio- 
logical or cultural description whereby the bacilli called by them vibrion 
septique and B. oedematis maligni respectively, could be subsequently 
identified with certainty. In every instance the source of their 
virus was the animal cadaver. In modern bacteriological text-books 
it is usually stated that B. oedematis maligni can be readily obtained 
by inoculating guinea-pigs with garden soil. The accuracy of this 
statement must, however, be questioned. A great deal of the 
confusion which exists at the present time as to what is and what 
is not vibrion septique of Pasteur or B. oedematis maligni of Koch, 
is probably due to this statement. The majority of the writers who 
followed upon Koch seem to have been content to regard as the 
bacillus of malignant oedema any sporing anaerobe obtained from the 
lesion caused by inoculating soil into animals, with the result that 
a considerable variety of characters has been assigned to this bacillus. 
Sanfelice and C. 0. Jensen, however, state that the bacillus of 
malignant oedema can only be obtained from garden soil with diffi- 
culty, while Passini never found it there. 

Von Hibler (1) in 1899 described three strains of malignant oedema 
which all liquefied serum ; later, in his monograph of 1908, he 
reaffirms that his strains of this bacillus all liquefied serum and 



90 



digested milk, as did those of Kitasato and Silberschmidt. The 
type considered to be the bacillus of malignant oedema by Theobald 
Smith and by C. 0. Jensen, digested milk. . Divergences of opimon are 
to be found in regard to the morphology of the bacillus, some descnb- 
ing the spore as more or less terminal (Eoux, Ohauveau and Arloing), 
others that it is chiefly central (Jensen and v. Hibler). There is also 
no agreement as to the fermentation reactions. Theobald Smith 
states that it may ferment glucose, lactose, and saccharose ; Jungano 
and Distaso consider that glucose and lactose are fermented but 
not saccharose, while Bahr found that strains collected from different 
sources showed varying fermentations. 

In consequence of these researches, characters were attributed to 
B.oedematis maligni (Koch) which we now know to belong to B. 
sporogenes, an entirely different organism. The confusion between 
these two bacterial types is perpetuated even in some of the most 
modern German writings on the anaerobic bacteria of war wounds^ 

In their elaborate and careful research Ghon and Sachs, assuming 
that the bacillus of malignant oedema was proteolytic, and finding 
that the organism which they had isolated was non-proteolytic, 
described it as a new bacillus of malignant oedema. There can, 
however, be little doubt now that the bacillus of Ghon and Sachs 
was the vibrion seftique of Pasteur. Grassberger and Schattenfroh 
were also of this opinion, but Ghon and Sachs could not agree with 
them because they believed that the mahgnant oedema bacillus 
of Koch was proteolytic as described, and they further accepted 
Koch's statement that the bacillus of malignant oedema was identical 
with vibrion septique. 

So far as we can ascertain at this distance of time the micro- 
organisms which were studied by Pasteur and Koch were identical ; 
they were both highly pathogenic for laboratory animals ; the 
lesions produced were identical, and smear preparations from the 
peritoneal surface of the hver of animals which had succumbed to 
the infection showed the characteristic long thread-forms from 
which the name vibrion septique was derived. 

To resume the history of these researches during the war, certain 
workers by the use of selective methods have succeeded in demon- 
strating the existence of some particular anaerobe and concluded 
therefrom that the anaerobe in question was one of the principal 
causes of gas gangrene. In this category may be placed the view of 
Bugen Praenkel with reference to B. pMegmonis emphysematodes 
[B welchii), of Conradi and Biehng with their B. sarcemphysematodes, 
of Wemberg with B. oedematiens and of Sacquepee with B. bellonensis. 

A certain unanimity was, however, reached by French and British 
bacteriologists and the following tables indicate the incidence of the 
respective types found : 

Weinberg's results, 91 cases of 



B. ferfringens (B. ivelchii) 
B oedematiens . 
B- sf orogenies . 
B.fallax . 
Vibrion se^tique 



gas gangrene. 



77% 

34 

27 

16-5 

13 



91 



Sacquepee's results. 

B. ferfringens (B. welckii) . . ... 
Vibrion seftique typical .... 
Vibrion septique atypical .... 
B. bellonensis ...... 

Mcintosh's results. Series A. 1914-18, 41 cases. 
B. welckii ...... 

B.sporogenes (including type XIII of Mcintosh) 
Vibrion seftique ..... 

Terminal oval sporing bacilli 
Unidentified types : 

Type No. XVIII .... 

Type No. XIX .... 

Series B. 1918. 52 cases. 
B. welchii .... 

B.sporogenes (including type XIII) 
Vibrion septique 

B. oedematiens .... 
Terminal oval sporing bacilli 
B. tetanoides .... 



82% 
28 
11 
35 



43-9% 
36-5 
19-5 
17-0 



4-8 

4-8 



67-3% 

38-7 

16-3 

4-0 
"8-1 

2-0 



Henry's results. 
B. welchii 
B. sporogenes 
B. tertius 



1917. 100 cases. 



3% 



Henry's results. 1918. 50 cases. 
5. welchii ........ 

Vibrion septique ....... 

B. oedematiens ........ 

Otter pathogenic anaerobes, of which 6% are probably 
B.fallax . . . . . . . 



48 
22 



80% 

16 

10 

10 



This last unpublished series of Henry's was derived from material 
obtained from cases of gas gangrene, sent to the Committee by 
corresponding members and medical officers serving with the British 
forces in Prance. It will be seen at once that the results differed from 
those which he obtained in 1917, but this can be explained by the 
fact that in his second series an entirely different method of analysis 
was employed the technical details of which were as follows : 

A series of preliminary experiments had shown conclusively that there 
was no antagonism between B. welchii, vibrion septique, and B. oedematiens, 
if these were introduced together into an alkaline meat medium kept under 
suitable anaerobic conditions. For- example, no evidence was obtained 
that B. oedematiens, reputed to give feeble growths, could be overgrown 
in culture either by vibrion septique or by B. welchii. Bach of the three 
organisms mentioned, either alone or in association with the others, 
reached its full development in 12 to 18 hours when grown in a meat 
medium. 

Pieces of infected muscle from the cases of gas gangrene were placed 
in tubes of meat medium, and the resulting cultures, 18 to 24 hours old, 
were used for inoculation into guinea-pigs. This ' whole culture ' method 
appeared to have certain outstanding merits : 



92 



(a) The transference of the micro-organisms from man to the g^^iea- 
pig occupied a very short time, and the loss of virulence was reduced 
to a minimum. vi v + 

(6) Any advantage of miorobic association would presumably be at 
a maximum in a mixed culture. 

Of the 50 specimens of muscle, 44 yielded cultures which were pathogenic 
to 250 grm. guinea-pigs in doses of 1 c.c, death resulting in 12 to 24 hours. 

A potent serum containing antitoxins to B. tetani, B. welchii, and vibrion 
septique protected guinea-pigs against 33 of these pathogenic mixed 
cultures. 

In 5 out of 11 cases in which the triple serum yielded no protection, 
the infecting organism proved to be B. oedematiens, and the infection 
could be arrested with an anti-oedematiens serum. Of the remaining 
6 cases 6 yielded anaerobes which were not B. welchii, vibrion septique, 
or B. oedematiens. Three of these five were probably pathogenic strains 
of B. fallax. 

A glance at the above tables of results is sufficient to show that 
there is a considerable degree of uniformity in the findings of the 
British and French workers. The German Tesults, however, are 
neither uniform nor do they agree with the above. To judge from 
the number of polemical articles which have appeared in the German 
medical press no finality has been reached, and it is apparent that 
this must be attributed to the fact that they have not yet succeeded 
in isolating the more important anaerobic bacteria in pure culture. 

Aschoff (3) considered that there were three main types which he 
called : 

1. ' Gasbrand Group '. 

2. ' Eauschbrand group '. 

3. ' Malignant oedema group '. 

and these were again subdivided into pathogenic and non-pathogenic 
types : 

Non-pathogenic types. 

B. saccharo-hutyricus 

immobilis 
B. amylobacter 
B. saccharo-butyricus 

mobilis 
B. paraputrificus 



1. Gasbrand 

2. Eauschbrand 



Pathogenic types. 
Welch-Fraenkel type. 



Conradi-Bieling type, 
Ghon-Sachs ' bacillus, 
vibrion septique Pas- 
teur, Kolmaitype. 



3. Malignant oedema B. putrificus 



Von Hibler's malignant 
oedema bacillus. Koch's 
malignant oedema ba- 
cillus ? 

In a later paper, Aschoff (4) republished the table with slight 

modifications m that the ' Gasbrand group ' was called 'immobile 

butyricus group ; ' Eauschbrand group ' being called the ' mobile 

butyricus group while the ' malignant oedema |roup ' was renamed 

putrificus group . » r 

E. Pfeiffer and Bessau came to the conclusion that there were four 
main types : 



93 

A. Non-putrejying anaerobes. 

1. Bacillus of Fraenkel. 

2. Bacillus of malignant oedema (Koch). 

B. Putrefying anaerobes. 

1 . ' Uhrzeiger ' or clock-hand bacillus. 

2. Par-oedema bacillus. 

Conradi and Bieling, however, considered that gas gangrene was 
due to one anaerobe, B. sarcemphysematodes, and that the Welch- 
Fraenkel bacillus, the bacillus of mahgnant oedema, the bacillus of 
Eauschbrand, &c., were merely different stages of B. sarcemphysema- 
todes. Two main types were fairly constant, viz. type A, which is 
a plump, non-motile rod, producing in milk an acid clot, which is 
later digested, as is coagulated serum ; type B which is thinner, 
actively motile and affects milk and coagulated serum in a manner 
similar to type A. They further state that type A can be changed 
into type B by a few passages on coagulated serum buttype B can 
only with dif&culty be changed into type A. They also claim that 
specific immune sera can be produced for each type. Conradi and 
Belling have evidently taken their ideas from Grassberger and 
Schattenfroh's conception of the mutability of certain anaerobes, 
but as in the case of these earlier workers the evidence produced is 
unconvincing. The work of Conradi and Bieling received a con- 
siderable amount of confirmation, but other German bacteriologists, 
Pfeiffer and Bessau, however, strongly opposed their view as they 
had not found any evidence in its favour. 

Klose (6) states that gas gangrene cannot be considered as due to 
any single species but must be ascribed to the action of a number 
of different bacilh. By serological tests he was able to sub-divide 
the anaerobes isolated, into four types. 

1. Welch-Fraenkel group. 

2. Putrificus group. 

3. Eauschbrand group. 

4. K;^ group (corresponding to Novy's bacillus). 

An analysis of 100 strains isolated from soldiers wounded at Verdun 
showed the following grouping : 

per cent. 

Group! 34 



2 . 
„ 3 . 
„ 4(K,) 
Unclassified 



24 

32 

6 

4 



Of these strains 36 were obtained by venepuncture. 

As regards these German results the position is not clear, and the 
discrepancies between the findings of individual workers are con- 
siderable. The descriptions of the biological and cultural charac- 
teristic which have been given do not enable us to identify with 
certainty the several species with which they were working. Their 
cultures have not been available for study in this country ; it is, 
therefore, impossible to dogmatize, but reading between the lines 



94 

it is possible to suggest an interpretation of the significance of their 
various names as follows : i. • i 

The Welch-Fraenkel group of the German workers can obviously 
be identified as B. welchvi. 

The BauscUrand group of Aschoff and of Klose presents greater 
difficulty. The true Eauschbrand bacillus (B. chauvoei) has not 
been isolated from war wounds by any British or French worker. 
Klose, however, states that in the early days of the war it was 
found that the Eauschbrand serum, prepared at the Hoechst works, 
agglutinated a considerable number of the strains of anaerobes 
isolated from wounds. It was, therefore, concluded that they were 
Eauschbrand bacilh. Later, he tested 12 Eauschbrand cultures 
of animal, as well as of human, origin and found them to be identical 
in their agglutinating reactions. 

The description of- these so-called Eauschbrand types by Klose, 
Zeissler (3), &c., shows that they failed to differentiate the organism 
from vibrion septique of Pasteur. This failure possibly arose from 
the fact that the Eauschbrand serum in question (Hoechst) was 
prepared not by the injection of B. chauvoei but from strains of 
vibrion septique which had been isolated from animals. It has been 
conclusively demonstrated by several writers (Markoff, Koeves,' 
Meyer, &c.) that vibrion septique may infect cattle, horses, and pigs 
with the production of symptoms similar to those of blackquarter. 
The supposition that Eauschbrand serum made at the Hoechst 
works was specific and might thus be used for differentiating B. 
chauvoei from other pathogenic anaerobes was not upheld by Pfeiffer 
and Bessau who stated that the clinical concept of ' Eauschbrand ' 
is not an aetiological entity. Further, H. Landau in a series of 
agglutination tests showed that the Hoechst Eauschbrand serum 
did not agglutinate two strains (A and B) of B. chauvoei isolated by 
Kitt one of the classical authorities on the disease in question. 
A serum prepared from Kitt's strain B did not agglutinate any of 
the ' Eauschbrand strains ' said to have been obtained from cases 
of human gas gangrene. In the Uncertain state of present knowledge 
it is not possible to dogmatize, but the Committee consider that the 
existence of B. chauvoei in human gas gangrene is not proven. 

The Putrificus group. The members of this group are proteolytic, 
liquefying coagulated serum and digesting milk. The pathogenicity 
apparently varies to a considerable extent. The ' Uhrzeiger ' or 
clock-hand bacillus of Pfeiffer and Bessau would be included here. 
In the present state of our knowledge of anaerobic bacteria there is 
little doubt that the putrificus group corresponds to the sporogenes 
group of British and French workers. The fact that the German 
bacteriologists found a certain number of the strains to be pathogenic, 
a result not observed elsewhere, is explicable on the assumption that 
•their cultures were contaminated with some of the pathogenic 
anaerobes. It is only this pathogenic effect which distinguishes the 
putrificus group from that of B. sporogenes. 

The close unanimity arrived at by the British and French workers 

has been recorded above. The only discordant result is to 

be found in the work of Sacquepee (11) who attaches great impor- 

:iance to an organism which he calls B. hellonensis. From the 



95 

descriptions which he has so far pubUshed this organism cannot 
be identified. 

In conclusion, the Committee are of opinion. that acute gas 
gangrene may result from at least three types of pathogenic anaerobic 
bacilli ; these in their order of frequency are as follows : 

1. B. welchii. 

2. Vibrion septique. 

3. B. oedematiens. 

In addition to these, many other anaerobes may be present 
in wounds. The exact part which these play in the production of 
gas gangrene is not clear. In pure culture they cannot be regarded 
as pathogenic for .laboratory animals. The more common forms 
of these are : 

B. sporogenes. 

B. parasporogenes. 

B. tertius (Hibler IX). 

B. Mstolyticus. 

B. fallax. 

10. The Isolation of Anaerobic Organisms of Wounds in 

Pure Culture. 

In the isolation of anaerobic bacteria no single method has been 
found which will meet every requirement. It cannot be too strongly 
insisted upon, that, whatever technique may be employed, pure 
cultures can be obtained only by the exercise of great technical skill 
and by the constant application of a critical attention. 

(i) Material Suitable for Examination. 

1 . Exudates from wounds. 

2. Pieces of infected muscle removed at operation. 

3. Fluid from haemorrhagic bullae. 

4. Blood from a vein. 

5. Post-mortem material. 

In the collection of the above material certain precautions are 
necessary. It is preferable that the bacteriologist should collect 
the material himself, so that it should be as little contaminated as 
possible. A sufficiently large sample should be taken to inoculate " 
several tubes of medium. Further, a Gram-stained smear preparation 
of the wound exudate or muscle juice should be made, as this 
is frequently found to give useful indications of the varieties of 
anaerobes present. 

(ii) Methods of Isolation. 

. Whatever the ultimate procedure adopted, the primary aim is 
to preserve all the anaerobes present in the sample taken, and for 
this, some medium as little selective as possible should be used. 
The alkaline meat medium or broth with a piece of fresh tissue are 
both- useful for this purpose, and part of the material should be 
inoculated into one or other of these media, which may be regarded 
as a repository. Surface growths on serum agar under very good 



96 

conditions of anaerobiosis may be made direct from the wound 
material if this is derived from a relatively uncontaminated source. 
The early mixed cultures obtained in the meat or tissue medium 
tubes should also be planted in series on agar slopes or plates and the 
colonies picked off. 

In addition to the above, a number of methods which are essentially 
selective in character have been found to be of great utiUty. 

They comprise : 

A. The separation by mechanical methods of the individual 

organisms in a mixture. 

B. The selection by appropriate heating of the spores contained 

in a mixture. 

C. The use of a selective media. 

D. The separation of a pathogenic anaerobe by animal experi- 

ment. 

A. Separation iy Mechanical Methods of the Actual Individual 
Organisms Present in a Mixture. 

Since the memorable discovery of Pasteur a great deal of attention 
has been paid to the cultivation of anaerobic bacteria but it is certain 
that prior to the war a large number of anaerobic species had not 
been obtained in a pure state. It is easy by some and possibly by 
all of the innumerable methods, which have burdened bacteriological 
literature, to cause anaerobes to multiply but few of the apparatus 
although mostly described as ' simple and effective ' will lead to 
the production of pure, isolated colonies on the surface of a sohd 
nutrient medium — a result obtained with ease in the case of aerobes. 
Some of the factors which militate against success with anaerobes 
have been repeatedly emphasized in various parts of this Eeport. 
Of these, cohesion of the material leading to a close association of 
two or more anaerobes in one and -the same colony is perhaps the 
most important and it must be further added that progressive 
dilutions may not necessarily overcome it. The high degree of 
motiUty of many anaerobes will also cause them to spread as a film 
oyer the surface obscuring others which may themselves have a pre- 
dilection to assume isolated colonial form. This is especially prevalent 
in the presence of excessive moisture. Sometimes the .medium is 
' not suited to the requirements of all the several bacteria in a mixture 
and the spores of some may lie dormant on the surface and may be 
subsequently picked up along with a well-developed colony and 
contmue with it in subsequent subculture. These are fundamental 
difficulties which must be expected and mastered if possible. The 
cohesion of the material may be overcome to a very considerable 
extent, as pointed out by Stoddard (1) if the original material and 
ail the subsequent dilutions are thoroughly shaken with glass beads, 
sea sand, or even saline solution, surface cultures being then planted 
out. A medium which will encourage the growth of as many of the 
anaerobes as possible should be employed. 

However interesting may be the fact that anaerobes can grow 
under apparently aerobic circumstances it is certain that separate 
colomfes on sohd media can only be obtained under conditions in 



97 

which oxygen is intentionally excluded and for surface growths 
the exclusion must be as complete as possible. 

The exclusion or removal of free oxygen can be carried out in 
a number of different ways each having a set of supporters. The 
general principles involved are : 

1. Cultivation in vacuo. 

' 2. Deep cultures as recommended originally by Hesse and Liborius 
and subsequently developed by Veillon and Zuber and by Burri. 

3. Cultivation in indifferent gases of which hydrogen and nitrogen 
are the chief. 

4. Absorption of oxygen by chemicals such as alkali in the presence 
of pyrogallol, or in atmospheres deoxygenated by combustion with 
hydrogen in the presence of palladium as originally suggested by 
Laidlaw and carried to perfection by Pildes and Mcintosh and 
Smillie. 

5. Combined methods in which evacuation in the presence of 
alkaline pyrogallol, or hydrogen in combination with alkaline 
pyrogallol with or without evacuation are the chief. 

Previous boiling of the medium — ^historically the oldest method — 
should be effected as far as possible but is inadmissible with many 
of the modern media containing coagulable protein. 

For some reason, imperfectly understood at present, anaerobes 
often show a great disinclination to grow as separate surface colonies 
and but few of the many methods will lead to such a development. 
It is quite evident that for such surface growths the anaerobiosis 
must be much more complete than in the case where colonies are 
embedded in the medium. It would also seem to be proved by the 
researches of Burri and Kiirsteiner that the anaerobiotic condition 
should be induced as rapidly and as completely as possible when 
once the anaerobes are implanted on the artificial nutrient medium. 

The exa,ct technical details of the various anaerobic apparatus 
should be consulted in the various periodicals and journals devoted 
to bacteriological literature. It suffices to say that in their extended 
and fruitful researches Weinberg and Seguin employed deep cultures 
after the manner of VeiUon and Zuber, Henry utiUzed, in particular, 
special dishes rendered anaerobic by alkaline pyrogallol while 
Miss Muriel Eobertson relied principally on thorough removal of 
oxygen by the air pump. Mcintosh was unusually successful with 
a simple apparatus in which the air was deoxygenated by hydrogen 
in the presence of palladium asbestos. In most cases it is found 
that test tubes are preferable to actual Petri plates for isolation 
of surface colonies. . 

A microbic emulsion may be diluted down to a pomt at whicn 
a minute volume sufficiently small to be examined microscopically 
can be shown to contain only one individual organism. Each minute 
fraction of the emulsion which contains a soUtary organism will if 
planted out into an appropriate medium give rise to a culture which 
is pure because it is derived from one individual. This method of 
obtaining a pure culture has been practised by a number of workers 
in the isolation of aerobes, and it has been apphed by several in- 
vestigators in the case of the anaerobes. The larger size of most 
anaerobe baciUi, as compared with that of aerobic organisras, 

G 



98 

a feature which greatly facihtates a microscopic count, appeared 
at first sight to be a factor which would enhance the chance of 
success. It was found, however, that of the sohtary baciUi isolated 
from a mixture of anaerobes only a very small percentage proved 
to be viable. This difficulty, taken in conjunction with the other 
Avell-known disadvantages that are inherent in any micro-inoculum 
technique, renders the method inapphcable to the isolation of 
anaerobes on an extensive scale. There are, however, occasions 
when the procedure in capable hands may prove to be of very great 
value. (Barber.) 

B. The Selection iy Appropriate Heating 0/ the Spores 
Contained in a Mixture. 

This method which has frequently proved to be very serviceable 
in the isolation of anaerobes has been practised with two objects 
in view. 

1. To separate spores from vegetative forms. 

A small volume of a mixed culture . sealed off in the capillary 
portion of a Pasteur pipette, is plunged into boiling water for fifteen 
seconds or subjected to a temperature of 80° C. for twenty minutes by 
immersion in a water bath. This heating is sufficient to kill off all 
vegetative forms so that only spores remain. By this means there 
can be removed completely from a mixed culture (a) non-sporing 
aerobes such as streptococci, B. proteus and coHform organisms, and 
(b) non-sporing anaerobes. The latter include B. welchii,. B. fallax 
and B. aer of etidus when the mixture has been grown in a carbohydrate- 
containing medium such as milk, glucose broth, &c. 

The heated portion of culture is inoculated into a liquid medium 
and the resulting subculture is used for plating out. 

2. To separate the more highly resistant spores in a mixture from 
those that are less resistant to heat. 

Von Hibler (3) practised this method and claimed good results for it. 

The use of heat to destroy non-sporing organisms is of proved 
value. On the other hand, we have not at the present time enough 
reliable evidence to determine the utility of the method when it is 
appUed to the differential separation of one species of spore from 
another. The almost ubiquitous appearance in cultures of B. sporo- 
genes the spores of which are particularly resistant to heat would 
appear to militate against success. 

C. Selective Media. 
If a mixed growth of anaerobes in a tube containing meat medium 
be exammed at repeated intervals, its bacterial population will be 
found to alter from day to day. The organisms which are the 
first to develop include such types as B. welchii, B. fallax and 
mbnow septique, all of ^^jhich reach their maximal growth within 
twenty-four hours of inoculation. After twenty-four hours, these 
organisms are replaced by others of a totally different character, 
such as B. sporogenes ; and this second developmental phase in the 

i'^rp'', f '- .f ^^ t^^^^ P^^^^' i^ ^hich end-sporers, of 
which B. tetani is the most -important example, predominate 



99 

The history of such a culture may thus be said to comprise three 
different periods or epochs, each of which is characterized by its own 
distinctive flora. The organisms of the first period have been found 
to be predominantly saccharolytic in character, while those of the 
second period are proteolytic. In the third period the medium is 
found to have reached a condition in which it is unfavourable to the 
growth of organisms of the first and second periods, but in which it 
is pre-eminently suitable for certain organisms of the end-sporing 
group. 

Under favourable conditions the same sequence of events can be 
traced in the bacteriological history of a wound infected with 
anaerobes. 

The information yielded by further investigation of the facts 
outlined above has proved of great assistance to the bacteriologist, 
for on it there is based the whole principle that controls and deter- 
mines the choice of selective or differential media in the isolation of 
anaerobes. 

(A) Selective media for the saccharolytic anaerobes. 

The first stage in the isolation of members of this group of anaerobes 
consists in producing from the original tissue broth or meat culture 
a subculture which will contain an aggregation of individuals which 
display some particular biochemical feature. For example, an 
anaerobe which has the capacity to ferment a particular carbohydrate 
will grow rapidly in a medium containing that sugar and will be found 
to outnumber all the organisms that do not ferment it. It is to be 
noted, however, that this numerical preponderance may be in 
evidence only at a certain stage in the history of the culture. There 
are, therefore, two factors to be borne in mind, viz., (1) the presence 
of a fermentable carbohydrate which will ensure the development of 
organisms with the capacity to ferment that carbohydrate, and 
(2) the period in the history of the culture at which the numerical 
superiority in special individuals exists. 

To ensure a further concentration of these individuals, all that 
is necessary is to make, in tubes of the same special medium, several 
subcultures in series, each separated from the preceding parent culture 
by an appropriate interval. 

"The second stage in isolation consists in making surface growths 
from the last subculture so as to get separate colonies which can be 
picked off and grown again in the selective medium. The agar 
employed for this purpose should be mixed with serum or alkaline 
egg medium. Where plain agar only is available, the admixture of 
the material to be inoculated with serum or alkaline egg before 
spreading it on the agar surface gives very good results. The agar 
may be used in Petri dishes or as slopes in test tubes, according to 
the choice of the individual worker. Many of the anaerobes tend to 
give continuous films of growth on a moist surface and it is therefore 
desirable to free the medium from condensation water as far as 
possible. The deep agar shake method which has been practised 
by many workers, particularly by the French, gives results that are 
distinctly inferior to the surface growth method. It is, however, 
ii useful method when surface growths cannot be obtained. 

G2 



100 

It is to be noted, that of the many methods which have been 
devised for obtaining anaerobiosis in the case of surface growths on 
sohd media, only a very few provide conditions sufficiently perfect 
to promote development of the more strict anaerobes. The method 
devised and described by Mcintosh and Pildes (1) has been tested 
extensively by a number of workers during the war and has been 
found to give excellent surface growths. The use of a metal or 
glass cylinder from which the air can be extracted by means of 
a pump and replaced by an atmosphere of hydrogen, the process 
being repeated two or three times to ensure the absence of oxygen, 
provides a method which has been severely tested by Miss Eobertson 
and others at the Lister Institute with equally good results. 

It is of great advantage to examine surface colonies with a pocjiet 
lens or under a dissecting microscope. It is only by this means that 
one can distinguish differences between colonies that look alike to 
the naked eye. 

A colony which appears to be isolated may be found to be sur- 
rounded by many others, or it may be seen to have developed in the 
midst of a continuous surface film, both of which conditions are diffi- 
cult to detect by the unaided eye. The examination may also reveal the 
growth of a contaminating organism in a culture presumed to be pure. 
The distinctive appearance in texture and outline shown by colonies 
of certain of the anaerobes has been described in that section of 
this Eeport which deals with the bacteriological features of the 
individual organisms. 

As specific instances of the application of selective methods in 
the isolation of carbohydrate-fermenting organisms, the following 
examples may be cited. 

1. B. welchii. 

B. welchii attains its maximal development in freshly boiled 
glucose broth tubes in three to six hours. It is thus possible to pass 
the same material through two tubes between the morning and 
evening of the same day. The second culture is plated out in the 
evening and left to grow overnight. Colonies picked off the next 
mormng may again be passed through glucose broth. The alternation 
of growth m glucose broth combined with plating out can be repeated 
as often as is necessary. It has been found that a growth which is 
reasonably certain to be-pure can be obtained in six days, i. e., after 
SIX agar platmgs and twelve rapid passages through glucose broth. 

Ma;ltose would seem to be more rapidly and more vigorously 
fermented by B. welchii than is glucose, so that with a 1 per cent, 
maltose broth or casein digest medium, it is possible to make from 
tour to SIX serial replants in the course of a day. 

2. Vibrion septique. 

This organism ferments salicin. The reaction takes longer to 
develop than does the fermentation of glucose or maltose by B. welchU 
and does not reach its maximum till twelve to twenty-four hours 
after inoculation A medium containing salicin may, therefore, be 
used to concentrate vibnon septique where it is associated with 
B welcUt and J3. sporogenes for neither of the latter are capable of 
attacking sahcm. The growth of such a microbic mixture fn 1 per 



101 

cent, salicin broth or casein digest for twenty-four hours .followed 
by plating on to agar containing 1 per cent, sahcin, the process 
being repeated several times, has been found to yield a good chance 
of obtaining vibrion s&ptique free from other organisms. 

In both of the instances just quoted organisms other than those 
specially indicated are likely to be met with. For example, in the 
isolation of B. welchii by rapid passage through glucose or maltose 
brothtubes, B. fallax or J3. aerofetidus may be found. Each of these 
organisms gives rise to colonies which after a little experience can 
be easily distinguished from those formed by B. welchii. Similarly 
the salicin method gives a good chance of obtaining B. tertius, but 
here again the colony can be readily differentiated from that of 
vibrion septique. 

(B) Selective media for the proteolytic anaerobes. 

The best known member of this group is B. sporogenes. It grows 
much more slowly than do the saccharolytic anaerobes. If a protein 
containing medium which is sugar free be inoculated with a mixture 
of anaerobes which includes B. sporogenes, it will be found that this 
organism outgrows the others. A medium made with tap water or 
saline containing bits of coagulated egg white offers no attraction 
to any anaerobe other than B. sporogenes. Small pieces of fish 
muscle or of crab muscle may be sutsstituted for the hard boiled egg. 
An alkaline egg medium is also serviceable in obtaining a concentra- 
tion of B. sporogenes. Material taken from a culture of two days or 
more is plated on to agar. After forty-eight hours' incubation colonies 
are picked off and sown into another tube of sugar free medium. As 
before, the process is repeated several times, until the .organism 
appears to exist in pure culture. 

(C) Exhausted media. 

A medium in which B. sporogenes has grown for some time is 
found to be suitable for the development of B. tetani, B. tetanomorphus 
and B. cochlearius, but not for the development of the saccharolytic 
anaerobes or of B. sporogenes itself. It would seem to be selective 
for the organisms mentioned, and was used by Tulloch for this 
purpose. 

D. Separation of a Pathogenic Anaerobe by Animal Experiment. 

Guinea-pigs or mice may be used for this purpose. 

1. In normal animals. 

Where a mixture of organisms contains only one pathogenic 
anaerobe, then this particular organism can be recovered after death 
from the heart blood of an animal which has succumbed to an intra- 
muscular inoculation of the mixture. If the pathogenic anaerobe 
be vibrion septique or B. oedematiens it may be obtained in pure 
culture from the animal's heart blood. It is to be noted, however, 
that certain strains of B. oedematiens may kill by intoxication 
without spreading into the circulating blood stream. Where the 
pathogenic anaerobe is B. welchii there is often a tendency for other 
organisms Such as B. sporogenes or streptococci or coliforms to 
appear in the blood along with B. welchii. 



102 

On the other hand, where two or all three of these pathogenic 
anaerobes are present in a mixture, any one of them or any com- 
bination of them may be found in cultures made fromheart blood. 

It follows, therefore, that the use of a normal animal for such 
a passage experiment is not likely to give a rehable result unless oiie 
can be reasonably certain that only one pathogenic anaerobe is 
present in the inoculated mixture. 

2. In protected animals. 

A guinea-pig which has been passively immunized by inoculation 
with antitoxic sera against B. welchii, vibrion sejptigue or B. oedema- 
tiens is capable of successfully resisting infection by the corresponding 
organism. For example, an animal adequately protected against 
B. welchii will never be found to develop a fatal B. welchiA infection 
though it may readily succumb to vibrion septique or to B. oedema- 
tiens if either of these be present in the inoculum. The same holds 
good for vibrion septique and B. oedematiens antitoxic sera, the im- 
munized animal being afforded absolute protection against infection 
by the corresponding organism but not against other pathogenic 
anaerobic organisms. 

Further, it has been possible by using appropriate mixtures of 
these antitoxic sera to protect animals against a combination of any 
two or of all three of these organisms. 

One may thus construct in a guinea-pig a sort of filtering mechan- 
ism which will retain or hold up certain organisms by inhibiting 
their development in the animal, and which will at the same time 
allow the growth and subsequent passage into the blood-stream of 
any pathogenic anaerobe against which no specific protection has 
been induced. This method was actually used in determining the 
pathogenic anaerobes present in specimens of muscle taken from 
fifty cases of acute gas gangrene in Prance. A combined serum 
containing 4,750 units of welchii antitoxin and 5,000 units of septique 
antitoxin per 10 c.c. was given intraperitoneally in a dose of 4 to 5 c.c. 
to guinea-pigs of 250 grm. weight, twenty-four hours before the 
moculation of the injecting dose of mixed culture. The latter 
consisted m each case of 1 c.c. of culture mixed with an equal volume 
of the same serum and left in contact with it for one hour at room 
temperature before mtramuscular inoculation. Of the protected 
gmnea-pigs which succumbed none died with a B. welchii or vibrion 
septique infection. The organisms which were recovered from the 
heart blood of these animals proved in every case to be B. oedematiens 
or some other different anaerobe. The full results wHl be discussed 
m detail elsewhere. This animal filtration method has been given 
an extensive trial and it can be safely recommended as providing 

Pn[w if r^'^fv ''^^^^^^ ^''^^ ^ ™^*^^^ «f anaerobes a purl 
culture of B. welch%%, vihnon septique or B. oedematiens 

in a case of urgency, three guinea-pigs or mice nroteeted with 

appropriate combinations of the three antlera (TeTchrvSrfon lep 

StTromth?; r 'I "^Y \' '"^^"^^*^^ -^*^ th^ wound Sarle 
n esent Zv bP^S. ■. ^ ^\'F^'^' °* *^^ pathogenic anaerobis 
over n gh^ ^ '^'^ ^^ *^' ''™1* «^*^i^e<i ^^ tbe animals 



103 

E. Summary. 

The above outline of the several methods which may be employed 
in the isolation of the anaerobes must be taken as affording no more 
than general indications in regard to the methods that are most 
likely to yield success. The individual researcher will develop them 
along the lines of his own particular need and will find that they 
are ca,pable of modification and adaptation in a number of useful 
directioiis. An active and resourceful manipulation of the methods 
of isolation available is rendered necessary by the great variety in 
the problems presented. 

The worker who sets out to obtain pure cultures of the anaerobes 
essays no easy task. There are fortunate occasions when he may be 
able to plant out the original, material directly on to plates, as, for 
example; when he is dealing with a post-mortem heart blood or with 
pieces of infected muscle taken at some distance from a wound. 
In the majority of instances, however, he is faced with material that 
is grossly contaminated, and in deahng with it the following points 
should be constantly borne in mind. 

1. Growth in a special selective medium is merely preparatory to 
and must alternate with plating out on serum agar. The anaerobic 
conditions employed must be rigorous enough to give good surface 
growth of the more exacting anaerobes. 

2. The criterion of purity of a culture is based entirely on its 
consistent behaviour when it is grown on a large range of different 
media and when its cultural reactions are observed over a long 
period of time. 

, 3. The worker need never hesitate to admit the impurity of one of 
his cultures when 'definite proof of the same is produ'ced before him, 
for the very best bacteriologists have been deceived by the anaerobes. 
It is only by the exercise of constant vigilance and the maintenance 
of a keen critical attention in regard to the possibility of contamina- 
tion that he can escape these pitfalls. 

(iii) Preservation of Cultures. 

A few general directions are given for the preservation of cultures 
when isolated. 

The desirable conditions under which to keep stock cultures are 
those in which the organism will form spores which will retain their 
vitality over long periods. As a general rule, media containing 
fermentable sugar and milk media should be avoided as the acid 
reaction produced by many anaerobes is unsuitable for the prolonged 
maintenance of the culture. 

The vast majority of the anaerobes may be subcultured on to meat 
medium, coagulated serum, or broth with a piece of coagulated 
egg white, and the tubes sealed in a blow-pipe after forty-eight to 
seventy-two hours' growth. The sealed cultures can be stored in 
the dark at the temperature of the laboratory for more than a year 
without losing their vitahty. When the sealed tubes are opened they 
should be subcultured on to meat medium, a large inoculum being 
used. The strain should then be grown on milk, meat medium, 
coagulated serum and gelatine, and the sugar reactions tested so as 



J 04 

to assure the worker that the culture is pure and typical. It is often 
at this period when a resting culture is once more subcultured after 
a long interval, that contaminations are found in a supposedly pure 
culture. 

Certain anaerobes require special treatment. 

B. welchii is best preserved in alkaline egg fluid or on coagulated 
serum as spores are more certainly formed in these media than in 
the others in use. 

B. Jallax is an organism which is apt to die out in stock cultures, 
spores are not formed readily and it is not always a prolific grower. 
It is best kept in alkaline meat medium or coagulated serum and it 
should be incubated for at least three days before the tubes are 
sealed. Cultures should be made in triplicate and the strain should 
be subcultured every four to six months. 

B. cochharms strains are also easily lost. Meat medium of an 
alkaline reaction, broth with a piece of living tissue or coagulated 
egg white, are the best media for stock cultures. In pure culture 
this organism is not a prolific grower. 



IV. SEEOLOGY. 

1. Toxins and Antitoxins. 

(i) Historical Introduction. 
A. B. welchii. 

Since the appearance in 1892 of the classical article by Welch and 
Nutta,ll on the gas bacillus, the study of the pathplogy of infection 
by this organism has attracted the attention of many workers. The 
exact significance of the pathological changes that are induced in 
the animal body as the result of infection have been variously 
interpreted. 

Eugen Fraenkel (1) looked upon the disease as being an intoxication 
due to the absorption of decomposition products formed in the tissues. 
Much the same view has been advanced in the course of the war 
by Conradi and Bieling (1) who considered that the various acids 
produced by the fermentation of carbohydrates were responsible 
for the tissue necrosis, and that this necrosed tissue favoured the 
development of saprophytic anaerobes, the resulting putrefactive 
products when absorbed giving rise to intoxication. Kamen in 1904 
t^i that 8-day old cultures contained a haemolytic substance 

riSoullnn '/''^'\ f "'°^'*^"^'/ ^^ ''''' and decided that 
although strong toxic substances could not be produced in vitro thev 
were probably formed in the animal body ^ 

Passim (1) found that the toxic substances in filtrates from old 
culures survived a temperature of 100° C. for 15 rninutes and 
further, that hey possessed no antigenic properties. ' 

McCampbell m 1909 attributed the lesions produced in animals 



105 

Stewart and West in 1916 came to the same conclusion. Of 
others who have investigated the toxic substances present in filtrates 
of B. welchii mention may be made of Korentchevsky, Herter, 
Simonds (1), Costa, and Troisier (3), and Ouranoff. 

In none of the papers cited, however, is there any adequate 
denaonstration that the toxic substances referred to possessed anti- 
genic properties. 

Eosenthal (1) appears to have been one of the first to attempt 
the preparation of a serum against B. welchii. He inoculated horses 
with gradually increasing doses of culture. The resulting antiserum 
proved to have only slight protective properties. 

In 1916 Weinberg (6) reported that he had been successful in 
preparing an antibacterial serum to B. perfringens, the horse being 
immunized with washed bacilli. This serum, in a dose of 0-005 c.c, 
neutralized one lethal dose of culture in guinea-pigs. There can be 
little doubt that this serum was really antitoxic but the difficulties 
experienced by Weinberg in obtaining a good toxin derived from 
B. welchii militated against any successful demonstration by him 
of its antitoxic value. The first clear exposition of B. welchii toxin 
is to be found in the communications of Bull and Pritohett (1). 
These workers proved that the necrosing and haemolytic properties 
of B. welchii filtrates were due to the presence of a real exotoxin, 
which could be destroyed by heating to between 60° and 70° C, 
and which was capable of stimulating the production of antitoxin 
when inoculated into animals. They further showed that this 
antitoxic serum when employed under proper conditions inhibited 
and arrested infection with B. welchii. 

B. Vibrion se-ptique. 

In contradistinction to our knowledge of B. welchii antitoxin, 
which is very recent, our information in regard to immunization 
against vibrion septique infection dates back to the publication of 
a communication by Eoux and Chamberland in 1887, Duensch- 
mann 1894, Leclainche and Morel 1901, M. Nicolle, Cesari and 
Baphael 1915, Eaphael and Prasey may all be cited as yielding proof 
of the production of protective sera against Pasteur's organism. 

0. B. oedematiens. 

This organism was described for the first time in the course of 
the war by, Weinberg and Seguin ((1) (2)). These observers have 
accurately defined the properties of the specific toxin, and have 
succeeded in producing high-value antitoxic sera in horses. There 
is no proof that the organism described in 1894 by Novy as B. oede- 
matis maligni II, and later by Migula as B. Novyi, produced toxin 
although Novy (p. 224) suggested that this was likely from the 
scarcity of bacilli in the oedematous areas. 

(ii) The Preparation of Toxin. 
A. Toxin of B. welchii. 

. . Bull (1) produced B. welchii toxin by growing the organism in 
glucose broth containing large amounts of sterile rabbit muscle. 
The aseptic removal of muscle from a rabbit presents considerable 



106 

.echnical difiaculties. To eliminate this particular ' complication 
has been the aim of all those who have since dealt with the problem, 
and more particularly perhaps of those workers who have had to 
prepare toxin on a sufficiently large scale for the immunization 
of horses. 

The preparation of toxin of B. welchii is beset with even more 
difficulties than is the production of the toxins of diphtheria and 
tetanus. There are so many variable factors to be considered in 
the process and many of these factors are so difficult of control that 
the issue can never be accepted as a certainty. With repeated 
experiment, however, and a fuller understanding of the essentials 
to success it becomes possible to establish the in vitro conditions 
which assure the probabihty of a fortunate result. 

The points which appear to be of special significance may be 
considered under the following headings : (a) the medium ; (b) 
the organism ; (c) the inoculum ; (d) the conditions of growth ; 
(e) filtration. 

(a) The Medium. — ^Experience has shown that the best medium 
for use on a large scale is a muscle broth which is prepared as follows : 

1. A mixture of minced horse muscle and tap water, in the propor- 
tion of 1 lb. muscle to 1 litre water, with 0-5% common salt, is boiled 
in a cauldron for 1 hour. 

2. The cooked meat is then removed and, to the broth, after 
filtration there is added 1 per cent, of peptone and 0-2 per cent, of 
glucose. 

3. The broth is then heated tiU the peptone is dissolved, and 
sufficient caustic soda is added to make it shghtly alkaline to litmus. 

4. It is then boiled for half an hour and filtered to remove the 
precipitate. 

5. Into a series of 4-Utre bottles (double Winchester quarts) 
there are placed about 2 handfuls of the cooked meat and 4 litres of 
broth. 

6. The bottles are plugged with cotton wool and autoclaved for 
3 hours at a pressure of 22 lb. per square inch. 

7. The end reaction of medium so prepared varies from 8 per cent, 
to 4 per cent, acidity when tested with phenol phthalein as the 
mdicator. 

PeptoTie appears to be a necessary constituent for the formation 
of toxin but it IS to be noted that the commercial brands differ 
considerably m value. 

Glucose has a powerful influence in promoting the toxicity of a 
filtrate, although its action in this respect must not be looked on 
as specific. It has a direct effect in encouraging the growth of 
B welchu and an indirect beneficial effect in that after fe^entation 
It provides the degree of acidity (phenol phthalein titration) that 
appears to be essential to the formation of toxin 

« SZfTrl^^ !v'^ •' •^*'*,^°'' °* ^ ^- "^'^^^ filtrate depends to 
carbohVdS ? "^T '"^'f'^^fr^oi the medium in fermentable 
statement b«; W 7if V^' ^''* to demonstrate this clearly and the 
statement has been fully borne out by .the findings of other workers 
Startmg with a sugar free medium b4 showed that the end Reaction 
taken at intervals from the first to the thirteenth day ofTnocuSon 



107 

lay between 2 per cent, and 2-5 per cent, acidity. By increasing the 
amounts of glucose in the muscle broth he found that the acidity 
rose, until, with a content of 3 per cent, in glucose an acidity was 
reached of 8 per cent. He further demonstrated that, although 
the presence of glucose in amounts up to 1 per cent, had little un- 
favourable influence on the toxicity of the filtrate, yet glucose in 
large amounts had a decidedly deleterious effect on toxin production. 
There is thus a distinct relationship between the end reaction of 
a filtrate and its toxicity. 

Experimen.ts conducted with 0-2 per cent, glucose muscle broth 
media in which the initial reaction has been arranged so as to vary 
from 8 per cent, acid on the one hand to 3 per cent, alkali on the other, 
have suggested that although the end reaction (after the growth 
of B. welchii) comes out at practically the same figure, viz. an acidity 
of 4 per cent, to 5 per cent., yet toxin production is most marked in 
media which were originally acid. It would seem therefore that 
B. welchii prefers an acid medium for the production of toxin and that 
if the organism be grown in an alkahne medium it expends its 
energies on the fermentation of the glucose present at the expense 
of toxin formation. Experiments dealing with this matter are still 
in progress. 

The buffering of a medium with sodium phosphate has not, up to 
the present, materially influenced the production of toxin. 

Fresh muscle. — The influence of fresh muscle in encouraging toxin 
formation is unquestionable. It has not been found necessary, how- 
ever, to use the large volumes recommended by Bull. A small 
piece of muscle about the size of a hazel-nut is sufficient for the 
purpose when introduced into a 4-litre bottle of medium. The 
muscle may be replaced by a few c.c. of a freshly prepared filtered 
muscle extract. The evidence as to the exact nature of the substance 
in muscle which is responsible for this beneficial effect is as yet 
incomplete. 

(&) The Organism. — There are considerable differences among the 
strains of B. welchii in regard to their toxin-producing capacity. 
Only a relatively small number appear to be good toxin producers 
and it is probable that further research is necessary to find a con- 
sistently good and reliable strain giving high-value toxin. Bull's 
strain, 61 7D, has been found to give very good results in the hands 
of other 'workers. There are, however, several strains available at 
the present time that are quite as good. The capacity of any given 
strain for producing toxin may be enhanced by increasing its viru- 
lence. This can be done by repeated animal passage, preferably 
through pigeons which are more convenient animals for this purpose 
than guinea-pigs. The inoculation of a lethal dose of culture into 
the pectoral muscles of a pigeon produces an illness which is fatal 
in 6 hours or less. The pectoral region in pigeons is readily accessible 
and portions of infected muscle can be easily removed under aseptic 
precautions. An emulsion of infected muscle can be introduced 
straightway into a pigeon without any intervening culture and thus 
one may carry out several rapid passages in a short space of time. 

It is to be noted that the virulence of a strain may increase by 
passage up to a certain point and then suddenly fall. This rhythmic 



108 

rise and fall in virulence has been noted several times. An organism: 
of ebbing virulence may be set aside in a culture tube in the hope 
that when next it is used- for passage the tide may have turned, 
A feature which all strains of B. welcUi seem to possess at some 
period or other of their growth in artificial media, is the tendency 
to produce a sticky gelatinous material, the exact nature of which 
is as yet uncertain. The phenomenon as it occurs in liquid cultures 
of B. welchii, and the peculiar characters of the surface colonies 
of an organism which is thus affected have been already noted. The 
tendency to produce this sticky material in no way interferes with 
the toxin producing capacity of an organism, but it does place a very 
formidable barrier in the way of successful filtration. On the 
assumption that an organism in this state might be ' sugar-sick ', 
attempts have been made to rid it of this ailment by long cultivation 
in carbohydrate-free media, but without any appreciable result. 
Miss Lacey has found, however, that if a sticky organism be washed 
repeatedly and kept in normal saline solution for a week in the ice 
chest, it loses the habit of forming glutinous cultures, 

(c) The Inoculum. — The inoculation of a medium may be conven- 
iently carried out by introducing a piece of infected pigeon muscle 
taken from a bird that has died as the result of infection. This method 
has been adopted during the war by the staff at Brockwell Hall 
in the preparation of toxin in large quantities, and has been found 
to yield consistently good results. The titre of the toxin may be 
still further improved by introducing a large dose of the weak broth 
culture (100 c.c. to a 4-litre bottle) obtained from the preceding 
pigeon passage. 

{d) The Conditions of Growth. — It is not necessary to adopt special 
anaerobic methods in growing B. welchii for the purpose of producing 
toxin. The meat medium already mentioned, if freshly autoclaved or 
re-steamed just before use, offers the conditions suitable for growth. 
The inoculum for each 4-htre bottle consists of a small piece of infected 
pigeon muscle plus 100 c.c. of culture. The cultures are allowed to 
mcubate at 37°-38° C. for 16 to 24 hours and then filtered. Cultures 
yield just as good toxin formation if grown at 42° C. 

(e) Filtration. Cultures are passed through layers of paper pulp 
m large porcelain cylinders until the filtrates are quite clear and 
translucent. The material is then passed through Berkefeld filters, 
after which it is tested for sterility. It should be stored in a cool 
place away from the light. 

B. Vihrion septique toxin ^ 
B. oedematiens toxin S 

The conditions which seem to determine the formation of these 
toxins are identical with those which obtain for B. welchii toxin. 
ihe same meat broth medium may be employed, the inoculum 
consisting as before of infected muscle plus culture 

Miss Eobertson has been able to prepare very high titre vihrion 
septique toxm using a plain broth or glucose broth medium which 
?.n.nW. 1 ^"I f •.*'' ^ .^t<=tion of Ph. 7-8 to 8-0, and which has been 
inoculated with bits of infected liver taken from guinea-pigs dead of 
vibrion septique infection. ^^ 



109 

Lastly, it cannot be too strongly emphasized that the factors 
which govern toxm formation in vitro are still beyond our control. 
However careful and exact one's preparations may be, the result 
appears still to be a matter of hazard. 

2. The Properties op the Toxins elaborated by B. welchii, 

ViBRION SEPTIQUE, AND B. OEDEMATIENS. 

(i) Physical Features. 

1. These toxins are all thermolabile and may be destroyed bv 
heating to 70° C. for 30 to 60 minutes. 

2. B. welchii toxin is reported by Bull as being non-dialysable. 
No similar evidence is as yet available in regard to the toxins of 
vibrion septique and B. oedematiens. 

Each of the toxins mentioned depreciates considerably in potency 
if passed repeatedly through Berkefeld candles and it may be con- 
cluded therefore that each is hkely to be non-dialysable. 

3. All the three toxins are very susceptible to low concentration 
of acid. B. oedematiens would appear to be thg most sensitive and 
vibrion septique toxin the least vulnerable, while the toxin of B. 
welchii would seem to occupy an intermediate position in this respect. 

In each case the toxin appears not to be precipitated out of solution 
by contact with acid but to be actually destroyed. There is no clear 
evidence that it is converted into toxoids. 

B. welchii toxin is said to be readily destroyed by free chlorine. 

(ii) Biological Characteristics. 
A. The toxin of B. welchii. 

1. It is haemolytic both in vitro and in vivo. The in vivo action 
is best seen in animals which have had an intravenous inoculation, 
though it can also be demonstrated constantly in mice which die 
as the result of an intramuscular dose of toxin. 

. Bull relates the case of a rabbit which received a dose of 1 c.c. 
intravenously at 10 a.m. and which had then a red corpuscle count 
of 5-4 millions. A progressive destruction of red cells occurred, 
until, 7 hours after inoculation, the count was only 1-0 million 
per c.mm. Also, a pigeon (quoted by Bull) which gave an initial 
count of 4-28 millions red corpuscles was found d^ hours after inocula- 
tion to have no more than 0-8 million red cells per c.mm. in its blood. 

The haemolytic action of the toxin is not due to the presence of 
acids, for it can be demonstrated equally well with toxins that have 
been carefully neutralized. 

Haematuria is a constant finding in mice ; it is less frequent in 
rabbits and in guinea-pigs. 

2. It is necrotic. This action is best seen in the case of muscle, 
though it can be demonstrated also in skin by intracutaneous 
inoculation. The muscle becomes oedematous, friable, and necrosed 
so as ultimately to form a difHuent pulp. 

3. It produces a marked serous oedema which may be found in 
large quantities in the subcutaneous tissues. This oedema fluid 



110 

when collected is found to be almost free from cells. It clots readily 
and yields a clear serum which is often of a bright yellow colour. 
In mice, as a rule, the fluid is blood-stained and gives after centri- 
fuging the spectrum of oxyhaemoglobin. 

4. It stimulates involuntary muscles. A rabbit which has received 
an intravenous dose of toxin exhibits violent intestinal peristalsis 
accompanied by diarrhoea. A similar reaction, though less common, 
occurs in mice and in guinea-pigs. The vomiting which, occurs in 
pigeons and which may be such a marked feature of the terminal stages 
of gas gangrene in man is probably explicable on the same grounds. 

5. Washed surface growths of B. welchii do not produce infection in 
laboratory animals even when inoculated in large doses. It has been 
found also that the washed bacilli separated out of a liquid culture 
are, by themselves, non-infective. If, however, an emulsion of washed 
bacilli be mixed with a sub-lethal dose of toxin, then there can be 
induced a typical gas gangrene infection which is rapidly fatal. 
The toxin therefore stimulates the development of B. welchii in the 
animal body and in this respect has aggressive properties. 

Only a very small number of bacilli under these circumstances 
is requisite for initiating infection. On the other hand the amount of 
toxin that is necessary for the combination to be successful is a con- 
siderable fraction of the lethal dose. 

It is probable that there are two contributory factors involved in 
the mechanism of the aggressive reaction, viz. (a) the necrosis of 
tissue which need not necessarily be looked upon as specific (Bullock 
and Cramer (1)) ; (b) the antiphagocytic property of the toxin which 
is specific. Both these effects can be completely inhibited by 
antitoxin. 

6. The toxin has antigenic properties. 

(a) By repeated injection of toxins into animals it is possible 
to set up an active immunity. 

(b) Under the same circumstances an animal produces an anti- 
serum which can be shown to possess the following properties. 

I. It neutralizes toxin in vitro, so that a mixture of toxin with 
the proper proportion of antiserum is innocuous when inoculated 
into animals. 

Such an antiserum is capable of neutralizing any B. welchii toxin. 

II. The antiserum when inoculated into animals produces a passive 
immunity to B. welchii toxin and also to B. welchii infections. 
Protection against the lethal effect of toxin was found by Bull to 
persist for two weeks after the giving of serum. 

The prophylactic use of the serum against actual infection was 
clearly demonstrated by Bull in the following experiment. Guinea- 
pigs which had received a small dose of serum were found on the first 
day a,fter inoculation to be capable of resisting at least 150 minimal 
lethal doses of culture. On the fifth day they survived 300 lethal 
^0^?^' oi^^*"e eighth day 60 doses, and on the eleventh day 20 doses. 

ill. ihe serum can be used therapeutically to treat the disease 
when this is already established. Experiments by BuU and others 
have proved that, if large guinea-pigs be used, a small dose of serum 
given 24 hours after the inoculation of a lethal dose of culture ensures 
complete recovery. 



Ill 

B. The toxin of vibrion septique. 

This toxiii is Kaemolytic, myolytic, and oedema-producing, but to 
a less degree than B. welchii toxin. On the other hand, it does not 
kill laboratory animals when given subcutaneously or intramuscularly 
except when inoculated in very large doses. Its lethal effect is best 
demonstrated by intravenous injection. 

Its antigenic properties are in general the same as those outlined 
for B. welchid, with this important difference. An antitoxic serum 
to vibrion septique does not appear to protect if it is given to an 
animal after the infection is established. 

C. The toxin of B. oedematiens. 

The distinctive feature of this toxin is the extensive and heavy 
gelatinous oedema it produces both in muscle and in subcutaneous 
tissues. It kills in small doses and gives rise in horses to an antiserum 
which is relatively more potent than those so far obtainable against 
B. welchii toxin or vibrion septique toxin. In this respect it more 
closely resembles the toxin produced by B. tetani. 



3. The Symptoms and Pathological Changes produced by 
THESE Toxins and the Estimation of their Minimal Lethal 
Dose in Animals. 

(i) The Toxin of B. welchii. 
A. Mice. 

Intramuscular inoculation of a multiple (5 to 10) of the lethal dose 
is followed by urgent dyspnoea and death with convulsions in 2 to 
3 hours. In these animals at post mortem, the local lesion may be 
very slight or absent, and the appearance of the viscera is normal. 
With a smaller multiple of the lethal dose (2 to 5) death occurs in 
6 to 8 hours. With one lethal dose there is produced an illness which 
lasts for about 48 hours. 

The progressive oedema which develops at the site of inoculation 
within a few hours and which extends rapidly upwards into the 
subcutaneous tissues of the abdominal wall is easily palpable. The 
inoculated limb is held stiffly with the toes firmly flexed. In 24 to 
48 hours the animal refuses food, loses weight, and has a staring coat. 
There is a moderate degree of diarrhoea, and haematuria is frequent. 
The limb and trunk muscles become paretic so that the animal 
totters in attempting to walk. Dyspnoeic periods with violent 
and laboured respiratory movements are an almost constant feature 
of the illness in its later phases. Ultimately the animal succumbs 
with a flaccid paralysis of all its limb muscles. 

The post-mortem findings in such an animal are as follows : A 
blood-stained oedema fluid is found extending from the site of 
inoculation into the opposite groin and upwards into both axillae. 
The muscles of the inoculated limb are pale, friable, and necrosed, 
but are never reduced to the liquid grumous state that is such 
a marked feature in cases of actual infection. The bladder is usually 
distendedwith urine which is deeply stained with blood. The kidneys 



112- 



are reddened and engorged and bulge from their capsules when cut 
into. The liver may show fatty degenerative changes either m 
patches or diffusely distributed throughout the whole viscus. There, 
may be an increased amount of fluid in the serous cavities, more 
particularly in the pleura and in the pericardium. The duodenum 
and part of the small intestine are distended with hquid contents 
which are heavily bile-stained. The rectum is often filled with gas 
and semi-fluid contents. The suprarenals may be reddened and the 
spleen may be much enlarged but both these changes are inconstant. 
The titration of B. welchii toxin as carried out by the intra- 
muscular inoculation of mice is exemplified in the following tabulated 
experiment which was undertaken to estimate the potency of the 
toxins produced by five different strains of the organism. 



strains. 



B. 85 

B. Petrie . 
B. King . 
B. Gorman 
B. Quirk . 



1-0 

+ + 
+ + 
+ + 
+ + 
+ + 



Amounts of toxin ex. 
0-8 0-6 0-5 0-4 0-3 

++ +— — 

++ ++ +— 

++ ++ ++ — 



0-2 



. lethal 
' dose. 



0-5 C.C. 
1-0 CO. 
0-6 CO. 
1-0 C.C. 
0-3 c.c. 



= death within 48 hours. 



— =recovery. 



Two animals were used in every case. 

Subcutaneous inoculation of B. welchii toxin into mice produces 
an illness which is exactly similar to that which follows on intra- 
muscular injection. 

The intravenous inoculation of mice with lethal doses of the same 
toxin would seem to give irregular results. Death may occur in 
a few minutes or after many hours, so that it becomes by no means 
easy to determine the exact minimal lethal dose. 

B. Babbits. 

One minimal lethal dose of toxin given intravenously to a full- 
grown rabbit produces death in 2 to 2J hours. A smaU percentage 
of rabbits may survive for a longer period, but as a rule, if death does 
not occur within 3 hours, the animal recovers completely. With 
multiples of a minimal lethal dose death may follow in J hour. In 
this case the symptoms are those of muscular excitation and restless- 
ness, with marked dyspnoea and convulsive seizures ending in death. 

The inoculation of one minimal lethal dose is followed immediately 
by a period of dyspnoea ; but this is usually transitory and passes 
off m a few minutes. Usually no symptoms manifest themselves 
for 1 to 11 hours. The animal then becomes restless and changes his 
position frequently. It is attacked with recurring periods of 
breathlessness, which, though at first slight, soon become longer in 
duration and severer in degree. At the same time it develops 
muscular tremors and paresis.. It hes down, slightly to one side, 
with all its hmbs fully extended. The intestines can be seen in 
violent peristaltic movement under the flaccid muscles of the belly 
wall. It defalcates frequently and may have actual diarrhoea 
with haematuria. The head is thrown back into the nape of the neck 
in a position of opisthotonos, and the nose is raised high so as to 



113 

relieve air-hunger. Towards the end, the flaccidity of the muscles 
becomes absolute. It can no longer raise its head, nor can it 
recover its position when this is altered by the observer. The 
respiratory embarrassment becomes accentuated. Periods of shallow 
rapid breathing give way to laboured efforts, separated by longer 
arid longer phases of apnoea. Finally there occur a few convulsive 
spasms and the animal is dead. 

With the exception of a small amount of blood-stained fluid in 
the peritoneal cavity, there is very little to be found that is charac- 
teristic when these animals are examined post mortem. There may 
be collections of fluid in the pleural and pericardial sacs. The urine 
is at times blood-stained, and exceptionally the serum may be tinted 
with haemoglobin. The rectum may have liquid and gaseous con- 
tents. Otherwise no naked-eye changes are to be observed. 

C. Guinea-pigs. 

The illness which results in guinea-pigs after the intramuscular 
inoculation of B. welchii toxin resembles in its main features that 
already described in mice. The post-mortem appearances are much 
the same, except that the effects of haemolysis are often less obvious, 
while reddening and enlargement of the suprarenals are probably 
more common. Numerous instances of the titration of B. welchii 
toxin by this method can be found in the experimental data recorded 
by Bull and Pritchett and by De Kruif and his colleagues. 

The intravenous inoculation of guinea-pigs does not appear to 
afford a reliable means of titration, because it is found difficult to 
fix a time limit for the recording of a lethal effect. 

D. Pigeons. 

Pigeons succumb to a lethal dose of toxin, given intramuscularly, 
in from 4 to 8 hours. A multiple of the lethal dose may produce 
death in 3 hours. When introduced by the intravenous route the 
toxin kills more rapidly (Bull and Pritchett). 

The titration of a stable B. welchii toxin, precipitated out of 
solution by absolute alcohol, is given in the subjoined table, the 
numbers given representing milligrammes of the dried product. 



Mice 


Guinea-pigs' 


Rabbits 


Pigeons 


intramuscular. 


intramuscular. 


intravenous. 


intramuscular. 


3-0+ + 


10+ 


50+ 


5-0+ 


2-0+ + • 


5+ + 


30+ 


2-5+ + 


1-0+ + 


3+- 


20+ + 


2-0+ — 


0-8 


2-5 


15- 


1-5 


. 0-5 


1-0 


10— 


10— 


0-5 








m.l.d.=l-0 


m.l.d. =5-0 


m.l.d. =20 


m.l.d. =2-5 



(ii) The Toxin of Vibrion septique. 

. It: would seem to be so difficult to prepare a vibrion septique toidn 
which when inoculated subcutaneously or intramuscularly will kill 
the animals that are commonly used for experiment, that most 
observers have had recourse to intravenous injections. Where mice 
and guinea-pigs are inoculated by this route, the problem of setting 
a satisfactory time limit in estimating the significance of a positive 



114 

result presents just as great a difficulty as that which occurs with 
B. welcUi toxin under similar circumstances. On the other hand 
where rabbits are used, there is less uncertainty in this matter because 
these animals either die in a few minutes or recover completely. 

The illness in rabbits has many features in common with that 
which follows the injection of B. welchii toxin. There are, however, 
certain characteristic points which distinguish it from the latter. 

1. The incubation period varies from a few seconds up to 20 or 
30 minutes. 

2. The clinical course of the illness is more fulminant in its mani- 
festations. The disease tends to be convulsive rather than paralytic 
in character. Within a few seconds or minutes of receiving an intra- 
venous injection, a rabbit may suddenly leap from the cage, perform 
a series of violent convulsive movements about the floor and drop 
dead in a corner of the room. Or again, the course of the illness may 
exactly resemble that produced by B. welchii toxin, except that it 
is compressed into a few minutes instead of being spaced over an 
hour or more. 

These animals show no special pathological changes post mortem. 
Exceptionally, where death has been delayed for several hours, small 
amounts of blood-stained fluid may be seen in the serous cavities. 

In view of the necessity for testing large numbers of toxins during 
the war and the difficulty of obtaining rabbits in sufficient numbers 
for this purpose, there has been elaborated at the Wellcome Physio- 
logical Research Laboratories a method of titrating vibrion septique 
toxin which depends on the estimation of the minimal oedema 
causing dose in mice. It has been found possible to determine this 
amount of toxin with sufficient accuracy to give reliable comparative 
values in the standardization of different antisera. The unit of 
vibrion septique antitoxin has been designated as that amount, of 
serurn which, when mixed for one hour at room temperature with 
10 minimal oedema-causing doses of toxin, will completely inhibit 
the production of oedema in the animal which is inoculated with 
the mixture. The method, however, is one that necessitates constant 
practice, and the results are open to wide differences in interpretation 
which depend entirely on the experience of individual observers. 

(iii) The toxin of B. oedematiens. 

This toxin kills laboratory animals when given intravenously, 
subcutaneously, or intramuscularly. It can be very easily titrated 
by the intramuscular injection of mice, the final readings in regard 
to a positive or negative result being recorded at the end of 48 hours 
as m the case of B. welchii toxin. 

Animals which succumb to B. oedematiens toxin show the same 
massive gelatmous oedema that characterizes a fatal infection by 
this organism. The muscles at the site of inoculation are much 
swollen with oedema fluid. They are pale, and often occupied by 

rCionSf>?™ haemorrhages, but"^ are not necrosed^ Smafl 
ettusions into the serous sacs are common 

tox?n^:S"out ifmic?"""*^ the .titration of a B. oe5.nu.ti.ns 



115 


Toxin LE 235 


c.c. 


0-01 + + 


0-005 + + 


0-002 + + 


0-001 + + 


0-0005+ + 


0-0002+ + 


0-0001 



4. The Standaedization of Antishea. 
(i) B. welchii antitoxic sera. 

Bull estimated the titre of his antitoxic sera by determining the 
smallest amount of serum that was requisite to neutralize one minimal 
lethal dose of toxin in a 250 grm. guinea-pig or a 300 grm. pigeon. 
This amount of serum he has defined as the unit of antitoxin. 

Where large numbers of different sera have to be tested this 
method becomes a very expensive procedure. To obviate the diffi- 
culty, the Wellcome Physiological Laboratories have used mice in 
place of guinea-pigs and pigeons. As has been already indicated the 
minimal lethal dose of B. welchii toxin can be readily estimated in mice. 
For testing purposes at the W.P.E.L. the unit of antitoxin has been 
fixed as being twice the amount of serum that is sufficient to neutralize 
the lethal effect of two minimal lethal doses, the mixture of toxin and 
serum being left in contact for one hour at room temperature before 
inoculation. The results obtained by this method are claimed to yield 
a figure-in unitage which within reasonable limits is identical with that 
obtained when the same sera are tested according to Bull's method. 

A. Titration of B. welchii toxin LW. 218. 





0-4 + + 








0-3 + + 








0-25+ + + + 






0-2 +-- 


— — 






0-15 








0-1 








0-1 








m.l.d. =0-250.0. 




B. Titration of antisera 


against two minimal lethal doses, viz 


0-5 c.c. 


SEBA 






115B 


25V 


14b 


810 


0-1 + + 


+ + 








0-01 + + 


+ + 








0-005 









0-004 









0-003 




+ - 




0-002 




+ - 




0-001 + + 


+ + 


+ + 





0-0008 









0-0006 









0-0005 






h 


0-0004 






+ + + + 


0-0002 






+ + 


O-OOOl 






+ + 


Unitage 





125 


1,000 


+ = death in 48 hours. 


— = 


= recovery. 




H2 







116 

The previous record shows an evaluation of some sera thus 
tested. The sera designated 115B, 25V, and 14b were German 
anti-gas gangrene sera captured on the Western Front, while 810 is 
a single horse serum prepared by Weinberg. The latter in addition to 
B. welchii antitoxin contains antitoxin to vibrion septique and to 
B. oedematiens. 



(ii) Vibrion septique antitoxic sera. 

French workers have defined as the unit of vibrion septique anti- 
toxin the smallest amount of serum which will neutralize the fatal 
effect of one lethal dose of toxin given intravenously to a full-grown 
rabbit, the mixture of toxin and serum being kept at room tempera- 
ture for one hour before inoculation. 

The following record illustrates the titration of Weinberg's serum 
810 by this method. 

A. Titration of toxin. 



20 CO. 
1-5 e.c. 
1-5 c.c. 
1-0 C.C. 
1-0 C.C. 



+ 7 min. 
+ 15 min. 
+ overnight 
recovery 
recovery 



B. Titration of serum against 5-0 c.c, i 
doses. 

serum. 
0-01 
0005 
0-004 
0-004 
0-003 
0-003 
0-002 
0-001 



e. 3 certain minimal lethal 



result. 
recovery 
recovery 
recovery 
recovery 
+ 20 mins. 
+ overnight 
+ 8 min. 
+ 3 min. 



The serum then is found to neutralize in such a way that 0-004 
gives complete protection against 3 m. 1. d., i. e. it contains approxim- 
ately 750 units. ^^ 



0-1 

0-01 

0-005 

0002 

0-001 

0-0008 

0-0006 

0-0004 

0-0002 

0-0001 

0-00008 

0-00006 

0-00004 

0-00002 

0-00001 



(iii) B. oedematiens antitoxic sera. 



25V 115B 124 

+ + — 

++ ++ ++ 

++ ++ ++ 

++ ++ ++ 

++ ++ ++ 



5 

+ =doath in 8 hours. 



410 



+ + 
+ + 

+ + 



411 



+ — 
+ + 
+ + 

+ + 
+ + 
+ + 



50 250 

— =reoovery. 



810 



+ ■+- 
-f + 
+ + 

S.OOO' 



117 

"Weinberg has defined as the unit of antitoxin the smallest amount 
of serum which will completely neutralize 100 minimal lethal doses 
of toxin in a mouse, the mixture being kept for one hour at room 
temperature before inoculation. The accompanying record illustrates 
an experiment in which various sera were titrated against a test 
dose consisting of 50 minimal lethal doses of toxin. 

5. Preparation and Standardization of Anti-gas gangrene 

Sera. 

(A Report by the Staff of the Wellcome Physiological Research Labora- 
tories). 

(i) Introduction. 

The investigation into the possibility of making anti-gas gangrene 
serum was first undertaken at the request of the Medical- Eesearch 
Committee, and the immunization of horses was commenced in 
October 1917. Dr. Mcintosh on behalf of the Committee, supplied 
suspensions of surface growths, and also broth cultures of the various 
anaerobes considered to be concerned in the production of gas 
gangrene. He added solutions of iodine to these cultures to produce 
attenuation. Two horses were immunized with these cultures over 
a period of some months. At one stage, when the horses had been 
under immunization for several months, and supplies of the bacterial 
suspension were not available, B. welchii toxin was injected. Prior 
to this time, the B. welchii antitoxic titre of sera from these two 
horses was low. What potency the sera of these horses possessed when 
issued for field trial in Prance in March 1918 was, therefore, the 
result of immunization with cultures and toxin in the case of B. welchii, 
and of cultures only in the case of vibrion septique, and other organisms. 
•In January 1918, Major Carrol Bull of the tfnited States Army had 
demonstrated in London the power of B. welchii antitoxin to neu- 
tralize the toxin of B. welchii. Bull presented a culture and a sample 
of antitoxin to the Laboratories. B. welchii toxin was rapidly 
prepared in large quantities for injection into horses. Antitoxin 
of sufficiently high titre to protect animals against large doses of 
living culture of B. welchii was available for field trial in March 1918. 

Meanwhile, evidence of the importance and frequency of the 
occurrence of vibrion septique was accumulating. At a War OfBce 
conference in March 1918 the pathologists present agreed that this 
organism was frequently present in wounds and was the cause of 
severe and fatal cases of gas gangrene occurring in soldiers. Work 
was immediately begun at the Laboratories with the aim of producmg 
an antiserum to vibrion septique. Cultures of vibrion septique were 
presented by Dr. Mcintosh, Miss Eobertson, and- Capt. Henry, 
and samples of toxin for testing purposes by Miss Eobertson. Large 
quantities of toxin were made at the Laboratories for injection into 
horses, and by May 1918 a serum had been produced which was of 
sufficiently high titre to protect an animal, which had received some 
hours previously 5 c.c. of serum, against many lethal doses of toxin 
or culture. Some of the horses yielding this serum had already been 
immunized with B. welchii. Thus, a double serum contammg 



118 

antitoxins to both B. welchii and vibrion septique became available, 
and was ready for trial in May 1918. , , ,, 

B. oedematiens toxin made by Captain Henry from cultures 
isolated by him, became available in quantity m June 1918. Serum 
of satisfactory protective value, as judged by results obtamed with 
laboratory animals, was produced in September 1918. 

The aim throughout the whole of the work was to produce as rapidly 
as possible a serum which, when injected into a soldier immediately 
after the infliction of a wound, might protect him against the onset 
of gas gangrene. 

It was decided at an early date to include the recognized pro- 
phylactic dose of tetanus antitoxin in the prophylactic gas gangrene 
serum. The laboratory staff commenced in February 1918 the 
preparation of T.W. serum, i.e. one containing sufficient antitoxins 
to both B. tetani and B. welcJiii to protect a laboratory animal against 
many lethal doses of cultures of these two organisms. 

T.W. serum was produced in quantity in March 1918. T.V.W. 
serum, containing antitoxin for B. tetani, B. welchii, and vibrion 
septique, was produced in August 1918. T.V.W.B. serum, containing 
antitoxin also to B. oedematiens became available only about the date 
of the armistice in November 1918. 

It may here be remarked that in 1918 almost the whole of the work 
was directed to the production of antitoxic, as opposed to anti- 
bacterial sera. It early became evident that the practical difficulties 
of preparing, in the very large quantities required, the ' whole 
cultures ' of bacilli to a constant standard of pathogenicity, so that 
the injections into horses would proceed smoothly and safely, were 
so great that the production of sera for field trial in large quantities 
would inevitably have been delayed for many months. It was found 
that antitoxins of high titre, when injected into animals, gave them 
complete protection against subsequent injection of many lethal. 
doses of living cultures. It was, therefore, considered advisable 
to concentrate the main effort on the production of purely antitoidc 
sera while investigating the methods of preparation >and evaluation 
of antibacterial sera. The termination of the war came before the 
experimental work with antibacterial sera was very far advanced. 

From the foregoing account it is clear that the sera actually 
tested m the British Army in France and used prophylactically and 
curatively were the first results of an attempt to produce three new 
immune sera. The nature of the toxins were only obscurely under>- 
stood and very serious difficulties both practical and theoretical 
had to be encountered. The sera were definitely not the final and 
most potent products possible, it is not even now known what titre 
should be considered to be a useful minimum. In judging the results 
therefore of the serumtherapy it must be borne in mind that the 
tests were made at the earhest possible moment and that rightly 
considered no really adequate therapeutic trial has been made of 
these sera. 

, ^ ,,.. (}^) Preparation of Antitoxins. 

A. B. welchii. 

ToKm.-Ordinary peptone meat broth or tryptic digestlbroth 
contaming 1 per cent, of glucose, was inoculated with a 9*-^onv 



119 

culture of B. welchii. The addition of sterile fresh muscle or of 
autoclaved meat seemed on the whole to be of advantage. The 
inoculated broth was kept at 37° for 24-30 hours, and then filtered 
through kieselguhr candles, carbohzed and kept in a cool place. 
It was not difficult to produce large quantities of toxin of which 
0-2 c.c. to 0-3 CO. would, when injected intramuscularly, kill a 250- 
grm. guinea-pig or a 20-grm. mouse. 

It was not found necessary to use any special methods for securing 
anaerobiosis : the broth was not covered with paraffin. The ino- 
culum was added to the broth a few hours after it had left the auto- 
clave and had cooled to about 40° C, 

Immunization of the Horse. 

The processes of immunization correspond almost exactly to those 
in common use for the preparation of diphtheria antitoxin. B. welchii 
toxin was injected intramuscularly at intervals of a few days, a repre- 
sentative series of doses being 1 c.c, 2 c.c, 4 c.c, 8 c.c. at intervals 
of 2 or 3 days. Ten weeks later the dose of toxin reached 400 c.c. 
and the antitoxic value of the horse's serum was 1,000 to 2,000 units. 
With the average horse a titre of 1,000 units was attained in about 
8 weeks ; at this point the animal was bled. 

Serum containing 6,000 units per c.c. was obtained from several 
horses, but'as the object of the work was to produce the maximum 
quantity of serum with a titre of 1,000 units in a given time, no 
opportunity was taken of continuing the immunization without 
bleeding to ascertain the highest titre attainable by this method. ' 

B. Vibrion septique. 

Toxin. — The method of making toxin was similar to that described 
for B. welchii. The inoculum was either a portion of a culture or 
the infected breast muscle of a pigeon injected the day previously 
with a culture of vibrion septique. 

The toxin produced in large quantities was sufficiently potent to 
kill a rabbit within ten minutes when a dose of 0-5 c.c. was injected 
intravenously. 

Immunization of the Horse. 

The methods closely resembled those used for the production of 
antitoxin to B. welchii, though the rate of increase of dosage was 
slower. An average horse would produce in 8 to 12 weeks serum 
containing 1,000 units of vibrion septique antitoxin per c.c 

0. B. oedematiens. 

Toxin. — ^By using the method above described in the B. welchii 
section, toxin was produced in large quantities of which 0-0001 c.c, 
injected subcutaneously, would kill a 20-grm. mouse within 
24 hours. The toxin is apparently rather unstable, the toxicity 
rapidly decreasing. Whether this old weak toxin is a good antigen 
is not at present known with accuracy. 



120 

Immunization of the Horse. 

The methods in use are exactly comparable with those in common 
use for the preparation of antitetanic serum. An average horse in 
3 months would yield a serum containing 5,000 units of B. oedematiens 
antitoxin per c.c. 

(iii) Methods of estimation of the value of anti-gas gangrene sera. 

A. Antitoxin of B. welchii. 

The unit in use was based upon a serum supplied by Bull which 
bore a label indicting the unitage: Bull used as his unit the quantity of 
serum which neutralized an amount of toxin equal to one lethal dose 
for the pigeon. We found that this quantity of serum was twice the 
amount necessary to neutralize twice the lethal dose for a mouse. The 
unit arrived at in this way was adopted for the titration of B. welchii. 

Toxins used for standardization would in a dose of 0-1 c.c. to 
0-25 c.c, when injected intramuscularly, kill a mouse within 24 hours. 

Serum, was produced containing 5,000 units per c.c. 

B. Antitoxin of Vibrion septique. 

The toxin produced by vibrion septiqiie differs from that produced 
either by B. oedematiens or B. welchii in that it does not cause death 
when inoculated intramuscularly or subcutaneously into mice. It 
can, however, be readily titrated by intravenous inoculation into 
rabbits, a high value toxin producing death in 3-15 minutes with 
a dose of 0-5 c.c. or less when tested by this method. 

A delayed reaction in which death occurs after 8 to 24 hours 
takes place with lesser doses. This delayed reaction is unsatisfactory 
from the point of view of a test as it is difficult to adjust the dose, and 
individual animals show a variable resistance. Since the intravenous 
test became impracticable when the work was carried out on a large- 
scale it was necessary to substitute some other method of titration. 

It was found that 0-05 c.c. of toxin injected intramuscularly into 
a mouse produces a recognizable oedema. This reaction is used to 
test the potency of the serum. The test dose of toxin used is ten 
times the amount, i. e. 0-5 to 1 c.c. according to the strength of 
the toxin. This test dose is equivalent to an intravenous lethal 
dose for the rabbit. A small number of antitoxins were titrated 
by : (1) the mouse-oedema method ; (2) intravenous injection into 
mice ; (3) intraperitoneal injection into mice ; (4) intravenous 
injection into rabbits. The results agreed sufficiently closely to 
justify the temporary use of the ' mouse-oedema method '. 

0. Antitoxin of B. oedematiens. 

The unit of antitoxin (which was already in use by Dr Weinberg 
at the Pasteur Institute) is that quantity of serum which when 
added to 100 minimal lethal doses of toxin injected intramuscularly 
mto a mouse will render the toxin inert and prevent the death of 
the animal. Sera were produced containing 20,000 units per c.d. 
We should hke to take this opportunity to express our thanks to 
Dr. Wemberg for his courtesy in sending us samples of his sera on 
several occasions. 



6. The Serum Therapy of Gas Gangrene^ 
(i) The Collection of data. 
The treatment of cases of gas gangrene by means of antisera was 
first carried out in British military hospitals in France during the 
spring months of 1918. Eeports on the results obtained by the use 
of these antisera were compiled by a number of E.A.M.C. officers 
and forwarded to the Medical Eesearch Committee. These reports 
form the basis of the following summary. 

The accompanying table (Table I) indicates the number of cases 
recorded by each observer, together with the number of deaths 
occurring in each series. 

Table I. 



Reported by 

Barling ....... 

Brenan ...... 

ElUs 

Hope ....... 

MoEwen 

McNee 31 

Stokes ....... 23 

Tytler 17 

Wyard ........ 2 

Totals 89 



No. of 
cases. 

1 
9 
1 
1 



No. of 



4 

1 

3 

1 

1 

12 

16 

11 

1 

50 



(ii) The Serum employed. 
The sera which were used in Prance were prepared by the staff 
of the Wellcome Physiological Eesearch Laboratories at Heme 
Hill. The titre of each serum, i. e., the number of units of B. welchii 
antitoxin and of vibrion septique antitoxin contained in each c.c. 
of serum is set out in Table II. 



Table II. 




B. welchii 


V. septique 


antitoxin 


antitoxin 


units. 


units. 


2,000 


— 






1,000 


300 






1,000 


300 






500 


300 






500 


300 






2,000 


— 






2,500 


1,600 




• 


1,000 


3,000 



G2 . 

G3 . 

G17 

G18 

G19 

G22 

G44 

G53 

It is to be noted that in the titration of the above sera the unit of 
B. welchii antitoxin is taken as being twice the amount of serum 
requisite to neutralize two minimal lethal doses of toxin in the mouse. 
This method of measurement has been found to yield a result which 
within reasonable limits is identical with the figure obtained by 
Bull's method of titration. 

The unit of vibrion septique antitoxin is estimated as being the 
amount of serum which will completely neutralize 10 minimal-oedema- 
causing doses of toxin in the mouse. This procedure yields a much 
Ijigher figure than that obtained by the French method of standardiza- 
tion, in which the antitoxin unit is taken as the amount of serum 
which will neutralize one intravenous minimal lethal dose of toxin 
in a full-grown rabbit. 



122 

The sera designated as G3, G17, G18, and G19 were prepared by 
the inoculation into horses of mixed cultures of anaerobes which had 
been attenuated with iodine. These attenuated mixed cultures 
were furnished by Mcintosh. The resulting sera proved to have 
a low antitoxic value. The amount of vibrion septique antitoxin did 
not pass beyond the 300 unit level, while the titre of B. welchii 
antitoxin remained in the neighbourhood of 100 units per c.c. It 
was only after treating these, same horses with B. welchii toxin that 
the content in antitoxin as set out in Table II was reached. The sera 
refeiTed to are claimed by Mcintosh to have distinct advantages 
over purely antitoxic sera in that they contain antibacterial bodies. 

The production of sera by the inoculation of attenuated whole 
cultures of anaerobes was found to present considerable diBficulties 
when carried out on a large scale, and the method was abandoned 
in favour of inoculation with sterile toxins. The sera designated 
G2, G22, G44, and G53, thus prepared, were therefore purely anti- 
toxic sera with no antibacterial content. 

None of the sera issued for use in the field contained antitoxin 
to B. oedematiens. 

The sera designated B. in section (vi) were B. welchii antitoxic 
sera prepared by Bull in the Eockefeller Institute and had a titre 
of B. welchii antitoxin varying from 750 to 1,800 units per c.c. 

(iii) Analysis of Becords. 

The serum therapy of cases of gas gangrene became available at 
a period in the history of the war when it had already been clearly 
demonstrated that the disease was amenable to rigorous surgical 
procedures. It so happened then that antisera came to occupy 
two separate roles in the treatment of gas gangrene. 

1. Where the focus of the disease could be extirpated or otherwise 
effectively dealt with by the surgeon the use of serum became 
an adjuvant to surgical measures. If, for example, symptoms of 
toxaemia persisted after complete removal of all anaerobe infected 
tissue, recovery could be hastened by the giving of serum. Or 
agaia, the giving of serum to a man suffering from serious toxaemia 
might improve his physical condition sufficiently to permit of opera- 
tion. In either instance, the beneficial effect of the serum was due 
to the neutralization of toxin which had already reached the circula- 
tion. The serum was not called upon to combat the infection. 

2. On the other hand, the serum came to be used in desperate 
cases, m which, for various reasons, surgical intervention was 
inadequate to cope with the disease. The task assigned to the serum 
in this particular class of case was a very severe one, for it had to 
deal not only with toxin in the circulation and in the infected tissues, 
but it had also to arrest the actual spread of infection and finaUy 
to overcome it. • ■' 

..f^^/ !f ''■^^ '^''!' ""^ ?^'^'' *^^ '^*^^^* f^at^J^es of which have been 
firllvIL''' section (yi), can be best analysed by considering, 
wbS'.i. A ^^7^ich death occurred ; and secondly, the casts 
beneficSr ""' ^^"'^ *^' ''™^ ^^' ^""^^'^ *« ^^^^ been 



A. Deaths. 

These may be classified as follows : 

1. Oases m which the wounded man recovered from gas gangrene 
and in which death may be reasonably attributed to some other 
conditions, such as a streptococcal septicaemia or a pneumonia. 

The following cases in section (vi) come under this category : 

Case number. Day of death. Cause of death. 

38 10 Lobar pneumonia. 

45 10 Lobar pneumonia. 

73 8 Broncho-pneumonia. 

63 9 Streptococcal septicaemia. 

74 6 Streptococcal septicaemia. 
88 21 Streptococcal septicaemia. 
56 — Streptococcal meningitis. 

All these cases had undoubted gas gangrene at the outset," and in 
each the disease was arrested by a combination of surgery and serum. 
Although it is impossible to ignore the influence of gas gangrene in 
establishing such secondary infections in a wounded man, yet it 
is obvious that these cases should not be looked on as failures in serum 
therapy. 

2. Cases in which a fatal issue is determined by the presence of 
an anaerobe other than those anaerobes against which the serum has 
been prepared. 

The serum therapy of gas gangrene was started at a time when 
there was still lacking much information as to the nature of the 
anaerobes that caused the disease in wounded men. It was known 
that B. welchii could be isolated from the majority of human cases. 
It was known too that vibrion septique could be found in a certain 
number of cases, but no figures from British sources were available 
as to the percentage incidence of this anaerobe in wounds. Further, 

B. oedematiens, which Weinberg claimed to have demonstrated in about 
one-third of his cases, had not been recorded by British workers, with 
the solitary exception of the case reported by Miss Dalyell. 

It became therefore a matter of prime importance, not only to 
determine in what percentage of cases of gas gangrene a welchii- 
septique antitoxic serum could be expected to give beneficial results, 
but also to investigate the cause of death in all instances in which 
liberal and fair trial of the serum proved unsuccessful. The first 
part of the investigation was completed, and provided a census of 
the anaerobes causing gas gangrene in man. It showed that in 
about 25 per cent, of all cases of gas gangrene the disease was due 
to anaerobes other than B. welchii and vibrion septique, i. e. in one- 
quarter of the cases a welchii-septique serum could not be expected 
to combat the infection. The second part of the investigation, 
which presumably would have yielded evidence complementary to 
the first part, broke down owing to the difficulty of procuring from 
Prance pieces of infected muscle from unsuccessfully treated cases. 

Of the 87 serum treated cases only 5 were proved to contain patho- 
genic anaerobes, infection by which in guinea-pigs could not be 
arrested by a welchii-septique serum ; and there can be no doubt, 
on the evidence available, that this number would have been con- 
siderably augmented if a thorough investigation had been possible. 



124 

Oases 48, 54, and 89 contained pathogenic strains of B. Jallax. 
Case 84 contained B. oedematiens, and an unidentified pathogenic 
anaerobe was isolated from Case 12. 

The details of Case 54 as reported by Stokes are as follows : 

Sa. Multiple wounds of thigh. Leg discoloured, crepitations round 
ankle. Oedema up to groin. Pulse very bad. 

28. 4. 18. Operation. Multiple incisions. Calf muscle good. Thigh 
muscles heavily infected. Given serum 30 c.c. Gr 17 and 30 c.c. G 22. 

29. 4. 18. Vomiting stopped. Feels better. Pulse better. Infection 
has not spread but crepitations obvious round knee and thigh. Later 
the foot became black and gangrenous. Given 60 c.c. of serum. He 
again felt better after the serum. In the evening again given 60 c.c. of 
serum. Amputation through upper third of thigh. Stood the operation 
well. He could not have done so 24 hours previously. 

30. 4j 18. In the morning fairly well but pulse very weak. Given 
40 c.c. G 17. In the evening became delirious and died about midnight. 
There was no recurrence of gangrene in the stump. 

Blood cultures. 

28. 5. 18. B. Jallax. 

29.5.18. B.fallax. 

29.5.18. {evemng) B.fallax. 

30. 5. 18. (after amputation) negative. 

The notes of Case No. 12 reported by Ellis are as follows : 

18. 10. 18. Wounded in right leg and ri^ht arm. 

19. 10. 18. Admitted to hospital. Through-and-through wound of 
anterior and external part of upper thigh. Also through-and-through 
wound of right knee-joint. Whole thigh tense, oedematous and tympanitic 
anteriorly up to Poupart's Hgament. No crepitation. OfEensive, efEerve- 
scent fluid bubbles from wound in knee-joint. 

Operation. Very incomplete excision of wound. Muscles discoloured 
but contractile, bleeding muscle reached in some places. Knee-joint 
flushed with saline and ether and closed. General condition fairly good, 
colour good, no vomiting, pulse small volume but not exceedingly rapid. 
Under anaesthetic pulse rapidly ran up to 100. 

Serum G 53, 100 c.c. in 1,000 c.c. salt solution intravenously. 

20. 10. 18, 3.30 p.m. Has been perfectly clear mentally until the last 
few hours. He is now slightly delirious. No vomiting, colour bad, sweating, 
intense thirst, pulse bad. No spread of gas to abdominal wall, but the thigh 
is more extensively involved and malodorous. 

Serum G 53, 100 c.c. in 1,000 c.c. saline intravenously. 7.0 p.m. death. 

Autopsy. 3 hours after death. 

Anterior muscles of thigh all putrid, posterior muscles full of gas. Gluteal 
and abdominal muscles not involved. 

Bacteriological examination. 

Muscle-direct smear, very numerous Gram-positive thick bacilh many 
showing spores. Spores oval, mostly subterminal but many central 

Meat culture— orgamsm described and B. sporogenes 

Blood culture, 9 hours before death, gave the organism described as 
occurring m direct smear from muscle. 

Bemarks. 

Very severe case of gas gangrene, not benefited by serum. Sporing 
anaerobe m blood culture. i" "S 



125 

In a letter accompanying cultures whicli were sent to the Medical 
Kesearcli Committee, Ellis says : 

' TMs is my first real failure with the serum. I should have laid down 
a serum barrage in the thigh and should have repeated serum at 9.30 
instead of 3.30. However, personally I do not think that in this case the 
serum was active.' 

3. Cases in which the serum was given either in too small amounts 
or at too late a period in the course of the disease. 

Experience with the antisera of which we have most reliable 
information, viz., those against tetanus and diphtheria, has shown 
conclusively that they must be given in large doses and at the earliest 
possible stage in the development of the illness. In the case of 
diphtheria, Behring and many other workers have demonstrated 
that with efficient serum treatment in the first two days of the disease 
the death-rate is reduced to about 7 per cent. The use of serum 
after two days gives less favourable results, and after the fifth day 
it is found to exercise little or no beneficial effect. The good fortune 
which attends the early serum therapy of diphtheria depends to a very 
large extent on the ease with which the condition can be diagnosed 
clinically, for the sore throat, which is such a well-marked clinical 
feature, attracts attention to the probable existence of the disease. 
Less happy results have been obtained in the treatment of tetanus, 
because here the disease may develop when it is least suspected, 
and a clinical diagnosis becomes possible only when grave symptoms 
are already manifest. It is for this reason that antitetanic serum 
best demonstrates its efficacy when it is given prophylactically. 
Now, both in diphtheria and in tetanus the march of events is 
measured in periods of days. In gas gangrene, on the other hand, 
the disease may rush relentlessly to a fatal issue in the course of 
a few hours, so that the wounded man may be practically moribund 
by the time that serum treatment becomes available. 

In searching through the records collected in France one finds 
a. large number of instances in which serum was first administered 
when the wounded man was really dying. It can be legitimately 
claimed that under these conditions the serum had no chance of 
effectively combating the disease. 

These moribund cases may be divided into two groups. 

Group A. 

Acutely fulminating types of the disease where death occurred 
within three days. The following may be taken as typical samples : 

Duration of life 







in hours after 


No. of case. 


Dai/ of death. 


giving of serum. 


17 


1 


G 


31 


1 


9 


?39 


1 


10 


11 


2 


4 


43 


2 


11 


55 


2 


7 


57 


2 


5 


58 


2 


2 


64 


2 


6 


65 


2 


8 


10 


3 


6 


49 


3 


6 



126 

Group B. 

Late cases of gas gangrene in which death occurred after the third 
day. 







Duration of life 
in hours after 


of case. 
68 
70 
69 
4 


Day of death. 

5 

6 

8 

11 


giving of serum. 
4 
6 

4 
6 



Group A were treated in forward medical units and represent 
a type of case which is unavoidable under modern conditions of 
warfare. Group B were base hospital cases and might conceivably 
have benefited if serum had been given earher. 

B. Recoveries. 

Of these there are 38 cases out of a total of 89 serum treated 
patients. They may be divided into three groups : 

1. Cases in which surgical treatment consisted in thorough 

cleansing of the wound, together with removal of foreign 
bodies, excision of infected muscle, &c. 

16 cases. 

2. Cases in which it was necessary to amputate a limb. 

17 cases. 

In groups 1 and 2 there are many instances in which beneficial 
effects are ascribed to antisera, J)ut in the absence of large statistics 
it is impossible to decide whether surgery plus serum accomplished 
more than surgery along could have done. 

3. Cases in which operative procedures were insufficient to arrest 

the disease, or in which the focus of infection was inaccessible. 
5 cases. 

Of these 4 were reported by Ellis and 1 by Tytler. It is to be noted 
that both workers gave large and repeated doses of serum, that the 
serum BlUs used for his series -was the best of the sera sent out to 
France, and that he used it locally in and around the infected tissues. 

One of the cases treated by Ellis and Tytler 's case are reported 
in full. 

Case No. 13. St. 

Multiple shell wounds, both legs and thighs. Double fractured femui. 
G.S.W. scalp. Wounded morning 17. 10. 18. 

19. 10. 18, 3.30. Patient in very bad condition, delirious and wildly 
excited ; colour definitely yellow. Pulse, bad quaUty, 140. Left thigh 
gangrenous with green discolouration below upper third. Left leg greenish 
black and absolutely cold to the knee. Very marked oedema and crepita- 
tion over whole thigh anteriorly extending up to groin with marked 
tympany on percussion, and marked reddening of the skin and bronzing. 
Posterior thigh muscles in upper third soft, glutei soft. Patient inoperabk. 
100 c.c. serum in 1,000 c.c. salt solution intravenously and 30 c.c. intra- 
muscularly in gluteal adductor muscles and in anterior abdominal wall. 

10.00. Remarkably improved, now perfectly clear mentally, pulse - 
112. Gas not spreading or very shghtly, some increase in skin redness 
especially over glutei. Oedema and crepitation if anything diminished 



127 

Spinal novocaine anaesthesia amputation in middle third of thigh immedi- 
ately above line of green discolouration. Adductors and vasti discoloured, 
contraction sluggish, other muscles appear fairly healthy. Collapse 
following amputation, pulse and respiration ceased, patient practically 
dead. Blood transfusion 700 c.c. with dramatic improvement. 

15.30. Sleeping quietly, pulse fair volume, 116. 60 c.c. serum intra- 
muscularly (40 c.c. in pectorals, 20 c.c. in thigh and buttocks). 

19.30. Blood transfusion 800 c.c. 

20. 10. 18. General condition good, full strong pulse, rate 112. tempera- 
ture rising, sweating. Mental condition absolutely clear. The area of 
skin redness is spreading and now extends some distance up left abdominal 
wall and up back over lumbar muscles. There is, however, no oedema or 
tenderness and the thigh feels quite soft though tympanitic. There is 
a bleb on anterior abdominal wall just above the area of redness. 15.50. 
80 c.c. serum intramuscularly (60 c.c. pectorals and 20 c.c. in abdominal 
wall and back). 

23.00. Vomiting since about 15.00, now vomiting almost continuously. 
Wound dressed, whole stump swollen with superficial crepitation, muscles 
all red, healthy and contractile except adductors, ends of which show 
superficial gangrene, beneath this muscle discoloured, soft, oedematous 
and non-contractile. 50 c.c. serum in 500 c.c. salt solution intravenously. 

21. 10. 18, 22.00. Much improved, vomiting eased ofi through the night 
and to-day he has not vomited at all. Colour good. Evening temperature 
99-6, pulse 90. General condition excellent. The redness over anterior 
abdominal wall and lumbar muscles has now entirely faded. Wound 
dressed, healthy except for adductors which are absolutely pulped, spongy, 
and full of gas. Dead muscle dissected away without anaesthetic. Other 
leg multiple wounds dressed, fractured femur put up in Thomas. 60 c.c. 
serum intramuscularly. 

Bacteriological Examination. 

Muscle 19. 10. 18, 8.30. Direct smear, non-sporing anaerobe, morpho- 
logically resembUng B. welchii. Fairly numerous diplococci culture, 
anaerobes probably B. welcMi and B. sporogenes. 

Muscle, 20. 10. 18, 23.00. Direct smear. Predominant organism a very 
large, very broad strongly Gram-positive bacillus with rounded ends, 
some show tendency to oat shape. Spore forms are fairly numerous and 
are chiefly subterminal but some are central. There is also a long 
slender bacillus with round terminal spore. Fairly numerous cocci in 
pairs and short chains. 

Bemarhs. 

Desperate case, successfully treated. Large doses required and recur- 
rence of signs of intoxication and local spread of gas when serum was not 
pushed, immediate improvement with increased dosage. All officers who 
have seen this case consider the recovery extraordinary and agree that 
they have never seen anything like it before. 

Case No. 76, Ke. 
24. 4. 18. G.S.W. Buttock. 

27. 4. 18. Admitted. P. 104, T. 103. 

28. 4. 18. X-ray shows large F.B. in left back. Vomiting. T. 104, 
P. 100. 

29. 4. 18. Operation. Wound excised and excision entended up over 
crest of ileum, which was sUghtly shattered. Large F.B. removed from 
body of psoas. Flavine gauze drainage. Probable gas infection. 



128 



Exploratory incision througli left rectus abdominis shows no lesion 
of bowel or peritoneum but some serous fluid. 

Vomited once, 3 p.m. 8.0 p.m. T. 99, P. 120. 

Urine shows considerable albumin and a good number of granular casts. 

30. i. 18. Condition fair. Not much sleep, a.m. T. 102, P. 120. 

No vomiting, p.m. T. 101, P. 114. 

1. 5. 18. Better night, but complains of pain in side, a.m. T. 97, P. 104. 

No vomiting, p.m. T. 103, P. 128, K. 32. 

Urine shows trace of albumin and occasional granular casts. 
. 2. 5. 18. Condition not so good. Very restless. No vomiting, a.m. 
T. 99-4, P. 78, E. 30. 

Left scrotum and inguinal canal show marked well demarcated swelling 
and tenderness. 

Operation. Inguinal canal opened. Coverings of cord contain pus and 
gas. Vessels of cord thrombosed throughout. Testicle and cord stripped 
out of scrotum. Incision extended upwards to abdominal wall and cord 
followed up retroperitoneaUy. Large abscess between ileo-psoas fascia 
and peritoneum extending into pelvis to base of bladder and upward in 
loin to former site of F.B. in psoas muscle. Deferential vessels are natural 
from internal inguinal ring to base of bladder. Spermatic vessels throm- 
bosed as far as renal vessels and perivesicular tissue is oedematous. Defer- 
ential vessels tied ofi at brim of pelvis and spermatic vessels at lower pole 
of kidney. The testicle and cord as removed are completely gangrenoiu? 
and show definite gas. Cultirres from this tissue yield streptococcus and 
B. welchii. 

3. 5. 18. Slept fairly well. Vomited in morning, a.m. T. 98, P. 120, 
E. 30. p.m. T ?102-6, P. 112, E. 28. 

2.30 p.m. 20 c.c. serum (W.G. 22) intravenously. Dressed. Showed 
some reaction with chill 30 minutes later. 



20 c.c. serum (W. 22) intramuscularly. 
20 „ „ 



4.30 p.m. 

6.30 „ 

7.30 „ . 20 

9.0 „ 20 

10.0 „ 20 

11.0 „ 30 

2.0 p.m. T. 103-4 P. 126 E. 26. 

6.0 p.m. T. 101-2 P. 90 E. 30. 

10 p.m. T. 102 P. 128 E. 30. 

4. 5. 18. Poor night. Very restless, irrational. Bowels moving con- 
stantly. Vomited after liquid food in morning. Seems distinctly improved 
generally, however. 

12 noon 10 c.c. serum (W.G. 22) intramuscularly 

2.0 p.m. 10 „ „ 

6.0 p.m. 10 „ 

9.0 p.m. 10 „ „ 



Slept at intervals during the afternoon. 
No vomiting. 



Takes noiirishment better. 



2.0 a.m. 


T. 


101 


P. 


120 


E. 30. 


6.0 a.m. 


T. 


102-4 


P. 


118 


E. 28. 


10.0 a.m. 


T. 


102 


P. 


120 


E. 28. 


2.0 p.m. 


T. 


102 


P. 


120 


E. 32. 


6;0 p.m. 


T. 


102-4 


P. 


120 


E. 36. 


10.0 p.m. 


T. 


101-8 


P. 


118 


E. 30. 



5.5.18 Eested more quietly. Slept at intervals 
of nourishment and has not vomited. 
1.0 a.m. 
6.0 a.m. 
8.0 a.m. 



Took a fair amount 



10 c.c. serum (W.G. 22) intramuscularly. 



12 noon. 

2.0 p.m. 

4.0 p.m. 

7.0 p.m. 
10.30 p.m. 

2.0 a.m. 

6.0 a.m. 
12 noon. 

6.0 p.m. 
lOp.m 



10 

10 

10 

10 

10 

10 

T. 

T. 

T. 

T 

T 



1044 

100 

100-8 

104 

103-7 



P. 
P. 
P. 
P. 
P. 



112 
114 
106 
130 



E. 
E. 
E. 
E. 
E. 



32. 
30. 
26. 
26. 



Vomited after food— 4.30 p.m. and 6.30 p.m. 

5. 18. Good night. Slept fairly well. Condition about the same. 



Takes nourishment well, 
infection 



Wound rather dirty externally. No evident gas 



10 c.c serum (W.G. 22) intramuscularly. 
T. 102-4 P. 120 E. 28. 
T. 97-8 P. 104 E. 26. 
T. 100 P. 106 E. 30. 



9.0 a.m. 
2.0 p.m. 
6.0 p.m. 
10 p.m. 

For two days following this the patient was in about the same condition, 
the morning temperature each day being down to normal and the evening 
temperature up to 104, with P. 110 to 140. No evidence of local gas 
infection spreading or of gas intoxication. Looks Hke a streptococcal 
infection. 

9. 5. 18. General condition the same or better, 
a.m. T. 98, P. 108, E. 36. p.m. T. 101, P. 120, E. 32. 

10. 5. 18. 6 a.m. T. 101-2, P. 116, E. 30. 
12 noon T. 97-8, P. 112, E. 32. 

Patient had 150 c.c. serum in first 10 hours from commencement of 
treatment. In the following 24 hours he had 40 c.c, and in the next 
24 hours, 90 c.c, i. e. 280 c.c. in all. Considering the extensive gas gangrene 
of the testicle and cord, its extension as far as the renal vessels, the size 
of the wound made in removal and the patient's general poor condition, it 
seems reasonable to ascribe his improvement in large part to the serum. 
At the time of the second operation the gas infection was well established 
and there was considerable infection in the retroperitoneal tissues of the 
iliac fossa. He was considered clinically an almost hopeless case. He now 
has a tremendous open wound, which when he lies on his opposite side 
exposes the muscular wall of the abdominal cavity from the bottom of 
the pelvis up to the lower pole of the kidney. The surfaces of this are 
covered with necrotic exudate but there is no evidence of deep infection. 
For the past week he has had the appearance of a streptococcal infection, 
but the last two days has definitely improved and now appears to have 
a fair chance of recovery. Since the first day's serum treatment he has had 
nothing to suggest spread of gas infection or intoxication. The morning 
after the second operation the patient had the appearance of gas intoxica- 
tion and in the opinion of his surgeon was a true gas intoxication. 



130 

(iv) Discussion. 
The serum therapy of gas gangrene has presented a problem of very 
great complexity. The reasons for this may be thus briefly outlined. ^ 

1. Gas gangrene is a polymicrobic disease. The chief orgamsmf| 
responsible for the condition are, in the order of their frequence and 
of their relative importance, B. welchii, vibrion septique, B. oedema- 
tiens, and JB. fallax. Each of these organisms, either by itself or in 
combination with the others, has been found to be the principal- 
pathogenic agent in fatal human cases. In addition, however, to 
these arch-criminals in the bacteriological syndrome, there are a 
certain number of anaerobes and of aerobes, non-pathogenic in 
themselves, which act as aiders and abettors in the production of 
infection. Not only is the number of possible bacterial combinations 
enormous, but it is also impossible for the bacteriologist, except 
after prolonged investigation, to determine the special combination 
of organisms which is responsible for the disease in any individual 
case. It follows therefore that, whereas in the case of diphtheria 
and tetanus we have to deal with a specific and well-defined entity, 
in the case of gas gangrene we are faced with a disease which is much 
more complicated, so that a successful therapy can be undertaken only 
with a serum that is polyvalent. The requisite constituents of such 
a serum are dealt with in the summary which concludes this report. 

2. The nature of the disease is such that it may be difficult to 
introduce the serum into the general circulation or to bring it into 
contact with the infected tissues. 

(a) The constriction of the peripheral veins, which is a marked 
feature in certain cases of gas gangrene toxaemia may render it 
difficult to carry out intravenous inoculation. Also, the depressed 
state of the circulation reduces to a minimum the chance of absorption 
when the serum is given subcutaneously or intramuscularly. 

(6) The infected tissues may be cut off from the general circulation 
owing to mechanical damage or occlusion of the vessels resulting 
from the wound itself. Or again, the vascular supply may be seri- 
ously reduced because of the oedema which is such a characteristic 
feature of the pathological process. 

The fortunate results reported by EUis,were, we beheve, due in large 
measure to the inoculation of serum into and around infected tissues. 

The short incubation period in man, and the ruthless rapidity with 
which the disease progresses when it is once estabhshed, are further 
factors which mihtate against successful treatment with serum. 

3. Of the anaerobes which produce gas gangrene in man, B. 
oedematuns is the only organism which yields a toxin comparable 
m valency to that given by B. diphtheriae or B. tetani. The toxins 
of B. welcUi and of vii)ri<>n septique, when tested on laboratory 
meSioned'^^ extremely low value in comparison with those just 

l,nLf ^^ ^^ accepted as a general principle in the immunization of 
Sirectlv S f^^^f '^^\'^' ^^'' "* the resulting antiserum varies 
ThefroblrJf? 7 ''' ? ^""T ?' *°^°^^ °* *he material inoculated. 
sepUque sera is identical with the problem of producing better toxin. 



131 

(v) Summary. 

1. Our information in regard to diphtheria and tetanus is based 
on countless laboratory experiments and a vast clinical experience 
extending over three decades. In contrast to this, our knowledge 
of the anaerobes is limited to investigations which have extended 
over a few months and which were frequently undertaken in un- 
favourable conditions. In this respect the work on anaerobes may 
be said to be still in its infancy, so that the serum therapy of the 
disease, as practised in the last stages of the war, was no more than 
an experiment. 

In view of the numerous factors which mihtate against success and 
which have been dealt within various sections of this report we consider 
that the results obtained in France have exceeded expectations. 

2. The sera which were issued for use in British hospitals in France 
contained antitoxin to B. welchii and vibrion septique only. Labora- 
tory experiments have shown that it is necessary to include antitoxin 
to B. oedematiens and B. fallax in addition, and the clinical reports 
from France add corroborative evidence to this finding. 

The value of antibacterial substances in a gas gangrene serum is 
as yet not clearly established. 

3. A gas gangrene serum should be given at the earliest possible 
stage in the illness. It is probable that prophylactic inoculations, 
if practised with suflBciently large doses, would give results which 
cannot be obtained when the serum is used therapeutically to check 
the already established disease. 

The production of an active immunity in man by the inoculation 
of carefully adjusted toxin-antitoxin mixtures is suggested as 
a procedure which might considerably curtail the incidence of the 
disease and diminish its morbidity rate. 

(vi) Table of Cases. 

Abbreviations, etc., used in the table giving details of cases treated 
with serum. 

Figures inserted in brackets in front of letterpress in the columns 
headed ' Surgery ' ' Serum ', ' Eesult ', and ' Bacteriology ' indicate 
days after admission. 

In the columns headed ' Serum ' the figures following on those 
in brackets indicate the number of cubic centimetres of serum 
given. The letters I.V., I.M., S.C. indicate intravenous, intra- 
muscular, or subcutaneous. A letter followed by a number indicates 
the batch of serum employed. 

For instance (6) 50 I. V. G17 iadicates that on the sixth day after 
admission 50 c.c. of serum were given intravenously, the batch of 
serum being G17. 

Other abbreviations occurring in this column are A.T.S. which 
indicates antitetanic serum, and when this is followed by a number, 
e. g. 750, it gives the number of units in the dose administered A.T.S. 
+W indicates antitetanic? serum with the addition of antiwelchii 
serum. 

In the column headed ' Eesult ' the sign + indicates the death of 
the patient, the figures in brackets, as mentioned above, giving the 
number of days after his admission to hospital ; the sign — indicates 
that the patient recovered. 



132 



Case. 
■1. Fe. 

2. Pi. 

3. Jo. 

4. Ey. 

5. Co. 

6. Ch. 

7. Sm. 

8. McI. 



9. Uh. 

10. Bo. 

11. Co. 

12. Le. 



13. St. 



U. Sa. 



Reported hy. 



Barling 
Barling 
Barling 

Barling 

Ellis 
Ellis 
EUis 
EUis 



Ellis 

Ellis 

Ellis 
EUis 



Ellis 



Hope 



Lesion. 



Bight leg and left thigh. 



Left shoulder and left 
foot. 



Bay of 
onset of 

gas 
gangrene. 




Left thigh. 



Left thigh, left shoulder, [ 1 
right arm. ! 



Right thigh. 1 

Buttock. I 1 

i 

Thigh and hand. i 1 



Thigh and thorax. 



Right knee. 

Right leg and right 
arm. 



Both legs and thighs. 



Neuk and shoulder 



1 



Multiple wounds thigh [ 1 
and leg. I 



Left thigh and right leg. 



13 



(1) Right leg cleaned up. 

(2) Left thigh cleaned up, removl 

of foreign body and bitB ^ 
clothing. 

(2) Foot amputated, foreign boi 
removed from shoulder. 

(6) Multiple incisions left arm and 

shoulder. ■ 

(1) Femoral vessels tied. 

(2) Amputation and excision of 

muscle. 

(7) Excision of part of femoral veil 

necrosed muscle removed. 

(1) Wound cleaned, foreign bodyl 

removed. 
(4) Left hip explored. 

(8) Left thigh opened up. 
(11) Amputation left arm. 

(1) Excision of wound and infected 
muscle. 

(1) Foreign body removed and in- 
fected muscle removed. 

(1) Amputation. 



(1) Amputation. 



(1) Amputation, reourrencein stump. 

' (1) Amputation. 

I (1) Amputation. 
(1) Excision of wound and muscle. 



(2) Amputation. 



(1) Excision of wounds, thyro-hyoid 
membrane sutured. 
(13) Ribs excised, stinking haemo- 
thorax drained. 




20 I.V. G 3. 



50 I.V. G 17. 



40 I.M. G 3. 



60 1.V. G 17 
10 I.V. G 22. 



60 I.M. G 53. 



60 I.M. G 53. 
60 I.M. 6 53. 

60 i.m; G 53. 
10 I.V. G 53. 

80 I.M. G 53. 
40 I.M. G 53. 
40 I.M. G 53. 
40 I.M. G 53. 



80 I.M. G 53. 
20 I.M. G 63. 

100 I.V. G 53. 



60 I.M. G 53. 

100 I.V. G 53. 
100 I.V. G 53. 



100 I.V 
30 I.M. 
60 1.M. 
80 I.M. 
60 I.M. 
60 1.M. 


. G53 
G53. 


A.T.S. 
A.T.S. 
50 I.V. 


500. 
750. 
G17. 



(6) + 



(7) + 



(7) + 



(11) + 



(3) + 

(2) + 
(2)_+ 



B. welchii, B. aporogenes, and strep- 
tococci. 



(16) + 



Wound gave B. welchii and 
tococoi. 



(2) Thigh gave B. welchii, B. sporo- 
genee, and streptococci. 

(2) Haemothorax gave B. welchii, 
B. sporogenes and strepto- 
cocci. 

(7) Haemothorax gave B. welchii 
and streptococci. 

Muscle gave B, welchii. 



(3) Blood gave B. welchii, B. sporo- 
genes and streptococci. 



Blood culture gave a sporing anae- 
robe. Muscle gave the same 
anaerobe together with B. sporo- 



Musole gave B. welchii, B. sporo- 
genes, and streptococci. 



Bemarl-s. 



Much relieved by serum ; it 
ought to have been repeated. 



Death 7 hours after serum. 



Serum had no effect on 
or local condition. 



Death a few hours after opera- 
tion and the giving of serum. 



Cured by serum. Fatal prog- 
nosis given by surgeon. 



Cure following serum therapy in 
spite of fatal prognosis. 

Death 6 hours after serum. 



Death 4 hours after serum. 

Antitoxic sera prepared against 
B. welchii, vibrion septique, and 
B. oedematiens did not protect 
guinea-pigs against this anae- 
robe. (Henry). Quoted in full 
in Section VI of report. 

A welohk-septique antitoxic se- 
rum protected a guinea-pig 
against the muscle culture 
(Henry). Quoted in full in 
Section VI of Report. 



Anaphylactic death. 



134 



Case. 
15. Sto. 



Beported by. 



McEwen 



16. Ta. I McNee 



17. Li 

18. McL. 

19. Co. 



McNee 

* 

McNee 
McNee 



20. Ho. McNee 

21. MoD. i McNee 

22. Co. I McNee 

I 

23. Ev. ; McNee 

24. Eok. : McNee 

25. Do. i MoNee 

26. Ea. j McNee 

27. Pr. i McNee 

28. Da. i McNee 

i 

29. Ki. i McNee 

30. Pe. McNee 

31. Cu. ! McNee 

32. Ro. ; McNee 

33. Ha. McNee 

34. An. j McNee 



Lesion. 



Day of 
onset of 

gas 
gangrene. 



Thigh. 



Right shoulder and arm. 



Both thighs, feet, 

wrist, &c. 
Right forearm. 



Thorax. 



Abdomen and both 
thighs. I 



Shoulder, back and 
both buttocks. 



Arm and thigh. 

Left arm. 

Left knee and ankle. 

Knee. 



Leg, posterior tibial' 
vessels torn. [ 

Left arm. 



Right thigh, left hand [ 

and back, left shbul- ! 

der. t 

i 

Right leg and thigh. 

Right buttock. 



Thigh and leg. 

Arm, brachial artery- 
divided. 1 

Thigh and leg. 



Right leg. 



(2) Ligature of femoral ve 

excision of muscle. 

(3) Amputation. 

(1) Wound cleaned and mm 
cised. 

(1) Wounds cleaned. 

(1) Wound cleaned. 

(2) Amputation. . 

(1) Aspiration. 

(3) Eib resection. 

(1) Wounds cleaned. 
(3) Further operation. 



(1) Shoulder, disarticulation. 



2 I (1) Wounds cleaned. 
j (2) Amputation. 

1 I (1) Infected muscle removed. 



1 I (1) Wound cleaned and in 
I muscle removed. 

2 (^) Amputation, 
(stump) I 

' Wounds cleaned up, ampt 

I later. 



1 
1 

1 
1 

1 
3 
1 



(1) Excision of muscle. 



(1) Excision of muscle. 



(1) Excision of wound. 



(1) Cleaning of wound and ex( 

of muscle. 

(2) Further excision. 

(1) Excision of muscle. 



(I) Wound cleaned up. 
(3) Amputation. 

(1) Excision of muscle. 
(3) Amputation. 

(1) Amputation through knee. 



Serum. 



12) 10 I.M. G 19. 


+ 


(1) 25 B. 1400. 


(2) + 


(1) 30 G 3. 


(1) + 


(1) 20 B. 725. 


- 


(3) 40 B. 1100. 
40 B. 1100. 


(5) + 


(1) 20 B. 1175. 
(3) 20 6 2. 


(4) + 


40 G 2. 




(1) 50 B. 1175. 
(2)20B. 1175. 
(3) 20 B. 1175. 


(3) + 


(2) 30 B. 1175. 

(3) 40 B. 1175. 


- 


(1) 40 B. 1400. 




(3) 40 B. 1400. 




(1) 40 G 3. 
(1) 40 G 3. 


- 



(2) 40 G 2. 



20 B. 



;i) 50 G 2. 
[2) 30 G 2. 

(1) 40 B. 1175. 

(2) 100 G 3. 



[1)A.T.S.+ W. 
[2)20B., 

[1) 50 G 2. 

[2) 50 G 3 . 



;i) 50 B. 1175. 

:i)A.T.S.+ W. 



;3) 80 B. 725. 
[i) 80 B. 1400. 



;i) A.T.S.+ W. 
;i)40G2. 



Sesult. 



(3) + 



Bacteriology. 



Remarks, 



'I Anaphylactic death. 



Death few hours after sornm. 



Haemothorax fluid gave pure B. 



P. M. oedema fluid gave B. wehhii, 
B. sporogenes, and ? B. oedema- 
tiens. 



Died 9 hours after operation and 
serum. 

Small prophylactic dose of serum 
only given. 

A welchii-septique antitoxic se- 
rum protected mice against 
whole culture (Henry). 



136 




.Bo. 

, Pi. 
. Th. 

Wa. 
Wo. 
Sm. 

Ma. 
Po. 

Or. 

We. 
. Br. 

Ba. 
Mo. 

Po. 
Re. 



MoNee 



MoNee 

MoNee 
MoNee 

McNee 
MoNee 
McNee 

McNee 
Stokes 

Stokes 

Stokes 
Stokes 

Stokes 
Stokes 

Stokes- 
Stokes 



Thigh and pelvis. 

Arm and thorax. 
Thigh. 

Thorax. 

Knee and thigh. 



DayoJ 
onset of 

gas 
gangrene. 



1 

(Thigh) 

1 



Lesion. 
Thigh and elsewhere. 

Multiple. 

Right arm. 
Multiple. 

Left thigh. 

Left leg. 

Left arm, brachial 
artery and median 
nerve torn. 

Arm. 

Thigh and buttock glu- 
teal artery torn, sci- 

. atic nerve damaged, 
fractured pelvis, rec- 
tum torn. 

Leg. 



Arm, trunk, and right ' 2 



Surgery. 



(1) Amputation of thigh- and excisi 
of other -wounds. 



(1) Amputation left leg. 
(1) Wounds cleaned up. 



(3) Amputation right arm and 1 
leg ; other wounds excif 
and foreign bodies remove( 



1 


(1) Amputation. 


1 


(1) Amputation. 


2 


(1) Wound cleaned, artery tied, ai 

nerve sutured. 

(2) Amputation. 


1 


(1) Disarticulation at the shoulde 


2 


(1) Wounds excised, foreign boi 
removed. 



(2) Circular amputation, somen 
crosed muscle left. 



(1) Partial amputation at field ai 

bulance. 
(3) Amputation. 

Too ill for operation ; forei( 
bodies and bits of bone i 
moved. 



1 (1) Wound cleaned upandhaem 

thorax aspirated. 

1 Inoperable, pulseless, and delirio 

on admission. 



(3) Amputation. 



r 

Serum 


Result. 


Bacteriology. 


Bemarks. 


1) 20 LV. B. 

2) 30 LV. B. 

3) 30 LV. B. 


— 




... 


2) 20 LV. B. 

2) 30 LV. B. 

3) 30 LV. B. 


— 






1) 20 B.' 

2) 30 B. 


- 






3) 20 B. 

4) 30 B. 
6) 30 B. 


(10) + 




Death from lobar pneumonia. 








1) 30 B. 


+ 




Death 10 hours after operation. 


1) 20 LV. B. 


- 






2) 30 B. 


— 







1) 30 B. 

2) 40 LV. G 18 

and G 19. 



!) 30 LV. G 19. 
!) 40 LV. G 19. 
1) 50 LV. G 19. 



t) 20 LV. G 19. 
i) 20 LV. G 19. 



) 20 LV. G 19. 
30 S.C. G 19. 
10 locally. 



) 20 LV. G 18. 
30 S.C. G 18. 



) 60 LV. G 3 
and G 22. 



) 30 LV. G 17. 

) 50 S.C. G 18. 



(10) + 



(3) + 



(2) + 



(3) + 



Blood culture gave streptococcus. 
Wounds gave B. welchii, B. spo- 
rogenes, and streptococci. 



Blood culture sterile. Wound gave 
B. welchii, B. sporogenes, another 
welchii-like anaerobe and strepto- 
cocci. 

Wound gave B. welchii, strepto- 
cocci, and staphylococci. 



(1) Blood culture nil. 

(1) Wound gave liirion septique 

and B. welchii. 

(2) P.M. heart blood gave vibrion 

septique and B. welchii. 

(1) Blood culture nil. 
(1) Wound gave B. welchii, B. fal- 
lax, and B. sporogenes. 

(1) Blood culture gave pure B. 

fallax. 
(1) Blister fluid gave B. fallax and 

B. vxlchii. 

Haemothorax fluid gave pure cul- 
ture of B. welchii. 

(3) Blood culture nil ; wound gave 

B. welchii, B. sporogenes, and 
another anaerobe not iden- 
tifled. 



Death 1 1 hours after operation. 



Death due to lobar pneumonia. 



Died 1 hour after serum. 



Died 6 hours after serum. 



138 



Case. 
[. Pr. 

!. To. 

I. Fo. 



Br. 



Du. 



Co. 



■ Sy. 



Wa. 



Ca. 



Ha. 



Ha. 



Reported by. 



Stokes 



Stokes 



Stokes 



Stokes 



Stokes 



Stokes 



Stokes 



Stokes 



Stokes 



Stokes 



Stokes 



Stokes 



Lesion. 



Right thigh, left thigh, 
and right groin. 



Dayo} 
onset of 

gas 
gangrene. 

1 



Multiple wounds both j 
legs and right arm. \ 



Both thighs and right 
arm. ] 



Thigh. 



Leg and thorax. 



Arm and head. 



Ijeft femur and right 
thigh. I 



Thigh. 



Multiple wounds of 
both thighs. 



Thigh. 



Buttock. 



Buttock. 



Surgery. 



(1) Wounds cleaned, femoral v 

tied. 
(12) Amputation for streptocoq 
infection of knee. 

(2) Wounds excised and forei 

bodies removed. 



(1) Wounds cleaned and forei 

bodies removed. 

(2) Amputation of left leg. 



(1) Multiple incisioiiB. 

(2) Amputation. 



(1) Wounds cleaned up. 

(2) Excision of infected muscle. 



(1) Amputation of arm. 

(2) Head operation. 



(1) Amputation left thigh. 

(2) Excision muscles right thigh. 



(1) Excision of wound and ligatu 

of femoral vein. 

(2) Amputation through thigh. 

No operation. 



(1) Excision of infected muscle ai 
removal of foreign body. 



(1) Excision of muscle andremovi 

of foreign body. 

(2) Further excision of muscle. 

(1) Foreign body removed and 
wound track excised. 



Serum 



35 I.V. G 19. 

15 S.C. G 19. 



(3) 25 I.V. G 19. 
25 S.C. G 19. 



(1) 20 I.V. G 19. 

30 S.C. G 19. 

(3) 40 S.C. G 19. 



(1) 30 I.V. 6 17. 
30 I.V. G 22. 

(2) 90 I.V. G 17 

and G 22. 



(2) 60 I.V. G 19. 



(2) 40 S.C. G 22. 



(2) 60I.V. G22. 



(2) 60 I.V. G 44. 



(1) 60 LV. G 44. 

(2) 60 IV. G 44. 



(1) 80 LV. G 22. 

40 S.C. G 22. 



(3) 60 I.V. G 22. 
60 S.C. G 22. 



(2) 60 I.V. G 22. 




(3) + 



(2)+. 



(2) + 



(2) + 



(2) + 



(2) + 



(2) + 



Bacteriology. 



Blood culture nil ; wound gave B. 
wdcMi, B. sporogenes, and another 
anaerobe not identified. 



Blood culture gave B. welchii and 
B. sporogenes ; wound gave B. 
welchii, B. sporogenes, strepto- 
cocci, diphtheroids, and coliforms. 

(2) Blood culture gave B. welchii. 

(3) Blood culture negative. 

(2) Wound gave B. welchii and 
other anaerobes. 

( 1) Blood culture gave B. fallax. 

(2) Blood culture gave B. fallax. 

(3) Blood culture negative. 

(1) Oedema fluid gave B. welchii 

and B. fallax. 

(2) Blood culture gave B. welchii. 



Cultures from the arm gave B. 
sporogenes, B. welchii, and pro- 
teus. Cultures from head gave 
B. welchii and B. sporogenes. 

(2) Blood culture negative. 

(2) Heart blood taken at post mor- 
tem gave B. welchii. Cul- 
tures from wound gave B. 
welchii, B. sporogenes, strep- 
tococci, and a coliform ba- 
cillus. 



(2) Blood 



culture negative ; B. 
and B. sporogenes 
isolated from the wound. 

(1) Cultures from the wound gave 
B. welchii, B. sporogenes, and 
a non-pathogenic tetanus- 
like bacillus. 

(1) Blood culture negative. 

(2) Post-mortem heart blood gave 

B. welchii and streptococci ; 
the wound yielded B. welchii, 
B. sporogenes, and strepto- 



B. 



Cultures from wound gave 
welchii and B. sporogenes. 



(1) Blood culture gave an aber- 
rant B. welchii. Wound 
' cultures gave this aberrant 
bacillus, B. welchii and B. 
sporogenes. 



Bemarke, 



Restless, pulseless, and pallid 
2 hours after serum. Death 
7 hours after serum. 

Late death from meningitis. 
Streptococci and staphylococci 
only found in cultures made 
from the meningeal exudate. 

Death 5 hours after serum. 



Death 2 hours after serum. 



Gas gangrene arrested by serum 



140 



'ase. ' Reported by. 



Stokes 

Stokes 

Stokes 
Tytler 

Tytler 

Tytler 

Tytler 

Tytler 

Tytler 

Tytler 

Tytler 



Tytler 




Left leg and shoulder. 



Bight thigh and but- 
tock, right arm, and 
head. 



Head. 



Left shoulder and back. 



Day of 
onset of 

gas 
gangrene. 



Surgery. 



(1) Excision of infected muscle. 



(1) Excision of muscles. 

(2) Eurther excision. 

No operation. 

(2) Amputation of leg. 
(5) Transfusion with blood. 

(2) Excision of infected muscles. 



(2) Amputation left thigh. 



(2) Wound excised. 
(6) Thigh incised. 

(5) Bullet removed from back, 
haemothorax fluid aspirated. 



(2) Amputation. 

(5) Excision of muscle from stump 
and buttock. 

(2) Excision of wounds and removal 

of foreign bodies. 

(3) Excision of buttock muscles. 



(2) Wound excised, skull trephined,, 
foreign body removed. 



Extensive muscle excision. 



'ol. Tytler Thorax. 



• Aspiration followed by rib resection. 



141 



Serum. 



1) 60 I.V. G 22. 

2) 60 I.V. G 22. 
40 S.C. G 22. 

2) 60 I.V. G 22. 

5) 10 I.V. G 3. 



4) 20 I.V. G 17. 
20 I.M. G 17. 

5) 20 I.V. G 22. 
20 I.M. G 22. 

8) 20 I.V. G 22. 



6) 20 I.V. G 22. 
40 I.M. G 22. 



5) 20 I.M. G 17. 

6) 20 I.M. G 17. 



3) 20 I.V. G 22. 
20 I.M. G 22. 



2) 20 I.M. G 19. 

3) 40 I.M. G 19. 

4) 20 I.M. G 19. 
20 I.M. G 22. 

5) 40 I.M. G 22. 

6) 40 I.M. G 22. 

7) 40 I.M. G 22. 

3) 10 I.V. G 3. 
20 S.O. G 3. 
20 S.G. G 3. 

4) 20 S.C. G 3. 



2) 20 I.V. G 22. 

20 I.V. G 22. 

20 I.V. G 22. 

10 T.V. G 22. 

10 I.V. G 22. 

10 I.V. G 22. 
?otal 90 c.c. in 

24 hours. 



Besidt. 



(9) + 

(2) + 
(2) + 



(5) + 
(8) + 

(6) + 



(3) + 



(8) + 



(6) 



(14) + 



Bacteriology. 



B. welchii and two types of B. 
sporogenes isolated from the 
wound. 

(2) Blood cxilture gave B. welchii 
and streptococci. 



Wound gave B. welchii and strepto- 
cocci. 

(4) Blood culture negative ; wound 
gave pure B. welchii. 

Wound gave B. welchii and strepto- 
cocci. 

Wound gave B. welchii and strepto- 
cocci. 

Haemothorax fluid gave B. welchii. 
Heart blood post mortem gave 
B. welchii. 

Smear from wound showed B. 
welchii, but cultures were nega- 
tive. 

(3) Blood culture gave B. welchii 

and a streptococcus. 
Wound at post mortem gave B. 
welchii and a streptococcus. 

(2) Wound gave B. welchii. 

(4) Wound gave B. welchii and 

streptococci. 
(6) Wound gave B. welchii and 
streptococci. 
Heart blood post mortem gave 
B. welchii. 

(3) Muscle excised at operation 

gave B. welchii and strepto- 
cocci. 

(4) Wound gave B. welchii, strepto- 

cocci, and some spore-bearing 
bacilli. 
(6) Heart blood post mortem gave 
B. welchii and a haemolytic 
streptococcus. Post-mortem 
cultures from pericardial 
fluid and liver yielded pure 
cultures of haemolytic strep- 
tococci. 



Remarks. 



Gas gangrene arrested by serum ; 
death from streptococcal septi- 
caemia. 

Death a few hours after serum. 
Died 8 hours after serum. 



Death 4 hoiu-s after serum. 



Death J hour after serum. 



Death 6 hours after serum. 



Death 3 hours after serum. 



Death probably resulted from 
streptococcal septicaemia. 



142 



Case. 
Ke. 



Reported hy. 



Tytler 



01. Wyard 



Con. 



Qu. 



Pu. 



Cu. 



Wyard 



MoNee 



MoNee 



McNee 



McNee 



Pro. 



Ip. 



Brenan 



Tytler 



Tytler 



Lesion, 



Day of 

onset of 

gas 



Left back. 



Back, neck, and tongue. 



Left buttock, groin, and 
pelvis. 

Buttock. 



Right elbow ; brachial 
artery and veins di- 
vided. 



Thigh. 



Both thighs; also chest 
with haemothorax. 



Haemothorax. 



Left buttock. 



Multiple wounds. 



(3) Wound excised and large forei] 
body removed from pso 
muscle. 



(1) Wounds excised and cleaned. 



(1) Wound laid open and infecte 
muscle excised. 



(1) Wounds cleaned ; vessels tied 

ulnar nerve sutured. 

(2) Amputation. 



(1) Excision of muscle. 
(2^ Excision of muscle. 



(2) Resection of rib ; evacuation of 
haemothorax fluid ; removal 
of foreign body and bits of 
clothing from the limg. 



(1) Amputation right thigh. 

(2) Excision wound left thigh. 



143 



Serum. 



)20I.V..G22. 
120 I.M. G 22. 
) 40 I.M. G 22. 
) 90 I.M. G 22. 

) 40 I.V. G 18. 

60 I.V. G 19. 

1) 20 I.V. G 19. 

I) 20 I.M. G 17. 
■) 100 I.V. G 17. 

) A.T.S+ W. 
I) G 44. 



40 B. 



I) A.T.S+W. 
J) 20 G 44. 



I) 20 G 44. 




(8) + 



(3) + 



30 B. 



W 100 I.V. G 44. 



5) 30 I.M. G 22. 

6) 30 I.M. G 22. 
6) 20 I.V. G 53. 



(2) + 




(5) + 



(14) + 



(6) + 



Culture from gangrenous testicle 
gave B. welchii and streptococci. | 



Cultures from wound gave no 
growth of anaerobes. 



Muscle from fost mortem showed 
? vibrion septique. 

Cultures from muscle gave B. wel- 
chii, B. sporogenes, streptococci, 
and another anaerobe. A wel- 
ohii-septique antitoxic serum did 
not protect a guinea-pig against 
the mixedmuscle culture(Henryf. 

Cultures from muscle gave B. wel- 
chii, B. sporogenes, a round end- 
sporer, and streptococci. A wel- 
chii-septique antitoxic serum pro- 
tected a guinea-pig against the 
mixed muscle culture (Henry). 

gave B. wel- 
, round end- 
■sporer, and 

A welchii- 
serum pro- 
against the 

(Henry). 



Cultures from muscle , 
chii, B. sporogenes, a 
sporer, an oval end 
another anaerobe, 
septique antitoxic 
tected a guinea-pig 
mixed muscle otilture 



Cultures from muscle gave B. wel- 
chii, B. sporogenes', a round end- 
sporer, an oval end-sporer, an- 
other anaerobe, streptococci, and 
a conform bacillus. A welchii- 
septique antitoxic serum pro- 
tected a guinea-pig against the 
mixed muscle culture (Henry). 

Cultures from haemothorax fluid 
gave B. welchii and B. sporo- 
genes. A welchii-septique anti- 
toxic serum protected a guinea- 
pig against this mixed culture 
(Henry). 

Culture from infected muscle gave 
B. welchii and B. oedematiens. A 
welchii-septique antitoxic serum 
did not protect a guinea-pig 
against this mixed culture 
(Henry). 

Culture from infected muscle gave 
B. welchii, streptococci, and an 
unidentified anaerobe. A wel- 
chii-septique antitoxic serum pro- 
tected a guinea-pig against this 
mixed culture (Henry). 



Semarks. 



Gas gangrene checked by serum. 
Case reported in full in Section 
VI. 



144 



Case. Reported hy. 



i. She. Tytler 



Du. 



Tytler 



Coo. Tytler 



Me. 



Tytler 



Day of 
onset of 



Lesion. 


gas 
gangrene 


Multiple, arm. 


3 


Thigh. 


2 


Buttock, torn gluteal 
artery. 


2 


• 





Swrgery. 



Thigh and buttookB. 



(3) Amputation. 



(2) Wounds cleaned up. 



(1) Wound cleaned,' gluteal arte 
tied. 



(1) Wounda cleaned. 

(4) Excision of hamstring muscle 



140 



Serum. 



J) 60 I.M. G 22. 
3j 20 r.V. G 22. 



2j 100 I.M, G 44. 



75 20 I.V. G 22. 
1) SO I.M. G 22. 
3) 90 I.M. G 22. 
)J 20 I.M. G 22. 
I) 20 I.M. G 22. 
J) 20 I.M. G 22. 

t) 100 I.M. G 22 
80 I.M. G 44. 
3) 40 I.M. G 44. 



Result. 



(21) + 



(9) + 



Bacteriology. 



Culture from infected muscle gave 
B. welchii and streptoeocoi. A 
welchii-septique serum protected 
a guinea-pig against this mixed 
culture (Henry). 

Cultures of material from the 
wound gave B. welchii, strepto- 
cocci, and an unidentified anae- 
robe. A welchii-septique serum 
protected a guinea-pig against 
this mixed culture (Henry). 

(5) Blood culture gave B. welchii 
and streptococci. 

(9) Blood culture gave strepto- 
cocci only. 



A welchii-septique serum did not 
protect guinea-pigs against mixed 
muscle cultures from this case ; 
nor did a high-titre oedematiens 
serum. An organism was ob- 
tained in pure culture from the 
heart blood of guinea-pigs which 
had succumbed to infection by 
mixed muscle cultures. This 
organism is believed to be a 
pathogenic B. fallax (Henry). 



Remarks. 



146 



BIBLIOGRAPHY 

The following bibliography (arranged alphabetically) contains 
the principal papers and books which have been written on the 
pathology aud bacteriology of anaerobic infections and in particular 
of gas gangrene. Papers dealing only with clinical symptoms and 
surgical treatment have not been included. With the exception of 
those marked * which were inaccessible all the papers have been 
consulted by Dr. William Bulloch, Chairman of the Committee. 
The bibliography is brought down to July 1919. 

AcHALME, P., (1) Examen bacteriologique d'un cas de rhumatisme articu- 

laire aigu moit de rhumatisme cerebral. Compt. rend. Soc. de biol., Par,, 

1891, 43, 651. 
AcHALMB, P., (2) Recherches sur quelques bacilles anaerobies et leur dif- 

ftrenoiatiou. Ann. de I'Inst. Pasteur, Par., 1902, 16, 633. 
Adami, Bowman, Adams and others. Combined inquiry into the presence 

of diphtheria and diphtheroid bacilli in wounds. Bull. Canad. Army 

Med. Corps, 1918, 1, 33. 
Adamson, Ji. S., On the cultural characters of certain anaerobic bacteria 

isolated from war wounds. J. Path, and BaeterioL, Cambridge, 1918- 

19, 22, 345. 
Adamson and Cutler, Note on a bacillus resembling B. tetani. Lancet, 

Lond., 1917, I, 688. 
Adye-Cueean, Gunshot wound, pericardium ; pyo-pericardium with gas 

infection. J. Roy. Army Med. Corps, Lond., 1918, 30, 599. 
Albeecht, p., (1) Ueber Infektionen mit gasbildenden Bakterien. Arch.f. 

hlin. Ghirurgie, Bar]., 1902, 67, 514. 
Albeecht, P., (2) Ueber Gasbrand. Deutsche med. Wchnschr., Berl. u. 

Leipz., 1917, 43, 445. 
Andbbs, Ueber pathologisch-anatomische Veranderungeu des Zentral- 

nervensystems bei Gasoedem. Munchen. med. Wchnschr., 1917, 64, 1600. 

Also Beitr. z. klin. Chir., Tubing., 1918, 109, 194. 
Andeewes, F. W., Report on spore-bearing anaerobic bacilli in the con- 
tents of the human intestine, with special reference to their concern with 

acute diarrhoea. Pep. Med. Off. Local Gov. Bd., Lond., 1896-7, 26, 

255. 
Aeloing, Lefons sur la tuberculose et certaines septicemics, recueillies par 

J. Courmont. Par., 1892, 520 pp., 8°. 
Aeloing, Coenevin et Thomas, Le charbon symptoraatique du bceuf; 

pathogenic et inoculations preventives. 2e Edit., Paris 1887 dd 281 

1 pi. • ' ' ^'^' ' 

Aemknbcht, W Beitrag zum Wesen und zur Therapie der Gasphlegmone. 
Munchen. med. IFcAmc/tr., 1915, 62 452 

^ Br:: \ll^'\llTZ isl ^^^P^'^^^-°-- ^-*-^- ^^^- Wohnschr., 

^Tri'9S 42:46^51? ^^'^°^°^^^ "'^^ "^^^^^^^''^ ^- ^-^'^-^ 



147 

AsoHOFF, (3) Ueber bakteriologisclie Befunde bei den Gasphlegmonen. 

lUd., 1917,43, 1468. 
AscHOFF, (4) Zur Frage der Gasodemerreger und ilirer Bedeutung fiir die 
Gewinnung eiaes Gasodemsclmtzserums. Veroffentl. a.d. Geh. d. Mil.- 
. Sanitdtswesens, Berl., 1918, Heft 68, 1. 
Bachmann, E., Beitrag zur Kennfcnis des Bacillus des malignen Oedems. 

Centralhl. f. Bakt&riol. [etc.], Jena, 1904, Orig. 37, 221 ; 353. 
Bahe, Einige Garungsversuche mit Bazillen der Odembazillengruppe. 
Ztschr. f. Infektionskr., 'parasit. Kranhh. und Hyg. d. Haustiere, Berl., 
1911, 9, 225. 
Baiee, Eduaed, Ueber Buttersauregahrung. Centralhl. f. Bakteriol. [etc.], 

Jena, 2*6 Abt., 1895, 1, 17 ; 84; 118. 
Bain, .1. B., A pseudo-tetanus bacillus. J. Boston Soo. M. Sc, 1901, 5, 

506. 
Baeber, M. A., The pipette method in the isolation of single microorganisms 
and in the inoculation of substances into living cells. Fhilippine J. of 
Sc, B. Trap, med., 1914, 0, 307. 
Barber and Dale, Note on a supposed soluble toxin produced in artificial 
culture by the bacillus of malignant oedema. Brit. M. J.^ Lond., 1915, 
II, 808. 
Baetoli, F., L'infezione gasosa delle ferite di guerra. Polidin., Eoma, 

1916, 23, sez. prat., 1493. 
Bashford, E. F., General pathology of acute bacillary gangrene arising in 

gunshot injuries of muscle. Brit. J. Surg., Bristol, 1916-17, 4, 562. 
Baughee, The Bacillus aerogenes capsulatus in blood cultures, with recoveries. 

J. Am. M. Ass., Chicago, 1914, 62, 1153. 
VON Baumgaeten, Kriegspathologische Mitteilungen. Miinchen. med. 

Wchnschr., 1918, 65, 175; 212. 

Eazy, Le serum de Leclainche et Vallee dans la prevention des infections 

post-operatoires. Bull, et mem Soe. de chir. de Par., 1917, n.s., 43, 824. 

Bbijeeinck, (1) Les orgariismes auaerobies obligatoires ont-ils besoin 

d'oxygfene libre? Arch, neerlandaises d. sc. exactes et naturelles. La Haye, 

1899, Ser. II, 2, 397. 

Beijeeinok, (2) Phenomenes de reduction produites par les microbes. Ibid., 

1904, Ser. II, 9, 131. 
Bbitzke, Zur Frage der Uebertragbarkeit des Gasbrandes. Berl. Min. 

Wchnschr., 1918, 55, 1143. 
Bell, F. M., Malignant oedema. Brit. M. ./., Lond., 1915, I, 843. 
*Bi:EAED, LuMiBEE et DuNBT, Phlegmon iii6tastatique gangreneux et gazeux 
d'origine streptococcique. Bull, mid. Par., 1918, 32, 146. Ref. BuU, de 
I'Inst. Pasteur, Par., 1919, 17, 129. 
Bbethelot, Alb., JEtude biochimique de deux microbes anaerobies du 

contenu intestinal. Ann. de VInst. Pasteur, Par., 1909, 23, 85. 
Wesson, A., Contribution k I'^tude du vibrion septique. Ibid., Par., 1895, 

9, 179. 
BiBNSTOCK, B., (1) Ueber die Bakterien der Faces. Ztschr. f. Mm. Med., 

Berl., 1884, 8, 1. 
BiBNSTOCK, B., (2) Untersuchurigen uber die Aetiologie der Eiweissfaulnis. 

Aroh.f. Hyg, Mnnchen u. Berl., 1899, 36, 335; 1901, 39, 390. 
BiENSTOCK, B., (3) Bacillus putrificus. Ann. de VInst. Pasteur, Par., 1906, 

20, 407. 
TiiBE, Aug., (1) Die Gasphlegmone im w^esentlichen eine Muskelerkran- 

kung. Med. Klinik, Berl, 1916, 12, 355. 
BiBE, Aug., (2) Anaerobe Wundinfektion (abgesehen von Wundstarr- 

irampf). ' Beitr. z. klin. Chir., Tubing., 1916, 101, 271. 

K 2 



148 

BiBEMANN, Zur Diagnose der Gasgangrari. Miinclien. med. Wchnschr., 1916, 

63, 1573. 
BiNGOLD, (1) Gasbazillensepsis. Deutsche med. Wchnschr., 1915, 41, 191. 
*BiNGOLD, (2) Kritisches uber Gasbazilleninfektionei). Beitr. z. Khnik d. 

InfeUionskranhh., Wiirzb., 1918, 6, 209. 
Bloodgood, Gas-bacillus infection ; surgical bacteriology. . Tr. Am. Swrg. 

Ass., Phila., 1916, 34, 122. 
BowLBY, Anthony, On British military surgery in tbe time of Hunter and 

in the Great "War. Brit. M. J., Lond., 1919, I, 205. 
BowLBY and Eowland. A report on gas gangrene. Brit. M. J., Lond., 

1914, II, 913. Lancet, Lend., 1914, II, 1265. J. Roy. Army Med. 

Corps, Lond., 1914, 23, 514. 
Bebdemann, Bacillus amylohacter A. M. at Bredemann in morphologiBcher, 

physiologischer und systematischer Beziehung, mit besonderer Beruck- 

sichtigung des Stickstoffbindungsvermogens dieser Species. Centralbl. 
/. Bakteriol. [etc.], Jena, 2*6 Abt., 1909, 23, 385. 
Beiegee und Ehrlich. Ueber das Auftreten des malignen Odems bei 

Typhus abdominalis. Berl. klin. Wchnschr., 1882, 9, 661. 
Beuce, Sir D., (1) Analysis of cases of tetanus treated in home military 

hospitals from August, .1914, to August, 1915. Brit. M. J., Lond,, 1915, 

11, 593; also Lancet, Lond., 19J5, II, 901; also J. Boy. Army Med. 

Corps, Lond., 1915, 25, 535. 
Betjce, Sir D., (2) Cases of tetanus treated in home military hospitals 

from August 1st, 1915, to July 31st, 1916. Lancet, Lend., 1916, II, 

929. 
Beuce, Sir D., (3) An analysis of cases of tetanus treated in home military 

hospitals during August, September, and part of October, 1916. lUd., 

1917, I, 986. 
Beuce, Sir D., (4) Fourth analysis of cases of tetanus treated in home 

military hospitals during part of October, November, and part of 

December, 1916. /Stci., 1917, II, 411. 
Beuce, Sir D., (5) Fifth analysis of cases of tetanus treated in home military 

hospitals during part of December, 1916, all January and February and 

part of March, 1917. Ibid., 1917, II, 925. 
Beuce, Sir D., (6) Tetanus ; analysis of 1,000 cases. Trans. Soc. of Trap. 
. Med. and Hyg., Lond., 1917-18, 11, 1. 
Beunnbe u. v. Gonzbnbach, Erdinfektion und Antiseptik. Centralbl. f, 

Ghir., Leipz., 1916, 43, 1019; 1917, 44, 546. 
Beunnee, t. Gonzbnbach und Eittee, Experimentelle Untersuchungen 

uber Erdinfektion and Antiseptik : ein Beitrag zur richtigen Einschatzung 

der chemischen Wundantiseptik. Beitr. z. klin. Chir., Tiibing., 1918, lU, 

57 A. 1^ 

Bull, 0. G., (1) The prophylactic and therapeutic properties of the anti- 
toxin for Bacillus Welchii. J. Exp. M., IS. Y. 1917 26 603. 
Bull, C. G., (2) Gangrfene gazeuse. Toxine specifique et antitoxine. C<mpt. 

rend. Soc. de biol.. Par., 1917, 80, 957. 
Bull, C. G., (3) Antitoxin for gaseous gangrene. iV. York M. J. retc.l 

1917, 106, 821. ' ■- ■^' 

^^^\^\'^r:' ^^] ^,^^"™^ <Jes resultats d'exp^riences faites avec des baeilles 
de Welch. Arch, de rried. et de pharm. milt. Par 1918 70 198 

Bull and Peitchett (1) Toxin and antitoxin of, and protekion against. 
Bacillus Welchii. J. Exp. M.,^.Y., I'^l? 26 119 

Pull and P^tchbtt (2) Identity of the toxin^ of different strains of 
Vo?7 o« fil*; ^^'*°'' '°fl"e"""g their production in vitro. Ibid, 



149 

Bullock, W. E., and Cramer, (1) On a new factor in the mechanism of 
bacteria] infection. Proc. Hoy. ,Soc., Lond., 8er. B, 1919, 90, 513. 

Bullock, W_. E., and Cramer, (2) On the mechanism of bacterial infections 
with special reference to gas gangrene. Sixth Sc. Rep. Imp. Cancer 
liesearch Fund, Lond., 1919, 23. 

BuKGE, R., Zur Aetiologie der Gasphlegmonen. Fortschr. d. Med. 1894 
12, 533. 

BuRCHARD, A., Gasabscess, Gasphlegmone und Gasgangran im Rontgeubild. 
Med. KUnik, Berl, 1916, 12, 744. 

BuRGHARD, Leishman, Moynihan and Wright, Memorandum on the treat- 
ment of the bacterial infections of projectile wounds. Brit. M. J., Lond., 

1915, I, 735. Lancet, Lond., 1915, I, 873. 

Burnet et Weissenbach, Classification des streptocoques. .Bull, de I'Inst. 

Pasteur, Par., 1918, 16, 657 ; 697. 
BuRRi u. KtJestbinbr, Ein experimenteller Beitrag zur Kenntnis der 

Bedeutung des Sauerstoffentzugs fur die Entwickelung obligat anaerober 

Bakterien. Centralbl. f. Bakteriol. [etc.], Jena, 1908, Abt. 2, 21, 289. 
BussoN und Gyoegy, Ueber anaerobe Wundinfektion durch Gasbrand- 

bacillen. Wien. klin. Wchnschr., 1916, 29, 737; 937. 
Canelli, II flemmone gangrenoso. Pathologica, Geuova, 1917, 9, 241. 
Cavina, Osservazioni sopra 20 casi di gangrena gazosa. Morgagni, Milano, 

1916, 58, pt. 1, 142. 

Cayrel, La flore miciobienne des blessures de guerre. Son ensemence- 

ment, sa nature, son evolution. Paris chirurg., 1916, 8, 81. 
Cazin, Traitemeut des plaies infectees par le serum de Leclainche et 

Vallee. Paris med., 1916, 6, 262. 
Chalier, Andre, (1) Gas gangrene. Dublin J. M. Sc, 1915, 140, 358 ; 

437. 
Chalier, (2) Donnees statistiques tirtes de 108 cas de gangrenes gazeuses 

vraies. Presse med., Pai-., 1917, 25, 390. 
Chalier, A. et Chalier, J., (1) La gangrfene gazeuse. Par., 1917. 
Chalier, A. et Chalier, J., (2) A propos de la bacteriologie et de la patho- 

g^nie de la gangrene gazeuse. Arch, de med. et depharm. mil., Par., 1918, 

69, 415. 
Chauvbau, De la septicemie gangreneuse. Bull. Acad, de med., Par., 1884, 

S6r. 2, 13, 1129. 
Chauvbau et Arloing, Etude experimentale sur la septicemic gangreneuse. 

lUd., Par., 1884, Ser. 2, 13, 604. 
Chavasse, p., Statistique des cas de tetanos observes dans la zone des 

arm6es du \^^ novembre 1915 au l^r fevrier 1917. Bull, et mim. Soo. de 

chir. de Par., 1917, n.s., 43, 1249. 
Chavigny, Gangrene gazeuse subaigue provoquc'e par un bacille special. 

Arm. de I'Inst. Pasteur, Par,, 1897, 11, 860. 
Chiari, H., Zur Pathologie des ' septischen Emphysems '. Prag. med. 

Wchnschr., 1893, 18, 1. 
Chudiakow, Zur Lehre von den Anaerobiose. Moskau, 1896, 8°, 116 pp. 

Ref. Centralbl. f. Bakteriol. [etc.], Jena, Abt. 2, 1898, 4, 389. 
CoENEN, (1) Ein Ruckblick auf 20 Monate feldarztlicher Tatigkeit mit beson- 

derer Berticksichtigung der Gasphlegmone. Beitr. z. klin. Chir., Tiibing., 

1916, 103 (Kriegschir. Hefte, VI), 397. 
CoBNEN, (2) Die Bosartigkeit des Gasbrandes in manchen Kampfgebieten. 

Berl. klim. Wohnsch/r., 1917,54:, 354: ; 379. 
Cole, !Note on a case of infection by Bacillus aerogenes capsulatus in 

which the organism was demonstrated in the circulating blood during 

life. Johns Hopkins Hasp. Bull, Bait., 1902, 13, 234. 



1.50 

CoNEADi uud BiBLiNG, (1) Zur Aetiologie und Pathogenese des Gasbrandes. 

Miinehen. med. Wchnschr., 1916, 63, 133; 178; 1023; 1068; 1561; 

160.8. 
CoNEADi und Sibling, (2) Ueber Gasbrand uiid seine Ui-Baohen. Berl. 

klin. Wchnschr., 1917, 54, 449. 
Costa et Teoisiek, (1) Sur I'association fr^quente du pneumocoque et du 

B. perfringens dans les blessures de guerre, notamment dans le syndrome 

' Gangrfene gazeuse'. Cornet, rend. Soc. de hidl., Par., 1915, 78, 283. 
Costa et Teoisibe, (2) Syndrome mortel d'oedfeme gazeux dans une blessure 

de guerre, provoqu6 par le ' Bacille neigeux '. Ibid., 1915, 78, 362. 
Costa et Tkoisiee, (3) Action h^molytique de certainesbacteries^naerobies 

des blessures de guerre. Ibid,, 1915, 78, 354. 
Costa et Teoisiee, (4) Sur un groupe de bacteries ana6robies des blessures 

de guerre intermediaire entre le ' B. peifringens ' et le 'Vibriou septique'. 

Ibid., Par., 1915, 78, 430. 
VAN COTT, Untersuchungen iiber das Vorkommen der Bacillen des malignen 

Odems in der Moschustinctur. Gentralhl. f. Bakteriol. [etc.], Jena, 1891, 

9, 303. 
CoTTET, Note sur les streptocoques dans les plaies de guerre. Frequence 

des streptocoques anaerobies. Gompt. rend. Soc, de biol.. Par., 1918, 81, 6. 
Ceambe, AV., (1) Further observations on the thyroid-adrenal apparatus. 

A histochemical method for the demonstration of adrenalin granules in 

the suprarenal gland. J. Physiol., Cambridge (Proc. of Physiol. Soc), 

1918, 52, p. viii. 
Ceambe, W., (2) Histochemical observations on the functional activity of 

the suprarenal medulla in different pathological conditions. Ibid., 1918, 

52, p. xiii. 
Ceamee, W., (3) Observations on the functional activity of the suprarenal 

gland in health and disease. Sixth Sc. Rep. Imp. Cancer Research Fund, 

Lond., 1919, 1. 
Cummins and Geaemb Gibson, An analysis of cases of tetanus occuriing in 

the British Armies in France between Nov. 1st, 1916, and Dec. 31st, 

1917. Lancet, Lond., 1919, I, 325. 
CuMSTON, Bacteriology of gas gangrene. N. York M. J., 1918, 108, 423. 
Dalyell, E. J., A case of gas gangrene associated with B. oedematiens. 

Brit. M. J., Lond., 1917, I, 361. 
Davids, H., (1) Malignes Odem. Ergebn. d. alia. Path. u. path. Anat. 

[etc.], Wiesb., 1901, Jahrg. VI 100. 
Davids, H., (2) Eauschbrand. Ibid., 1901, Jahrg. VI, 116. 
Dean and Mouat, The bacteria of gangrenous wounds. J. Roy. Army Med. 

Corps, Lond., 1916, 26, 189; 349; also Brit. M. J., 1916, I, 77. 
De Angelis, Sulla flora batterica della gangrena gassosa. Gazz. d. osp., 

Milano, 1916, 37, 246. 
Debono, p.. On some anaerobical bacteria of the i;orraal human intestine. 

Gentralhl. f. Bakteriol. [etc.], Jena, 1912, Orig. 62, 229. 
Degtjy, Sur les infections putrides avec embolies g'angreneuses multiples. 

Pans med., 1916, 6e annge, 15 juillet, 62. 
Dblbet, p., (1) Quelques observations de plaies de guerre traitees par 

1 auto-vaccm iod6 total de Weinberg et Seguin. Gompt. rend. Soc. de biol. 

Far., 1916, 79, 22. ° ^ 

Delbet, p., (2) Un cas de gangrene gazeuse, traits par le s6rum mixte anti- 

gangreneux (Weinberg et S6guin). Ibid., 1918, 81, 798 

Tq^, a o^^t^^"^' ^'^ Gasphlegmone im Kriege. Wim. klin. Wchnschr. 
Deeganz, D«r Gasbrand. Wien. klin. Wchnschr., 1916, 29, 7. 



151 

Dickson, E. C. Botulism. A clinical and experimental study. Roolcefeller 
Inst. Med. Research, 117 N. Y., 1918, pp., 8o, 10 pi. 

DiSTASo, (1) Sur les microbes proteolytiques de la flora intestinale de 
1 homme et des animaux. CentralU. f. Bakteriol. retc.l Jena, 1911, Orio' 
59, 97. L J > . s- 

DiSTASO, (2) Contribution a I'^tude sur rintoxication intestinale. lUd. 

Jena, 1912, Orig. 62, 433. 
DiSTASO, (3) Flora of wounds and flora of putrefaction. Lancet. Lond.. 

1916, I, 74. 1- > , 

Donaldson, R., Character and properties of the 'Reading' bacillus on 
which a new method of treatment of wounds has been based. J. Path, 
and Bacterial., Cambridge, 1918, 22, 129. 

Donaldson and Joyce, A method of wound treatment by the introduction 
of living cultures of a spore-bearing anaerobe of the proteolytic group. 
Lancet, Lend., 1917, II, 445. 

Douglas, Fleming and Colbbrook, (1) Studies in wound infections; on 
the question of bacterial symbiosis in wound infections. Ibid., 1917, 
I, 604. 

Douglas, Fleming and Colebeook, (2) Studies in wound infections ; on 
the growth of anaerobic bacilli in fluid media under apparently aerobic 
conditions. Ibid., 1917, II, 530. 

Douglas, Fleming and Colebeook, (3) Unpublished report to Med. Res. 
Committee. 

Doyen et Yamonouchi, La flore bact6rienne et le traitement des plaies de 
guerre. Compt. rend. Soc. de hiol. , Par., 1916, 79, 228. 

Dudgeon, Gas gangrene. Lancet, Lond., 1914, II, 1384. 

Dudgeon, Gaednbk and Bawtree, On the bacterial flora of wounds pro- 
duced during the present war. Ibid., 1915, I, 1222. 

DuENSOHMANN, Etude experimentale sur le charbon symptomatique et ses 
relations avec I'oedfeme malin. Ann. de I'Inst. Pasteur, Par., 1894, 8, 
403. 

DuHAMEL, Gasphlegmone. Deutsche med. Wchnschr., Berl. u. Leipz., 1916, 
42, 1126. 

V. DuNGERN, Ein Fall von Gasphlegmone unter Mitbeteiligung des Bac- 
terium Coli. Miinchen. med. Wchnschr., 1893, 40, 747. 

Dunham, (1) Report of five cases of infection by the Bacillus aerogenes cap- 
sulatus (Welch). Johns Hopkins Hos^. Bull., Bait., 1897, 8, 68. ■ 

Dunham, (2) Observations to determine the motility of the Bacillus aero- 
genes capsulatus under anaerobic conditions. Ibid., 1897, 8, 74. 

DuPBRii, (1) Recherches bacteriologiques sur quelques cas de gangrene 
gazeuse. Presse mid.. Par., 1915, 23, 309. 

DuPBRiB, (2) Recherches bacteriologiques et hematologiques sur quelques 
cas de gangrfene gazeuse. Gaz. hebd. d. sc. med. de Bordeaux, 1915, 36, 89. 

Dupont et BiLLAUDBT, Gangrene gazeuse du membre inf^rieur gauche, 
Amputation. Injection du serum de Leclainche et Valine. Gu6rison. 
Bull, etmern. Sac. de chir. de Par., 1916, 42, 2793. 

Eisbnbbeg, p., Des leucocidines et hemolysines chez les ana^robies. Ann. 
de I'Inst. Pasteur, Par., 1908, 22, 430. 

Elliot and Henry. Infection of haemothorax by anaerobic gas-pro- 
ducing bacilli. Brit. M.J., Lond., 1917, I, 448. 

Emery, W. D'E., Some factors in the pathology of gas gangrene. Lancet, 
Lond., 1916, I, 948. 

Emeys-Robbets and Cowell, Gas gangrene of muscle with special refer- 
ence to its morbid anatomy and histology. J. Path, and Baoteriol., 
Cambridge, 1917, 21, 473. 



153 

VAN Eembngem, (1) Untersuchungen iiber Falle von Fleischvergiftung mit 
Symptomen von Botulismas. Centralbl. f. Bakteriol. [etc. J, Jena, 1896, 

19 443. 
VAN 'Ebmbngem, (2) Ueber einen neuen anaeroben Bacillus und seine 

Beziehungen zum Botulismus. Ztschr. f. Hyg. u. Infektionskrankh., 

Leipz., 1897, 26, 1. 
VAN Ebmbngem, (3) Der Baoillus botulinus und der Botulismus. Handb. 

d. path. Mikro-organismen, von KoUe-Wasseimann. 2^ Aufl., 1912, 4, 

909. 
Ernst, P., Ueber einen gasbildenden Anaeroben im menscblicheu Kbrper 

und seine Beziehungen zur ' Sohaumleber '. Arch. f. path. Atiat. [etc.], 

Berl., 1893, 133, 308. 
EscHEBiCH, T., Die Darmbakterien des Sauglings und ihre Beziehung zur 

Physiologie der Verdauung. Stuttg., 1886, 2 pi. 
Fasiani, G. M., Ricerche batteriologiche sui j^errni anaerobi delle gangrene 

gassose (Nota preventiva). Sperimentale, Firenze, 1918, 72, 37. 
Fauntleeoy, Gas bacillus infection. Ann. Surg., Phila., 1916, 68, 1, 2 pi. 
Feemi u. Bassu, Untersuchunfjen iiber die Anaerobiosis. Centralbl. f. 

Bakteriol [etc.], Jena, 1904, Orig. 35, 563; 714, and 1905, Orig. 38, 

138; 241; 369. 
Fesslee, (1) Die Gasphlegmone, Mimclten. med. Wchnsehr., 1915, 62, 

1025. 
Fbsslee, (2) Weitere Erfahruugen iiber die 'Gasphlegmone. Ibid., 1915, 

62, 1581. 
FiCKBE, Ueber ein Toxin des aus Gasbrandfallen isolierten Bacillus oedematis 

maligni. Med. Klinik, Berl. 1917, 13, 1181. 
FiESSiNGEE, N., (1) La defense leucocytaire dans la plaie de guerre. Arch. 

de med. exper. et d'anat. path., Par., 1916, 27, 270. 
FiESSiNGBE, N., (2) Serotherapie et autovaccinotherapie antigangreneuse. 

PressemM., Par., 1917, 25, 106. 
FiESSiNGBE, N., (3) La biologic de la gangrene gazeuse. Bev. gBn. de cli/n. et 

de therap., Par., 1917, 31, 20; 38. 
FiESSiNGBE et Baeeieu, Action pathogfene de certains bacilles aerobies 

sporules dans les plaies de guerre. Oompt. rend. Soc. de bioL, Par., 1918, 

81, 1052. 
FiESSiNGBE et GouBAULT, BaciUes aerobies sporules dans les plaies de 

guerre. Ibid., 1917, 80, 492. 
Fiessingee et Metbe, Reaction myeloide au cours des infections ana6robies 

des plaies de guerre. Ibid., 1918, 81, 1055. 
Fiessingee et Montaz, Contribution k I'etude des exudats de la plaie de 

guerre : I. Les premieres heures ; II. La periode de d^tersion (lOe heure 

ilOejour). /Sid, 1916, 79, 495. 
Fiessingee et Vignbs, Septicemic k ana^robies, ictfere pl6iochromique et 

siirreiialite h^morragique au cours d'un phlegmon gangreneux de la cuisse. 

Bull, et mem. Soc. med. d. Mp. de Par., 1916, 3^ s. 40, 470 
Fitzgekald, J. G. and Robertson, D. E., 'Report of an" outbreak of 

diphtheric wound infection among returned soldiers. J. Am. M. Ass>, 

Chicago, 1917, 69, 791. 
Fitzgerald, Mabel P., The induction of sporulation in the bacilli belonging 

to the aerogenes capsulatus group. J. Path, and Bacterial., Cambridge, 

1911, 15, 147. ° 

Flechtenmacher, Foudroyanter Gasbrand nach Herniotomie. Wien. Mm. 
Wchnsehr., 1918, 31, 964. 

Fleming, (1) Some notes on the bacteriology of gas gangrene. Lancet, 
Loud., 1915, II, 376. o„ o a e 



153 

Fleming, (2) On the bacteriology of septic wounds. Ibid., 1915, ]I, 638. 
Plesch, M., Verwendung von Trypaflavin als Wundantiseptikuin bei Gas- 

phlegmone. Munclien. med. Wehnschr., 1918, 65, 970. 
FlOeokbn, Beiti-age zur Pathologie und Klinik der Gasphlegmoiie. Beitr. 

zur. klin. Chir., Tubing., 1917, 106, 485. 
FoED and Laweencb, Haemolytic substances in heated milk and in milk 

cultures of Bacterium Welchii. Johns Hopkins Hosp. Bull., Bait., 1917, 

28, 245. ^ > > . 

Feaenkel, Eenst, (l)Beitiage zur Atiologie und Prophylaxe desGasbdems. 

Verofentl. a.d. Geh. d. Mil.-Sanitatswesens, Berl., 1918, Heft 68, 53. 
Feaenkel, Eenst, (2) Zur Entstehung und Verhutung der mensohlichen 

Gasodemkrankungen. Centralbl. f. Bakteriol. [etc.], Jena, 1918, Orig. 

81, 447. 
Feaenkel, (Eenst,) Feankenthal und KOnigsfeld, Zur Aetiologie, 

Pathogenese und Prophylaxe des Gasodems. 2Ied. Klinik, Berl, 1916, 12, 

689; 716. 
Feaenkel, Eugen, (1) tjber die Aetiologie der Gasphlegmonen (Phlegmone 

emphysematosa). Ihid., 1893, 13, 13. 
Feaenkel, Eugen, (2) Ueber Gasphlegmone, Schaumorgane und deren 

Erreger. Ztschr. f. Uyg. u. Infektionshrankh., Leipz., 1902, 40, 73. 
Feaenkel, Eugen, (3) Ueber die Aetiologie und Genese dor Gasphlegmonen, 

Gascysten und der Sohaumoigane des menschlichen Kbrpers. Ergebn. d. 

allg. Path. u. path. Anat, Wiesb., 1904, Jahrg. VIII, 403. 
Feaenkel, Eugen, (4) Ueber Gasgangriin. Miinchen. med. Wehnschr., 1914, 

61, 2217. 
Feaenkel, Eugen, (5) Kritisches fiber Gasgangran. Ibid., 1916, 63, 476. 
Feaenkel, Eugen, (6) Ueber malignes Oedem. Deutsche med. Wehnschr., 

Berl. u. Leipz., 1916, 42, 1405. 
Feaenkel, Eugen, (7) Ueber Gasbrand. Ibid., 1916, 42, 1533. 
Feaenkel, Eugen, (8) Ueber bakteriologische Befunde bei den Gasphleg- 
monen. Ibid,, 1917, 43, 1612. 
Feaenkel, Eugen, (9) tJber die Reinziichtung der Krankheitserreger des 

malignen Odems und Gasbrandes aus infizierten Wunden. Centralbl. f. 

Bakteriol. [etc.], Jena, 1918. Orig. 81, 13. 
Feaenkel, Eugen, (10) Die blutschadigende Wirkung des Fraenkelschen 

Gasbazillus. Deutsche med. Wehnschr., Berl. u. Leipz., 1919, 45, 317. 
Feaenkel, Eug. und Wohlwill, Das Zentralnervensystem bei Gasbrand. 

Ibid., 1918, 44, 508. 
Feaenkel, Eugen, und Zeissler, (10) Die Differenzierung pathogener 

Anaerobier. Miinchen. med. Wehnschr., 1919, 66, 39. 
Feanchini, Achille, Sulla rapiditi di sviluppo della cancrena gassosa. 

Gazz. d. osp., Milano, 1917, 38, 193. 
Feankau, Deummond and Neligan, The successful conservative treatmisnt 

of early g.is gangrene in limbs by the resection of infected muscles. 

J,. Roy. Army Med. Corjys, Loi.d., 1918, 30, 608. 
Feanz, E., Ueber den Einfluss der Witterung auf die Gasbrandinfektion 

der Wunden. Miinchen. mM. Wehnschr., 1916,63, 1830. 
Feanz, R., Ueber die Gasentziindung. Beitr. z. klin. Chir., Tiibing., 1917, 

108. 443. 
Feouin, Sur le microbisme latent des plaies et du tissu cicatriciel des 

blessures de guerre. Compt. rend. Soc. de biol., Par., 1916, 79, 752. 
FeOnd, Kriegchirurgiaehe Erfahrungen bei Gasgangran. Beitr. z. klin. 

Chir., Tubing. Kriegsehir. Hefte, III, 1916, 98, 447. 
FtaTH, Bfeitrag zur Kenntnis der Gasbranderreger. Miinchen. med. 

Wchnsehr.i 1916, 63,1169. 



154 

Gakhtgens, Vergleichende Untersucliungen liber die Eireger des Gas- 
braiids und des malignen Oedems. CentralU. f. Bakteriol. [etc.J, Jena, 

1917, Orig. 80, 166. 

Gaffky, Experimentell erzeugte Septicamie mit Eiicksicht auf progressive 

Virulenz und accommodative Ziichtung. Mitth. a. d. kais. Gesundheitsamte, 

Berl., 1881,1, 80. 
Gaudier, Fiessingbe et Montaz. Importance du terrain dans le d6ter- 

minisme des grands accidents infectieux par les anaerobies et en particulier 

par le ' B. perfringens '. Gomjjf. rend. Soc. de hiol., Par., 1916, 79, 

851. 
V. Gaza, Die operative Behandlung der Gasphlegmone in ihrem progre- 

dienten Stadium. Beilr. z. klin. Chir., Tubing., 1916, 98, 426. _ 
Geeingbe, Ueber Nebennierenveranderungen bei Gasbrand. Wien. klin. 

Wchnschr., 1917, 30, 943. 
Ghon, Gasbrand. Aetiologie und pathologische Anatomic. Wien. klin. 

Wchnschr., 1917, 30,389. 
Ghon u. Sachs, M., (1) Beitrage zur Kenntnis der anaeroben Bakterien 

des Menschen. Zur Aetiologie des Gasbrandes. Gentrdlhl. f. Bakteriol. 

[etc.], Jena, 1903, Orig. 34, 288; 609. 35, 665. 36, 1 ; 178. 
Ghon u. Sachs, M., (2) Beitrage zur Kenntnis der anaeroben Bakterien des 

Menschen. VII, Zur Aetiologie der Schaumorgane. Ibid., 1909, Orig. 

48, 396. 
Glynn, E. E., The relation between Bacillus enteritidis sporogenes of Kleia 

and diarrhoea. Thompson Yates Lab. Hep., Liverpool, 1901, 3, part II, 

131. 
GoADBY, K., (1) An inquiry into the natural history of septic wounds. 

Lancet, Lond., 1916, II, 89; 585; 851. 
GoADBY, K., (2) Bacteriology of septic war wounds. J. Roy. Microseop. 

Soc, Lond., 1917, 269. 
GoADBY, K., (3) The bacterial flora of war wounds ; influence of treatment. 

Brit. M. J., Lond., 1918, I, 581. 
GOEBEL, Ueber den Bacillus der ' Schaumorgane '. CenfraJhl. f. allg. Path. 

u. path. Anat., Jena, 1895, 6, 465. 
GoLLA, F., An analysis of recent tetanus statistics. Lancet, Lond., 1917, 

n, 960. 
GooDALB, A note on anaerobes in soil-infgcted *ounds. Ibid., 1918, 

I, 768. 
GooEMAGTiGH, (1) Sur le fonctionnement de la capsule surrenale humaine 

dans les gangrfenes gazeuses. Compt. rend. Soc. de biol., 'Pa.r., 1918, 81, 14. 
Gooemaghtigh, (2) Contribution k I'etude du fonctionnement de la capsule 

surrenale humaine k I'etat normal et dans les etats infectieux en particulier 

dans les gangrfenes gazeuses. Arch, de med. exp. et d'anat. path.. Par., 

1918, 28, 277. r ' > 

Gould, Alpeed G., A case of malignant oedema. Ann. Sura., Phila., 1903, 
38, 481. " ' 

GovAEETs, Precede d'etude de la topographie microbienne dans les plaies. 

Compt. rend. Soc. de biol., Par., 1917, 80, 606. 
Geassbeegbe Ueber Buttersauregiihrung. Morphologie des Eauschbrand- 

ba^cillus und des Oedembacillus. Arch.f. Hyg., Munchen u. Berl., 1904, 

Geassbeegbe u. Schattbnfeoh, (1) Ueber Buttersauregahrung. Zur 
Moiphologie des beweglichen Buttersaurebacillus. Ibid. 1902 42 219 
^^6^40"" ^'^^™^'^^°^' (2) Ueber Buttersauregahrung. 'ibid., 

Geassbeegbe u. Schatteneeoh, (3) Das Eauschbrandgift. Handb. d. 



155 

Technik und Methodik d. Immunitatsfoischung, von Kraus und Eevaditi. 

Jena, 1908, 1, 161. 
Geoss, Q., (1) Gangrfene gazeuse.. (Documents statistiques.) Bull. Acad, de 

med., Par., 1916, 3^ s., 76, 586. 
Gross, G., (2) A propos de 134 cas de gangrfene gazeuse. Bull, et mem. Soc. 

de chir. de Far., 1917, n.s,, 43, 636. 
Geosso, Weitere Untersucliungen uber die Unteisoheidungsraerkmale 

zwischen Rauschbrand- malignen Oedem- und Bradsotbacillen. Zugleieh 

ein Beitrag zur Aetiologie der Bradsot und des Gebuitsiauschbrandes. 

Gentralhl.f. Bahteriol. [etc.], Jena, 1913, Orig. 70, 156. 
GuEEMONPEEZ, Fe., Gangifene gazeuze pendant la guerre de 1914-1916. 

26 edit., Paris, 1916. 4 vols. 
Habeeland, Gefassbefund bei Gasbrand. Deutsche med. Wehnschr., Leipz., 

1919, 45, 491. 
Hagbmakn, (1) Gasphlegmone. Beitr. z. klin. Chir., Tiibing., 1916, 98, 

617. 
Hagemann, (2) Ueber Gasphlegmone. Deutsche med. Wehnschr,, Berl. u. 

Leipz., 1916, 42, 178. 
Hanasiewicz, Zur Patbogenese des Gasbrandes. Miinohen. med. Wehnschr., 

1916, 63, 1030. 
Haede, E. S., (1) Gangrene gazeuse a B. perfringens. Compt. rend. Soc. 

de biol, Par., 1915, 78, 134. 
Haede, E. S., (2) Milieu de culture pour I'obtention des anaerobies des 

exudats par I'enrichissement en germes. Ibid., 1917, 80, 661. 
Haeeis, J. E. G., Contribution to the biochemistry of anaerobes. VIII. 

The biochemical comparison of microorganisms by quantitative methods. 

J. Path, and Bacterial., Camb., 1919 (see Wolf, and Wolf and Harris). 
Haetlby, (1) Metastatic gas gangrene; five cases, with recovery in three. 

Brit. M. J., Lond., 1917, I, 481. 
Haetlbt, (2) Five cases of metastatic gas gangrene — three followed by 

recovery. J. Boy. Army Med. Corps, Lond., 1918, 31, 172. 
Hatch, W. K., Gas gangrene and tetanus. Brit. M. J., Lond., 1915, 1, 545. 
Hatjsbe, G., tjber Fauhiissbacterien und deren Beziehungen zur Septicamie, 

ein Beitrag zur Morphologic der Spaltpilze. Leipz., 1885, pp. 94, 15 pi. 
Hautefbuille, Bacteriologie des plaies de guerre. Compt. rend. Soc. de 

biol, Par., 1918, 81, 1068. 
Heddaeus, tJber Serumbehandlung des Gasbdems. (Bericht iiber 94 

Falle.) Veroffentl. a. d. Geb. d. Mil.-Sanitatswesens, Berl., 1918, Heft 

68, 77. 
Heidlee, Gefassschuss und Gasbrand. Wien. klin. Wehnschr., 1916, 29, 

254. 
Hbim u. Knore, Anaerobiotische Anreicherung zur Eeinzuchtving des Gas- 

brandbazillus. Miinchen. med. Wehnschr., 1917, 64, 1246. 
Hbitz-Boyee, Hematome et gangrfene gazeuse. Fresse med., Par., 1916, 

24, 394. 
Hempl, Hilda, Some proteolytic anaerobes isolated from septic wounds. 

J. ffyg., Cambridge, 1918-19, 17, 13. 
Hbnet, H., (1) An investigation of the cultural reactions of certain anaerobes 

found in wounds. J. Fath. and Bacteriol, Cambridge, 1916-17, 21, 344, 

11 Pl- 
Heney, H., (2) On some anaerobes found in wounds and their mode of 

action in the tissues. Brit. M. J., Lond., 1917, I, 806. 

Hbetee, C. a., On bacterial processes in the intestinal tract in some cases of 

advanced anaemia, with especial reference to infection with B. aerogenes 

capsulatus (B. welchii). Jo'im: Biol. Chem., N.Y., 1906-7, 2, 1. 



156 

Hksse, W., u. E., Tiber Ziichtung der Bacillen des malignen Odems, Deutsche 

med. Wchnschr., Berl. u. Leipz., 1885, 11, 214. 
Hewlett, R. T., Probable identity of Achalme's bacillus of acute rheumatism 

and Klein's B. enteritidis sporogenes. Tr. Path. Soc , hood., 1901,52, 114. 
Hetde, Ueber Infektionen mit anaeroben Bakterien ; ein Beitrag zur 

Kenntnis anaerober Staphylokokken und des Bacillus fundiiiiformis. 

Beitr. z. klin. Chir., Tiibing., 1910, 68, 642. 
VON HiBLEE, (l)'Beitrage zur Kenntnis der durch anaerobe Spaltpilze 

erzeugten Infektioiiserkrankungen der Tiere und des Menschen sowie 

zur Begriindung einer genauen bakteriologischen und pathologisch- 

anatomisohen Differentialdiagnose dieser Prozesse. Gentralhl. f. Bak- 

teriol. [etc.], Jena, 1899, 25, 513; 593; 631. 
VON HiBLBR, (2) Ueber die Differentialdiagnose der pathogenen Anaerobien. 

Verhandl. d. deutsch. jpath. Gesellsch., 1905, 9, 118. 
VON HiBLEE, (3) Untersuchungen iiber die pathogenen Anaeroben. tJber 
. die anatomischen und histologiachen Veranderungen bei den durch sie 

bedingten lufektionserkrankungen des Menschen sowie der Tiere und 

iiber einige nicht-pathogene Anaerobenarten. Jena, 1908. 
VON HiBLEE, (4) Rauschbrand. Handb. d. path. Mikroorganismen von 

Kolle-Wassermann. 2te Aufl., 1912, 4, 788. 
HiTSCHMAKN u. LiNDENTHAL, (1) Ueber die Gangrene foudroyante. Sitz- 

ungsh. d. k. Akad. d. Wissensch., mat/iemat-naturw- KL, Wien, 1899, 108, 

3. Abt., 67. 3 pi. 
HiTSCHMANN u. LiNDENTHAL, (2) Ein weiterer Beitrag zur Pathologic und 

Aetiplogie der Gangrene foudroyante. Wien. klin. Wchnschr., 1 900, 13, 

1057. 
HixON, The gas bacillus in a brain injury. Brit. M. J., Lond., 1915, 11, 

905. 
HoPKES, Ueber Gasbrand. Deutsche Ztschr.f. Chir., Leipz., 1917, 140, 1. 
Houston and McCloy, The relation of the enterococcus to 'trench fever' 

and allied conditions. Lancet, Lond., 1916, II, 632. 
Hull, The gangrene of war ; gaseous cellulitis or emphysematous gangrene. 

Ibid., 1916,11, 144. 
IvENSjM. H. F., (1) A clinical study of anaerobic wound infection, with an 

analysis of 107 cases of gas gangrene. Brit. M. J., Loud.. 1916, II, 872. 
IvENS M. H. P., (2) A clinical study of anaerobic wouud infection, with an 

analysis of 107 cases of gas gangrene. Froc. Roy. Soc. Med., Lond., 

1916-17, 10, Sect. Surg., 29. > " ■> 

Jablons et Peask, Teclmical note : Prelimiuary communication on the use 

ot a simple me,hum for the early identification of streptococci and certain 

anaerobic bacteria present in wound-. Compt. rend. Sjc. de hiol, Par., 

J-y J- o, 8X, 1072* 
Jacobsohn Euo., Einiges uber Gasphlegmone und Gasgangran. DentscU 

med. W^cArasc/ir., Berl. u. Leipz. 1917 43 685 

^^^^^: W ^■' ^'"^"1 * ^fT- ^"^^^"^^ ^- P^tJ^- Mikro-organismen, von 
Kolle-Wassermann, Ite Aufl., 1903, 2 619 s . 

'z%fz:, Sr^9r7-i?ir2i^ '-''''''''' '' '^ -"-^^^- ^— 

^""n^: ^' ^"^ Ob^^-^y^ations on gas-bacillus infection in France. Swrq 
ffynec, awrf OJsi!., Chicago, 1917, 25 113 ^ 

JuNGANO et DisTAso, Les ana^robies. ' Paris (Masson), 1910 
JuNGLiNO, Otto, Histologische und klinische Bsitkge zur anaeroben 
Wundinfektion. Beit,: z. klin. Chir., Tiibing., 1917 107 411 *^^^''°''^° 



157 

Kamen, Zur Aetiologie dei- Gasphlegmone. Centralbl. f. Bakteriol. [etc.], 

Jena, 1904, Orig. 35, 554 j 686. 
Katzenstein, Die BeriicksichtiguDg der luheiiden Infektion bei der Opera- 
tion Kriegsbeschadigter. Deutsche med. Wch'nschr., Ber]. u. Leipz., 1910, 
43, 1536. 
Kausoh, Ueber die Gasphlegmone. Beitr. z. Bin. CJiir., Tiibing., 1916, 97 

(Kriegschir. Hefle, II) 7. 
Kebrowski, Ueber die Bedingungeu, unter welchen anaerobe Bakterien 
auch bei Gegenwart von Sauerstoff existiren Konnen. Ztschr. f. Hyg. u. 
Tnfektionskr., Leipz., 1895, 20, 358. 
*Kehl, Ueber metastatisclie Gasphlegmone. Deutsche Ztschr. f. Chir., 

Leipz., 1917, 142, 302. 
Kempnek, Weiterer Eeitrag zur Lehre von der Fleischvergiftung. Das 
Antitoxin des Botulismup. Ztschr. f. Hyg. u. Infehtionskrcmlch., Leipz., 
1897, 26, 481. 
Kkeey, Ueber die Zersetzung des Eiweisses durch die Bacillen des malignen 
Oedems. Sitzungsh. d. k. Akad. d. Wissensch., math.-naturw. CI., Wien, 
1889, 98, Abt., 3. 445. 
Kebet, E., u. Fbankbl, S., (1) tJber die Eiuwirkung der Baoillen des 
malignen Odems auf Kohlehydrate. Ihid., 1890, 69, Abt. 2, 
b.,-297. 
Keery, R., u. Feankel, S., (2) tJber die Einwiikung der Bacilltn des ma- 
lignen Odems auf Kohlehydrate und Milchsaure. Hid., 1891, 100, Abt. 2, 
b., 304. 
KiTASATO, (1) tJber den Eauschbrandbacillus und sein Kulturverfahreii. 

Ztschr. f. Hyg. u. Infektionskrankh., Leipz., 1889, 6, 105. 
KiTASATO, (2) Ueber den Tetanusbacillus. Ibid., 1889, 7, 225. 
KiTASATO, (3) tJber das Wachsthum des Rauschbiandbacillus in festen 

Nahrsubstraten. Ibid., 1890, 8, 55. 
KiTT, Th., Eauschbrand. Handb. d. path. Mikro-organismen von Kolle- 

Wassermann. 1*6 Aufl., 1903, 2, 601. 
Klein, E., (1) Ein neuer Bacillus des malignen Oedems. Centralbl. f. 

Bakteriol [etc.], Jena, 1891, 10, 186. 
Klein, E.. (2) Ueber einen pathogenen anaeroben Darmbacillus, Bacillus 

enteritidis sporogenes. Ibid., 1895, 18, 737. 
Klein, (3) Eepoit on an inquiiy into the relations of Asiatic cholera and 
cholera nostras, or English cholera. Sep. Med. Off. Local Gov. Bd., Lond , 
1895-96, 25, 173. 
Klein, E., (4) Report on the morphology and biology of Bacillus enteritidis 
sporogenes ; on association of this microbe with infantile diarrhoea and 
with cholera nostras, and on its relations with milk, with sewage, and 
with manure. Ibid., 1897-8, 27, 210; 312. 11 pi. Tia,ns. Centralbl. 
f. Bakteriol. [etc.], Jena, 1. Abt. 1897, 22, 113 ; 577 : 23, 542. 
Klein, (5) Further report on the Bacillus enteritidis sporogenes. Sep. 

Med. Off. Local. Gov. Bd., Lond., 1898-99, 28, 312. 
Klein, E., (6) Zur Kenntniss und Differentialdiagnose einiger Auaerobier. 

Centralbl. f. Bakteriol. [etc.], Jena, 1901, 29, 991. 
Klosb, (1) Ueber Toxin- und Antitoxinversuche mit dem Friinkelschen 

Gasbrandbazillus. Miinchen. med. Wchmschr., 1916, 63, 723. 
Klose, (2) Bakteriologische und serologische Untersuchungen mit dem 
Frankelschen Gasbrandbacillus. Ztschr. f. Hyg. u. Infektionskrankh., 
Leipz., 1916, 82, 197. 
Klose, (3) Toxin und Antitoxinversuche mit einem zur Gruppe der 
Gasoedembazillen gehorenden Anaeroben. Miinchen. med. Wchnsclw., 
1917, 64, 154L 
Klose, (4) Bakteriologisch-serologische Grundlagen zur Frage der Herstel- 



158 

lung eines Gasbi'andserums. Verofentl. a. d. Geh. d. Mil.-Sanitatmesem, 
Berl., 1918, Heft 68, 17. 
Klose, (5) Zur Frage der Blutinfektion mit Gas-Oedem-Bazillen bei der 
Gas-Oedem-Erkrankung. Ztschr. /. Hyg. u. Infektionskrankh., Leipz., 

1918, 85, 223. 
Klose, (6) Die anaerobe Mischinfektion bei der Gasoedemerkrankung. 

Med. Klinik, Berl., 1918, 14, 33. 
Klose, (7) Bakteriologische und serologische Untersuchungen rait einem 

zur Gruppe der Gasbdembazillen gehbl-enden Anaeroben. Ztschr. f. Hyg. 

u. Infektionskrankh; Leipz., 1918, 86, 213. 
Klose, (8) Ueber die Bezieburigen in der Aetiologie der menschlichen 

Gasbdemerkrankung und des tierischen Eauschbrandes. Munohen. med. 

Wchnsclw., 1919, 66, 66. 
Klose, (9) Bakteriologisch-serologische Grundlagen zur^Bekampfang und 

Behandlung der Gas-Oedem-Erkrankungen mittelsteines polyvalenten 

Gas-Oedem-Serums. Ihidi, 1919, 66, 266. 
Klose, (10) Der Rauschbrand und verwandte Erkrankungen der Thiere. 

Berl. klin. Wchnschr., 1919, 56, 292. 
Klose, (11) Die Herstellung und Prilfung der Gasodemserums. Ibid., 1919, 

56, 517. 
Klotz and Holman, Infection by the gas bi^cillus in coal mines. J. Inf. 

Dis., Chicago, 1911, 9, 251. 
Koch, Zur Aetiologie des Milzbrandes. Mitth. a. d. k. GesimdheitsanUe, 

Berl., 1881, 1, 49. 
KoLACZEK, Ueber larviert verlaufenden Gasbrand bei Schussverletzuugen, 

Beitr. z. klin. Chir., Tubing., 1916, 103 {Kriegsehir. Hefte, VI), 202. 
KoLLB, RiTZ u. ScHLOSSBBEGEE, Untersuchungen tiber die Biologie der 

Bakteriender Gasbdemgruppe. Med. Klinik, Berl., 1918, 14,281; 854. 
KoLLB, Sachs u. Geoegi. Serologische und serotherapeutische Studien 

bei Gasoedem. Deutsche med, Wchnschr., Berl. u. Leipz., 1918,44, 257. 
KOEENTCHEVSKT, VV., (1) Contribution k I'etude biologique du 'B. 

perfriugens ' et du 'B. putrificus '. Ann. de V Inst. Pastewr, Par., 1909, 

23, 91. 
KOEENTCHEVSKY, W., (2) Experimentelle Beitrage znr Lehre von dergastro- 

intestinalen Autolntoxikation. Centralbl. f. Bakteriol. fetcl 1911, Orig. 

59, 526. 
KOVES, Rauschbrand- und Bradsot-ahnliche Krankheit der Schweine. 

Ihid., 1917, 80, 40. 
Db Keuif, Adams and Iebland, The toxin of Bacillus Welchii. I. Toxin 

production by various strains. J. Inf. Bis., Chicago, 1917, 21, 580. 
De Keuip and Bollman, The toxin of Bacillus Welchii. II. The mechanism 

of infection witli B. Welchii. Ihid., 1917, 21, 588. 
KiJMMBLL, Wundinfektionen, iiisbesondere Wuudstarrkrampf und Gas- 
brand. Verhiitung durch primare Wundversorguue. Beitr. z. klin. Chir., 

Tiibiiig., 1915, 96, 421. 
KiJESTEiNBE, Beitrage zur Unteisuchungstechnik obligat anaerober Bak- 

terien, sowie zur Lehre von der Anaerobiose uberhaupt. Gentrodbl.f 

Bakteriol. [etc.], Jena, Abt. 2, 1907, 19, 1 ; 97 ; 202 • 385 
Lachowicz und Nbncki, Die Anaerobiosefrage. Arch, f d. ges. Physiol, 

1884, 33, 1. Nencki, Bemerkungen zu vorstehenden Abhandlung. 

Ibid., 1884, 33, 10. * 

Laidlaw, Some simple anaerobic methods. Brit. M.J., Lond. 1915 T 497. 
Landau, Hans, Untersuchungen tiber Gasbrand- und RauschbrandbazillcDi 

mit besonderer Beriicksichtigung ihres serologischen Verhaltens und ihrer 

Veianderhchkeit. GentralU. f Bakteriol. [etc.], Jena, 1917, Orig. 79, 



159 

Lardbnnois et Badmel, (1) Les infections malignes des plaies de guerre 

par microbes ana^robies. Les processus turadfiants, gangreneux et gazeux 

Presse med., Par., 1916, 24, 506. 
Laedennois et Baumbl, (2) Les infections gangreneuses des plaies de guerre 

par germes anaerobies. Comp. rend. Acad. d. sc, Par., 1916, 163, 616. 
Laedennois et Baumel, (3) Les formes anatomiques des infections gan- 
greneuses des membres consecutives aux plaies de guerre. Arch, de med. 

et depharm. mil., Par., 1918, 70, 1. 
Lautenschlagbe, L., Toxikologische Untersuchungen des M. Fickerschen 

Gasodemtoxins und seines spezifischen Antitoxins. Arch. f. exp. Path. 
■ u. Pharmahol., Leipz., 1919, 85, 1. 
*Laval, Les infections gangrtoo-gazeuses. Bull, mid , Par., 1915 29, 121 ■ 

165. 
DE Lavekgne, Gangrene gazeuse ; presence frequente du vibrion septique 

et du B. hellonensis. Compt. rend: Sos. de biol., Par., 1918, 81, 705. 
Lawen und Hesse. "Weitere Beitrage iiber Bakterienbefunde bei frischen 

Kriegsschusswuuden, besondeis bei Lungeiiverletzungen (anaerobe 

Pleura-infektion). Deutsche Ztschr. f. Chir., Leipz., 1918, 144, 330. 
LecSne et Feouin, (1) Nouvelles recherches demontrant la r(5alite du 

microbisme latent dans les plaies de guerre cicatrisees. Compt, rend. 

Acad. d. sc, Par., 1916, 162, 722. 
LECiiNB et Feouin, (2) Kecherches experimentales sur le mecanisrae de 

I'enkystement des corps etrangers et du microbisme latent. Ibid., 1916, 

162, 798. 
Leclainchb etlMoEBL, La serotherai)ie de la septicemie' gangreneuse. Ann. 

de VInst. Pasteur, Par., 1901, 15, 1. 
Leclainchb et Vallee, (1) B,eoherolies experimentales sur le charbon 

symptomatique. Ihid., 1900, 14, 202; 613. 
Leclainchb et Vall^ie, (2) Etude compar^e du vibrion septique et de la 

bacterie du charbon symptomatique. Ibid., 1900, 14, 590. 
Leclainchb et Vallee, (3) Traitement serique specifique des plaies. 

Bull, et mSm. Soc. de chir. de Par., 1916, n.s. 42, 1804; 'trans. J. Gomp. 

Path, and Therap., Edinb. and Lond., 1916, 29, 283. 
Leclainchb et Vallee, (4) Le traitement serique specifique des plaies et 

des infections consecutives. Presse med., Par., 1917, 25, 187. 
*Legeos, G., (1) Recherches bacteriologiques sur les gangrenes gazeuses 

aigues. Paris, 1902. Eef Gentralhl. f. Bakteriol. [etc.], Jena, 1903, 

Ref. 33, 525. 
Legeos, G., (2) Recherches histologiques sur les gangrfenes gazeuses aigues. 

Arch, de mid. exper. et d'anat. path., 1903, 15, 1. 
Legeos, G., (3) Un cas de gangrfene gazeuse association de B. perfringens et de 

B. oedematiens. Presse med., Par., 1917, 25, 101. 
Legeos, G., (4) Gangrene gazeuse de la region scapulaire avec destruction 

musculaire causee par le B. histolyticus (Weinberg et Seguin). Compt. 

rend. Soc. de bid., Par., 1918, 81, 96. 
Lehmann und Neumann, Atlas und Grundriss der Bakteriologie. 5. Aufl., 
. Munchen, 1912. 
Lehndoefp und Stiefler, Zytologische Blutuntersuchungen bei Gasbrand. 

Wien. klin. Wehnschr., 1917, 30, 1258. 
Leishman and Smallman, An analysis of recent cases of tetanus in the 

British expeditionary force, with special reference to their treatment by 

antitoxin. J. Roy. Army Med. Corps, Lond., 1917, 28, 295. Also: 

Lancet, Lond., 1917, I, 131. 
*Leo, Un cas de gangrene gazeuse k vibrion septique avec septicemie 

streptocoque. Paris chirurg., 1916, 7, 511. 



160 

LssNii et Phocas, Presence de microorganismes vivaiits et virulents k la 

surface de projectiles inclus dans des tissus cicatrises. Compt. rend. Acad 

d. sc, Far., 1916, 163, 174. 
Leuchs, J., Bacillus botulinus. Handb. d. path. Mikroorganismen von 

Kolle-Wassermanu, 2*6 Aufl., 1912, 4, 939. 
Lbvaditi et Dbleez, (1) Sur I'origine cutan6e des streptocoquea adapt^s 

dans les plaies de guerre. Compt. rend. Acad. d. so., Par., 1917, 165, 

444. 
Levaditi et Dbleez, (2) Le role de la flore cutanee dans la containinaticm 

des plaies de guerre. Ambulance de 'I'Ocian ', 1917, 1, 277. 
Liivy, CoTTB et Lataeget, Un cas de septicemie k Bacillus perfringens 

compliquant une blessure de guerre. Faris niM., 1915, 5« anuee, 311. 
Lett, Fouecade et Bollack, Sur la frequence du Bacillus perfringens dans 

les plaies do guerre. Gompt. rend. Soc. de biol., Par., 1915, 78, 284. 
LiBOBius, P., Beitrage zur Kenntnis des Sauerstoffbediirfnisses der Bac- 

terien. Ztsohr. f. Hyg. u. Infektionsh'ankh., Leipz., 1886, 1, 115. 
LiCHTENSTEiN, Stefanib, Neuere Untersuchungen liber den Erreger des 

Gasbrandes und malignen Odems. Med. Klinik., Berl., 1919, 15, 121. 
LiEBLEiN, Kriegschirurgische Erfabrungen uber Gasbacilleninfektionen. 

Wien. Mm. Wchnsclw., 1917, 30, 199. 
LttFFLBR, F., Zur Immunitatsfrage. Mitth. a. d. k. Gesundheits'amte, Beri., 

1881,1,134. 
Loeis-Melikov, (1) Etudes des spores de B. perfringens. Compt. rend. Soc, 

de UoL, Par., 1909, 67, 806 
Loeis-Melikov, (2) Les anaerobies dans la fievre typhoide.* Ann. de I'Inst. 

Pasteur, Par., 1913, 27, 541. 
LtJDEKiTZ, Zur Kenntnis der anaeroben Bacterien. Ztschr. f. Hyg. u. In- 

fektionskrankh., Leipz., 1889, 5, 141. 
LuMiiiRE, A., Sur les tetanos post-seriques. Ann. de I'Inst. Pasteur, Par., 

1917, 31, 19. 
McGampbbll, The toxic and antigenic properties of Bacterium Welchii. 

J. Inf. Dis., Chicago, 1909, 6, 537. 
McGlannan, The use of Bull's serum in the treatment of wounds infected 

with the Bacillus aerogenes capsulatus welchii. Surg., Gynec, and Obst, 

Chicago, 1918, 26, 336. 
MoIntosh andFiLDES, (1) A new apparatus for the isolation and cultivation 

of anaerobic micro-organisms. Lancet, Lond., 1916, 1, 768. 
McIntosh et FiLDES, (2)Nouvelle m6thode d'isolement'et'de culture pour les 

microbes anaerobies. Gompt. rend. Soc. de biol., Fa,r., 1916 79 293 
McIntosh and Fildes, (3) The classification and study of the knaerobic 

bacteria of war wounds. Med. Research Committee, special report series. 

No. 12, Lond., 1917, pp. 74. ^ 

McLbod and McNee, On the anaemia produced by the injection of the 

haemolysm obtained from streptococci and on the question of natural and 

acquired immunity to streptolysin. J. Path, and Bacterial., Cambridge, 

McNee and Dunn, The method of spread of gas gangrene into living muscle. 

Brit. M. J., Lond., 1917, 1, 727. 
Malonb, E. H a simple apparatus for isolating single organisms. /. Pdth. 

and Bacterial., Cambridge, 1918-19, 22 222 
Malonb and Ehea, Studies on streptococci 'recovered from sick and wounded 

soldiers in France. Ibid., 1918-19, 22, 210 
Maekobe Vergleichende bakteriologische und serologische Studien titer 

Kauschbrand und Pseudorauschbrand. CentralU. f. Bakteriol. retc.l 

Jena, 1911, Grig. 60, 188. ^ L I' 



ibi 

Maequhabdt, Zwei Falle von Gaspblegmone. Miinchen. med. Wchnschr., 

1916, 63, 142. 
Maequis, Les infections anaerobies des plaies de guerre. Ann. d'hva.. Par. 
1916,46 8.26,65. ^^ . • 

Maewedel, (1) Ueber ofifene und rubende Gasinfektion. Deutsche med. 

Wchnschr., Berl. u. Leipz., 1917, 43, 771 ; 803; 836. 
Maewedel, (2) Ueber die Infektionen von Sohusswunden iiach Beobach- 
tungen an Verwundeten des belgisch-friinzbsischen Kriegssobauplatzes 
1914-1917. Beitr. z. klin. Chir., Tubing. , 1918, 113, 433. 
Maewedel u. Wehesig, Ueber Gasbraud durch anaerobe Streptokokken. 

Munchen. med. Wchnschr., 1915, 62, 1023. 
Matti, Der Gasbrand. Deutsche med. Wchnschr., Berl. u. Leipz., 1916, 

42, 43. 
Medical Eeseaech Committee, B,eports of the Anaerobe Committee. 
No. 1. The demonstration of anaerobes in wounds of recent date. Lond., 
1918,. 15 pp., 8°. 
Meneeeul, Gangrfene gazeuse prodiiite par le vibrion septique. Ann. de 

VInst. Pasteur, Par., 1 895, 9, 529. 
Mbtchnikoep, Etudes sur la flore intestinale. Ibid., 1908, 22, 929. 
Meyee, K. F., The etiology of 'symptomatic anthrax' in swine, 'specific 

gas phlegmon of hogs '. J. Inf. Dis., Chicago, 1915, 17, 458. 
Migula, System der Bakterien. Handbuch der Morphologie, Entwicklungs- 

geschichte und Systematik der Bakterien. Jena, 1900. Bd. 2. 
MoEGAN and Vilvandee. The X-ray diagnosis of gas in the tissues. Brit. 

Jif. J,, Lond., 1917, I, 8, 1 pi. 
MoEGBNEOTH u. BiELlNG, Ueber experimentelle Chemotherapie der Gas- 

brandinfektion. Berl. klin. Wchnschr., 1917, 54, 723. 
MoEO, E., (1) Morphologische und biologische Untersuchungen fiber die 
Darmbakterien des Sauglings. I. Die Bakterienflora des normalen 
Frauenmilchstuhles. Jahrh.f. Kinderh., Jena, 1905, 61, 687. 
MoEO, E., (2) Morphologische und biologische Untersuchungen fiber die 
Darmbakterien des Sauglings. II. Die Vertheilung und die Sohicksale 
der normalen Bakterien im Sauglingsdarm. Ibid., 1905, 61, 870. 
Most, Zur Frage der rezidivierenden und ruhenden lufektion bei Kriegs- 

verletzungen. Miinchen. med. Wchnschr., 1915, 62, 1161. 
Mullal-x and McNee, A case of gas gangrene exhibiting unusual proofs 
of a blood infection. Brit. M. J., Lend., 1916, 1, 478. Also /. Roy. Army 
Med. Corf 8, Lond., 1916, 27, 763. 
Muscatello e Gangitano, Eicerche sulla gangrena gassosa. Biforma 

med., Napoli, 1898, 14, pt. 3, 471. . 

. Naociaeone. Flemmone e gangrena gassosa. Contributo batteriologico 

eclinico. iei/twwa msd, Napoli, 1917, 33, 772; 790." 
Nekam, Deux cas d'oedeme malin chez Thomme. Ungar. Arch. f. Med , 

Wiesb, 1892-3,1, 440. -, , -n r 

Nencki, M., (1) Die Anaerobiose und die Gahrungen. Arch. f. exp. Fatfi. u. 

Pharmakol., Leipz., 1886, 21, 299. . 

Nencki, M., (2) Untersuchungen fiber die Zeisetzung des Eiweisses durch 
anaerobe Spaltpilze. Sitzungsber. d. k. Akad. d. Wissensck, Wien, 1889, 

Abt. 2, b. 397. , • o i 

Nbncki, Maeceli, (3) Opera omnia, 1904. Braunschweig. 2 vol. 
NicOLAiEE, A., Ueber infectibsen Tetanus. Deutsche med. Wchnschr., 1884, 

10, 842. , ., . .• i 1 

NicOLLE, M., Cbsaei et Eaphael, Etudes sur le vibrion septique et le 

'Bacterium chauvoei'. Ann. de VInst. Pasteur, Par., 1915, 29, 165. 
Noble, "W., Experimental study of the distribution and habitat of the 

tetanus bacillus. /. Inf. Dis., Chicago, 1915, 16, 132. 

L 



i6» 

Novy, Ein neuei- aiiaerober Bacillus des malignen Oedenls. Ztschr.f. Hyg. 

u. Infehtionshrankh., Leipz., 1894, 17, 209. 
VAN Oettingen, Anaerobie und Symbiose. Ihid., 1903, 43, 463. 
Oliva, La gangrena gassosa delle ferite di guerra ; studio clinico-terapeutico. 

Riforma med., Napoli, 1917, 33, 172; 202; 233; 256. 
Olsen, O., Ueber Vergarung und Saurebildung der Gasbazillen in ihrer 

Beziehung zui- Sporulation. Med. Klinik, Berl., 1917, 13, 99. 
*Ombeedan]>je, L'infection gangreneuse des p]aies de guerre. Paris med., 

1915, 5e annee, 360, 374. 
Omelianskt, Die Zuchtung anaerober Kleiiilebewesen. Handh. d. techn. 

Myhologie, von Lafar, Jena, 1907, 1, 576. 
Obsos, Die Form der tief liegenden Bakterien- und Hefekolonien. Centralhl. 

f. Bahteriol. [etc.], Jena, 1910, Orig. 54, 289. _ _ . 

Oeticoni, Sur la presence du Bacillus perfringens et d'un bacille pyogene 

dans les plaies gazeuses de chiiurgie de guerre. Compt. rend. Soc. de 

hiol., Par., 1915, 78, 126. 
OuEANOFP, Sur rbemotoxine du B. Welcliii (B. perfringens). Ihid., 1917, 

80, 706. 
Owen and Glynn, A case of unusual infection with a gas-forming bacilluR. 

Liverpool M.-Chir. J., 1910, 30, 347. 
Passini, (1) Ueber Giftstoffe in den Kulturen des Gasphlegmonebazillur. 

Wien. Klin. Wchmchr., 1905, 18, 921. 
Passini, (2) Studien fiber faulnisserregende anaerobe Bakterien des normalen 

menschlichen Darmes und ihre Bedeutung. Ztschr. f. Hyg. u. Infekiions- 

hrank., Leipz., 1905, 49, 135. 
Passini, (3) Ueber anaerobiscb wachsende Darmbakterien (mit spezieller 

Berucksichtigung der streng anaerobischen Buttersaurebazillen). Jahrb. 

f. Kinderh., Leipz., 1911, 73, 284. 
Passini, (4) Altera Erfahrungen fiber die Auaerobier des Gasbraudes. 

Wien. klin. Wchnschr., 1917, 30, 203. 
Pastede, (1) Animalcules infusoires vivant sans gaz oxygene libra et deter- 
minant des fermentations. Compt. rend. Acad. d. sc, Par., 1861, 52, 344. 
Pasteue, (2) Experiences et vues nouvelles sur la nature des fermentations. 

Ibid., Par., 1861, 52, 1260. 
Pastetje, (3) Nouvel exemple de fermentation determinee par des animal- 
cules infusoires pouvant vivre sans gaz oxygfene libre et en dehors de tout 

contact avec I'air de I'atmosphfere. Ibid., Par., 1863, 56, 416. 
Pasteue, (4) Eecherches sur la putrefaction. Ibid., Par., 1863, 56, 1189. 
Pasteue et Joubeet, (1) Charbon et septicemie. Bull. Acad, de mid.. Par., 

1877, 2e serie, 6, 781. 
Pasteue et Joubeet, (2) Charbon at f epticemie. Compt. rend. Acad. d. sc, 

Par., 1877, 85, 101. 
Pasteue, Joubeet et Chambbeland, La theorie des germes et ses applica- 
tions a la medecina et la chirurgie. Ibid., 1878, 86, 1037. 
Pa YE, Ueber Gasphlegmone im Kriege. Miimhen. med. Wchnschr., 1915, 

62, 57. 

Penhallow, Latent gas bacillus infection in a healed bullet wound. Lancet, 

Lond., 1916, I, 866. 
Petei, Spontanes Auftreten von malignem Oedeni bei Kaninchen sowie 

einer Septikaemie bei Giinsen, Enten und Huhnern. Centralhl. f. d. med. 

Wissensch., Berl., 1884, 22, 833; 849. 
Peannee, Zur Frage der sogenannten Gasphlegmone. Med. Klinik, Berl., 

1915, 11, 1100. > ' 

Ppeiepee u. Bessau, Ueber bakteriologische Befunde bei den Gasphleg- 

monan Kriegsverletzter. Deutsche med. Wchnschr., Bei-1. u. Leipz., 1917, 

43, 1217; 1255; 1281. l^ > > 



Phocas, Le reveil de rinfection des plaies par I'acte chirurgical. Bull, et 

mim. Soc. de chir. de Par., 1915, n. s. 41, 1933; 1959. 
PiCQuH;, Frequence et prophylaxe de la gangrene gazeuse. J. de mkl. de 

Bordeaux, 1915, Anuee 87, 12. 
"PiEEi, La cangrena d'ospedale (infezione Gangrenosa delle ferite da guerra). 

Biv. osp., lloma, 1915, 5, 312. 
Plaut, H. C, (1) Agglutinatioiisteclmik und Wert der Dunkelfeldbeleuch- 

tung beim Stadium der AnaSrobier. Deutsche med. Wchnschr., Berl. u. 

Leipz., 1917, 43, 302. 
Plaut, H. C, (2) Beitrag zur Geschossuntersuchuiig auf aerobe und 

anaerobe Bakterien. Gentralbl. f. Bakteriol. [etc.], Jena, 1918, Orig. 

81, 41. 
PoLiCAED, Associations microbiennes dans les plaies de guerre en voie de 

cicatrisation. Compt. rend. Soc. de biol., Par., 1916, 79, 273. 
PoLiCAED et Desplas, Tolerance du tissu de bourgeonnement des plaies de 

guerre en voie de cicatrisation pour des corps strangers de dimensions 

microscopiques. Mecanisme du microbisme Intent de certaines cicatrices 

cutanees. Compt. rend. Acad. d. sc. Par., 1917, 164, 249. 
PoLicAED, Desplas et Ph^ilip, Eecherches biologiques sur les plaies de 

guerre. La flore microbienne et ses. rapports avec revolution clinique 

et les caractferes de la blessure. Ibid., 1916, 162, 181. 
PoLiCAED et Phelip, Les premiers stades de revolution des lesions dans 

les blessures par projectiles de guerre. Ibid., 1915, 161, 15. 
Peazmowski, a., Untersuchungen tiber die Entwicklungsgeschichte und 

Fermentwirkung einiger Bakterien-Arten. Leipz., 1880, pp. 55, 2 Taf. 
Peibeam, (1) B. 0., Gasbrand und Anaerobensepsis. Wien. klin. Wchnschr., 

1917, 30, 947. 
Peibeam, B. O., (2) Zur Frage der Gasbrand-Metastasen. Ibid., 1919, 32, 

310. 
Peibeam, H., Der Tetanus in den Kriegsjahren 1914 bis 1915. Med. Klinik, 

Berl., 1917,12, 1094; 1124. 
*E.AMALHAO, A bacteriologia das feridas de guerra. Revista de semiotica 

laboratoridl, Porto, 1.918, 1, fasc. 1, 113, 2 pi. Kef. Bull, de I'Inst. 

Pasteur, Par., 1919, 17, 118. 
Eanpt, Zur Frage der Metastasenbildung bei Gasgangran. Miinchen. med. 

Wchnschr., 1916, 63, 1682. 
Eaphael et Fbasby, Toxine du vibrion septique et antitoxine corre- 

spondante. Compt. rend. Acad. d. sc, Par., 1915, 161, 361. 
PiBiNKE und Beethold, Die Zersetzung der Kartoffel durch Pilze. Unter- 

sueh. aus d. hotan. Laborat. d. Univ. Gottingen, Berl., 1879, Heft 1. 
Rbttgee, (1) Studies on putrefaction. J. Biol. Ghem., N. Y., 1906-7, 2, 

71. 
Ebttgee, (2) Further studies on putrefaction. Ibid., 1908, 4, 45. 
Eevel, De la gangrfene gazeuse. Etude clinique et therapeutique. Bull, et 

mim. Soc. de chir. de Far., 1915, n. s. 41, 1716. 
Eeveechon et Vauohee, Constance et precocite de la presence du bacille 

' perfringens ' dans les lesions de gangrene etc. Compt. rend. Soc. de biol, 

Par., 1915, 78, 145. 
EiCKEE und Hakzbe, Beitrag zur Ken.itnis der odem- und gangranerzeu- 

genden Wirkung anaerober Bazillen bei den Versuchstieren und beim 

Meuschen. Beitr. z. Bin. Chir., Tubing., 1918, 112, 289. 
RiTTBE, Ueber Gasbrand. Beitr. z. klin. Chir., Tubing., 1916, 98 {Knegs- 

chir. Hefte, III), 47. , . ^ j 

HoBEETSON, Mueibl; (1) Notes upon certain anaerobes isolated from wounds. 
J. Path.' and Bacterial., Cambridge, 1915-16, 20, 327. 

L2 



164 

ROBEKTSON, MuEiEL, (2) Notes on the vaccination of guinea-pigs with B. 

perl'ringens. Lancet, Lond., 1916, II, 516. _ _ 

EOBEBTSON, MuBiEL, (3) Notcs on the occurrence of B. tetani in wounds. 

Tr. Soc. Trop. M.and Hyg., Lond., 1917, 11, 56. 
EoBBETSON, MuEiBt, (4) Notss on vibrion septique. Brit. M. J., Lond., 

1918, I, 583. 
RoccHi, Giuseppe, Sulla gangrene gassosa. Riforma rned., ^apoll, 1917, 

33, 387. .. . , o.. ,• . ,, 

Rodella, a., (1) Uber anaerobe Bakterieu im norinalen Sauglingsstuhle. 

Ztschr.f. Hyg. u. Infektionskrankk, Leipz., 1902, 39, 201. 
Rodella, A., (2) Sur la differenciation du ' Bacillus putrificus ' (Bienstock) et 
des bacilles anaerobies tryptobutyriques (Achalme). Ann. de I'Inst. Pas- 
teur, Par., 1905, 19, 804. 
Rodella, A., (3) Ueber die Bedeutung der streng anaeroben Faulnis- 
bacillen fiir die Kasereifung. Centralhl. f. Bakteriol. [etc.], Jena, 2*6 Abi 
1906, 16, 52. 
RoGEK, Gr. H., Quelques efFets des associations microbiennes. Gompt. rend. 

Soc. de biol, Par., 1889, 41, 35. 
Rosenthal, G., (1) Le serum anti-perfringens, le Wright- vaccin anti-per- 
■fringens, dans la medication des infections- graves anaerobies. Ibid., 
1910,68,1044. , 
Rosenthal, G., (2) A propos du Wriglit-vaccin antiperfringens. IMd.. 

1915,78,365. 
Roux, (1) Sur la cultuie des microbes anaerobies. Ann. de I'Inst. Pasteur, 

Par., 1887, 1, 49. 
Roux, (2) Immunity contre le charbon symptomatique conferee par des sub- 
stances solubles. Ibid., 1888, 2, 49. 
RoTJX et Chambeeland, Immunite contre la septicemic conferee par des 

substances solubles. Ibii/., 1887, 1, 561. 
Rumpel, Bericht iiber die praktischen Erfahrungen mit Serumbehandlung 
der Gasphlegmone. Veroffentl. a. d. Geb. d. Mil.-Sanitatswesens, Berl., 
1918, Heft 68, 35. 
Rumpel, Klose, von Wasseemann, etc., Gasodemschutz und -bekampfucg. 

Beitr. z. Min. Chir., Tubing., 1918, 113, 76. 
*RuNEBBEG, B., Studien iiber die bei peritonealen Infektionen appendicu- 
laren Ursprungs vorkommenden Sauerstofftoleranten, sowie obligat 
anaeroben Bakterienformen, mit besonderer Beriicksichtigung ibrer 
Bedevitung fiir die Pathogenese derartiger Peiitonitiden. Arb. a. d. path. 
Inst. d. Univ. Ilelsingfors, Berl., 1908, 1, 271. 
Rupp, Ueber einen Fall von Gasgangran mit Metastasenbildung. Miinchen. 

med. Wchnschr., 1916, 63, 919. 
Sacquepee, (1) Le bacille de I'oedeme gazeux malin. Compt. rend. Soc, de 

biol. Par., 191,5, 78, 316; 540; 588. 
Sacquepee, (2) A propos d'une note de M Weinberg et P. Seguin infcitulee : 

Le B. oedematiens et la gangrfene gazeuse. Ibid., 1915, 78, 547. 
Sacquepee, (3) Une modalite de la gangrene gazeuse; Toedfeme gazeux 
malin et son agent pathogene. Bull, et mem. Soc. de chir. de Par., 1915, 
n.s. 41, 1132. 

Same title. Presse med., Par., 1915, 23, 183. 
Sacquepee, (4) Sur la gangrene gazeuse. Septicemic gazeuse et oedfeme 

gazeux malin. Ibid., 1915, 23, 218. 
Sacquepee, (5) La septicemic gazeuse et I'ced^me sazeux malin. Ibid.. 

-1915, 23, 227. 
Sacquepee, (6) Demonstration exp^rimentale des lesions des gangrenes 
gazeuses. Ibid., 1915, 23, 242 



165 

Sacquepee, (7) Sur la gangrene gazeuse. Bull, et mem. Soc. med. d. Mn. 

de Par., 1915, 3e s., 39, 965. 
Sacquepee, (8) Sur le bacille de I'oeclfeme malin. Gompt. rend. Soc. de bioL 

Par., 1916, 79, ,11 5. 
Sacquepee, (9) Etudes sur la gangrene gazeuse; le bacille de I'asdfeme 

malin. Ann. de I'Inst. Pasteur, Par., 1916, 30, 76. 
Sacquepee, (10) Recherches sur la gangrene gazeuse des plaies de guerre. 

Presse med., Par., 1916, 24, 194. 
Sacquepee, (11) Sur le Bacillus bellonensis (ancien Bacille de I'oedfeme 

gazeux malin). Preparation de serums specifiques ; quelques propii^tes 

essentidles des serums. Gompt. rend. Soc. de bid., Par., 1917, 80, 850. 
Saoquep^ie, (12) Preparation d'un serum theiapeutique iinti-bellonensis. 

Presse med.. Par., 1918, 26, 17. 
Sacquepee, (13) Principes dune methode de trjiitement serolherapique 

antitoxique de la gangrfeue gazeuse. Ibid., 1918, 26, 18. 
Sacquepee, (14) Recherches sur la gangrfene gazeuse des plaies de guerre. 

lUd., 1918, 26, 197. 
Sacquepee, (15) Traitemeut serotherapique de la gangrfeue gazeuse. Gompt. 

rend. Gonfir. chir. interalliee p. I' etude d. plaies de guerre. Par., 

1918, 18. 
Sacquepee, (16) Etudes bactetiologiques sur les plaies de guerre (Blessures 

osteo-musculaires des membres). J . de physiol. et de path, gen., Par., 1918, 

17, 621. 
Sacquepee, (17) La flore initiale habituelle et la Acre de passage dans la 

gangrene gazeuse. Gompt. rend. Soc. de biol., Par., 1918, 81, 526. 
Sacquepee, (18) Quelques precedes d'isolement des bacteries pathogenes. 

Signification pathog^nique des resultats dans la gangrene gazeuse. Ibid., 

1918, 81,529. 
Sacquepee, (19) Sur le traitemeut preventif et curatif de la gangrene 

gazeuse. Arch, de med. et de pharm. mil., Par., 1918, 70, 201. 
Sacquepee, de Laveegne et Dbhoenb, Infections k, Proteus dans les plaies 

de guerre. Gompt. rend. Soc. de biol., Par., 1918, 81, 944. 
Salus, G., Zur Biologie der Eaulnis. Arch. /. Hyg., Munchen u. Berl., 

1904, 51, 97. 
Sanselicb, Untersuchungen uber anaerobe Mikro-organismen. Ztschr.f. 

Hyg. u. Infektionshrankh., Leipz., 1893, 14, 339. 
Saktoet et Spillmann. Sur la bacteriologie de la gangrfene gazeuse. 

Gmnpt. rend. Acad. d. sc, Par., 1915, 160, 210. 
ScHATTENFEOH und Geassbeeger, (1) Ueber Buttersauregahrung. Arch. 

f. Hyg., Munchen u. Berl., 1900, 37, 54. 
SciiATTENPEOH u. Geassbeegbe, (2) Ueber Buttersaurebacillen und ihre 

Beziehungen zu der Gasphlegmone. Miinehen. med. Wchnschr., 1900, 47, 

1033. ,. , 

ScHATTENFEOH u. Geassbeegee, (3) Die Beziehungen der unbeweglichen 

Buttersaurebacillen zur Eauschbrandaffection. Ibid., 1900, 47, 1733. 
ScHLOSSBBEGBE, Die Differenzierung der anaeroben Gasodembakterien. 

/6i(i., 1919,66, 348. ' _ ,^. „^ , , 

SCHMID, Tod durch Gasembolie bei Gasphlegmone. Wten. klin. Wchnschr., 

1915,28, 1317. ^ , . , „^, , 

*ScHOCK, Drei Falle von Gasbdem beim Pferde. Berl. tierarxtl. Wchnschr., 

1918, 34, 392. . , . , . , , 

SCHOLTZ, Ueber das Wachsthum anaerober Baktenen bei ungehmderten 

Luftzutritt. Ztschr.f. Hyg. u. Infektionskr., Leipz., 1898, 27, 132. 
SchOnbaube, Ein Beitrag zur Frage der Anaeiobeusepsis beim Gasbrand. 

Wien. klin. Wchnschr., 1917, 30, 110. 



166 

ScHULTZE, W. H., Zur Kenntnis der pathogenen Bedeutung des Bacillus 

phlegmones emphyBematosae. Arch. f. path. Anat. etc., Berl., 1908, 193, 

419. 
ScHtTZB, H., Iodine and sodium hypochlorite as wound disinfectants. Brit. 

Jlf.J,, Lond., 1915, II, 921. 
Segen, Ein Beitrag zur Prophylaxe der Gasphlegmone und des Tetanus. 

Wien. klin. Wchnschr., 1917, 30, 16. 
Sbhet, Der Gasbiand. Med. Klinik, Berl., 1916, 12, 1056 ; 1081. 
EB Sejoue, Contribution a I'etude de la septic6mie gazeuse. Bull, et mem. 

Soc. de chir. de Far., 1917, n. s., 43, 325. 
Seltee, Zur Aetiologie der Gasphlegmone. Deutsche med. Wchnschr., Bar]. 

u. Leipz., 1915, 41, 1189. 
Senechal, Etude histologique sur un cas de gangrene gazeuse au debut ; 

considerations therapeutiques. Progres med.. Par., 1916, 3^ s., 31, 

137. 
Sheen and Klein, B. G., A case of war wound infection by an anthracoid 

bacillus, the wound being caused by a shattered trigger-guard. Lancet, 

Lond., 1915,1, 1292. 
SiLBEESCHMiDT, Bakteriologischcs uber einige Falle von ' Gangrene fou- 

droyante ', von Phlegmone und von Tetanus beim Menschen. Ztschr. f. 

Hyg. u. Infektionskrankh., Leipz., 1902, 41; 427. 
Silbbestein, Gasbrandtoxine und Antitoxine. Wien. klin. Wchnschr., 1917, 

30, 1642. 
SiMONDS, J. P., (1) Studies in Bacillus Welchii with special reference to 

classification and to its relation to diarrhoea. Monog. Rockefeller Inst. 

M. Research, K Y., 1915, No. 5, 130 pp. 
SiMONDS, J. P., (2) Classification of the Bacillus Welchii group of bacteria. 

J. Inf. Bis., Chicago, 1915, 16, 31. 
SiMONDS, J. P., (3) A propos des effets de I'oxygene sur le Bacillus per- 

fringens. Compt. rend. Soc. de hiol., Par, 1916, 79, 904. 
SiMONDS, J. P., (4) Studies in gas gangrene. Tr. Chicago Path. Soc, 1916^ 

17, 10, 167. 
SiMONDS, J. P., (5) Eecherches sur la presence de spores de bacteries 

anaerobies dans les uniformes militaires (armee beke). Presse m^d.. 

Par., 1917, 25, 100. ' . 

SiMONDS, J. P., (6) Studies on Bacillus Welchii with special reference to gas 

gangrene. J. Exper. M., N. Y., 1917, 25, 819. 
Smillie, W. G., New anaerobic methods. J. Exp. M., N. Y., 1917, 26,59. 
Smith, J. P., The bacteria of gangrenous wounds. Brit. M. J., Lond 1916 

I, 184. ' 

Smith Theobaid, Ueber einige Kulturmerkmale des Eauschbrandbazillus. 
Ztschr. f Infektionskrankh. . . . d. Haustiere, Berl., 1905 1 26, 

Smith (Theobald), Beown and Walkee, The fermentation tube in the 
study of anaerobic bacteria, with special reference to gas production and 
the^use of milk as a culture medium. J. Med. Research, Bost., 1905, 14, 

Spallanzani, L, Opusculi di fisica animale e vegetabile. Modena, 1776, 
Vol. I, Chap. VII, 117. 

^'f^TS' ^''''''=^^™"^ ^"d Gasbrand. Miinchen. med. Wchnschr., 1915, 

STmBE (1) Die Differentialdiagnose des Gasbrandes. I. Die putride 

1918 143 350 " ' Behandlung. Deutsche Ztschr. f CAzV., Leipz., 

Stemmlee, (2) Die Diiferentialdiagnoee des Gasbrandes. Ibid., 1918, 144, 



167 

Stewaet (Maey) and West, E., Study of a strain of B. Welchii isolated in 

J; ranee, together with some notes on gastric ulcers. J. Immunol., Bait. 

and Cambridge, Eng., 1916, 1, 189. 
Stewaet, Matthew J. A study of the coliform organisms infecting the 

wounds of war. J^. iT^/gr., Cambridge, 1917, 16, 291. 
Stoddard, J. L., (1) Points in technic of separating anaerobes. J. Am. M 

Ass., Chicago, 1918, 70, 906. 
Stoddakd, J. L., (2) Bacillus egens. A new pathogenic anaerobe. J. Exper. 

Jf., N. Y., 1919, 29, 187. ■' 

Stoddard, J. L., (3) B. multifermentans tenalbus. Zanc«<, Lond., 1919,1. 12. 
Stokes, A., (1) A note on some cases of blood infection by an anaerobic 

organism secondary to wounds. J. Roy. Army Med. Corps, Lond.. 1915 

24, 379. ^ ^ J • &■/ • 

Stokes, A., (2) Some cases of blood infection by an anaerobic organism 

secondary to wounds. Lancet, Lond., 1915, I, 746. 
Stokes, A., (3) A note on some cases of blood infection by an anaerobic 

organism secondary to wounds. J. Soy Army Mfd. Corps, Lond. 1916 

27, 361. 
Stokes and Tytler, Eeport on the bacteriology of wounds. Brit. J. Surg., 

Bristol, 1918, 6, 111. 
Stolz, a., Die Gasphlegmone des Menschen. Beitr. z. klin. Chir., Tubing., 

1902, 33, 72. 
Straub, W., Toxikologische Untersuchungen des M. Ficlserschen Gasodem- 

toxins ijnd Antitoxins. Miinchen. med. Wchnschr., 1919, 66, 89. 
Sturgbs and Eettgee, Methods for the isolation and cultivation of Bacillus 

putrificus and other obligate anaerobes. J. Bacterial., Bait, and Cam- 
bridge, 1919, 4, 171. 
SuDECK, Ueber Gasphlegmone. Deutsche med. Wchnschr., Berl. u. Leipz., 

1915, 41, 483. 

Swan and Goadby, On the recrudescence of local sepsis in completely healed 
wounds as the result of some surgical interference or passive movement. 
Brit. M. J., Lond., 1915, II, 741. 

Swan, Jones and McNee, The occurrence of acute emphysematous gan- 
grene (malignant oedema) in wounds received during the war. Lancet, 
Loud., 1914, II, 1161. 

Tavbl, Ueber den Pseudotetanusbacillus des Darmes. Centralbl. f. Bak- 
teriol. [etc.], Jena, 1898, 23, 539. 

Taylor, H. D., and Austin, The action of antiseptics on the toxin of 
Bacillus "VV«lchii. J. Exper. M., N. Y., 1918, 27, 375. 

Taylor, K., (1) Note on a case of self-inoculation with the Bacillus aerogenes 
capsulatus. Lancet, Lend., 1915, II, 977. 

Taylor, K., (2) Tissue fragments and wound infection. Ann. Surg., Phila., 

1916, 64, 641. 

Taylor K., (3) Observations on the pathology and bacteriology of gas gan- 
grene. J. Path, and Bacteriol., Cambridge, 1916, 20, 384. 

Taylor, K., (4) La gangrene gazeuse. Evolution et traitement. Arch, de 
med. et de pharm. mil., Par., 1916, 65, 838. 

Taylor, K., (5) Note sur deux cas mortels de gangrfene gazeuse metasta- 
tique. Ibid., 1916, 66, 353. 

Taylor, K., (6) Factors responsible for gaseous gangrene. Lancet, Lond., 
1916, I, 123. 

Taylor, K., (7) Two fatal cases of, metastatic gas gangrene. Ibid., 1916, 
11, 1057. 

Taylor, ,K., (8) Gas gangrene —its course and treatment. Tr. Amer. Surg. 
Ass., Phila., 1916, 34, 128. 



168 

Taylob. K., (9) Specificity in antiseptics. Lancet, Lond., 1917, I, 294. 
Thibs, Ueber zwei Hauptfoimen der Gasinfektion. Beitr. z. klin. Glnr., 

Tubing., 1918, 109, 157, 3 pi. . , . • 

TAN TiBGHBM, Identite du Bacillus amylobactei- et du Vibnon butyrique de 

M. Pasteur. Compt. rend. Acad. d. sc, Par., 1879, 89, 5. 
*TiETZE und KoEBSCH, Ueber Gasphlegmone. Deutsche med. Wchnsch., 

1914, 40, 2004.' 
Tjesiee, (1) Recherches sur la flore intestinale des nournssons (etat normal 

et pathologique). Thfese de Par., 1900, 253 pp. 
TissiEK, (2) Repartition des microbes dans I'intestin du noumsson. Ann 

de I'Inst. Pasteur, Par., 1905, 19, 109. 
TiSSiEE, (3) Recherches sur la flore bact^rienne des plaies de guerre. Bull. 

Aca^de med.. Par., 1916, 76, 337. 
TissiEE, (4) Recherches sur la flore bacterienne des plaies de guerre. Ann. 

de rinst. Pasteur, Par., 1916, 30, 681. 
TissiEE, (5) Recherches sur la flore bact^rieune des plaies de guerre. Ibid., 

1917, 31, 161. 
TissiER, (6) Application de donnees bacteriologiques k la chirurgie de 

guerre. Bull, de I'Inst. Pastev/r, Par., 1918, 16, 273. 
TissiBR et Martblly, Recherches sur la putrefaction de la viande de 

boucherie. Ann. de I'Inst. Pasteur, Par., 1902, 16, 865. 
TizzoNi, Cattani, and Baquis, Bakteriologische Untersuchungen fiber den 

Tetanus. Beitr. z. path. Anat. u. z. ally. Path., Jena, 1890, 7, 568. 
Tbectjl, Production de plantules amylifferes dans les cellules .vegetales 

pendant la putrefaction; chlorophylle cristallisee. Compt. rend. Aead.d. 

sc, Par., 1865, 61, 432. 
Trifaud, De la gangrfene gazeuse foudroyante. Mev. de chir., Par., 1883, 

3, 776; 850; 958. 
Teillat, Des conditions de transport des germes morbides par les vetements 

et les objets. Bull. Acad, de med.. Par., 1915, 73, 609. 
TuLLOCH, W. J., (l)On the bacteriology of wound infections in cases of tetanus 

and the identification of Bacillus tetani by serological reactions. J. Roy. 

Army Med. Corps, Lond., 1917, 29, 631. 
Ttjlloch, W. J., (2) The part played by concomitant infection with anaerobic 

organisms (other than B. tetani) in the causation of tetanus. Brit. M. J., 

Lond., 1918, I, 614. 
TuLi.oCH, "W. J., (3) The isolation and serological differentiation of B. tetani. 

Proe. Roy. Soc, Lond., 1919, B. 90, 145. 
TuLLOOH, W. J., (4) The distribution of the serological types of B. tetani 

in wounds of men who received prophylactic inoculation and a study of 

the mechanism of infection in and immunity from tetanus. Ibid., 1919, 

B. 90, 529. 
(JppENHEiMEE, Ein neuer gaserregender Bacillus. (B. aerogeues aerophilus 

agilis, nov. spec.) Beitr. z. path. Anat. u. z. alia. Path., Jena, 1902, 31, 

383. 
Unteestjchtjngen fiber Serumschutz bei Gasodem, zusammengestellt iiu 

Sanitats-Departement des K. Preuss. Kriegsministeriums. Veraffenil. 

a. d. Geb. d. Mil.-Sanitatswesens, Berl., 1918, Heft 68, 140 pp. See 

Asohoff, Klose, Rumpel, v. "Wassermann, Ernst Frankel, and Heddaeus. 
Vaillaed et RouGET. Note au sujet de I'etiologie du tetanos. Ann. de 

I'Inst. Pasteur, Par., 1893, 7, 755. 
Vauchee, (1) Le B. oedematiens dans un,cas de gaugrfene gazeuse. Compt. 

rend. Soc. de biol. Par., 1917, 80, 251. 
Vauchee, (2) Essais de terotherapie preventive antigangreneuse. Ibid. 

1917,80,956. ^ ^ 



169 

Vaucher et GuEEiN, Action du Bacillus sporogeiies sur quelques hydiates 
de carbone et sur matiferes proteiques des milieux de culture. Ibid., 
1918, 81, 362. 
Vkillon, Utilisation de la bacteriologie dans la chirurgie de guerre. Aroh. 

de med. et de pharm. mil., Par., 1918, 69, 15. 
Veillon et ZuBEE, Eecherches sur quelques microbes stiictemeut aiiaerobies 
et leur r61e en pathologie. Arch, de med. exper. et d'anat. path.. Par., 

1898, 10, 517. 
Vincent et Stodel, (1) Influence du traumatisme sur la gangrfene gazeuse 

exp^rimeutale et sur le reveil de cette infection. Compt. rend. Acad. d. 

so.. Par., 1917, 164, 870. 
Vincent et Stodel, (2) Sur un serum preventif et curatif de la gangrene 

gazeuse. Ibid., 1918, 167, 137. 
Vincent et Stodel, (3) Sur la serothempie autigangreneuse par un serum 

multivalent. Ibid., 1918, 167, 245. 
VoGEL, Ueber den Nachweis von Gasbraudbazillen im Blut bei einem 

Fall von Gasbrandmetastase. Milnchen. med. Wchnsehr., 1917, 64, 

294. 
VoGT, K. E., Ueber die Metastaseubildung bei Gasinfektion. Beitr. z. klin. 

Chir., Tiibing., 1918, 109, 203. 
VoTTBLER, Utber die Differentialdiagnose der pathogenen Anaeroben durch 

die Cultur auf Schiagagar nnd durch ihre Geisseln. Ztschr. f. Hyg. u. 

Infektionsh:, 1898, 27, 480. 
Wallace, C, (1) The colour changes seen in skin and muscle in gas gan- 
grene. Brit. M. J., Lond., 1917, I, 725, 4 pi. 
Wallace, C, (2) Gas gangrene as seen at the casualty clearing stations. 

/. Roy. Army Med. Corps, Lond., 1917, 28, 528. 
v. Wasseemann, (1) Experimentell-theiapeutisclie Studien aus der Gruppe 

der Gasbraiiderreger. Med. Klinik, Berl., 1916, 12, 437. 
Y. Wasseemann, (2) Bemerkungen zur Frage der Herstellung eines Gasodem- 

Serums. Veroffentl. a. d. Geb. d. Mil.-Sanitdtsivesens, Berl., 1918, Heft 

68, 45. 
AVedeehake, Ueber Bekampfung und Behandlung anaerober Wundinfek- 

tion. Deutsche med. Wchnsehr., Berl. u. Leipz., 1917, 43, 1135. 
Weinbeeg, (1) Recherches bacteriologiques sur la gangrene gazeuse. Gompt. 

rend. Soo. de bid.. Par., 1914, 77, 506. 
Weinberg, (2) Premiers essais de .vaccinotherapie des infections gazeuses. 

Ibid., 1914, 77, 543. 
Weinbeeg, (3) Un cas de ganarene gazeuse k vibrion septique. Ibid., 1915, 

78, 141. 
Weinberg, (4) Sur la bacteriologie de la gangrene gazeuse. Ibid., 1915, 

78, 286. 
Weinbeeg, (5) Recherches sur la gaugiene gazeuse. C'ornpt. rend. Acad. d. 

sc.,Par., 1915, 160, 325. 
Weinbeeg, (6) Bacteriological and experimental researches on gas gangrene. 

Froc. Boy. Soc. Med., Lond., 1915-16, IX (Occas. Lect.), 119. 
Weinberg, (7) Gas gangrene in the present war. Glasgow M. J., 1916, 

85 241. 
AVeinbeeg, (8) Le B. oedematiens et le bacille de I'oedfeme gazeux nialin 

(Sacquep^e) sont deux especes differentes. Cumpt. rend. Soc.de biol, Par., 

1916, 79, 174. . 

Weinbeeg et Seguin, (1) Notes bacteriologiques sur les infections gazeuses 

76ici., 1915, 78, 274. 
Weinbeeg et Seguin, (2) J^e B. wdelnatiens et la gangrene gazeuse. Jbid., 

1915, 78, 507. 



170 

Weinberg et Seguin, (3) Reponse k M. Sacquep^ie. Ibid., 1915, 78, 550. 
Weinbeeg et Seguin, (4) Du serum anti-oedematiens. Ibid., 1915, 78, 

552. 
Weinberg et Seguin, (5) Floie microbienne de la gangrene gazeuse. Le B. 

fallax. Ibid., 1915, 78, 68.JB. 
Weinberg et Seguin, (6) Deux cas de gaugrfene gaeeuse consecutifs 'a la 

ligature des gros vaisseaux. Ibid., 1915, 78, 736, 
Weinberg et Seguin, (7) Un vibrion septique k aspect atypique en g^lose 

profonde, etc. /6i(?., 1915, 78, 738. 
Weinberg et Seguin, (8) Formes pseudo-graves d'infections gazeuses. Ibid., 

1916, 79, 116. 
Weinberg et Seguin, (9) Le B. fallax et la gangrfene gazeuse. Ibid., 1916, 

79, 581. 
Weinberg et Seguin, (10) Bacillus sporogenes des plaies de guerre. Ibid., 

1916, 79, 1028. 
Weinberg et Seguin, (11) Eeproduction exp6rimentale des formes putrides 

de la gangrene gazeuse. Ibid., 1916, 79, 1136. 
Weinberg et Seguin, (12) Contribution a I'^tiologie de la gangrfeue gazeuse. 

Gompt. rend. Acad. d. sc. Par,, 1916, 163, 449, 
Weinberg et Seguin, (13) Etudes sur la gangrene gazeuse. Ann. de V Inst. 

Faxteur, Par., 1917, 31, 442. 
Weinberg et Seguin, (14) Essais de serotherapie de la gangrene gazeuse 

chez I'homme. Compt. rend. Acad. d. So., Par., 1917, 165, 199. 
Weinberg et SiiGUiN, (15) Demonstiatiou des lesions provoquees chez le 

cobaye par le B. histolyticus. Quelques observations sur la toxine de 

ce microbe. Gompt. rend. Soc. de biol., Par., 1917, 80, 157. 
Weinberg et Seguin, (16) Quelques documents sur la preparation de la 

toxine et de I'antitoxine du vibiion septique. Ibid., 1917, 80, 715. 
Weinberg et Seguin, (17) Evolution des idees sur les infections gazeuses. 

Conceptions d'avant la guerre. Bull, de I'Inst. Pasteur, Par., 1918, 16, 

1;41. 
Weinberg et Seguin, (18) La gangrfene gazeuse. Paris, Masson, 1918, 

pp. 444, 8 pi. 
Weinberg et Seguin, (19) Un cas de gangifene gazeuse a B. cedematiens 

cliez uncivil. Gompt. rend. Soc. de biol., 'P»,r., lQ\i, 8\, 1?,^. 
Weissenbach, Un bon caractere differentiel entre le Streptocoque pyogfene 

et I'Enterocoque : resultat de Tensemencement en eau pepton^e glucosee 

k la bile. Ibid., 1918, 81, 559. 
Welch, Morbid conditions caused by Bacillus aerogenes capsulatus. Johns 

Hophins Hosp. Bull., Bait., 1900, 11, 185. 
Welch and Flexner, Observations concerning the Bacillus aerogenes cap- 
sulatus. J. Exper. M., N. Y., 1896, 1, 5. 
Welch wnd Nuttall, A gas-producing bacillus (Bacillus aerogenes capsu- 
latus, nov. spec), capable of rapid development in the blood-vessels 

after death. Johus HopUns Hasp. Bull., Bait., 1892, 3 81 
VON Werdt, F., ( 1 ) Malignes Oedem. Handb. d. path. Mikroorganismen von 

Kolle-Wassermann, 2*^ Aufl., 1912, 4 837 
VON Werdt, F., (2) Der Gasbrand und seine Erreger. Handjj. d. path. 

Mikroorganismen von Kolle-WaEserinann, 2te Aufl 1912 4 878 
Werner, G. Die Agglutination bei Gasphlegraonebazillen. ' ArcKtHuq. u. 

Jnfe/cttonskrankh., Munchen u. Berl., 1905 53 128 
^SrTsgf 'l£''7 ^^^^' ^°" Oasgangran. Arch.'f. path. Anat. (etc.), 

WiETiNG Die Pathogenese und Klinik der Gasbazilleninfektion (G.RL). 
J->eutsche Ztschr. f. Ghir., Leipz., 1917,14,1 1. 



171 

Wilson W. James The plugocytic response to the introduction of bacteria 

into clean wounds. Brit. M. J., Lond., 1918, I, 533 
Wilson, W. James, and Stebb, Points in the technique employed in the 

isolation and cultivation of anaerobic bacteria. lUd., 1918 II 568 
\VraoGEADSKY, Clostridium pastoriunum, seine Morphologic und seine 

^igenschaften als Buttersaureferment. Centralhl.f. Bakteriol. retcl, Jena 

2te Abt. 1902, 9, 43; 107. L J. > 

^v'^'r^l ^' ^''^^^ Contributions to the biochemistry of pathogenic anaerobes. 

y._ The biochemistry of vibrion septique. J. Path, and Bacteriol.. Cam- 
bridge, 1918, 22, 115. 
Wolf, C. G. L., (2) Contribution to the biochemistry of anaerobes. VI. 

The proteolytic action of Bacillus sporogenes (Metchnikoff) and Bacillus 

welchii. Ibid., 1919, 22, 270. 
WOLJ?, C. G. L., (3) Contribution to the biochemistry of anaerobes. VII. 

The biochemistry of Bacillus proteus. /6?c?., 1919, 22, 289. 
Wolf, C. G. L., (4) Contribution to the biochemistry of anaerobes. IX. 

Biochemistry of Bacillus oedematiens. Ibid., 1919, unpublished. 
Wolf and Harris, (1) Contributions to the biochemistry of pathogenic 

anaerobes. I. The biochemistry of Bacillus welchii and Bncillus sporo- 
genes (Metchnikoff). Ibid., 1916-17, 21,386. 
Wolf and Harris, (2) Contributions to the biochemistry of pathogenic 

anaerobes. III. The effect of acids on the growth of Bacillus welchii 

(B. perfringeus) and Bacillus sporogenes (Metchnikoff). Bio-Chem. J., 

Liverp., 1916-17, 11, 213. 
Wolf and Harris, (3) The conditions of growth of Bacillus welchii in the 

presence of oxygen. Lancet, Lond., 1917, II, 787. 
Wolf and Harris, (4) Contributions to the biochemistry of pathogenic 

anaerobes. IV. The biochemistry of Bacillus histolyticus. J. Path, and 

Bacteriol., Cambridge, 1918, 22, 1. 
Wolf and Telfee, Contributions to the biochemistry of pathogenic 

anaerobes. II. The acid production of Bacillus welchii (B. perfringens) 

and Bacillus sporogenes (Metchnikoff). Bio-Ohem. J., Liverp., 1916-17, 

11, 197. 
WoLFSOHN, G., Bericht uber 100 Falle von Gasbdem. Beitr. z. klin. Chir., 

Tubing., 1918, 112, 560. 
Weight, A. E„ (1) Conditions which govern the growth of the bacillus of 

' gas gangrene ' in artificial culture media, etc. Lancet, Lond., 1917, 1, 1 ; 

also Proc. Roy.Soc. Med., Lond., 1916-17, 10, Occas. Lect. 1-32. 
Weight, A. E., (2) Des conditions qui gouvernentledeveloppementdubaoillede 

la ' gangrene gazeuse ' dans les milieux artificiels de culture dans le serum 

sanguin ' in vitro ', dans le cadavre et dans I'animal vivant. Compt. rend. 

Confer, chir. inleralliee p. T etude d. plaies de guerre. Par., 1918, 22. 
Weight and Fleming, Acidaemia in gas gangrene and on the conditions 

which favour the growth of its infective agent in the blood-fluids. Lancet, 

Lond., 1918, I, 205. 
WuRCKER, Ueber Anaerobiose, zwei Faulniserreger und Bacillus botulinus. 

Sitzungsber. d. physikalisch-nied. Sozietdt in Erlangen, 1909, 41, 209. 
AVtTTH, O., Die Konstitution des Oedemlysins. Milnchen. med. Wchnschr., 

1919, 66, 175. 
Zacheel, Zur Differentialdiagnose der Gasbranderieger. Wien. klin. Wchn- 
schr., 1917, 30, 517; 904. 
Zbisslee, (1) ZurZtichtung des Bacillus phlegmonis emphysematosae Eugen 
Frankel. Deutsche med. Wchnschr., Berl. u. Leipz., 1917, 43, 878. 



172 

Zeisslee, (2) Ueber die Eeinzuchtung pathogenei- Anaerobier. Ztschr. f. 

Hyq. u. Infektionskrankh., Leipz., 1918, 86, 52. , „, . 

Zeisslee, J., (3) Der Rauschbrand und verwandte Erkrankungen der Thiere. 

Berl. kiin' WchmcJir., 1919, 56, 107 ; i88. ,, ,. , ^ 

ZiNDEL, (1) Worauf beruht der Unterschied in der Afortalitat des C^as- 

biandes im Frieden uiid Kriege? Munchen. med. Wchnschr., 1916, 63, 

Zindel' (2) Die neueren Arbeiten uber Gasphlegmone. Beitr. z. klin. Chir., -\ 
Tlibing., 1917, 105 (Kriegschir. Hefte, VII) 257. 



ADDENDUM 

Beebe, Un cas de gangrene gazeuse toxique a B. perfringens. Compt. 

rend. Soc. de biol, Far., 1919, 82, 992. 
Chutbo, Traitement s(!iotherapique curatif de la gangrfene gazeuse. Bull. 

et mem. Soc. de chir. de Par., 1918, 4.4:, 1527. 
DoNGES u. Elfbldt, Beitrage zum Befunde von Diphtheriebazillen in 

"Wunden. Deutsche med. Wchnschr., Berl. u. Leipz., 1919, 45, 545. 
Dtjval et Vauchee, Essai de sdrotherapie preventive antigangreneuse. 

Bull, et mem. Soc. de chir. de Par., 1918, 44, 1 187. 
Duval et Vauchee, A propos de la seroth^rapie de la gangrfene gazeuse. 

Ihid., 1918, 44, 1387. 
Duval et Vauchee, Premiers resultats des essais systematiques de sero- 

tli6rapie preventive antigangreneuse. Ibid., 1918, 44, 1535. 
*Geoegi, W., Ueber Saureagglutination bei Erregern der Gas6dems, Arh. a. 

d. Inst.f. exp. Therapie, Frank, a. M., 1919. Heft 7. 
Klosb, F., Zur Frage der Toxinbildung von Gas-Oedem-Bazillen. Gentralhl. 

f. Bakteriol, [etc.], Jena, 1919, Orig. 83, 306. 
KoLLE, Sachs u. Georgi Experimentelle Untersuchungen iiber die Wirkung 

des Gasbdemseruras. Ztschr. f. Hyg. u. Infektionskr., Leipz., 1918, 86, 

113. 
Maequis, Courboules [et al.], La serotherapie antigangreneuse par les 

injections de serum antibellonensis et antivibrion septique. Bull, et 

mem. Soc. de chir. de Par., 1918, 44, 1522. 
Nevin, Maey, Experimental gas gangrene ; the protection by antiserum 

and antiserum mixtures. J. Inf. Bis., Chicago, 1919, 25, 178. 
NiGST, P. F., Ueber die putriden Phlegmoneu (Gasphlegmonen) des Krieges. 

Got: Bl.f. Schweiz. Aerzte, Basel, 1918, 48, 438; 479 ; 508. 
*IliTZ u. "ScHLOSSBEEGEE, Ueber die Wirkung chemischer Mittel auf 

Gasbrandbakterien in vitro und in vivo, Arh. a. d. Inst. f. exp. Therapie, 

Frankf. a. M., 1919. Heft. 7. 
EouviLLOis, GuiLLAUME-Louis [et al.], Traitement de la gangrtoe gazeuse 

par I'emploi des serums antigangreneux (m^thode de Saoquepee). BvM. 

et mem. Soc. de chir. de Par., 1918, 44, 1226. 
*ScHLossBBEGEE, H. , Die Hamotoxine der Gasbrandbakterien. Arb. a. d. 

Inst. f. exjj. Therapie, Frank, a. 51., 1919, Heft 6 and 7. 



173 



APPENDIX 

THE HISTOPATHOLOGY OF GAS GANGRENE 

By E. H, Kettle, M.D. Lond. 

A SYSTEMATIC study of the histopathology of any infection should 
proceed along certain definite lines, and sliould include investiga- 
tions into 

1. The distribution and dissemination of the organisms in the 
body. 

2. The changes taking place at the site of infection. 

3. The lesions produced in the other tissues and organs of the 
body. 

A. From the action of circulating toxins formed by the organisms. 

B. As the result of the presence of the organisms. 

4. Reactions taking place in various organs in the form of — 

A. Attempts to combat the infection. 

B. Changes accompanying the process of healing and repair. 
Although the anaerobic bacilli do not lend themselves very readily 

to investigation along these lines, I have adopted this scheme as 
a working basis since it is desirable to bring these organisms into 
comparison with other pathogenic microbes. 

For several reasons, the material derived from cases of gas- 
gangrene in the human being are often unsatisfactory. The lesions 
due to the presence of anaerobic bacilli in the body are essentially 
degenerative, and it may be impossible to distinguish them from 
similar changes due to other causes; and another source of con- 
fusion is the jpost-inortem degeneration which must elapse between 
the death of the patient and the autopsy. Further, there is frequently 
a coexisting infection with other organisms which again complicates 
the analysis of the histological findings. The human material to 
which I had access suflPers in a varying degree from these defects, and, 
in addition, the greater part of it was quite uncontrolled by clinical 
or bacteriological data. 

The lack of satisfactory human material may be compensated for 
tor some extent by the use of laboratory animals. Rabbits, guinea- 
pigs, rats and mice, can all be infected by subcutaneous or intra- 
muscular inoculation, but it is not easy to reproduce the disease as 
it occurs in man. Either the dose is too small and the inoculatiouH 
fail to take efiect, or the local lesion is so severe that the animals 
die rapidly from toxaemia before the disease becomes generalized. 
Similarly, I have found that intravenous inoculation with cultures- 



174 

of the bacilli produce death without the formation of localized 
lesions. Of the experimental material which had been placed at 
my disposal, very little was of value, and, under these circumstances, 
this investigation is incomplete in many particulars. In respect to 
section 4 of the scheme my observations are insuflBcient to admit 
of conclusions being drawn, and I shall therefore omit any dis- 
cussion of this part of the problem. 

1. The Distribution and Dissemination of the organisms in the body. 

The severity and extent of the lesions at the site of inoculation 
have so dominated the clinical picture that gas-gangrene has come 
to be looked upon as a local disease of muscle. This, liowever, 
is not correct. The organisms are not confined to the neighbourhood 
of the wound but are distributed widely throughout the body, 
producing lesions which are identical with those occurring in muscle. 
It must be recognized, of course, that the bacilli may proliferate 
extensively in the body after death, and that their distribution in 
tissues obtained at autopsy is not always representative of their 
ante-mortem dissemination. But when due allowance is made for 
this, an examination of human and experimental material shows 
that septicaemia is of common occurrence. 

In human material, organisms have been found in association 
with definite lesions in the heart muscle, the liver, the kidney, the 
spleen, the lymph glands, the supra-renal glands, and the meninges 
of the brain. Apart from lesions, I have found organisms widely 
distributed in veins and capillaries. In many instances this dis- 
semination must be regarded as a post-mortem phenomenon, but 
cases remain in which the only possible explanation is a terminal 
septicaemia. 

In animals, probably because of the rapidity with which they 
succumb to toxaemia, septicaemia is uncommon. I have, however, 
been able to demonstrate organisms in the circulating blood and in 
the bone marrow of animals killed at varying times after intra- 
muscular inoculation. 

Dissemination of the organisms occurs in three ways : 

(a) By direct extension in loose areolar and connective tissue. 

(b) By growth along lymphatic vessels. 

(c) By invasion of the blood-stream. 

(a) Dissemination by direct extension in loose areolar and con- 
nective^ tissue. This is probably the most important mode of 
dissemmation since it is the way in which infection spreads from 
the primary wound till the whole limbs and even the trunk become 
involved. The muscles are only of importance in defining the path 
ot infection masmuch as the areas enclosed by the epimysium and its 
prolongations, the perimysium, form, for all practical purposes, 
v?"^?i- i''' ':'^f".°el3 la which the organisms can advance. The 
bacilli flourish in loose areolar tissue and tend to spread along 
tissue spaces and connective tissue planes with great rapidity. 

{b) Dissemination by groivth along lymiAatic vessels. The bacilli 
grow very readily within the lymphatic vessels as distinct from the 



175 

tissue lymph spaces, a condition of permeation occurring which 
IS very similar to that seen in the spread of a malignant growth 

Lj-mphatic permeation by the bacilli can be demonstrated in tlie 
neighbourhood of the site of inoculation in animals, and I have also 
tound It taking place in human material. A particularly strildno- 
instance was seen in a deep cervical gland of a cbild who died o'f 
gas-gangrene of the arm. Areas of the gland showed early localized 
degenerative changes, and a main afferent lymphatic vessel was 
plugged with a thick column of bacilli. 

(c) Dissemination by invasion of the bloodstream. This is not 
so easy to demonstrate. Occasionally, in the local lesion, a few 
organisms can be detected in the capillaries or veins, but this 
invasion of the blood-stream is quite overshadowed by the luxuriant 
growth of the organisms in the tissues and the lymphatics. Never- 
theless, it would appear that visceral lesions are due to a blood-borne 
infection. Bacilli may frequently be demonstrated in the sinusoids 
of the liver, even though they may be absent in other oi-gans ; and 
I have found them in considerable numbers in the capillaries of the 
glomeruli. I have never, however, found bacilli in the lungs under 
conditions in which post-mortem change could be excluded ; and 
I have never seen pulmonary lesions which could be ascribed 
to them. 

2. The changes taking place at the site of infection. 

The changes in muscle following infection with anaerobic bacilli 
have been fully described, and it is unnecessary to deal at any 
length with this aspect of the questiou. It is important, however, 
to realize that the infection of muscles is purely an accidental 
phenomenon. Anaerobic bacilli have no specific action on muscles, 
nor do they find in them any substances which are especially 
necessary to their metabolism. It is in deep penetrating wounds 
that the organisms thrive best, and in the limbs such wounds involve 
muscle. Wounds of the thoracic or abdominal viscera are fatal, or 
receive operative treatment, from considerations othei' than the 
occurrence of gas-gangrene, but were these organs of less immediate 
importance to the life of the individual there is no reason to suppose 
that the incidence of gas-gangrene would not be as high in them as 
it is in wounds of muscle. 

The characteristic lesion of gas-gangrene is necrosis brought 
about by the toxins elaborated. by the organisms. The changes are 
most obvious in the muscle fibres which pass through the stages of 
cloudy swelling, and gradual loss of striation, to coagulation necrosis 
and solution. The most highly organized structures, the muscles, 
nerves, and the epidermis and its derivatives are affected earliest, 
but eventually the blood-vessels and the connective tissues are also 
destroyed. The organisms, which are usually present in enormous 
numbers, lie in the connective tissues, and do not invade the other 
structures except, perhaps when necrosis is far advanced. In an 
infected muscle they tend to remain within the limits of the 
epimysium ; and they extend to neighbouring muscles by spreading 
in the subcutaneous tissue, o'r the deep connective tissue trabeculae. 



176 

The most striking feature of the lesion is the entire lack of any 
inflammatory reaction. Some proliferation of the sarcolemma nuclei 
can sometimes be seen in front of the advancing margin of the 
infection, but this is all. The muscle fibres are absolutely quiescent, 
and, moreover, there is a complete absence of wandering cells. In 
the subcutaneous and connective tissues, on the other hand, there 
occurs a leucocytosis which may be very pronounced ; the leucocytes 
are of the polymorphonuclear variety, and are actively phagocytic. 
There is little or no reaction of the fixed connective tissue cells. 

The blood-vessels show no very characteristic lesions. In advanced 
cases the constituents of tlieir walls lose their staining reactions 
and undergo necrosis in common with the rest of the tissues, but in 
the early stages they present no constant changes. 

Thrombosis of veins and capillaries frequently occurs but is not 
invariable. I have not been able to detect any alteration in the 
endothelium to account for this, but there is clear evidence of a 
toxic destruction of red blood corpuscles which in itself would tend 
to cause conglutination thrombi. In advanced lesions haemorrhage 
and haemolysis frequently occurs. 

I have not been able to come to definite conclusions as to possible 
diiferences in the lesions produced by the various anaerobic bacilli. 
B. oedematiens and V. sejitique appear to give rise to greater 
oedema and more advanced vascular changes than B. wdchii, and 
haemorrhage and thrombosis appear to be pai-ticularly associated 
with V. septique, but my observations on this point are not 
conclusive. 

3. The lesions produced in the other organs and tissues of the body. 
A . From the action of circulating toxins. 

The finer details of cell degeneration are often obscured by post- 
mortem changes ; experimental material is thus particularly 
necessary for this portion of the investigation. By compai-ing the 
tissues of animals, killed some twenty-four hours after being infected 
with cultures of anaerobic bacillij with human organs obtained at 
autopsy, it is possible to arrive at some idea of the changes which 
usually take place. It was my intention to work also with animals 
which had been inoculated with toxins alone, but I was forced to 
abandon this line of research since most of the material at my 
disposal proved to be infected with pyogenic cocci. 

The Blcod is afiected very considerably by the toxins of the 
anaerobic bacilli. Thrombosis is common, not only in the primary 
locus, but also in the smaller vessels throughout the body, the 
thrombi being of the hyaline, conglutination type. A varying 
degree of haemolysis seems to occur quite early in the infection, for 
the spleen olten contains a large quantity of haemosiderin, the 
granules lying tree in the pulp, or within macrophages and poly- 
nuclear leucocytes. In advanced cases granules ot" pigment may be 
found in quantities in the endothelial cells lining'^capiUaries and 
veins, in the cells of the hver, and in the renal epithelium, especia^Uy 
m the cells of Henle's loops. I have not been able to m^ke any 
observations on the condition of the cells and haemoglobin content 
of the circulating blood during life, and an examination of the bone 



177 



marrow m a few cases has not revealed any process of regeneration. 
In some cases the contents of the vessels in the organs examined 
suggests a slight leucocytosis, but this condition is not constant. 
The organs are frequently congested, but there is nothino- to show 
that this reaction is at all specific. ° 

The Liver. The liver cells in human organs constantly show 
cloudy swelling which may be extreme, and occasionally some fatty 
degeneration. The fat has no distinctive distribution, and neither 
of these changes can be looked upon as specific. 

In experimental animals the hepatic lesions are much more severe. 
Within twenty-four hours of subcutaneous or intramuscular in- 
oculation the liver cells show extreme changes. There is no fatty 
degeneration, but the cytoplasm becomes entirely disorganized and 
appears to undergo partial solution, nothing but an irregular 
granular debris remaining within" an unusually prominent cell wall. 
The nuclei tend to become hypochromatic but are not otherwise 
affected. Similar changes occur in animals inoculated intravenously 
with cultures of bacilli, and in one mouse which lived some forty- 
eight hours after intravenous inoculation there was, in addition, 
advanced coarse fatty degeneration. 

The Kidney. Here again the changes are constant and severe. 
The convoluted tubules are chiefly affected. Fatty degeneration 
never occurs, but the cells undergo advanced cloudy swelling 
and very often disintegrate. The mitochondria become swollen and 
irregular in size, and no longer maintain their normal longitudinal 
arrangement. The cell protoplasm shows an increased affinity for 
acid dyes, and in some of the experimental animals the appearances 
almost suggest poisoning with mercury or uranium, so severe are 
the lesions. There is little change in the nuclei beyond a diminution 
in the amount of chromatin. It is not uncommon to find coagulum 
or granular debris in the tubules, and though there is no definite 
change in the glomeruli, Bowman's capsules sometimes contain 
a little coagulum. The cells of Henle's loops and the collecting 
tubules are only slightly affected. 

'The Spleen. The most constant change in the spleen is due to 
the altered blood condition, and consists of oedema, occasional 
thrombosis in the capillary vessels, and pigmentation. Some of 
the organs have been congested but in others the pulp is empty. 
An increased number of macrophages has been noted in several 
specimens, and sometimes these cells are multinucleated. 

In two specimens there is a very interesting change in the 
Malphigian corpuscles. They are considerably hypertrophied and 
exhibit a central proliferation of endothelial cells, the result being 
a condition which strongly recalls the germinal node of the lymph 
gland. The change appears to be analogous to that which occurs 
in scarlet fever and diphtheria. In some of the nodes the endo- 
thelial cells seem healthy, but in others there is a considerable 
degree of nuclear fragmentation and degeneration of the cytoplasm ; 
and. occasionally poly nuclear leucocytes can also be detected among 
the cells. This condition appears to be a reaction to the toxaemia 
alone, for I have not been able to demonstrate the presence of 
bacilli. 



M 



178 

Dymph glands. These are congested, but d-o not exhibit amy 
reaction comparable with that seen in* the- spleen, 

Musele. Slight fatty degeneration sometimes occurs intheheart 
muscle, but I have found no change in the voluntary muscles apart 
from the presence of bacteria. In one pigeon I found advanced 
fatty degeneration of tlie pectoral muscles into which the toxin of 
5. welchii had been injected some hours previous to death. 

The Supra-renal glands. My material was unsuitable for the< 
demonstration of adrenalin, and I have not detected any specifici 
changes in such glands as I have been able to examine. 

The Thyroid gland. My material is insufficient to allow of 
definite conclusions being drawn, but in one case the parenchyma 
cells show advanced degeneration and desquamation, with an 
almost complete disappearance of the colloid substance. 

B. As the result of the presence of the organism. 

Muscle. I have frequently examined portions of the- voluntary 
muscles from different parts of the body in human and experimental 
cases of gas-gangrene but I have never found in them localized 
lesions, or toxic changes indicating any specific action of the 
anaerobic bacilli. 

In one case a lesion was present in the heart muscles, the muscle 
fibres had lost their striafcion and presented the typical 'ground 
glass ' appearance, and their nuclei no longer stained. Numbers of 
bacilli were lying free among the desquamated fibres. There was 
a complete absence of any cellular exudate. 

Liver. The liver appears to be a favourable site for the develop- 
ment of lesions. My material contains numerous examples, and 
I have found typical lesions in the liver of a child who died forty 
hours after an injury to the arm in which gas-gangrene developed. . 

The organisms reach the liver by the blood-stream, and can be 
seen lymg free in the sinusoids. I have never seen definite bacterial 
emboli, nor are the bacilli ingested by the endothelial or parenchyma 
cells, but isolated organisms appear to be held up mechanically, and 
eventua ly small colonies are developed. In a wide zone arouid 
these CO onies the liver cells are killed. The" process appears to be 
extremely rapid; and of the nature of a coagulation necrosis. At 
the periphery of the lesion the cells may show cloudy swelling, and 
the nuclei may be hypochromatic or pycnotic, but for the most part 
they are peculiarly quiescent. The cells suddenly lose their staining 
power, but their outlmes are retained, and their nuclei can still be 
distinguished and should fatty. degeneration happen to be present, 

lln / f^ ,r?- '*' ^ ^' ""^^'^ ""*• 0"ly i^ t^« centre of the 
«nTLl ' ' disintegrate. Here they undergo partial solution 
dWp^T T^'^'^'^t into irregular masses of copulated and semi- 
S^hl J T-'" ^t""'^^' proliferate extensively in the centre 

them X'^'J^^'^ f^J'"^''^^^ *^" ^^^^ ««"«' but not invading 
contain ?T"*^™'''' ''^ t^^f periphery of the lesion the rinusoids maf 
anv celhJr ™ono and polynuclear leucocytes, but there is never 
devplni ! .t^' '^^"^ *^^ ^"'^°^- I*^ the late stages gas 

of fh^^!;nPPr'''°^'^ under pressure, for the cells at the.pfriphfry 
of the vacuoles are compressed. Most of these gas vacuoles are 



179 

empty, but occasionally they contain cell debris and bacilli ; and 
ba'ciUi are always found in numbers at their margins. 

The bile ducts and supporting tissues of the organs show no 
change unless they are involved in an unusually large focus when 
they undergo degeneration in a similar way. 

The Kidney. Lesions have been found in the kidney in a few 
cases. In all essentials they are exactly similar to the hepatic 
lesions. The presence of colonies of organisms is accompanied by 
widespread degeneration, and, as might be expected, the changes 
are first seen in the cells of the convoluted tubules. As in the 
livei-, gas bubbles develop in late stages. 

The Spleen. Here the lesions are not uncommon and are of the 
usual type, though the appearances are complicated by the vascular 
and toxic changes which have been described. It is noteworthy 
that even here there is never any cellular exudate, and the bacilli 
are never phagocytosed. 

Lymphatic glands. Infection of the lymphatic glands is probably 
much more common than is thought. The organisms reach the 
gland by permeation along the lymph vessels, and give rise to 
the usual quiet necrosis. Here, again, the lack of any reaction 
is very striking. 

The Supra-renal glands. In one case I found a lesion which 
conformed in every respect with the lesions in other organs. 

Nervous System,. The peripheral nerves in the local lesions show 
changes similar to those in the muscles and other tissues. I have 
not found lesions in the central nervous system, but in one case 
there was a condition of cerebral meningitis, apparently due to one 
of the anaerobic bacilli. The lesion, in this case, had the character 
of an ordinary acute meningitis. The membrane was congested 
and covered with a polynuclear leucocytic exudate, and there was 
active phagocytosis of the bacilli by the leucocytes. Many bacilli 
were present in the vessels of the underlying cerebral cortex, but 
the nerve cells showed no change. 



h 2 



180 



DESCRIPTION OF PLATES 

1. B. welchii. Film preparation, 18-liour-old culture on alkaline egg. (Henry, 

Pig. 36). 

2. B. welchii. Film preparation, 3 days' growth on alkaline pieat, showing short 

cocooid forms. (Henry, Fig. 37.) 

3. B. welchii. Film preparation from infected haemotliorax fluid showing strepto- 

bacillary form. (Henry, Fig. 38a.) 

4. B. welchii. Filamentous and chain forms grown on alkaline egg agar for 24 

hours. (Henry, Fig. 38b.) 

5. B. welchii. 24-hour culture on alkaline egg agar to show involution forms. 

(Henry, Pig. 39.) 

6. B. welchii. 2 days' growth on human serum to show spore formation. (Henry, 

Pig. 40.) 

7. B. welchii. 24-hour culture in starch broth to show spore formation. (Original.) 

8. Vibrion sepiigue. 24-48 hour meat culture, x 1500. (Robertson, B. M. J., 

Fig. 1.) 

9. Vibrion septiqm. Noguchi tube culture showing rods, spores, and a 'citron' 

type. X 1500. 

10. Vibrion septique. Post-mortem specimen from muscle of guinea-pig ; citrons 

and club-shaped types are shown, x 1500. 

11. Vibrion septique. From liver of guinea-pig showing 'citron', 'bulb' and oval 

forms. X 1500. 

12. Vibrion septique. From liver of guinea-pig showing long filaments, &e. xlOOO. 

13. Vibrion septique. ' Citrons ' from haemorrhagic blister fluid of patient suffering 

from gas-gangrene, x 1500. 

14. Vibrion septique. 48 hours' growth on serum agar, x 1000. (Mcintosh, Plate I, 

Fig. i). 

15. Vibrion septique. Noguchi culture. X 1000. (Mcintosh, Plate I, Pig. 5.) 

16. Vibrion septique. Smear from peritoneum of a mouse. x 500. (Mcintosh, 

Plate I, Fig. 6.) 

VI , 1&. Vibrion septique. Broth cultures, x 1000. (Mcintosh, Plate I, Figs. 7 and 8.) 

19. B. oedematiens. Egg broth culture. (Henry, Fig. 45a.) 

20. B. oedematiens. From surface colony. (Henry, Fig. 45b.) 

21. B. chauvoei. Four days' broth culture, x 1000. (Mcintosh, Plate II, Pig. 1.) 

22. B. cJmmoei. 24-hour Noguchi tube, x 1000. (Mcintosh, Plate II, Fig. 2.) 

23. B. clmmoei. Smear from muscle of guinea-pig. x 1000. (Mcintosh, Plate II, 

24. B. histolyticus. 16 hours' growth in alkaline meat, x 1500. (Original). 

25. 26. B. botulinus. Broth culture, x 1000. (Mcintosh, Plate III, Figs. 4 and 5.) 

27. B. sporogefks. Noguchi tube culture, 3 days' growth, x 1000. (Mcintosh, 

Plate IV, Fig. 4.) ^ 

27a. B. sporogenes. 48 hours' agar. (Mcintosh, Plate II, Fig. 4). 

28. B. tertius. 5-day-old colony on alkaline egg agar, x 1300. (Henry, Fig. 42.) 

29. B. cochlearius. Broth culture, x 1000. (Mcintosh, Plate II, Fig. 9.) 

30. B.tetanomorphus. 48 hours' growth on agar, x 2000. (Mcintosh, Plate II, Fig. 4.) 

31. B. tertius. (Original.) 

32. B. cochlearius. (Original.) 

33. 34. B. tetanomorphus. 48 hours' growth on agar. (Original.) 

35. B.fallax. Broth culture two days old. x 1300. (Henry, Fig. 43.) 

36. B. aerofetidus. 3 days' culture in alkaline egg broth, x 1300. (Henry, Fig. 44.) 

37. B. bifermentans. 48 hours' growth meat medium, x 1500. (Original.) 

38. B. sphenoides. 48 hours' growth starch broth. (Original.) 



181 

39. B. memi^s. 48 hours' growth potato broth. (Original.) 

40. J ^t^^. Broth culture, x 1000. (Mclnto.h, Kate II, Fig. 8.) 

42 J If. St°"-l----''l'^aline egg agar.' x6. (Hen ;, Pig. ^2.) 

42. B u^emn 48.hour growth colony in agar. (Mcintosh, Plate V, Fig 1 r 

43. ^.^<i.^„fe.„.. 3 days old colonies on egg agar. x6. (Henry, Fig. 34.) 
U. Bmstdytuue. 8 days old colonies on egg agar, x 6. (Henry, Fig 35 ) 
4u. r^hr^on septzque. 24 hours' culture. (Mcintosh, Plate VI, Fig 3 ) 

46. Vibrion septi^. 24 hours' surface colony. (Mcintosh, Plate VII, Fi. 2 ) 

47. B. sporoge,ies. Surface colonies 3 days old. x 6. (Henry, Fi» 28 ) "' 

48. 5. ^-^-- ^^^-^ace^coW : egg agar. Slow-growing coronie's 2 days old. 

49. B. teriius. Surface colonies on egg agar, x 6. (Henry, Fig. 29.) 

50. B./aUax. 3 days old colonies : egg agar, x 6. (Henry, Pig. 30.) 

51. B. aerofilidus. 3 days old colonies : egg agar, x 6. (Heni-y, Fig. 33 ) 

52. B. wdchii. 24 hours' growth on serum agar slope. (McIntosh| Plate XII, Fig 5 ) 

53. B.^orogenes. Growth on serum agar slope 1 week. (Mcintosh, Plate XII, 

54. B. teriius. 48-hour growth on serum agar slope. (Mcintosh, Plate XIII, Fig. 5.) 

55. VibHm septique. 4 days' growth on serum agar slope. (Mcintosh, Plate XIII, 

56. B. welchii {StTa.m 82). Milk culture, 24 hours old, to show stormy fermentation 

JVote the disappearance of the cream layer, the disruption of the casein clot, 
and the displacement of shreds of clot upwards through the paraffin layer. 

57. B. welchii (Strain Pasteur). Alkaline meat culture, 1 week old, and showing 

reddening of the meat. Contrast this with the normal meat in Plate XXVII, 

58. B. welchii (Strain 82). Culture in alkaline egg, 5 days old. Note the diffuse 

opacity due to fine friable clot. There is no shrinkage and no sign of 
digestion. Contrast this with normal alkaline egg in Plate XXVII, Fig. 21. 

59. B. sporogenes (Strain Pasteur). Milk culture, 1 month old. To show undis- 

turbed cream layer and complete digestion of the milk. 

60. B. sporogenes (Strain Pasteur). Culture in alkaline milk, 1 week old. To show 

digestion and blackening. 

61. B. sporogenes (Strain Harwood). Culture in alkaline meat, 1 week old. To show 

marked blackening of the upper layers of meat. 

62. B. sporogenes (Strain Pasteur). Culture in alkaline egg. To" show cylindrical 

coagulum which has undergone gradual shrinkage and partial digestion. 

63. B. fertius (Strain 68). Milk culture, 6 weeks old. To show solid clot fissured 

by gas bubbles. 

64. B. tertius (Strain Benson). Culture in alkaline meat, 14 days old. To show 

pinking of meat and commencing bleaching on the surface. 

65. B.fallax (Strain Trucol). Milk culture, 1 month old, showing strong solid clot 

disrupted by bubbles, and with orange colour at the surface. 

66. B. fallax (Strain 92). Milk culture, 5 days old, showing separation into a turbid 

whey and a soft friable clot which is slightly disrupted by gas. There is no 
dislocation of the cream layer. Bubbles have collected on the air surface of 
the paraffin layer. 

67. B. aero/elid'us (Weinberg). Culture in milk, 36 hours old, and showing disrup- 

tion of casein clot by gas. The fermentation is not so vigorous as, and 
develops later than, the reaction given by B. welchii. 

68. B. aerofetidus (Weinberg). Growth in alkaline meat, 7 days old, showing 

reddening of the meat and slight blackening of the upper layers. 

69. B. oedematiens (Strain Domange). Growth in milk, after 4 months' incubation, 

to show slight separation of whey from subjacent casein flocculi. 

70. B. oedematiens (Strain Domange). Culture in alkaline meat, after 15 days, to 

show pinking of meat and extensive bleaching from surface downwards. 

71. B. histolyiicus (Weinberg). Milk culture, 2 days old, showing complete digestion, 

with no disturbance of the cream layer. 

72. B. histolyiicus (Weinberg). Culture in alkaline meat, 4 days old, to show 

digestion and formation of glistening white balls of tyrosin. 



183 

73. Ordinary milk under paraffin. 

74. Alkaline meat medium under paraffin. 

75. Alkaline egg medium under paraffin. 

76. Lesions produced in a guinea-pig by the injection of a culture of Vibrion s^tique 

into the muscles of the thigh. The subcutaneous tissues over the abdomen 
and thorax are the seat of a clear, deep red, sero-sanguinolent oedema in 
which there are extensive collections of gas bubbles. The abdominal muscles 
and inguinal fat are intensely congested. 

77. Lesions produced in a guinea-pig by the injection of a culture of BaciUus ■welchii 

into the muscles of the thigh. In comparison with those of V. septique the 
oedema is only slightly blood-stained, the formation of gas is usually less 
marked, and the muscles themselves are pale and looked bleached. 

78. Lesions produced in a guinea-pig by the injection of a culture of Bacittus 

oedematiens into the muscles of the thigh. The characteristic feature is a clear 
colourless gelatinous oedema, while the abdominal muscles are for the most 
part unaltered. In general, gas formation if present at all is trivial. 

(Pigs. 56-75 from Henry (Figs. 1-9 and 11-21).) 



Printed under the authority of His Majebtv's Stationery Ofpioe 
By Frederick Hall, at the University Press, Oxford. 



PLATE I 



^~X J 



/I - /x 






y ^^ -Nj..^ 



Fig. 1 



% 
t « 






• 



# - 



» 



Fig. 



PLATE II 




Pig. 3 



Fig. 4 



i 
I 



^/ 



I 



< 



A , 



Fig. ,5 



PLATE III 



/' 



-/ 






49 



O \ 



Fig. 6 



O 



V 




PLATE IV 






^ 



\ / 



V/ 






Fig. 8 



^% '^ 










•0 



Fig. 9 



.€ 



A 



^r? 




/ «» 






^ \ 



Fig. 10 



/? 







Fig. 11 




Fig. 13 






PLATE V 






Pig. 14 



« • 




Fig. 15 



Fig. 16 



i». 



«* I' 



Fig. 17 



J 



Fig. 18 



PLATE VI 



/ 



■' s 



\ 

V 

V* 

i 

Fig. 19 




Fig. 20 



. '' : 



r>^^ 



\ 



/ ' 



L_ 



Fig. 21 



Fig. 22 



Fig. 23 



PLATE Vn 




Fig. 27 a 



Fig. 24. 



_/ 



Fig. 25 



— ' ^ " Fig. 27 

Fig. 26 






nL 



/ "v^ 



t:-^ 



i... 



■-^ \ 






^ 



r. 



\ 



/- 



^ 



\> 



^\ 



^ 



/• 



Fig. 28 



i'LATj<: • vn I 



v.' 






i 


/ 


> 


1 











Fig. 29 




rIG. 60 




Fig. 31 




Pig. 32 



PLATE IS 




Fig. 33 




Fia. 34 



PLATE X 



**^ "/* <^ /^" 






, / 



\ f 



" _' ' '''^ 



FiQ. 85 



/ ^ 



_^x 



/ I 






"V 



^ ^ 






// 









«\ 



/ 









/i'K-n" 



Fig. 36 




PLATE XI 



Fig. 37 




Fig. 38 




Fig. 39 



fLATE XII 




Fig. 40 



Fig. 41 




Fig. 42 



• 



-r 



'^. 






"aw 




i^liH 




T^'-'-V-. 


s 


y 


--^ 


^ 










..^^ 




1- 


7 **"-- 






"^- 





•'id. 43. 



PLATE XIII 




Fr<^. 44 



^^f" 







Fic4. 45 



Fig. 46 




A-K ^lA*w«^l_ 



Fig. 47 



PLATE XIY 




Fig. 48 




Fig. 49 



PLATE XV 




Fig. 50 




Fig. 51 



PLATE XVI 





Fig. ^2 



Fig. 53 





Fia. 54 








ITT 




"°'5^^5^ 



<f > 



"'^W^-*'''**'**^ 






3 
&H 





-..^btk 



I \{w ._^V^'/ fe 



T~~TT! 







WPseWOTBJBH?!!*" 





M 



~vsmi 











F" 








Figure 76 




Figure 77 




Figure 78 



xiospwai, jiasiar, jLyio-iB. 

IV. Eeport upon combined Clinical and Bacteriological Studies of Dysen- 
tery Cases from the Mediterranean. Price 4s. 6cl, post free 4s. 8cl 
No. 7. V. Report upon 2,860 Enteritis 'Convalescents' received at Liverpool 
from various Expeditionary Forces. Price 6s., post free 6s. 2^d 

No. 15. A Study of 1,300 Convalescent Cases of Dysentery from Home Hos- 
pitals : vs^ith special reference to the Incidence and Treatment of Amoebic 
Dysentery Carriers. Price Is. Sd., post free Is. 4d. 

No. 29. A Contribution to the Study of Chronicity in Dysentery Carriers. 
Price,9d., post free 10|d. 

No. 30. An Investigation of the Flexner-Y group of Dysentery Bacilli. 
Price Is., post free Is. l^d. 

No. 40. Studies of Bacillary Dysentery occurring in the British Forces in 
Macedonia. Price 3s. , post free 8s. 2d. 

No. 42. A Study of the Serological Eaces of the Flexner Group of Dysentery 
Bacilli. Price 2s., post free 2s. 2d. 

Venereal Diseases : 

No. 23. An Analysis of the Results of Wassermann Reactions in 1,435 Cases 

of Syphilis or Suspected Syphilis. Price 2s., post free 2s. l^d. 
No. 41. I. An Investigation into the Ultimate Results of the Treatment of 

Syphilis with Arsenical Compounds. 

II. A Clinical Study of the Toxic Reactions which follow the 
Intravenous Administration of ' 914 '. Price 2s., post free 2s. l^d. 
No. 44. Reports of the Special Committee upon the Manufacture, Biological 

Testing, and Clinical Administration of Salvarsan and of its Substitutes. 

No. I. Price Is., post free Is. IJd. 
No. 45. Unsuspected Involvement of the Central Nervous System in Syphilis. 

Price Is., post free Is. l^d. 
No. 47. The Accuracy of Wassermann Tests, applied before and after death, 

estimated by. Necropsies. I. The Wassermann Test applied before death. 

Price 2s. 6d., post free 2s. 8d. 
No. 55. I. Results of the Examination of Tissues from Eight Cases of Death 

following Injections of Salvarsan. 

II. The Influence of Salvarsan Treatment on the Development and 
Persistence of Immunity, as indicated by Measurements of Agglutinins. 
Price 3s. 

(Vide infra. Reports of tlte Committee upon Pathological Methods, Nos. I-IV.) 

Alcohol: 

No. 31. Alcohol — Its Absorption into and Disappearance from the Blood under 

different conditions. Price Is., post free Is. Ifd. 
No. 34. The Influence of Alcohol on Manual Work and Neuro-muscular 

Co-ordination. Price 2s., post free 2s. l-|d. 
No. 56. The/ Effects of Alcohol and some other Drugs during Normal and 

Fatigued Conditions. Price Is., post free Is. l^d. 

Eeports of the Committee upon Pathological Methods : 
No. 14. I. The Wassermann Test. Price 3d., post free 4id. 
No. 19. II. The Laboratory Diagnosis of Gonococcal Infections. 

III. Methods for the Detection of Spirochaetes. Price Is., post free 
Is. lid. 

No. 21. IV. The Diagnostic Value of the Wassermann Test. Price Is., post 

free Is. 1-d. 
No. 35. Y^. The Reaction of Media. Price Qd., post free 7|d. 
No. 51. VI. The Laboratory Diagnosis of Acute Intestinal Infections, including 
the Principles and Practice of the Agglutination Tests. Price 4s. 6d., post 
free 4s. 8|d. 
Eeports of the Special Investigation Committee on Surgical Shock and 
Allied Conditions : 
•\T„ OK T .A7-TT WniirKl-Sbnck and Haemorrhase. Price 4s., post free 4s. 6d. 



eports of the Air Medical Investigation Committee : 

No. 37. VIII. The Effects of Diminished Tension of Oxygen, with especial 
reference to the Activity of the Adrenal Glands. 

IX. The Ear in relation to certain Disabilities in Flying. Price Is., 
post free Is. l^d. ' 

No. 53. The Medical Problems of Flying (including Eeports Nos. I- VII). Price 
6s., post free 6s. 4|d. 

eports of the Committee upon Accessory Pood Factors (Vitamines) : 
No. 88. Eeport on the Present State of Knowledge concerning Accessory Food 
Factors (Vitamines). Price 4s., post free 4s. 3d. 

Reports of the Committee upon Anaerobic Bacteria and Infections : 
No. 89. Report on the Anaerobic Infections of Wounds and the Bacteriological 
and Serological Problems arising therefrom. Price 6s., post free 6s. id. 



No. 8. Report upon Soldiers returned as Cases of 'Disordered Action of the 

Heart ' (D.A.H.), or 'Valvular Disease of the Heart ' (V.D.H.). Price Is., post 

free Is. l^d. 
No. 9. A Report upon the use of Atropine as a Diagnostic Agent in Typhoid 

Infections. Price Is., post free Is. l^d. 
No. 10. The Mortalities of Birth, Infancy, and Childhood. Price Is. 6d., post 

free Is. 8d. 
No. 11. The Causation and Prevention of Tri-nitro-toluene (T.N.T.) Poisoning. 

Price Is., post free Is. 2d. 
No. 12. The Classification and Study of the Anaerobic Bacteria of War Wounds. 

Price 2s., post free 2s. 2^d. 
No. 13. An Enquiry into the Composition of Dietaries, with special reference 

to the Dietaries of Munition Workers. Price 9d., post free lO^d. 
No. 16. A Eeport on the Causes of Wastage of Labour in Munition Factories. 

Price Is. 6d., post free Is. 7^d. 
No. 20. A Study of Social and Economic Factors in the Causation of Rickets, with 

an Introductory Historical Survey. Price 2s., post free 2s. 2d. 
No. 24. A Report on the Investigation of an Epidemic caused by Bacillus 

aertrycke. Price 9d., post free 10|d. 
No. 32. The Science of Ventilation and Open Air Treatment. Part I. Price 10s., 

post free 10s. &d. 
No. 36. Studies of Influenza in the Hospitals of the British Armies in 

France, 1918. Price 3s. 6d., post free 3s. S^d. 
No. 43. Albuminuria and War Nephritis among British Troops in France. 

Price 3s. 6d., post free 3s. 8^d. 
No. 48. A Report on the probable Proportion of Enteric Infections among 

Undiagnosed Febrile Cases invalided from the Western Front since October, 

1916. Price 3s., post free 3s. 2d. 
No. 49. On the Destruction of Bacteria in Milk by Electricity. Price 9d, post 

free lO^d. 
No. 62. The Science of Ventilation and Open Air Treatment. Part II. Price 

6s., post free 6s. 5d. 
No. 57. Studies in Wound Infections. 

'he following were published under the direction of the Medical Research Com- 
mittee : 

Milk and its Hygienic Relations, by Janet E. Lane-Claypon, M.D., D.Sc. (Lond.). 

Price 9s. net. [Longmans, Green & Co.] 
The Amoebae living in Man, by Clifford Dobell, F.R.S. Price 7s. 6d. net. 

[Bale, Sons & Danielsson, Ltd.] 

n addition to the publications contained in the list given above, numerons memoirs upon 
work aided hy the Medical Eesearch Cowncil have appeared in Scientific Journals.