WllXlAMS & WIIXINS CO, BALTIMOHE ,^,,,,^«Wfl!«'««S««»»»S^ "ru ! r-=I ; nj CD o io m THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA This book is copyright. It may not be reproduced by any means in whole or in part without permission. Application with regard to copyright should be addressed to the Publishers. THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA BY K. A. BISSET, D.Sc. Reader in Systematic Bacteriology, University of Birmingham SECOND EDITION E. & S. LIVINGSTONE, LTD. EDINBURGH AND LONDON 1955 Acknowledgements I WISH to acknowledge with gratitude the assistance given to nic in the production of this book by the Editors of the Joiinml of Hygiene, the Journal of General Microbiolo^^y, the Journal of Patliolo(^y and Bacteriolojiy, Experimental Cell Research, and Biocliiniica et Biopliysica Acta, all ot whom lent blocks of illustrations ; by the Camhridq^e University Press, Oliver & Boyd Ltd., and the Elsevier Press who provided the blocks in question, and especially by Dr. Emmy Klieneberger-Nobel, Dr. Woutera van Iterson, Professor J. Tomcsik, Dr. A. L. Houwink and Drs. Birch-Anderson, Maaloe and Sjostrand, whose admirable micrographs are the subject of many of these illustrations. I wish also to thank the Editors o( the Proceedings of the Royal Society, the Journal of Bacteriology, the Annales de I'Institut Pasteur, the Proceedings of tlie Society for Experimental Biology and Medicine and Cold Spring Harbor Symposia, Professor R. H. Stoughton, Professor R. J. V. Pulvertaft, Dr. J. Lederberg, Dr. R. R. Mellon, Dr. E. O. Morris, Professor A. R. Prevot and Drs. Chapman and Hillier for permission to reproduce illustrations, and Dr. Joyce Grace, Miss P. E. Pease, Miss B. G. Fewins and Mr A. A. Tuffery for their helpful observations and material. Lastly, I wish to thank Professor J. F. D. Shrewsbury tor the facilities which he has placed at my disposal ; and Miss C. M. F. Hale for her skilled and patient work in the preparation ot the cytological material upon which this monograph is based. K. A. BISSET, Department of Bacteriology, University of Birmingham. Preface to Second Edition THE second edition of this monograph inckides a considerable amount o{ new information which serves, for the most part, to confirm and expand the general theories of cytological structure and behaviour in bacteria which I advanced six years ago. hi particular, greatly improved demonstrations have been achieved of such, formerly rather problematical processes as gonidial reproduction, nuclear reduction, and the development of the flagella, and of such structures as the blepharoplast or the startlingly complex cross-walls which sub-divide a staphylococcus internally. I have continued to place the main weight of my arguments upon my own observations or upon information which I have been able personally to confirm, but have welcomed several remarkable contributions to knowledge in the form, for example, of Chapman and Hillier's electron micrographs of sections o( Bacillus ccrens, which prove, contrary to my previous belief, that the cross-wall is indeed a centripetal ingrowth, and the consummately skilful phase contrast studies of Tomcsik which, while demonstrating the profundity of my former ignorance of the nature of bacterial capsules, provide a gratifying confirmation of my hypothesis, based upon entirely different evidence, of the development of cell wall. Since the first edition was published there has also been a notable increase in the amount of corroborative evidence provided by studies in genetics, biochemistry and biophysics. The single, reductionally dividing chromosome of the vegetative nucleus, which has been the subject of the most lucid cytological demonstrations, has been entirely vindicated by the genetical studies of Witkin and o( Cavalli-Sforza and Jinks, after a period in which multiple, or even branched chromosomes, for the existence of which there is no acceptable cytological evidence, were at various times postulated by geneticists. And in a similar manner, the cytological and physico-chemical evidence upon the behaviour of cell envelopes and flagella have been in exceedingly close accordance, and have even begun to shed some light upon the problems of antigenic structure. Since it is now the subject of a separate monograph [Bacteria, Livingstone), the chapter upon bacterial systematics has been omitted, but the allied problem, more cogent to this study, of cytological evidence of the evolutionary relation- ships o£ bacteria, has been discussed at greater length. Because the text is intended to be a synthesis of available information, the practice of relegating references to the head of each sub-section, except in cases of argument or of historical interest, has been continued. There appears to be a consensus o{ agreement that what may be lost in ease of tracing a reference to a single point is more than gained in clarity and brevity. April 1955 < K. A. BISSET. ^ Preface to First Edition THIS book does not attempt to review the literature upon bacterial cytology, of which the bulk is very great and the value, in many cases, difficult to assess. The bibhography is confined to a relatively small number of works, almost all recent. No attempt has been made to supply references for analytical discussion or general information. The purpose is rather to present a reasoned case for regarding bacteria as living cells with the same structure and functions as other living cells, and to correlate the available information upon the various types of bacteria. Bacteria, as living creatures, have been little studied. It is their activities as biochemical or pathological agents which have received almost undivided attention. Even these problems, however, cannot fail to be clarified by a better knowledge o{ the organisms responsible. It is also hoped that biological workers in other fields may profit by contact with this, largely unknown, body of evidence, and may find the comparisons and analogies useful and stimulating in their related studies. I have attempted, as far as possible, to base my arguments upon my own observations, or upon such information as I have been able personally to confirm. Where I have not had die opportunity to do so, I have tried to indicate clearly the status of the argument. K. A. B. December 1949 Contents CHAP. I. Introduction III. Technique (a) The Staitiitig of Bacteria (b) Hydrolysis Techniques PACE I (c) Differentiation Techniques ......... 8 (d) The Romanou'shy Stains ......... 8 (e) Simple Dyes ........... 9 (f) Use of Enzymes, etc 9 (g) Classical Cytoloj^ical Procedures . . . . . . . . lO (h) Cell Wall Stains I2 (i) The Mounting of Material ......... 14 (j) The Stainin(^ of Fla^ella ......... 16 (k) Electron Microscopy . . . . . . . . . .16 (1) Phase-contrast Microscopy ........ (m) Colony Preparations ......... (n) Summary .......... Surface Structures (a) The Cell Wall . (b) The Cell Membrane . (c) Development of Envelopes . (d) The Cell Wall of Myxohacteria (e) The Spore Coat (f ) The Nuclear Membrane (g) Slime Layers and Capsules . (h) Flagella .... (i) Summary 17 21 21 25 33 36 40 41 41 41 44 47 IV. The Bacterial Nucleus (a) Historical 51 (b) The Restin^T Nucleus 53 (c) The Primary Vegetative Nucletis 62 (d) Germination of the Resting Stage 64 (e) The Spherical Vegetative Nucleus 65 (f ) The Nucleus in Complex Vegetative Reproduction 67 (g) The Nature of the Chromosome 72 vii - '^5213 CONTENTS IV. The Bacterial Nucleus — coitti.. (h) The Secondary Nucleus (i) The Rod-Uke Nucleus (j) Formation of the Restin(^ Nucleus (k) Sutnntary .... V. Reproduction (a) The Growth Cycle . (b) Simple Vegetative Reproduction (c) Post-fission Movements (d) Complex Reproductive Methods (e) Fission in Mycobacteria and Corynehacteria (f ) Branchinfi and Budding (g) Swnmary .... VI. Sexuality in BACThniA (a) The Existence of Sexuality . (b) Sexuality in Sporing Bacilli . (c) Syngamous Vegetative Reproduction . (d) Microcyst Formation in Myxohacteria and Euhacteria (c) Sexual Fusion in the Secondary Nuclear Pliase (f) Sexuality in Mycobacteria (g) Sexuality in Streptomyces (h) Sexual Fusion in Proteus and Streptobacillus (i) Sununary ..... VII. Life-Cycles in Bacteria (a) General ...... (b) The Life-Cycle in Myxohacteria (c) The Life-Cycle in Euhacteria (d) The Life-Cycle in Streptomyces (e) The Life-Cycle in Chlamydobacteria and Caulobacteria (f ) Gonodia as a Stage in the Bacterial Life-Cycle (g) L-Organisms ...... (h) Summary ...... VIII. Macroformations (a) The Myxobacterial Fruiting Body (b) The Myxobacterial Swarm .... 99 99 103 105 109 III 113 115 115 118 118 119 121 121 124 125 129 134 CONTENTS IX CHAP. PAGE VIII. Macroformations — COIItd. (c) The Swarm of Proteus . . . . . . . . -135 (d) Chlaniydobacterial Aggregates . . . . . . . • i3<^ (e) The " Medt4sa-Head " Colony . . . • • - • -137 (f) Smooth Colonies 141 (g) Rough and Smooth Colonies of Streptococci . . . . . ■ H^ (h) Colonies of Streptomyces ......... 142 (i) Colonies of Caulobacteria ......... 143 ( j) Summary ........... 144 IX. The Evolutionary Relationships of Bacteria (a) Morphological Evidence ......... i4<^ (b) Previous Classifications . . . . . . . . .146 (c) The Ancestral Bacterium . . . . . . . . -150 (d) Flagellar Pattern 151 (e) Aerial Distribution . . . . . . . . . 153 (f) Autotrophic Bacteria I55 (g) Summary . . . . . . . . . . .156 X. The Genetics of Bacteria (a) Genetical Confirmation oj Cytology . . . . • • -157 (b) Genetical Evidence . . . . . . . . . 159 (c) Summary ........... 160 Index t6i List of Illustrations FIG. I. 2. 3- 4- 5- 6. 7- 8. 9- 10. II. 12. 13- 14. 15- 16. 17- 18. 19. 20. 21. 22. 23- 24. 25. 26. 27- 28. 29. 30. 31- 32. 33- 34- 35- 36. (37. The Morphology of C. diphtheria; . . . , Bacilhis Fixed and Stained .... Cytological Staining of Cocci Cytological Staining of Azotobacter . Granules in the Cell Envelopes of BaciUus Demonstration of Capsules by Tomcsik's Method Sections of Bacteria ..... Cell Division in Bacteria .... Behaviour of the Cell Wall in Dividing Streptococci Cell Division in Cocci ..... Cell Envelopes in BaciUus .... Electron Microscopy of Cell Envelope Material Cell Envelopes observable by Electron Microscopy Bacterial Flagella ...... The Bacterial Blepharoplast .... Development ot Flagella in the Germinating Microcyst Cytology of Oscillospira ...... The Vegetative Nucleus ..... Microcysts o( Bacteriacecv ..... Effect of Hydrolysis on the Spore Nucleus The Spore Nucleus ...... Appearances of the Nucleus ..... Sections of Bacterial Nuclei The Germination of the Resting Stage The Spherical Vegetative Nucleus . Complex Vegetative Reproduction in Bacteriuttt The Primary Nucleus and Vegetative Fusion Cells in various Bacteria Tracings of Photomicrographs of Vegetative Fusion Cells Fate of Chromosomes in Cell Division . Nuclear Division in Azotobacter The Vegetative Nucleus in Bact. coli The Secondary Nuclear Phase in Bact. malvacearum Types of Rod-like Nucleus .... xi PAGE I II 12 13 15 19 20 26 28 29 30 34 39 42 44 (45 146 52 54 56 57 58 61 63 65 (68 I69 70 71 74 75 76 77 80 xii LIST OF ILLUSTRATIONS FIG. NO. PAOE 38. Alternative Modes of Division in Mycobacteria and Coryncbacteria ... 92 39. The Cytology of Corynebacteria and Mycobacteria 93 40. Branching in Bacteria 95 41 I , , I 100 Maturation of the Spore .......... 42. I I loi 43. Life-Cycle of ATofrtr^Z/rt 104 44. Nuclear Cycles in Bacteria .......... T05 45. Maturation of the Microcyst in Brtrr. fo// . ....... 106 The Cytology of Myxobacteria ' 48. The Nuclear Reduction Process no 49. Maturation of the Resting Cell in Brtff. ;//<;/i'(;f<'(7n/;// . . . . . .111 50. Maturation of the Resting Cell in M. f///)tTn(/('5/5 112 51. Life-Cycle in Af//;('»/y«'5 Bci'i.s- . . . . . . . .114 52. Maturation of Resting Cell in 5/;//tT('/)/;('n(.s- . . . . . . .115 ^^■j Lifc-CvcleofCc»//()/wftT 54.) ' I 123 55. The L-Stage in the Bacterial Life-Cycle 126 Bacterial Gonidia ............ S7-I I 130 58. Life-Cycle of /l^oro/wf/er 131 59. Myxobacterial Fruiting Bodies . . . . . . . . -135 60. Stages in the Growth of a Medusa-head Colony . . . . . .138 61. Growth of a Rough Colony .......... 139 62. Smooth and Rough Colonies 140 63. Colonies of Streptococci .......... 142 64. Chlamydobacteria ............ 143 65. Caulobacteria ............ 143 66. Relationships of Cocci and Bacilh ......... 148 67. Evolution of Flagella . . . . . . . . . . .152 68. Diploid Forms of the Nucleus i$8 CHAPTER I Introduction MUCH of what has in the past been written of the morphology of bacteria has been based upon the assumption that, because of their small size, and the difficulty, by the methods usually employed, of observing the complexities of their structure, they may be regarded as simple in form and primitive hi philogeny. The temptation to regard small size, and simplicity of structure, whether real or apparent, as criteria of a primitive condition, has often proved the cause of error and confusion in the classification of other groups of living organisms. As more information becomes available it is almost invariably discovered that the simplest creatures exhibit characters which suggest a relationship with others, much more complex, or may themselves prove to be less simple than they had been believed. This has proved to be true of bacteria also. Although for long believed, in spite of much evidence to the contrary, to be almost structureless cells, reproducing by simple fission, they have proved to possess an intricacy of structure rivalhng that of any other type of living cell, and to undergo life-cycles of considerable complexity. There is little doubt that the reason why so much more has been learned of the physiology of bacteria than of their morphology, is their very great importance in medicine, industry and agriculture. The immediate, practical problems ot bacteriology have overshadowed the more academic questions of their biological nature. The techniques which were devised for the solution of these problems have been notable, in almost every case, for their failure to provide even a minimum ot basic, biological information. Indeed it may be said that much of the information o( this nature, accumulated since the commencement of systematic bacteriology, has tended rather to obscure than to clarify the underlying truths. ( Especially is this true of the staining techniques employed tor routine 2 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA examination of bacteriological material and cultures. The distorted vestiges o( bacteria which survived the technique of drying and heat-fixation were accepted as truly indicative of the morphology of the living organisms. And while, from time to time, satisfaction has been expressed at the fact that bacteria will survive, undistorted, treatment which produces the most obvious damage in larger cells, the vaHdity of the assumption that they do, in fact, survive such treatment has seldom been called to question. Staining methods have also been devised, almost without exception, for the purpose of identifying cHnically important species of bacteria, and are often most admirably suited to this task. It is surprising to find, however, that much time and labour has been directed to the elucidation of the appear- ances observed by these methods, and the explanation, in cytological terms, of the artefacts which they produce. Even with this disability a great deal of accurate information has in fact been obtained, but has tailed to carry conviction. In many cases this has been because of inadequate illustration, which alone can make such studies com- prehensible, except to the initiate. Probably the reason has been an unduly pessimistic view of the possibihties of photomicrography, and a certain timidity in the submission of drawings and diagrams, due perhaps to a fear of misinterpreting such tiny structures, and a corresponding fear of ridicule. It is also remarkable that many workers in the field ol bacterial cytology appear to have been almost entirely ignorant ot the parallel studies of others, and have tailed to receive the stimulus which such knowledge can afford. Conversely, there has been no lack of reviews of the subject, but these have often been made by authors whose lack ot practical knowledge ot the structures described has disqualified them tor the task ot correlating the available information, which is otten obscure and mutually contradictory. The artiticiality ot contemporary or recent views upon bacterial morpliology has thus served to widen the gap between bacteriology and other biological sciences, as well as to confuse and retard the advance of bacteriology itself In the evolution of modern cytological methods, much is owed to the interest taken by mycologists in the myxobacteria. These micro-organisms do not respond well to the techniques of hcat-fixation and Oram's stain, most INTRODUCTION usually employed in routine bacteriology, and the necessity for the employ- ment of more refined methods of examination has encouraged the study of eubacteria in a similar manner. The readily-demonstrable nuclear structures and beautiful and complex life-cycle of myxobacteria stimulated the search for the truth concerning the parallel structures and processes in those bacterial genera more commonly encountered in the laboratory. The studies of biochemists upon the nucleoproteins of bacteria have also contributed greatly to the increase in our knowledge of, and interest in, the problems of bacterial cytology. One of the most useful staining techniques tor the demonstration of the bacterial nucleus is a direct adaptation of a microchemical test, the Fculgen reaction, which has itself given much information upon the subject. Bacteria have recently come to be regarded as suitable material for genetical studies, and although little has so far been done to correlate genetical and cytological information, a gratifying degree of mutual support has already been achieved (Chapter X), and it is to be hoped that the interchange of information between these two branches of bacteriology may, in the future, prove as helpful to both as it has done in other biological fields. The information compiled in the following chapters has been obtained by classical microscopic methods, in most instances, but a considerable advance m the techniques of electron and phase-contrast microscopy, as applied to this subject, has in the last few years provided valuable confirmatory evidence on several points, and promises to do more. It should be emphasised that a reasonable degree of correlation between the results obtainable by different techniques must always be sought before too much weight is placed upon any one of these. The disagreements which have arisen in bacterial cytology have been surprisingly few. But almost all of these have been caused by the uncritical reliance of a single worker, or a small group, upon a single methoci. ' ^-fl CHAPTER II Technique THE STAINING OF BACTERIA (8,11,53) THE progress ot our knowledge ot bacterial morphology has, m the past, been considerably retarded by the fact, which may at first have appeared advantageous, that recognisable microscopic preparations of bacteria can be made by the technique of the heat-fixed film. A small quantity of a bacterial culture, or of: pus, or similar pathological material, is thinly spread upon a slide, dried, and then heated strongly with a naked flame, in order to fix it firmly upon the slide. Bacteria fixed in this manner and stained by Gram's method, or simply with a strong solution of a basic dye, dried once more and examined directly under the oil-immersion lens of the micro- scope (the oil serving also as a clearing agent), preserve an appearance which enables them to be recognised as bacteria, and even classified within rather broad limits. Their appearance under this treatment has become famihar to generations ot bacteriologists, and is usually that which is recorded in the descriptions o£ species. Little or no detail can be perceived in such a pre- paration, and it has thus become, and until recently has remained a dogma that no detail exists to be seen. This opinion is fortified by the fact that equally little structure can, as a rule, be made out in unstained, hving bacteria, especially as these are seldom at rest, either because of their own motility or from the efirct of Brownian movement. It is true that, from time to time, valuable observations upon the structure of bacteria have been made, either by the cytological techniques already employed in other biological sciences, or by a careful study of unstained material, but little attention has been paid to these findings by the great TECHNIQUE 5 majority o( bacteriologists, and the interpretation ot heat-fixed material has not been questioned seriously. The main reason lor the unilorm appearance of stained bacteria is that their affinity for the basic dyes which are commonly employed is so great that the strongly stained cytoplasm and cell membranes mask the underlying structures. This masking effect is accentuated by the shrinkage which results from drying. This shrinkage is often very considerable, reducing the bacterium to as little as half or a third of its natural size, and manifesting itself typically in the appearance of the anthrax bacillus or of related chain-forming bacilli, 11 {Reproduced from the Journal of General Microbiology) Fig. 1. THE MORPHOLOGY OF C. DIPHTHERIAE. A . True morphology. B. " Typical appearance" in heat-fixed material. The cell contents are shrunken and the cell wall unstained. in which considerable gaps are seen between the visible bacilli, actually the shrunken protoplasts. The rigid cell wall remains unstained and invisible, holding the chain together. Drying and shrinkage are an essential part of many staining procedures, notably those intended to demonstrate the " typical morphology " o£ Corynehactcrium diphtherici\ The metachromatic granules cannot be demonstrated in undried preparations, and are, in tact, artefacts produced by the specific staining of a dried aggregate of nuclear and other basophilic material. 6 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA Many bacteria are multicellular, and their appearance is much altered by drying. The granular appearance of the tubercle bacillus is due to the shrinkage of the contents of the small, almost spherical cells which make up the bacillus, so that unstained gaps appear between them. In this case also the cell wall remains unstained, but retains the dried cells in their original relationship. It will thus be seen that three main problems must be solved in the demon- stration of the true morphology of bacteria. Distortion due to drying must be avoided, the masking effect of the strongly staining protoplasm and cell membranes overcome, and those structures demonstrated which, like the cell wall, are difficult to stain. The first is simple and entails merely the avoidance ot drying at all stages of preparation. The second and third present more difficulty. The problem of overcoming the masking effect of the surface structures was solved, as so often happens, by accident. B: HYDROLYSIS TECHNIQUES FOR NUCLEAR STAINING (13, 15, 16, 19, 24, 30, 32, 40, 41, 44, 46, 48, _so, SI, 5^. 5i, 57) The Feulgcn reaction is a microchemical test which depends upon the formation of a purple compound when aldehydes react with Schiff's reagent, A positive Feulgen reaction is given by deoxyribose nucleic acid, after its purine bases have been removed by acid hydrolysis. Ribonucleic acid does not give a positive reaction. The hydrolysis is performed in Normal hydro- chloric acid at a temperature of 60° C, and the subsequent staining with Schift 's reagent reveals the nuclear structures o( bacteria with reasonable clarity. This was one ot the first methods to give a true picture of the bacterial nucleus, and it was later discovered that if the final staining was pertormed with Giemsa's solution, instead of Schiff's reagent, a much clearer picture was obtained. This was the acid-Giemsa stain, which has been the basis of nearly all recent work upon the bacterial nucleus, although the information which it pr(widcs can be verified by other methods. The purpose of the preUminary treatment with hydrochloric acid is two-fold. The nucleoproteins of the underlying structures are partially hydrolysed so that the aldehyde group of the associated pentose sugar is T E C H N I Q U 1- 7 released and combines with the stanung agent. At die same time the stainable material ot the outer layers o{ the cell is more completely hydrolysed, so that its masking eftect is reduced. This dift'erentiation is made possible by the fact that the nuclear structures are composed largely of Fculgeii-positivc deoxyribose nucleoproteins, whereas the cell membrane and surface layers of the cytoplasm usually contain a higher proportion of ribose nucleoproteins. To pcrtorm the stain, smear preparations are made upon shdes or cover- slips. They may be unfixed, although these tend to wash off, or they may be fixed in osmic acid vapour. Most fixatives should be avoided as they may completely alter the appearance of the nucleus. Hydrolysis in Normal HCl should be conducted at a temperature, approxi- mately, of 60° C. Staining, in dilute Giemsa, is best performed at 37° C. The periods required tor hydrolysis and staining arc exceedingly variable and may be different at different ages of the same culture. It is often necessary to examine the preparation with the microscope, in order to determine whether it is suitably stained, and for this purpose a water-immersion lens is a great convenience. Most bacteria require from ten to twenty minutes hydrolysis, and thirty minutes in the staining solution. Some require longer periods or stronger solutions. A properly stained preparation is bright pink in colour, the nuclear structures staining more intensely than the cytoplasm, which may stain bluish or purple in some cases. Inadequate hydrolysis is indicated by a uniform purple colour, and excessive hydrolysis by a pale pink colour and blurred outline. Inadequate or excessive staining periods are self-evident in the appearance of the preparation. It is important to use fresh reagents, and otherwise inexplicable failures may be found to be due to neglect to do so. Other methods of staining give comparable results, and may be useful ill the case ot bacteria which do not stain well by the classical method. Cold perchloric or trichloracetic acid, or even weak alkalis, may be substituted for hydrochloric acid. A variety of different dyes may be used instead of Giemsa. Thionin gives good results and has been widely used, but is much less specific than Giemsa, and stains the basophilic elements of the cell envelopes as well as the nucleus, which is liable to cause confusion in interpretation. 8 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA C; DIFFERENTIATION TECHNIQUES (2, 10, 13, 20, 48, 66) The mcthylcnc-bluc-cosin method has been used to demonstrate the nuclear material of bacteria which will not stain readily by acid-Gicmsa, it is unfortunately irregular in its results and may be liable to produce artefacts. Basically the method is exceedingly simple. The preparation is stained with aqueous methylene blue and diftcrentiated with eosin. The cytoplasm stains pale blue, and the nuclear structures dark blue or purple. In practice, however, it is a diliicult technique to perform, and is not suitable for all strains of bacteria. The fdm should be made thick and stained until dark blue throughout. It is then washed in water, differentiated for a few seconds in cosin and immediately washed again. The action of the eosin is very rapid, and it will entirely remove the blue colour if it is allowed to act tor too long. It was noted previously that this technique may usefully be employed upon bacteria which resist staining by acid-Giemsa, and the converse is also true. For this reason, methylene-blue-eosin is best regarded as a useful adjunct to acid-Giemsa, and is not recommended as a routine cytological method. Similar results arc obtainable by the use of crystal violet, with nigrosin as a differentiating agent. D: THE ROMANOWSKY STAINS (3^ 47) The methylene-blue-eosin technique differs from the better-known staining methods of the Romanowsky type in that the combination of the acidic and basic dyes is permitted to take place during the period of the stainmg reaction. The more orthodox methods arc often of considerable value, however, and simple staining with Giemsa will often prove of value in the case of bacteria, such as myxobacteria and some members of other orders, whose surface structures lack the strong affinity for dyes exhibited by many. Valuable observations have been made in a variety of bacterial groups by the use ot these methods. TECHNIQUE 9 E: SIMPLE DYES (i, T4, 58, 59) Even the simple dyes, especially basic fuchsin and methylene blue, may be ot value upon occasion, it the errors of heat-fixation and drying, which usually accompany their use, can be avoided. The affinity of bacterial cytoplasm for the basic dyes is so great that a short treatment will often produce an appearance ot negative staining of the nuclear structures, which appear pale and refractive against the stained background. This phenomenon is well known, and is usually described as bipolar staining. Accumulations of basophilic material at the poles are also associated with the growth of the cell. Reagents and even displaced nucleic acids may form aggregates in these areas and appear as granular artefacts. An interesting refinement in the use oi a simple dye, which has been employed with considerable success, consists in permitting a thin film of carbol fuchsin to dry upon a slide. The bacteria are suspended in a drop of water upon the coverslip which is inverted upon the slide and sealed at the edges. The dye is taken up gradually by the bacteria and the process may be followed under the microscope. This method has proved of value in the description o( certain of the complex processes which precede the formation of the resting nucleus, but appears to have failed to demonstrate the active, vegetative condition of the nucleus in the same species of bacteria. F: THE USE OF PROTEIN MATERIALS (17, 25, 60, 6r, 62, 63, 64, 65, 68) As the surface material, the affinity of which for basic dyes tends to obscure the internal structures of bacteria, is composed mainly of ribose nucleic acid» it has been found possible to digest away this material with the enzyme ribonuclease. This leaves unharmed the deoxyribose nucleic acids ot the nucleus itself, which can then be demonstrated without difficulty. 10 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA Valuable results have been achieved by the digestion of surface structures with lysozynie and by various combinations of lysozynie, trypsin and other proteins at specific values of pH, and by phase-contrast studies of the effects upon the cell of antibodies active against cell-envelope components. (Section L, below.) G: CLASSICAL CYTOLOGICAL PROCEDURES (3, II, 14, 18, 43, 51, 67) The cytological staining techniques which have been employed for plant or animal cells arc often of value also in the case of bacteria. These are too numerous to be dealt with in detail. Iron alum hacmatoxyhn and borax carmine have both proved useful in demonstrating the bacterial nucleus. Cytochenncal techniques for the demonstration of polysaccharide food reserves, fat globules and similar materials have been extensively utilised in the investigation of bacteria, but the results achieved have been marred by the absence of any attempt, in most cases, to preserve the natural appearance of the cells. It is also the opinion of the author that these supposedly specific staining reactions, of which the use of osmic acid vapour or the naphthol dyes for lipids are fair examples, arc much less reliable than has been supposed. The enigmatic nature of the majority of demonstrable granules in bacteria is tacitly admitted by the practice of coining for them such titles as " meta- chromatic granules " or " volutin " ; they have even been claimed to be mitochondria. These granules are rarely apparent except in dried material, and are often artefacts, although it is not denied that reserve foodstuffs, in the form ot polysaccharides or lipids, may normally be present in the bacterial cell. It is also beyond question that the nuclear bodies of bacteria have many times been described in circumstances which have led to their being confused with these and other unidentifiable granules. TECHNIQUE II (Reproduced from Experimental Cell Research) Fig. A GROUP OF BACILLUS FIXED AND STAINED BY VARIOUS METHODS (1) Diagram of the cytological structure of a group of bacilli at various stages of cell division. The top-left bacillus is divided into four cells by complete cross-walls ; fission is commencing. The top-right bacillus is divided by a complete cross-wall and subdivided by cvtoplasmic septa alone. The lower bacillus has recently divided and has only two cells. f.ZiV., cell wall ; ^.Z?., growing point ; 5., cytoplasmic septum preceding cross-wall ; s.e., septum in earlv stage ; the darkly-stained junctions of the septa are growing points ; n, nucleus. (2) The same group demonstrated by a simple, basic dye. (3) Stained by Gram and over decolorised. (4, 5) Artefacts caused by unsuitable fixation. (6) Sporulation appearances. CYTOLOGY LIFE-HISTORY B A C T K R I A H: CELL IV ALL STAINS (S, 9, i6, 1 8, 20, 26, 27, 2S, 37, 3S, 39, 45, 53, 56, 66) Bacteria arc enclosed in a rigid cell wall which normally resists staining. It may be rendered visible by mordanting in tannic acid, phosphomolybdic acid or cationic detergents. These agents serve the dual purpose of mordanting the cell wall and so altering the protein material ot the cell that it is rendered unstainable and does not obscure the details ot the transverse septa, where these occur. Tannic acid also forms a stainable complex upon the surface of the cell wall, and it it is stained before the mordanting process is complete, this complex may produce an outline picture of the wall, but fail to show internal details. Using 5-10% tannic acid, the wall can be stained with 0-2 °o crystal violet ; with 1% phosphomolybdic acid, i°o methyl green gives the clearest results. The times required for staining and mordanting vary from a few seconds upwards. Fig. 3 THE CYTOLOGICAL STAINING OF COCCI (1) Micrococcus cryophiltis by Hale's method for cell walls. Showing two- or four-celled cocci. (2) A similar group stained by trichloracetic acid and (jiemsa. Such arrange- ments of nuclear material have been misconstrued as mitotic figures by some {Reproduced horn the Jounuil of llaiU-nology) observers. The uncierlying cell membrane is not easily demonstrated. Transverse septa derived from it, containing a large protein component, are stainable by simple, basic dyes or by acid-Giemsa, and are sometimes rendered more obvious by fixation with Bouin's solution or similar agents. Bacteria which have been slightly plasmolysed by such fixatives can be stained by 0-050;^ Victoria blue in such a manner as to demonstrate the cell wall and cell membrane simultaneously. TECHNIQUE 13 Bacterial cell mcinbrancs arc also stained by dyes of the Sudan, fat soluble group. This probably indicates the presence of a lipoid or lipoprotein com- ponent, which is also bchcvcd to exist in the cell membranes of other organisms, but probably does not mean that the cell membrane should be regarded as predc^minantly lipoid in constitution. In addition to the well-known methods of Hiss's and Muir's stains, capsules and slime layers are also stainable by the tannic-acid-violet technique, but these methods give little hint oi the remarkable structure which can be discerned in bacterial capsules by phase-contrast microscopy. (Reproduced from the Journiil of General Mi Fig. 4 CYTOLOGICAL STAINING OF AZOTOBACTER Appearances produced by various staining tecnhiques. (1) Nucleated cells stained by lithium carbonate and Giemsa. (2) As (1), by magnesium sulphate and thionin. (3) As (1), by nitric acid and thionin. Comparison of these figures shows positive and negative differential staining of the large perinuclear granules. (4) Phase-contrast studies of a, vacuolated and b, nucleated cells. (5) Vacuolated cells by nitric acid and thionin. The outlines of the lipid globules alone are stained. (6, 7) As (5), stained for lipids by Ziehl-Neelsen and Sudan IV. respectively. In the latter case the lipids have become displaced towards the periphery ; possibly because ♦ of partial solution. 14 THE CYTOLOGY AND LI FK-HISTO R Y OF BACTERIA /: THE MOUNTING OF MATERIAL (K, 9, 10, ir, 12, 22, 24, 29, 42, 53) Cytological preparations should be mounted upon the thinnest available slides and coversHps, and it is a distinct advantage to prepare the smear upon the CO verslip, so that the part of the preparation which is firmly adherent to the glass is nearest to the objective of the microscope. It is also simpler to transfer a coversHp from one reagent to another without the necessity oi employing large volumes of fluid. The author has found sputum tubes and watchglasses to be very suitable as containers, and the small volumes ot reagents which they contain may be renewed at frequent intervals. Coverslips should be sealed to the slide, at the edges, with wax or vaseline, unless the preparations are dehydrated and mounted, to which there is no theoretical objection. In practice, however, it will be found that some shrinkage and distortion will usually result, and the clarity ot the fniished preparation will compare unfavourably with that of a simple water mount, although possessing the advantage ot being permanent. It the edges are carefully sealed a water mounted preparation will last for several days, in the refrigerator, although it may deteriorate rapidly at room temperature. It is worth emphasising that far more detail can usually be made out in a good photomicrograph, with all the advantages of colour filtration, than can be discerned, by the most experienced observer, by direct microscopic examination. Impcrmancncc ot preparations is thus ot little importance provided that interesting appearances are photographed. It is also true that appearances which cannot be reproduced, more or less at will, are unlikely to be cither true or important, and their impermanencc is not to be regretted. Fig. 5 BASOPHILIC GRANULES IN THE CELL ENVELOPES OF BACILLUS (1,2) Partially acid-hydrolysed bacteria, stained with Giemsa, showing nuclear bodies and granula and diffuse basophilia respectively in the cell envelopes. (3) Over-hydrolysed specimens showing occasional granules. (4, .5, 6) As (3) with added, extraneous DNA which has adhered in the form of stainable granules, especially at the poles and cross-walls. This indicates that the granules appearing naturally in (1) and (3) may well consist of similar nucleic acids translated from the nucleus and cell envelopes by the hydrolysis procedure. Such granules have frequently been mis- interpreted as nuclear bodies, mitotic centrioles, mitochondria, etc., etc. TECHNIQUE I5 ft//' T^ iKcprndii, ,,/ // 'III /■ 1/', I : I Reproduced from the Journal ofGemral M uroh„,to^v, hy /.,■ Fig, 6 „, „f F,ul,-,s„r I. lomrsik) THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA SECTIONS OF BACTERIA Diagrams drawn from electron micrographs of Bacillus cereus are compared. Diagram A is taken from the study by Chapman and HilUer, and B from material prepared in this laboratory. The former has been treated with 2°,, osmium tetroxide which has had the effect of rendering the cell wall very clearly visible but may also have coagulated the softer structures. Section B has been accorded minimal treatment before embedding. The cell envelopes are much less clearly seen, but the arrangement of the various structures is in closer accordance with that seen in stained preparations. A — the cell wall has an outer (A) and an inner {A^) layer ; the latter may or may not correspond to the cell membrane. The cross-walls are seen complete (B) and developing by ingrowth {B^) ; an important observation. The material (C) between the inner boundaries of the developing cross-wall, and (C^) at an early stage of their formation, corresponds to the basophilic septum which precedes the true cross-wall. Its present appearance may well be an artefact arising from the coagulation of a much thinner, continuous structure across the cell. Similarly, the extension of the matrix (/->'), in which the nuclei (D) are embedded now extends across the former cell boundaries, at {B^) and (C^). B — The cell wall (E) and cross-walls (F) have thickenings (G) at their junctions, such as can be observed by phase-contrast by the method of Tomcsik, and in some stained pre- parations. The cross-walls are indistinct but continuous ; their mode of development cannot be discerned. The matrix (H^) of the nuclear bodies (//) does not extend across the cell boundaries where these have remained complete. TECHNIQUE 21 M: COLONY PREPARATIONS (5, 6, 7) Although a bacterial surface colony upon solid medium is a scmi-artificial formation, the study of its minute structure is often of biological interest and diagnostic importance. It is capable of providing evidence of the natural relationship of the bacteria to one another, and, especially in the study of dissociation, may indicate differences in structure and behaviour which are not always obvious by other methods of examination. Entire colonies may be embedded and sectioned, like portions of tissue, but as the colonies are usually exceedingly thin and flat (much more so than they appear to the unaided eye), whole mounts may be made upon slides or coverslips. These are usually termed impression preparations. They are best made from very small colonies, although quite large ones can be mounted if the growth is sufficiently tough. A small piece of medium bearing the desired colonies, is cut out with the point of a knife and placed, face downward, upon a slide or covershp. Surface tension will suffice to keep the medium, and the attached colonies, firmly pressed to the glass. It is then fixed, in its entirety, preferably in Bouin's solution, until the medium is blanched throughout, and can be peeled away from the glass, leaving the colonics adhering to the surface of the covcrslip. The preparations may then be washed, stained and mounted. Sometimes they are of great beauty. The best medium for this purpose is blood agar. It is very adhesive when fixed, and becomes firmly attached to the glass, so that it can be peeled away without danger of sliding the medium laterally and destroying the colony. Plates should be inoculated with the rounded tip of a glass rod, to avoid scratching the surface, they should be perfectly dry, and free from bubbles and other irregularities. N: SUMMARY The examination of bacteria in dried, heat-fixed smears, stained by the i. # ^ y ^:-t ,j4 6 ^^7 ((5, 6, 7J n-prodiicai Jmm Lxpci uncntul L.ll AV^un//) Fig. 8 SURFACi; STRUCTURES 27 cocci to branched tilanicnts, a single bacterium may contain from two to a dozen relatively tiny cells. The partitions between these cells may break down in the course ot the autogamous processes which accompany sporulation, and at other times, but are usually found in the vegetative cells. The cell wall is permeable, it does not grow, but is secreted, in certain well-marked areas, by the semi-permeable cell membrane which underlies it. Measurements of electron micrographs of sectioned bacteria suggest that its thickness varies from lOO to 250 Angstrom units. The cell wall is difficult to demonstrate and is seldom observed in pre- parations stained by the usual methods of routine bacteriology. Under such conditions, the existence of the cross-walls is liable to pass unsuspected. The cross-walls are laid down, in the dividing cell, by cytoplasmic septa which stain well with basic dyes, and when the multicellular structure of: such bacteria goes unrecognised these basophilic septa are liable to be confused with nuclear structures and other cytoplasmic inclusions. The complex cellular structure of many bacteria has long been known but seldom adequately appreciated until Robinow (1945, „l Mi. A Fig. !» BEHAVIOUR OF THE CELL WALL IN DIVIDING STREPTOCOCCI A. Long-chained streptoccus. Division is by the production of cross-walls, each parallel to the last, exactly as in rough bacilli. B. Short-chained streptococcus. Division is by constriction of the cell wall. All stained by tannic-acid-violet • 3000. A multicellular structure, sometimes with as many as twenty cells, separated by cross-walls and cytoplasmic septa, in each bacterium, has been demonstrated in nearly all Gram-positive bacteria. An extreme multicellularity is found in the giant bacteria such as Caryophanon and Oscillospira. That the multicellularity of these bacteria is fundamental, and by no means a superficial subdivision of filaments by the irregular growth of septa, is shown by the observation of Tomcsik (195 1), that in Bacillus anthracis the characteristic division of the bacillus into four small cells extends also to the polypeptide capsule, in which lines of demarcation can be seen, corresponding to the positions o^ the cross-walls internally. SURFACE STRUCTURES 29 Until recently, very little was known of the composition of bacterial cell walls, although the evidence suggested that they resembled the cell walls of plants in containing a large polysaccharide fraction, together with lipid and nitrogenous components. Some parts of the lipid component probably belongs to the cell membrane. Holdsworth (1952(7, 1952/)) isolated a protein- carbohydrate complex from the cell wall of Corytiehacteriiiin diphthcriac, in mmxxmw 12 3 4 <50 XS) m 1 ® («)<£Si®( (Reproduced from the Journal of Genera! Microbiology.) CELL DIVISION IN COCCI A. Long-chained streptococcus, corresponding approximately to a rough bacillus. Division is by the production of transverse septa. B. Short-chained streptococcus, corresponding to a smooth bacillus. Division is by constriction of the cell wall, and each nuclear unit represents a chromosome pair (Chapter IX). C. Septate staphylococcus. Resembles A, except that each septum is produced at right-angles to the preceding one. This morphology is also typical of the Gram-negative " diplococci." D. Unicellular coccus. Division is by constriction, but the organism possesses a central, spherical nucleus (Chapter I\'). which the protein component differed in constitution from the intracellular proteins. Holdsworth claims that the carbohydrate fraction is an oligo- saccharide and polysaccharides have also been described from the same source. Histochemically the cell walls and cross-walls of C. diplitlieriac and many other species of bacteria can be demonstrated to contain polysaccharides, by the use of periodic acid. ((2, :i, 4) reproducftl from the Journal «/ (.cnernl M icrobiology). Fig. 11 surfac:e structures 31 Pcrlinps surprisingly, tlic lipid clement is an important factor in the maintenance of the ct>nsicierable structural rigidity <-^t the cell wall. If it is removed the strength of the structure is greatly reduced. Diftcrences exist between the nature and composition of die cell walls ot Gram-positive and Gram-negative bacteria. Those from Gram-negative genera have a higher lipid content and a much fuller range of amino-acids in the protein traction ; and their cytochemical behaviour more closely resembles that ot intact wool proteins, whereas Gram-pc^sitive bacteria resemble degraded wool proteins in this respect. These observations may serve partially to explain the greater apparent rigidity of the cell walls of Gram-negative bacteria, which can form relatively long filaments without the necessity ot the cross-walls which Gram-positive genera produce under the same circumstances. They are also very interesting from the immuno- logical viewpoint ; the great complexity of the cell wall proteins of Gram- negative bacteria is almost certainly correlated with their more satisfactory behaviour as antigens, whereas in the rough variants of Bactcriaccae the loss oi antigenic specificity is associated with a change to the septate morphology and mode of cell division characteristic of Gram-positive bacilli. In addition, Fig. 11 CELL ENVELOPES IN BACILLUS In " rough " septate bacilli the bacillus is usually divided centrally by a complete cross cell wall and subdivided by developing cross-walls at varying stages. The latter may stain as cell walls or cell membranes {i.e. basophilic) according to the stage of development. (1) B. cereus, cell walls by Hale's method. Demonstrating only mature cross-walls x 3000. (2) B. megaterium, cell walls by tannic-acid-violet. Cross-walls at various stages can be seen ; the most mature appear double, possibly because of tannic-acid complex deposited on their faces. < 4,500. (3) B. megaterium stained by haematoxylin. Developing cross-walls show as dark bars ; the cell wall does not stain. Such basophilic areas have frequently been mistaken for nuclei or portions of nuclear structures, x 4,500. (4) B. megaterium stained by acid-thionin. This dye is less specific for the nuclear structures than Giemsa and stains also the basophilic areas of developing cross-walls, which appear either as readily recognisable bars or else as dots between the nuclei. In the latter form they have been confused with mitotic centrioles. x 4,500. (5), (6) Cell walls and cell membranes stained simultaneously by the Victoria blue method of Robinow and Murray (1953). The bacilli {B. megaterium) are slightly plasmolysed and the cell membrane is retracted from the mature cross-walls. The developing cross-walls show as dark bars, apparently composed of, or surrounded bv, the same material as the cell membrane. *x 3,000. 32 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA Gram-negative cell walls alone show birefringence, and they cannot so readily be freed from the cell membrane by plasmolysis. Cytological studies of the cell wall, especially those by Robinow, Klieneberger-Nobel, Bisset and Morris, have owed much to the technique of Eisenberg (1910), perfected by Robinow (1945) of mordanting with tannic acid, which renders it possible to stain the otherwise resistant cell wall with basic dyes, and at the same time destroys the stainability of the cell proteins which enclose and mask the cross-walls and other internal structures. Probably, in addition, the cytoplasmic proteins are precipitated to some extent on the interior of the cell wall, which assists in rendering the outline of the walls stainablc by basic dyes. An even better staining method (Hale, 1953), employing phosphomolybdic acid as a mordant, probably acts in the same way, since these agents are alike in being protein precipitants. The fact that the mature cross-walls resist their action, whereas the basophilic septa are destroyed, constitutes soinc degree of confirmation for the hypothesis that the latter consist of protein whereas the former contain a large polysaccharide component. The cross-walls can be seen very easily in bacteria which have been crushed or disrupted and partially emptied of their cell contents. By electron micro- scopy it is only in such material, or in sections, that the septate structure of bacteria can be discerned. However, the desiccation of material prepared for this method of examination may cause the cellular structure of such markedly septate organisms as Mycohacterintn to manifest itself in the form of a row of granules, rather like their beaded appearance in Zichl-Ncelscn stained pre- parations. In electron microscope studies by workers apparently unaware of the septate structure of Mycohacterinw, these granules have been identified with a variety of intracellular structures, for example, nuclei, sap-vacuolcs, and mitochondria. It is not suggested that none of the structures described form part of these granular aggregates, indeed the nuclei must necessarily do so ; but the identification of a structure which comprises so high a proportion of any one of the relatively tiny component cells of these multicellular bacteria with any single cytoplasmic component, on the evidence of a fancied resemblance in an electron micrograph, is exceedingly difficult to justify. S U R h A C i: S T 1< U C T U R li S 33 By comparison of electron nncrographs of crushed or sectioned bacteria of septate and non-septate morphology an independent verification of these observations, made upon stained material, by classical microscopy, has now been estabHshcd. B: THE CELL MEMBRANE (2, 3, 4, 8, 9, II, 15, 17, iS, 20, 21, 22, 26, 27, 31, 39, 41, 42, 43, 46, 57, sS, 65, 66, 67, 71, 72, 77, 81, X3, 86, 87, 88, 91, 98, 102, 103, 104) Little is known of the structure of the cell membrane. Extensive studies have been made upon its physico-chemical behaviour, but diese are beyond the scope ot the present work. A few observations have been made upon its chemical constitution, but fewer than in the case of the cell wall. It appears to be a semi-permeable membrane about lOO Angstrom units in thickness. Unlike the cell wall, the underlying cytoplasmic membrane, which constitutes the osmotic barrier of the cell, is embarrassingly easily stained by almost any method. This applies especially to the cytoplasmic septa and the growing points, which are associated with the membrane. These structures are highly basophilic, so much so that they will stain well with many dyes supposedly specific for chromatin, mitochondria, reticulocytes, etc. The entire cell membrane has a high content o( nucleoproteins, but these are especially concentrated in the growing points, and it is interesting to consider the observation of Pijper (1938) that somatic agglutination occurs by the adherence of the bacteria at the tips of the cells. Not only is the cell wall very thin at this point, being in the process of formation, but one of the main aggregations of nucleoproteins in the cell membrane occurs immediately underneath, so that it would appear that the growing points, and presumably also the almost identical cytoplasmic septa, may be a major somatic antigen. The nucleic acids in the cell membrane are almost unquestionably res- ponsible for the ease with which the entire bacteria can be stained, and when tliis stainable cortex is viewed through the tips of the cell, the well-known optical illusion of " bipolar staining " is observable, sometimes reinforced by the appearance of the basophilic growing points. The nucleic acids are also claimed to cause the phenomenon of Gram-positivity, where it occurs (Henry SURFACE S T K U C T U U i: S 35 and Sraccy, 1943 ; Henry, Staccy and Tcccc, 1945). According to these authorities it is the dit^erence in prc^portion of ribose and deoxyribose nucleic acids which determines the Gram-reaction, the former being preponderant in Gram-positive bacteria. On the other hand Mitchell and Moyle (1950) attribute Gram-positivity to the presence ot a phosphoric ester, the occurrence of which is independent of the concentration of ribose nucleic acid. Hoftman (195 1 ) suggests that a difference in the tyrosine content of the pentose nucleo- proteins ot Gram-positive and Gram-negative bacteria might account for the difterence. As in other types oi cell, the cell membrane contains an element which stains with lipid-soluble dyes, and there is little reason to doubt that it is similar in most respects to the cell membranes which are, perhaps, the most important organs of any living cell, and which serve to insulate the cell contents from the effects of variation in the outside medium by the exercise oi a regulatory function upon the passage of dissolved material through their surtace. The actual demonstration ot the cell-membrane as a separate structure has been achieved in lysozyme digested bacteria by Weibull {1953 b,c) and Tomcsik and Guex-Holzer (1954) . In the last stages of dissolution, the protoplasts appeared as " ghosts," and the empty membrane survived momentarily. The behaviour of the cell membrane at cell division provides additional evidence that it is a positive structure, and not merely an altered surface or interface. The septa which initiate cell division are clearly resoluble, although surrounded on both sides by cytoplasm. The large quantity of nucleic acids contained in, or associated with the cell membrane may be a cause ot contusion. It is observed that, as in the case ot plant cells, the effect ot certain cytological procedures, for example formalin fixation or acid-hydrolysis, may be to increase the basophilia of the nucleus Fig. 12 ELECTRON MICROSCOPY OF CELL ENVELOPE MATERL\L Spirillum sp., wall of crushed cell. The outer layer is represented by the striated pattern in the upper part of the plate, the inner by the pattern of regular globules below, .\lthough not conclusive in itself, this appearance is in accordance with the view that the envelope 36 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA at the expense of that of the cell envelopes. ExperimentaUy, it can be shown that DNA of extrinsic origin may become adsorbed upon the bacterial cell envelopes in the form of granules of very deceptive appearance. The possibility of a transfer of nucleic acids between the nucleus and cell envelopes must always be taken into consideration whenever small, basophilic structures are under examination. Septa of a different and specialised type, with an active basophilic component are found in the mother-cells of Rhizohiimi swarmers. These fundamentally resemble the polysaccharide-complex cross-walls of Bacillus species and are lined on both sides with basophilic material resembling the normal, membran- ous septa, but greatly thickened. They appear to provide a secretory lining to the lumen of the cells wherein the tiny swarmers are formed. The mis- conceptions which have arisen from the appearance of these " barred cells " are discussed in a later section : (Chapter VII). A recent and very important study on the difterentiation of the cell wall, membrane and cytoplasm by Robinow and Murray (1953), should be read in the original by all who are interested in this subject. C; DEl'ELOPMENT OF THE CELL ENl'ELOPES (3, Ch 7, f<, 9, IS, i'"^, 19, ^9, 30, 64, 71, 72, Si, S6, S7, 92, 93, 94. 9S, 96, 97. 9^) The fundamentals of bacterial cell division described by Schaudinn (1902) 1903) have been well substantiated by later work. The two main morpho- logical types of bacteria, which correspond to the " smooth " and " rough " colony forms, divide by constriction of the cell wall and by the formation of cross-walls respectively. In the " rough " type, which is especially typical of the large. Gram-positive bacilli, the division of the cells by cross-walls proceeds more rapidly than does their complete separation, so that the coiled bands of filaments are formed which give the well-known " Medusa-head " colony appearance. The actual separation of r(uig]i bacilli docs ncn necessarily occur immediately after the completion of cell division. Nor, wlien it does occur, is it invariably the most longstanding cross wall which is split, although this is usuallv the case. Separation may occur in such a manner as regularly to S U R F A C E S r 1< U C T U R E S 37 produce one large and one small daughter bacillus, i.e. containing more and less cells. Filaments of a different type, without cross-walls, arc formed in a more elaborate, probably syngamous type of cell division, which occurs in many difterent bacterial groups. These filaments eventually fragment to form new bacterial units. In all forms of cell division, the new cell wall material, whether it appears as a constriction or as a complete cross wall, is secreted by a basophilic septum derived from the cell membrane, which appears across the cell as the first sign ot incipient division, exactly as in certain types of plant cell. The con- strictive ingrowths are secreted at the points of junction of the cell wall and cross-walls or septum, and in the septate bacilli and cocci it is here that the main growth oi the cell wall proceeds. The new surface is, as it were, passed outwards around the edge ot the internal division between the cells to the (Hitside surface. The work of Tomcsik (1954) confirms that the main growth of the capsule is also at this point, where a distinct thickening of the cell wall can be observed by his methods. A similar thickening can be observed also in sections of bacteria prepared in this laboratory. In the unicellular, non- septate types the new cell wall is formed at the point of division and also at the tips of the cell ; usually only at one tip, the growing point. The point of division becomes the growing point of one or both of the daughter cells, so that there is no fundamental difference between the septa and the growing points. In very young cultures division may be so rapid that a single bacillus is subdivided into three or four cells, by septa derived from the cell membrane, while the process of ingrowth of the wall is incomplete. The evidence of the growth of " smooth " bacteria, and the formation of the cell wall at a growing point at one tip is derived from several sources. The first is the arrangement of the surface structures and especially the flagella in electron micrographs of growing and dividing bacteria and germinating microcysts. In such cells the flagella appear progressively shorter towards one pole where the wall is relatively thin, flexible and electron-transparent. Frequently one daughter cell may be provided with a full quota of well- ' developed flagella, whereas the other has very short flagella or none. It 38 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA appears that newly formed cell envelopes develop their flagella by outgrowth over an appreciable period of time, and the progressive immaturity of the flagella towards the pole which exhibits a thin area of cell wall indicates that the wall itself is more recently formed at that end of the cell, and that in effect it grows from that pole. It is, in any case, logical to assume that a rigid yet " dead " structure, such as a cell wall, would form in this manner. It could not grow all over its surface, but ccuild approximate to doing so only by constantly tearing and reforming, of which there is no evidence. In those multicellular bacteria which, as described above, form the cell wall at a number of growing points at the junctions of the cell walls and cross-walls [i.e. at the poles of the individual cells), an appearance of uniform growth is given. Further information upon the subject of growing points is derived from the researches of Bergersen (1952) upon the effect of sub-lethal concentrations of chloramphenicol on Bact. coli. The bacteria developed typical basophilic concentrations in the cell membrane at various points, and from these grew irregular side-branches. As these branches enlarged the growing points remained at the tip so long as growth persisted, retaining their basophilic character indicative oi the presence of metabolically active nucleoprotcins. The growing points, like all such metabolically-active areas, are also centres of oxidation-reduction activity, and for this reason have been equated with mitochondria by some authorities. However, they bear nc^ resemblance to Fig. 13 CELL ENVELOPES The cell envelopes observable by electron microscopy. Owing to the opacity of the bacterial cell, little can be seen in most intact specimens under the electron microscope ; however, in sectioned or accidentally disrupted material some detail can be observed. (1) Bacillus cereus, section showing one complete cross-wall and a second at an earh' stage of ingrowth. The inner edge of the incomplete septum is lined with material which probably represents the cytoplasmic septum preceding and secreting tlie new cell wall material, but which may be coagulated by the osmium tetro.xide with which the preparation has been treated. .10,000. (2) Cell wall of partially disrupted Spirillum sp., showing striated structure and blepharo- plasts at the bases nf the compound flagella. ■: 20,000. (3) Cell envelopes of jiartially disrupted Pseiidomonas sp. ; the cell wall is transparent and apparently simple ; the blepharoplasts are visible at the bases of the flagella ; the emergent protoplast still appears to be covered by a membrane. ■' 10,000. (4) Where the structure is sufficiently small to be electron transparent the cross-walls can be demonstrated in the intact organism, as in the stalks oiCaulobactcy, which are extensions of the body wall, x 15,000. ( (1) KtpHiduced from the Journal of Bacteriology by permission of Drs. G. B. Chapman and J. Hilliet) ( 1-2.) Reproduced by permission oj Miss P. E Pease) Fig. 13 40 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA mitochondria in form, arrangement, or, so far as is known, in function, and this interpretation has been severely criticised. D: THE CELL IVALL OF MYXOBACTERIA (54, 62, Sy, 90) The myxobacteria dither from most other bacteria in that they lack the rigid cell wall, and are independent of flagella for motility. No cell wall whatever can be detected in myxobacteria by ordinary staining methods, but its existence, and some of its structural characteristics, can be interred from other considerations. Although the cells exhibit some degree of flexibility, their structure is sufficiently rigid to enable them to retain the bacillary form when immersed in fluid. Such strength would unquestionably not be possessed by the improtected cell membrane. Myxobacteria are motile only when in contact with a surface, whether a solid surface c^r the surface film of a fluid. They show a marked inclination to move along the lines of physical stress in the surface, a phenomenon which has been described as elasticotaxis. Their mode of progression has been variously described, but appears to the author to be a worm-like action analogous to peristalsis. Flexion is occasionally shown but probably is not a necessary function of locomotion. This implies a muscular activity in the cell wall, which must be capable of contracting circumferentially, to extend the cell, and also longi- tudinally, to shorten and expand the cell. Muscular action in so small an organ is not exceptional and is obvious in the locomotory cilia and flagella of many small creatures, including bacteria. It may therefore be presumed that the cell wall o{ myxobacteria comprises a longitudinal and circular system of contractile fibres, which enable it to perform the flexion and the peristaltic action which can be observed. It may be regarded as reasonably certain that such contractile fibres are composed of protein, which serves to explain the difference between the chemical reactions of the myxobacterial cell wall and of the polysaccharide structure which is common to most other bacteria. The microcysts of myxobacteria possess a rigid cell wall which more closely resembles that of eubacteria. SURFACE STRUCTURES 4I E: THE SPORE COAT (23,60,81,82) The resistance of the bacterial spore to inimical agencies has been attributed by some authorities to the impermeable spore coat, and by others to the peculiar condition of the cytoplasm of the spore. While the spore coat is difficult to stain, it is no more so than the cell wall of vegetative bacteria, from which it docs not appear to differ markedly in any respect. Like the cell wall, the spore coat may be partially destroyed by hydrolysis with hydrochloric acid, and it is probably composed mainly of polysaccharide material. Electron micrographs ot sections ot spores have shown that the spore coat is a single structure in Bacillus ccrciis, whereas in B. iiwoatciitini at least two layers can be discerned (Robinow, 1953). The ejection of the turgid nuclear material from hydrolyscd spores indicates that the spore coat possesses a small area ot weakness, possibly a germination pore. F: THE NUCLEAR MEMBRANE (5, 13, 14, 30) The apparent absence of a nuclear membrane in bacteria has many times been remarked upon in cytological literature. This is probably due to the fact that bacteria have been examined mainly in their period of active, vegetative growth, when the nucleus is permanently in the chromosomal condition. The resting nucleus has a membrane ot normal appearance. Electron micrographs of sections ot bacterial nuclei have shown signs ot a fibrous layer surrounding the denser portion of the nucleus. G; SLIME LAYERS AND CAPSULES (34, 56, 88, 92, 93, 94, 95, 96, 97, 98) Many bacteria possess a surface layer of mucoid material, and occasionally a well-defnied capsule. The two structures have often been confused, but ' are distinct and may be found simultaneously. Fig. U BACTERIAL FLAGELLA Spirillum serpens Gold-shadowed electron micrograph, showing flagella passing through the cell wall to the protoplast. suRi'ACH stru(;turi:s 43 The phasc-Lontrasr studies ot TcMiicsik, and Toincsik and Ciuex-Holzer have revohitioniscd the previous concept of bacterial capsules as an amorphous layer of polysaccharide or polypeptide mucus, by demonstrating that they may possess an exceedingly ccMiiplex structure of alternate striations of poly- saccharicie and polypeptide, perpendicular to the cell wall ; that is parallel to the cross-walls, which are marked by exceptionally massive lamina, as arc the poles of the bacilh. The growth of the capsule, like that of the cell wall, occurs mainly at the junctions ot the cell wall and cross-walls. H: FLAGELLA (15, 16, 21, 22, 3S, 47, 4.S, 50, SI, S3, <^'3, 77, 7^, 79, 99, roo, loi, 103, 104) Motility in the great majority of bacterial groups is by means ot flagella. These number from one to several hundred. They may be arranged singly or in small groups at the poles of the cell, in which case they render their possessor very actively motile in a fluid mcdiimi, or they may be arranged in larger numbers peritrichously, which is probably an adaptation to movement in viscous media or on moist surfaces (Chapter IX). The polar flagella are approximately 26 mi in diameter, in I 'ibrio and Psciidoniotias ; the peritrichous flagella may be as small as 19 ni// in Proteus. Because of their small size their mode of action is difficult to determine. They have been described as lashing, but more probably act by waves of contraction passing down their fine coils. The wavelength of these coils varies from 2 to 3 /x and is constant for any bacterial species, but variants in a single strain may have flagella c^f double the normal wavelength. Flagella ponit away from the direction of motion, and the rearmost may become twisted together into a spiral thread. Cast-oft flagella, in fixed preparations, also tend to knit up into whips in this manner. The flagella originate m the cell membrane or surface cytoplasm and pass outwards through the cell wall. When the cell wall is removed by digestion with lysozyme, the flagella remain attached to the protoplast. Their point of origin is a basal granule or blepharoplast, approximately spherical and rather larger in diameter than the flagellum which arises from it. In most bacterial < genera each flagellum arises from a separate granule, but the flagellar fibrils 44 THE CYTOLOGY AND LIFE-HISTORY OF ACTERI A of Spirilla, each of which is about half the diameter of a typical, iinfibrillar flagellum, arise in bundles from single granules, so that each bundle constitutes a compound flagellum. This is considered to represent a primitive condition, intermediate between that in typical bacteria and in the flagellate protista (Chapter IX). Fig. 15 BACTERIAL BLEPHAROPLASTS A. Electron microfi;raph of Spirillum sp. showing two flagella attached to a blepharoplast. B. V ihrio cholerae with monfibrillar flagellum and its blepharoplast. [Sec also figs. 13, 16,57). The microcysts and spores of bacteria are devoid of flagella. These commence to grow at germination ; usually at the pole remote from the growing-point of the cell. Because of their muscular activity, and because they are complete antigens, flagella are almost certainly composed of protein in all cases. In Proteus they have been shown to consist of a fibrous protein resembling myosin. By comparison with the contractile muscle protein actomyosin, however, it lacks the sulphydryl groups which play an important part in this complex. The possession of flagella is the most important single factor suggesting relationship between bacteria and the flagellate protista, rather than with the blue-green algae, as is often suggested, and much of the argument concerning evolutionary relationships between bacteria is based upon the evidence of these structures. R,t»aduic-d JioiH the J^unud ,,t hc,ui.:I Mi Fig 16 {See Legend on page 47) Fig. 17 (Reproduced from Hu- Journal of General Microbiology). SURFACE STRUCTURES 47 /. SUMMAR Y Bacteria possess a cell wall oi great strength and rigiditv, overlving, and secreted, in certain well-marked areas, by the underlying, senii-pernieable cell membrane. Bacteria may be subdivided by cross-walls into a number of cells varying from two to twenty or more. In unicellular bacteria the main growth of the cell wall is at the tips ; in septate forms, the main growth is at the junction of cell wall and cross-walls. These sub-divisions and this mode of growth extends also to the capsule, which may be verv complex, with polysaccharide and polypeptide lamina. The cell wall contains polysaccharide, lipid and protein elements. The chemical composition of the cell wall is more complex in Gram-negative than in Gram-positive bacteria. The cell membrane contains lipid, protein and nucleo-protein elements. It gives rise to the flagella. In cell division a septum derived from the cell membrane precedes and secretes the cross-wall or ingrowth of the cell wall. These septa and the growing points which secrete the cell wall are basophilic and physiologically active. There is evidence that they constitute a major somatic antigen. They Figs. 16 and 17 DEVELOPMENT OF FLAGELLA Development of flagella in the germinating microcyst. The resting cells of flagellated bacteria are devoid of flagella ; on germination these develop first at the poles of the cell, especially at the pole remote from the growing point. Electron micrographs, gold-shadowed. (1, 2, 3, 4) Salmonella typhi, stages in the germination of the microcyst. (I) Microcyst, without flagella. x 30,000. (2, 4) Young vegetative cells with short flagella concentrated towards one pole of the cell. ■ 9,000 and x 7,000. (3) Germinating microcyst with very short flagella towards both poles. ■ 27,000. (5, 6, 7) Bacfeniim coll, stages in the germination of the microcyst. (5) Germinating microcyst with two very short flagella originating at one pole of the cell from an obvious blepharoplast. > 20,000. (6, 7) Two polar flagella further developed. ■ 16,000. (8, 9, 10) Development of flagella demonstrated by silver impregnation stain. 3,000. (8) Proteus, germinating microcyst. (9, 10) Sal. typhi, very young cells w^ith sub-polar flagella. (II) Germinating microcysts of Pseudomonas fluorescens. The flagella emerge more closely together than in the case of the foregoing examples, which will e\entually become 16,000. THE CYTOLOGY AND LIFE-HISTORY OF ;acteria may also be confused with cellular inclusions and have given rise to a variety of misconceptions. The cell wall of myxobacteria is flexible and is an organ of motihty. Spore coats resemble the cell walls and may be single or multiple. The flagella are organs ot motility, they are spiral and composed of fibrous protein. They arise from blepharoplasts in the cell membrane. In the germinating microcyst the flagella develop first at the pole remote from the growing point of the cell. BIBLIOGRAPHY (i) Angelico, R., Calo, a., D'Amore, A., Mariani, A., Mariani-Marelli, O., ScANGA, F. (1952) Rend. Inst. Sup. di Sanita. 15. 627. 2) Bartholomew, J. W. and Mittwer, T. (1951)). Gen. Microbiol, s- 39- 3) Bergersen, F.J. (1952) Proc. Univ. Otago Med. Sch. 30. 3. 4) Bielig, H. J., Kausche, G. A. and Haardick, H. (1949) Zs. f Naturforschung. 46. 80. 5) Birch-Andersen, A., Maaloe, O. and Sojstrand, F. S. (1953) Biochim. Biophys. Acta. 12. 395. 6) Bisset, K. a. (i93S)J. Path. Bact. 47. 223. 7) Bisset, K. A. (1939) J. Path. Bact. 48. 427. 8) Bisset, K. A. (1947) J. Gen. Microbiol. 2. 83. 9) Bisset, K. A. (1948a) J. Gen. Microbiol. 2. 126. o) Bisset, K. A. (1948b) J. Gen. Microbiol. 2. 248. i) Bisset, K. A. (1948c) J. Hyg., Camb. 46. 173. 2) Bisset, K. A. (1949) J. Gen. Microbiol. 3. 93. 3) Bisset, K. A. (1950a) Exp. Cell. Res. i. 473. 4) Bisset, K. A. (1950b) J. Gen. Microbiol. 4. 413. 5) Bisset, K. A. (1951) ibid. 5. 155. 6) Bisset, K. A. (1952) Bacteria. Edinburgh. Livingstone. 7) Bisset, K. A. (1953a) Nature, Lond. 171. iiiS. 8) Bisset, K. A. (1953b) Symp. Bcut. Cytoi l-lonic. VI. hit. Cong. Microbiol. 9) Bisset, K. A. (1954a) J. Bact. 67. 41. 20) Bisset, K. A. (1954b) Exp. Cell. Res. 7. 232. 21) Bisset, K. A. and Hale, C. M. F. (1951a) J. (len. Microbiol. 5. 150. 22) Bisset, K. A. and Hale, C. M. F. (1951b) J. Gen. Microbiol. 5. 592. 23) Bisset, K. A. and Hale, C. M. F. (1951c) J. Hyg., Camb. 49. 201. SURFACE STRUCTURES 49 BissET, K. A. and Hale, C. M. F. (19.S3) Exp. Cell. Res. 5. 449. BissET, K. A. and Moore, F. W. (1949) J. Gen. Microbiol. 3. 3S7. Booij, H. L. (1949) Dis. Faraday Soc. 6. 143. BuRDON, K. L. (1946) J. Bacr. 52. 66s. Cassel, W. a. (195 1 ) J. Bact. 62. 239. Chance, H. L. (1952) Stain Tech. 27. 2_S3. Chapman, G. B. and Hillier, J. (1953) J. Bact. 66. 362. Chayen, J. and Norris, K. P. (1953) Nature. 171. 472. Dawson, I. M. (1949) Symposium: "The Nature of the Bacterial Surface," Oxford, Blackwcll. Dawson, I. M. and Stern, H. (1954) Biochim, Biophys. Acta. i}. 31. DuGUiD, J. P. (1951) J. Path. Bact. 63. 673. Eisenberg, p. (1910) Zbl. f. Bakt. I. 53. 481. Fischer, R. and Larose, P. (1952). Canad. J. Med. Sci., 30, cS6. Grace, J. B. (1951)]. Gen. Microbiol. 5. 519. Grace, J. B. (1954)). Gen. Microbiol. 10. 325. Harris, J. E. (1948) New Biology. 5. 26. Hale, C. M. F. (1953) Lab. Practice. 2. 115. Henry, H. and Stacey M. (L943) Nature. 151. 671. Henry, H., Stacey M.and Teece, E. G. (1945) Nature. 156. 720. Hoffman, H. (195 i) Nature. 168. 464. HoLDSWORTH, E. S. (i952a) Biochim. Biophys. Acta. 8. no. HOLDSWORTH, E. S. (1952b) Biochim. Biophys. Acta. 9. 19. HouwiNK, A. L. (1953) ibid. lo. 36. Houwink, a. L. and van Iterson, W. (1947) ibid. i. 527. HouwiNK, A. L. and van Iterson, W. (1950) ibid. 5. 10. Hurst, H. (1952)}. Exp. Biol. 29. 30. Iterson, W. van (1947) Biochim. Biophys. Acta. i. 527. Iterson, W. van (1953) Symp. Bact. Cytol. Rome. IV. Int. Cong. Microbiol. Jarvi, O and Levanto, A. (1950) Acta. Path. Microbiol. Scand. 27. 473. KiNGMA BoLTjES, T.J. (1948)}. Path. Bact. 60. 275. Klieneberger-Nobel, E. (1947a) J. Gen. Microbiol, i. 22. Klieneberger-Nobel, E. (1947b) J. Gen. Microbiol, i. 33. Klieneberger-Nobel, E. (1948) J. Hyg., Camb. 46. 34s. Knaysi, G. (1942) J. Bact. 43. 365. Knaysi, G. (1946) J. Bact. 51. 113. Knaysi, G. and Baker, R. F. (1947) J. Bact. 53. 539. Knaysi, G. and Hillier, J. (1949) ibid. 57. 23. Knaysi, G., Hillier, J. and Fabricant, S. (1950) J. Bact. 60. 423. Krzemieniewski, H. and S. (1928) Act. Soc. Bot. Pol. 5. 46. Leifson, E. (195 1 ) J. Bact. 62. 377. Martens, P. (1937) La Cellule. 46. 357. Mitchell, P. (1949) Svmposium : " The Nature ot the Bacterial Surface." Oxford. Blackwell. 50 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA (73 (74 (75: (76: (77: (7« (79 (8o (,Si (^2 («3: (84' (^s: (86 («7' (88 (89 (90 (91 (92 (93 (94 (95 (96 (97 (98 (99 (too (toi (102 (103 (104 (105 Mitchell, P. and Moyle, J. (1950) Nature. 166. 21S. Mitchell, P. and Moyle, J. (1954) J. Gen. Microbiol. 10. 533. Morris, E. O. (1951a) J. Hyg., Camb. 49. 46. Morris, E. O. (i9_sib) J. Hyg., Camb. 49. 175. Morris, E. O. (1952) Chem. and Indust. 120. MuDD, S., Brodie, a. F., Winterscheid, L. C, Hartman, P. E., Beutner, E. H. and McLean, R. A. (1951) J. Bact. 62. 729. MuDD, S., Winterscheid, L. C, DeLamater, E. D. and Henderson, H. ]. (1951) J. Bact. 62. 459. Murray, R. G. E. and Robinow, C. F. (1952) J. Bact. 63. 298. Pennington, D. (1949)}. Bact. 57. 163. Pennington, D. (1950)}. Bact. 59. 617. Peshkoff, M. a. (1940) J. Gen. Biol. (Riiss.) i. 598. Pijper, a. (1938) J. Path. Bact. 47. i. PijPER, A. (1946) J. Path. Bact. 58. 325. Pijper, A. and Abraham, G. (1954) J. Gen. Microbiol. 10. 452. Pringsheim, E. G. and Robinow, C. F. (1947) J. Gen. Microbiol, i. 267. Robinow, C. F. (1945) Addendum xo : The Bacterial Cell. Dubos, R. f. Harvard Univ. Press. Robinow, C. F. (1953)]. Bact. 66. 300. Robinow, C. F. and Murray, R. G. E. (1953) Exp. Cell. Res. 4. 390. Salton, M. R.J. (1952) Biochim. Biophys. Acta. 9. 334. Salton, M. R.J. and Horne, R. W. (1951) Biochim. Biophys. Acta. 7. 177. Schaudinn, F. (1902) Arch. Protistenk. i. 306. Schaudinn, F. (1903) Arch. Protistenk. 2. 421. Stagey, M. (1939) Symposium : " The Nature of the Bacterial Surface. " O.xford, Black well. Stanier, R. Y. (1942a) Bact. Rev. 6. 143. Stanier, R. Y. (1942b) J. Bact. 44. 405. Thornton, H. G. and Gangulee, N. (1926) Proc. Roy. Soc. B. 99. 427. ToMCSiK, J. (1951) Experientia. 7. 459. ToMCSiK,J. (1954) Mod. Prob. Paed. i. 410. ToMCSiK,J. and CiuEX-HoLZER, S. (1951) Schvveiz. Zeit. f. Allg. Path. u. Hakt. 14. 515. ToMCSiK,J. and Guex-Holzer, S. (1952) ibid. i_s. 517. ToMcsiK, J. and Guex-Holzer, S. (1953) ibid. 16. 882. ToMCSiK, J. and Guex-Holzer, S. (1954a) J. Cien. Microbiol. 10. 97. ToMCSiK, J. and Guex-Holzer, S. (1954b) ibid. 10. 317. Weibull, C. (1948) Biochim. Biophys. Acta. 2. 351. Weibull, C. (1948) ibid. 2. 351. Weibull, C. (1950) Act. Chem. Scand. 4. 268. Weibull, C. (1953a) J. Bact. 66. 137. Weibull, C. (1953b) ibid. 66. 688. Weibull, C. (1953c) ibid. 66. 696. WoN(;, S. C. and Tung, T. (1940) Proc. Soc. Exp. Biol. Med. 43. 749. CHAPTER IV The Bacterial Nucleus A: HISTORICAL (i, 3, 5, 10, 14, 15, t6, 17, 12, 24, 2S, 42, 43, 44, 45. 50, sr, 5^, 57, 5>^, 59) THE existence ot the bacterial nucleus has long been denied, mainly upon the evidence that it cannot readily be demonstrated in prepara- tions fixed and stained according to standard bacteriological procedures (Chapter II). Good descriptions of the nuclear apparatus, as it is now believed to exist, have been published from time to time, but have been ignored by almost all bacteriologists. It is significant that the observations upon eubacteria, made in the last decade, were preceded by an interest, in widely separated parts ot the globe, in the cytology of myxobactcria. These micro-organisms cannot readily be studied in heat-fixed smears, and so received the cytological treatment usually denied to eubacteria. Also their nuclear material can clearly be demonstrated by simple staining techniques, which that of eubacteria often cannot. The application, to other fields of bacteriology, of the information obtained from the study of myxobactcria, gave considerable stimulus to those minds which had difficulty in accepting the defeatist views upon bacterial cytology, which had been current for so long. A revolution in the study of the bacterial nucleus resulted from the adoption of the technique of acid hydrolysis, as a preliminary to staining. This was originally applied in the process o{ the Feulgen reaction for nucleic acid, which was used successfully by Stille and by Piekarski in 1937, and by many others at about the same time. Piekarski also discovered that after acid hydrolysis the nuclear structures stained clearly with Giemsa. This technique was adopted by Robinow (1942), whose beautiful photomicrographs attracted V % % (Reproduced frim the Jnurnal of Ceneral Microbiology hy permission of Mr A. A. Tuffery). Fig. 18 CYTOLOGY OF OSCILLOSl'IRA Because of its relatively great size, an exceptionally clear picture of the bacterial nucleus is afforded by the giant bacterium Oscillospira. In (1) the twin, rod-shaped nucleoids are shown both in profile and endwise. (2) Shows their arrangement within small, disc-like cells. In (3) and (4), a, b and c represent progressively earlier stages in the maturation of the spore and condensation of its nuclear content. (Acid-Giemsa, •: 1,000). TH E B A CTE R I A L N U C L K U S 53 considerable attention. Robinow's description of the paired, chromosome- like nuclear bodies was not the first to be made, but was the first to obtain general credence. An easily available but little known paper by Paillot (1919) shows the paired bodies clearly stained by Giemsa, and other examples might be quoted. Observations were also made by ultra-violet light and by electron micro- scopy which, it less striking in themselves, served to confirm those made by the acid-Gicmsa technique. Another valuable staining method, the methylene-blue-eosin technique was devised by Badian (1933) and used with success by subsequent workers. Two interesting studies were performed by a method of vital staining with fuchsin, by Stoughton (1929, 1932) and Allen et al. (1939). These papers stand rather apart from the main lines of discovery in this subject, because, although they include observations ot great interest, fully substantiated by photomicrographs, the cytological processes described arc unlike those more commonly found, in some particulars. These papers will be discussed in Chapters VI and VII. Claims have from time to time been made to demonstrate a classical mitotic process in the bacterial nucleus, but these have never proved acceptable to experienced bacterial cytologists. Almost without exception they have been based upon an exceedingly small range of observations, and the most recent (DeLamater and Mudd, 195 1 et scq.) has been supported by the constant (unacknowledged) republication of a single " metaphase spindle," at different magnifications, and sometimes inverted. This work has also been heavily criticised upon technical grounds. B: THE RESTING NUCLEUS (2, 4, 5, 10, II, 12, 13, 18, 19, 20, 21, 25, 26, 27, 31, 32, 33, 34, 35, 36, 40, 41, 42, 47, so, 52, 53, 59, 60) The bacterial nucleus, like those of other types o( cell, may appear in a variety of different guises. It is probably even more protean than most, but the changes of form which it undergoes are paralleled by similar processes 3 V. •^^% 6r« ♦ > it >♦ THE BACTERIAL NUCLEUS S5 wliich have been observed in algx and tungi, or even in more complex creatures. The form ot nucleus which is usually regarded as the standard equipment oi" a cell, a roughly spherical, vesicular structure, is found in most bacteria at some stage of their life-history, and in some at all stages. This form of nucleus was not the fu-st to be described ni bacteria, nor is it the easiest to demonstrate. Frequently it occurs in resting conditions of the cell, when metabolic activity is low, and the nucleoprotein content, upon which " nuclear " staining reactions depend, is considerably reduced. Hence, in the Bactcriacea', its presence was undetected tor some years after the appearance of the active nucleus, in this type oi bacterium, was well recognised. In those bacteria which possess spherical nuclei in the active condition it is more readily demonstrable, stains clearly and apparently contains its full quota of: nticlcoproteins. It is found in the active form in some, although by no means all cocci, in the small cells which comprise the bacillary forms of corynebacteria and mycobacteria, and in Azotohacter. In the small bacteria it appears spherical and homogeneous, but in A^orohactcr, which is considerably larger, it may be seen to possess a vesicular structure, consisting of an unstained vacuole surrounded by chromatinic granules. It may be supposed that the same structure would be found in the nuclei of the smaller cells, were it possible to resolve them with the microscope. There is evidence that not all these granules are cytochemically identical. Fig. 19 THE VEGETATIVE NUCLEUS The vegetative nucleus of bacteria appears typically in the form of paired rods, dividing reductionally, and usually lying at right-angles to the long axis of the bacterium. (1-4) Stained by Acid-Giemsa. (1) Shigella flexneri, x 5000. (2) Bacillus wiegaterium, x 3000. (3) Bacillus cereits, very slightly hydrolysed, showing the relatively small appearance of the nuclei and residual basophilia in the cell envelopes. ■ 3000. (4) Caryophanon latiiiii, ■; 3000. (5) Micrococcus cvyophilus, a markedl\- multicellular Gram-positive coccus, with one nuclear body in each cell. These appearances have been confused with mitotic figures by workers ' xmaware of the septate structure of the coccus. Trichloracetic acid and Giemsa, x 5,000. « t . t 1^ ^-^ m •% '%'■ »^Qft. -.4 10 {Rcproiiuci-d Iron, llir Jiiunuil nf Ccnnal Microbiology). Fig. 20 ACTERIAL NUCLEUS 57 The nuclei of mycobacteria were described as Feulgcn-positive granules, regularly arranged along the length of the bacillus, before it was realised that the bacillus is multicellular, and that each granule was a cell nucleus. Some confusion has also resulted from identification of these granules with those Fig. 21 EFFECTS OF HYDROLYSIS ON THE SPORE NUCLEUS a. Spore with nucleus in natural position. b. The " crescentic nucleus." The nuclear material forms a pool between the cytoplasm and spore coat. c, d. The " peripheral nucleus." The spore coat is bulged outwards by the ejected nuclear material. e. Complete ejection of nuclear material. which appear in the well-known granular or beaded effect seen in heat-fixed preparations of M. tuberculosis. The latter arc, in fact, merely the shrunken cell contents. The metachromatic granules of C. diplitlieria' have also been identified with nuclei by some workers, and disproved by others. They are not seen except in dried preparations, and are artefacts consisting of an aggregate of stainable material within the larger, terminal cells of the bacillus. Fig. 20 MICROCYSTS OF BACTERIACEM The appearance of the resting cell and resting nucleus may be very distincti\e, even in bacteria of which the vegetative stages are similar. Acid-Giemsa, < 3000. (1, 2) Bacterium coli. Small, oval cells with an eccentrically staining nucleus. Proteus and most Salmonella are similar. (3, 4) Bacterium aerogenes, much larger, with a small, central nucleus. ((4) is stained by methylene-blue-eosin) . (5-7) The large microcysts of 5. typhi. < (8) Shigella schmitzii, large oval cells with a central nucleus. * % » jr H> ^4* • «^ 2 Cv5es<^j »; !'>/'.<.♦* *'0 THE BACTERIAL NUCLEUS 59 Coccal genera which possess this type e>f nucleus may often form short filaments ccMitaining hnir or five nuclei. These filaments are entirely distinct from the reproductive filaments formed by many bacteria, including strepto- cocci, which possess nuclei of the chromosomal type (Sections C and D below). Where the spherical nucleus is found only ni the resting stages of the bacterium it is often more obviously vesicular and stains eccentrically. The main body of the nucleus stains poorly or not at all, and may be difficult to distinguish from the cytoplasm of the cell. The stainable portion of the resting nucleus is sometimes merely a crescentic portion of the outline, but may take the form of a single or double spherical body at one edge. Often this is the only portion of the nucleus which can be resolved, and the appearance is of a small, double, spherical nucleus lying eccentrically in the cell. In the case of myxobacteria and cytophagas the stainable portion may be tadpole-shaped. The spore nucleus resembles other resting nuclei, but appears to be in a condition of turgour (possibly due to concentration of the protein materials, Chapter IX). The effect of acid-hydrolysis is to weaken the spore-coat, so that the nuclear material may be forced from its natural position to lodge as a pool of basophilic material at the periphery of the cell. The appearances THE SPORE NUCLEUS The bacterial endospore has a vesicular resting nucleus, typical of such nuclei, except that it appears to be in a condition of turgour when mature. Under the influence of acid- hydrolysis processes the nuclear material may be partly or completely ejected, giving the various appearances which have, in the past, been described as " peripheral " or " crescentic " nuclei. The immature spore nucleus does not behave in this fashion. The weak spot, through which the nuclear material may be ejected, possibly represents a germination pore. (1) Spores of Bacillus sp., treated with N/1 nitric acid for 5 minutes, stained Giemsa and re-stained tannic-acid-violet, to demonstrate that the ejected nuclear material is outside the spore coat. One spore has retained it within the spore coat and shows the " peripheral nucleus." -5000. (2) Spores of B. siibtiHs. showing the spore nucleus in its natural condition. Acid-Giemsa. X 3000. (3, 4) Spores of Clostridium u\=lchii. Nitric acid for 10 minutes, stained crystal violet. All types of appearances seen ; a, nucleus in natural condition ; b, " crescentic nucleus " ; £, " peripheral nucleus." Other sporeg are in intermediate stages, x 3000. (5) Maturing spores of CI. welchii, as in (3, 4), showing no change of position of nucleus. (6) As (1), after several hours hydrolysis. (7) As (I), electron micrograph, gold-shadowed. ,-, 16,000. 60 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA resulting from this reaction have been variously described as " extracyto- plasmic," " peripheral " or " crcscentic " nuclei, but they are now generally agreed to be artefacts. The identity of the spherical nucleus, as found in active cultures of some genera, with that which is confined to the resting stages of most types of bacteria, whose active nucleus is chromosomal, is not certain. Bacteria, whose active nucleus is spherical but small, may possess a resting nucleus which is larger and more obviously vesicular (Chapter VI). In the cells of most classes of living organism the nucleus returns from the chromosomal to the resting condition between each division, but in bacteria the mitotic condition of the vegetative cell may be retained throughout the period of active reproduction, and the resting nucleus is restored only when active reproduction ceases. The active condition of the nucleus is so much more readily demonstrable that it has been supposed that the nuclear material preserved an organised form only in yoimg cultures, and became disintegrated and distributed throughout the cytoplasm when cultures were more than a few hours old. This, however, is a fallacy. Fig. 23 APPEARANCES OF THE NUCLEUS (1) Vesicular vegetative nucleus in a Gram-negative coccus. Methylene-blue-eosin. (2) False appearance of vesicular nucleus in multicellular coccus, .\ctually the nuclear material of several cells is condensed centrally. Methyl-violet-nigrosin. (3) As (1) in Sarcina sp. (4), (5) Nuclear bodies in Mycobacterium lacticola and in Xdcui'diii sp. These appear spherical, but this may be because they are too small to be resolved. (6) Appearance of vesicular nucleus in acid-hydrolysed cells of Aerobacter sp., probably due"Ho laking of the stain. All plates at / 3000. # # ^ ■«• •• • % * \- 1 :, ••- \ « -.• • /^^, > ft • 7' Fig 23 62 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA C; THE PRIMARY I'EGETATIVE NUCEEUS (5, 6, 7, 9, 10, 12, 14, 27, 30, }2, 33, 3S, 43, 44, 45, 46, 4S, 50, 51, 52, 60, 61) In the young cultures of most cubacteria, niyxobacteria and such chlaniy do- bacteria as have been described, as well as in the primary mycelium ot strepto- myces, the nucleus is found in what is sometnnes described as the primary form. It consists of a pair of short rod-like bodies, sometimes slightly broader at the ends than the centre, lying transversely to the long axis of the bacterium, and occupying almost its entire width. They are termed chromatinic bodies or chromosomes, although their exact identity with tlie chromosomes of plant and animal cells is dubious. They divide with the cell, splitting longi- tudnially in the manner of chromosomes. In consequence of this mode of division the pairs may he parallel to one another or at a shght angle. They arc sometimes so close together as not to be resc»luble separately by the microscope, and sometimes quite widely separated. The chromosomes were originally described as single, spherical bodies, and this description is still applied to their appearance in electron micrographs, and by ultra-violet light. Although the bodies appear rod-shaped, they show a marked tendency to present the long axis of the rod to the observer, so that it has been suggested that they are, in fact, disc-shaped or in the form of a short, spiral band. This is to some extent supported by the observed form of the nuclei of the large micro-organism Caryopluvioii liitiii}i, which resembles bacteria in many of Its morphological attributes. Its nuclei are ring or disc-shaped, and lie, as a rule, in a plane transverse to the long axis of the bacillus-like organism. An even more remarkable demonstration of a primary vegetative nucleus m tlie form of unequivocal transverse rods is given, however, by Oscillospira, another member of the same group, and even larger than Caryophamvi. The con- sensus of opinion is that the chromatinic bodies are, as they appear to be. Figs. 24 and 25 SECTIONS OF BACTERIAL NUCLEI Electron micrographs of ultra-thin sections of Bacterium coli. The material has probably suffered some distortion in the process of fixation in osmium tetroxide solution and embedding in synthetic resin, but nevertheless shows the absence of cross-walls in this type of bacterium, and the nuclear bodies in the form of short rods in both longitudinal and transverse sections. Exactly as in stained preparations, the nuclei appear singly or in pairs towards each end of the cells, or in larger numbers in the filaments. ^ '^ t/4 [HtpiiiduLtd In, in lliiulnnii,,! ,7 J Imf'livsici Act^i. hv pri Fig 25 64 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA short rods. And in fact they occasionally appear as though viewed troni the end, although not so frequently as might be expected in such a case. The figures of Robinow and Kliencberger-Nobel occasionally show the pairs of rods crossed. In the experience of the author this is exceedingly uncommon, but if the chromosomes are in fact rods there is no theoretical reason why they should not appear crossed. Genetical evidence also supports this concept of a reductionally dividing nucleus (Witkin, 195 1). The slightly dumbbell-shaped appearance of the chromosomes is usual, although not invariable. It is not indicative of division in the transverse plane, as might, perhaps, be expected. Division of the chromosomes is invariably by longitudinal splitting, and it is obvious that if these bodies carry the genes of the cell, arranged in a linear manner, as from genetical considerations must necessarily be true, then their division cannot take place hi any other way. The nuclear unit of the vegetative cell is a single pair of chromosomes. In bacteria of unicellular, smooth morphology a pair is disposed at each end of the cell, but it is probable that both pairs are of identical genetical con- stitution (Chapter X). In very young cultures of bacteria of this morphology the cells often contain only a single pair of chromosomes, and the bacteria mav contain from one to four or six cells (Chapter III). Each of the four cells of a rough bacillus contains one pair of chromosomes, but the arrange- ment of the cells, and the method of cell division in the two, morphological types, are perfectly distinct (Chapter III). It is probable that these multicellular smooth bacteria occur mainly in cultures which are in process of germination. D: GERMINATION OF THE RESTING STAGE (27, 30, 31, 32, 3S, 36, 39, 55, 61) The germinaticMi of the spore, microcyst or resting cell is accomphshed ill exactly the same manner in each case. The cell commences to elongate, and in the case of some spores and microcysts, casts the outside wall. The vesicular nucleus is transformed into a single, large transverse rod, which almost immediately divides into two, and afterwards, in the case of smooth types, into four. Normal cell division then commences. THE 13 A C T E R I A L N U C L H U S 65 During the process ot genninaticMi the cell increases in size, except in the case of myxobacteria which tend rather to diminish, the nucleic acid content increases, and the nuclear material becomes large and readily stainable. This is the period of the lag phase of the culture. In the logarithmic phase, which immediately follows, the bacteria at first divide by simple fission alone, but later this method is accompanied by others more complex, and eventually the resting nucleus is restored. Fig. 26 THE GERMINATION OF THE RESTING STAGE A . The spore of a rough bacillus. B. The microcyst of Cytophaga sp. C. The resting cell of Bad. colt. The process is similar in each case. The wall may be shed or absorbed. The vesicular nucleus is transformed into one or more bar-shaped bodies and these divide to give the vegetative nucleus. E. THE SPHERICAL VEGETATIVE NUCLEUS (7, 10, II, 17, 21, 26, 42, 50) The exact behaviour of the nucleus of the corynebacteria, mycobacteria and cocci at cell division has not been recorded. It appears to be a simple sphere, even when it attains to a size comparable with that at which the double rod-shape of the primary nucleus of eubacteria may distinctly be ,resolved by the microscope. In the reduction process which precedes the 66 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA formation of the resting nucleus in M. tuberculosis the nuclear material appears in the form of a pair of chromosomes or chromosome complexes (Chapter VI), but these are not normally evident at cell division. In those cocci which possess a spherical, vegetative nucleus it may be seen to elongate with the cell in the course of division, but the details of the 2rr# # • • (Reproduced from the Journal of Hygiene.) v^ ## t V •• ♦- • •••. •• ':•^• •••••* .;♦ M A **^*> ••••••. (Reproduced from the Journal of General Microbiology.) Figs. 27 and 28 THE SPHERICAL VEGETATIVE NUCLEUS Fig. 27 Fig. 28 1. Coccus (not all are of this type). A unicellular coccus. The nucleus 2. Corynebacterium. appears to elongate on division. Acid-Giemsa ■ .3000. (Compare Fig. 7 (12). process are entirely obscure. Most cocci arc multicellular, however, and in some of these, misinterpreted as single cells, entirely spurious " mitotic figures " have been described (r./. Bisset, 1954). Upon this point as upon others the condition in Azotohactcr, which because ot its larger size may more easily be studied, provides the best available information. The vesicular nucleus ot Azotohactcr undergoes a process which resembles a simple mitosis. The nuclear material becomes concentrated in four chromosome complexes, two of which migrate into each half oi the dividing cell and there divide and subdivide, reforming the chromatinic THE BACTERIAL NUCLEUS 67 granules which surround the nucleus and which probably represent the elementary chromosomes. The germination o{ the cyst in Azotohacter is also entirely comparable with the similar germination processes in other bacteria. The vesicular resting nucleus divides into two chromosome complexes and then into the four of the active nucleus, at which stage the cyst is ruptured and the vegetative cell escapes. F: COMPLEX VEGETATIVE REPRODUCTION (8,9, 10, 27, 31,41, 52) This mode of reproduction is common, although not invariably present, in many types of bacteria, but does not appear to be found in spore-bearing cubacteria. The nuclear appearances which accompany it are very striking. Filamentous forms arc common in fdms made from such cultures, but in contrast to the fdaments which occur in rough cultures, which are multi- cellular, repeating in each unit of the chain the nuclear pattern ot the individual, the reproductive fdaments are unicellular and their nuclear material is arranged in a distinctive manner. Such fdaments may, however, occur in cultures of both smooth and rough morphology, and even in cocci. Although there are recognisable differences between the various types the general plan is simdar in all of them. The slightly spiral morphology of the individual bacterium is more clearly shown in these elongated cells, which, because they are curved in two dimensions are difficult to photograph in such a manner as to keep their entire length in focus. In smooth cultures, the shortest fdaments, which are about twice the length of a single bacterium, have their chromosomes packed together at the centre of the cell. Where they can be distinguished separately they are invariably found to consist of three pairs. Filaments of slightly greater length contain six pairs of bodies, which may be together at the centre ot the cell, or at a later stage, distributed in pairs throughout the length ot the fdament. This increase represents a single nuclear division within the fusion cell. A < second nuclear division occurs after the chromosome pairs have been .*// r ^\ ^4 8 :? A. 4 ^Wa 11 Fig. 29 COMPLEX VEGETATIVE REPRODUCTION (]-r>) Streptococcus faecalis, x 3000 ; 16-11) Shigella flexneri. x 3000. (I) Large cell with elongated fu.sion nucleus. (2-4) Development of filament. (5) Fragmentation of filament. (6, 7) Trinucleate pre-fusion cells. (8, i>) Fusion nuclei. In (8) compare fusion cell {left) with rlivif ■mal cell (ri'^ht). (10, 11, 12) Development of fusion nucleus. (13, 14, 15) Redistribution of nuclear elements in growing filament, (IH, 17) Fragmentation of filament. (16) stained for nuclear structu 1 V- ' il2| '14 15 (Reprotluccit from the Journal of Hygiene.'^ ing cell {coitir) and ■es, (17) for cell walls THE BACTERIAL NUCLEUS 69 redistributed in the filament, and the latter then fragments into nidividual bacteria, each containing two pairs ot chrcMiiosomes. Thus each chromosome of the original six becomes the parent of the entire complement of two pairs in one daughter bacterium. The occurrence ot the fusion process has now been confirmed by genetic studies (Section G, below). ^ ^ /n r\ '" 13 14 «* ^» •• J* • •» ♦** ^» i ^ '•t« f fit -AN%tl7 f rt 19 {Fhot„„utrosr„plis hy Dr. E. Klu-nfbergfr-Sohfl. Reproduced from the Jourmil o] General Microbiology). Fig 47 SEXUALITY IN BACTERIA lOp the nucleus may or may not be accompanied by complete division of the cell. If the gametic nuclei fuse within the original cell, then the process is autogamous, if the two gametes separate entirely and conjugate with other partners, the process is sexual. Cytological and genetical evidence agree that the former is most common but that the latter may occur upon occasion. \n eubacteria and cytophagas the division and refusion of the nucleus occurs in an elongated cell, often with a marked, central constriction, but in most myxobactcria the cell is already spherical. A reduction process almost identical with that found in sporing genera occurs in myxobactcria, cytophagas, non-sporing eubacteria and Nocardia, as well as in Actinomyces (Section G below). The nucleus divides into two unequal parts and the smaller of these is ehminated. As in the case of sporers, sometimes more than one " polar body " is ehminated. E: SEXUAL FUSION IN THE SECONDARY NUCLEAR PHASE (8, II, 23, 26, 27, 28) One of the first records of sexual fusion in a non-sporing bacterium was made upon an organism in the secondary nuclear phase by Stoughton (1929, 1932). The condition is quite frequently adopted, in the latter stages of vegetative growth, especially by bacteria of rough morphology (Chapter IV). The bacterium contains a single, central, nuclear unit, probably a pair of chromosome complexes, and may often produce the appearance of end-to-end Figs. 46 and 47 {See pp. 107 and 108) THE CYTOLOGY OF MYXOBACTERIA (1), (3), (4) Myxococcus fulvus, germinating microcysts, Giemsax3000. (2) Chondrococcus exiguus (as above). (5), (6) (8). M. fulvus, young vegetative cells, showing chromosomes, GiemsaxSOOO. (7) Ch. exiguus (as 5). |9), (10) M. fulvus, burst microcysts, Giemsa x 3000. 11) M. virescens, maturing culture, bacilli gathering to form microcysts, Giemsa x 3000. ;i2) M. fulvus, maturing culture, Giemsa x 3000. ;i3), (14), (15) M. fulvus, nuclear fusion, Giemsa X 3000. ' ;i6), (17), (18), (19) M. fulvus, microcyst formation, Giemsa X 3000. no THE CYTOLOGY AND LIFE-HISTORY OF ACTERIA conjugations, with the nuclear material fused at the point of contact. Usually only a single pair of bacteria will conjugate thus, but there seems little reason to doubt that the star-hke clusters which have been reported in Phytomonas tumejaciens and in cytophagas, as well as in many other genera, are exactly similar, multiple conjugations. In the case of P//. tumefaciens the concentration of Feulgen-positive material at the centre of the cluster has been described. Fig. 48 THE NUCLEAR REDUCTION PROCESS (1) In Rhizobiuin ; a, b, c, d represent stages in the maturation of the resting nucleus, and the ehmination of a small daughter nucleus. (Reproduced frn)ii Cold Spring Harbor Symposia) . (2, 3) The elimination of the rejected daughter nucleus in Bacillus ; x, y shows two suc- cessive stages in the same organism. Phase-contrast photographs in the living state. [Re- produced from Experimenta .Cell Research, by permission of Prof. R. J . ]'. Pulvertaft). SEXUALITY IN BACTERIA III The resting nucleus may be formed directly from the secondary nucleus, with or without the intervention of a sexual process, hi the case of Bad. ftiahaccamm, described by Stoughton, the microcyst is extruded from the point oi contact of the conjugating cells, or from the side of the bacterium when conjugation is not apparent. As the microcyst grows the mother cells shrink, and eventually disappear. This lateral extrusion of the microcyst has been reported in Bacteriacea' by Mellon (1925), but the cytological processes {After Stoughton. Reproduced from the Proceedings of the Royal Society.) Fig. 49 MATURATION OF THE RESTING CELL IN BACT. MALV ACE ARUM The microcyst is formed from the secondary nuclear phase, and is extruded laterally from the mother cell. which accompanied it were not described. It is apparently no more than a variant o£ the commoner process of direct transformation o^ the vegetative cell. F: SEXUALITY IN MYCOBACTERIA (21, 22, 24, 25) Among the many accounts oi morphological development cycles in M. tuberculosis, only that of Lindegrcn and Mellon (1932, 1933) gives a con- vincing description of the accompanying cytological phenomena. It is interesting to observe, however, that it includes a sexual process, in the 112 CYTOLOGY LIFE-HISTORY OF BACTERIA formation of the resting nucleus, which accords very closely with those which have already been described in other bacterial groups. The process of formation of the resting nucleus commences by the con- jugation of two of the small cells produced by the fragmentation of the vegetative bacillus. Their small, spherical nuclei fuse, giving rise to a typical, eccentrically-staining, vesicular, resting nucleus. fi ® ® {Modified after Lindegren and Mellon.) Fig. 50 MATURATION OF THE RESTING CELL IN M. TUBERCULOSIS Upper line. — Sexual conjugation of cells derived from the bacillary form ; formation of the vesicular nucleus. Lower line. — Reduction division. Two large chromosomes are formed from the vesicular nucleus and one passes to each daughter cell. The vesicular nucleus is reconstituted in each ; lacking the typical eccentric granule. Reduction follows conjugation, as in all other cases where it has been described in bacteria, and resembles a mciotic cell division. The resting nucleus develops a number of granular threads, which contract, forming two chromosomes. The cell divides, and one chromosome passes to each daughter cell. Beyond this stage the process ceases to resemble that which is found in eubacteria. The mature cell germinates by fission, into tetrads, and thus the small, elementary cells of the vegetative culture are restored (Chapter V). This isolated description gains greatly in force by its remarkably close resemblance to the condition in most other groups, in the maturation of the resting nucleus, and especially to the more recent descriptions by Morris of the processes of initiation and maturation of the secondary phase in Actinomyces bovis (Section G below). 'SEXUALITY IN BACTERIA TI3 G; SEXUALITY IN STREPTOMYCES AND ACTINOMYCES (17, 24, 27) The descriptions of these genera are taken ahnost exckisively from the admirable descriptions of- Kheneberger-Nobel (1947a) and Morris (1951a) respectively. The nuclear material ni die primary mycelium of streptomyces is in a form which resembles the chromosome complexes of young cultures of eubacteria. The secondary mycelium, from which the spores arise, contains rod-shaped, fusion nuclei. The sexual process which brings about this trans- formation is unhkc any of those which have so far been described, in that a special, sexual organ is produced for the purpose. Branches arise from the primary mycelium, which may be much ramified or tightly curled at the ends. These become entangled with other, similar branches, forming " nests " of filamentous cells, within which the initial cells of the secondary mycehum arise. The initial cells are spherical or oval and contain a central nucleus. They germinate to form the secondary mycelium, with its rod-shaped fusion nuclei. Spores are formed by the fragmentation of the fusion nuclei within special, aerial hypha; arising from the secondary mycelium. The chromatinic material at first forms paired chromosomes which are transtormed into the simple spherical spore nuclei. The primary mycelium, which arises trom the germination of the spores, is apparently haploid, like the spores themselves, and the secondary mycelium, diploid, but the details of the intervening sexual process, which presumably occurs in the mycelial " nests," are not described. The nuclear material of the primary mycehum ot the anaerobic actinomyces is also similar to that of vegetative eubacteria, but in this case the branching of the filaments is impermanent, and the mycelium less complex than in the streptomyces. The initial cells of the secondary phase arise by the conjugation of bacillary units from the primary mycelium, and germinate to produce an apparently diploid mycelium, which is branched but non-septate and coenocytic. The spores arise singly on short branches, and the reduction process which 114 THE CYTOLOGY AND LIFE-HISTORY OF ; A C T E R I A precedes their maturation is similar to that already described for cubactcria (Sections B, D above). A nuclear division gives one large and one small daughter nucleus, of which the latter is eliminated. ,. 31a 1 JJ®®QJ^ 2N ^«®®®\*" J.8 17 16 15 {Reproduced from the Journal of Hygiene). Fig. 51 THE LIFE-CYCLE OF ACTINOMYCES BOVIS (I) Spore. (2-4) Germination. (5-8 and 33-36) Vegetative reproduction in the primary mycelium. (9-17) Formation of " initial cell " by fusion of units from primary mycelium. (18-21) Germination of " initial cell " to give coenocytic secondary mycelium. (22-24 and 37-41) Growth and branching. (25-31) Development and maturation of spore, nuclear reduction. (According to Morris). SEXUALITY IN BACTERIA II5 H: SEXUAL FUSION IN PROTEUS AND STREPTOBACILLUS (13) It has been stated that Strcptobacilltis moniliformis is capable of reproduction by the formation of spore-hkc bodies arising from swollen sporangia upon the filaments of the bacteria. These bodies are clanned to be sexual in origin, forming in other bacteria, including Proteus, at the edges o^ swarms where these make contact with other, similar swarms. This interpretation will be discussed in Chapter VII. ® (5 © Qi Formation and maturation of the resting stage in the anaerobic actinomycete Spherophorus, according to Prevot. The resemblance between this process and that described independently for Actinomyces bovis is most striking (compare Fig. 51). /.• SUMMARY A well-marked process ot autogamous or sexual fusion accompanies the the formation ot the resting nucleus in all groups ot bacteria. In sporing bacilli and streptomyces a longitudinal, rod-shaped fusion nucleus is formed, and from it haploid spores arise. In the case of the sporing bacilli the reduction process is very obvious, and precedes spore maturation. In myxobacteria and non-sporing eubactena the nucleus divides and reconjugates. The two gametes contain spherical, central nuclei, and fuse to form a microcyst containing a vesicular, resting nucleus. The vegetative bacterium is haploid and the reduction process precedes the maturation of the microcyst. A similar process has been recorded in the anaerobic actino- myces, and a reduction division occurs during maturation. Il6 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA Sexual fusion may also occur in the secondary nuclear phase in eubacteria, and has been reported to take place at the edges of Proteus swarms. Cytological evidence of the nuclear cycle indicates that the vegetative form is haploid, and that fusion is immediately followed at a varying interval of time by a reduction process with the elimination of one or more daughter nuclei. The gametes are homothallic. Fusion is occasionally sexual but more frequently autogamous. These conclusions are confirmed bv genetical evidence. The complex method ot vegetative reproduction is also probably sexual. It is analogous to the formation and fragmentation of a symplasm. BIBLIOGRAPHY (i) Allen, L. A., Appleby, J. C. and Wolf, J. (1939) Zbi. f Bakt. II. 100. 3. (2) Badian,J. (1930) Act. Soc. Bot. Pol. 7, 55. (3) Badian,J. (1933a) Arch, f Mikrobiol. 4. 409. (4) Badian,J. (1933b) Act. Soc. Bot. Pol. 10. 361. (5) Beebe, J. M. (i94i)J. Bact. 42. 193. (6) BissET, K. A. (i94Sa)J. Gen. Microbiol. 2. 248. (7) BissET, K. A. (i94.Sb) J. Hyg., Camb. 46. 173. (S) BissET, K. A. (1949) ibid. 47. 182. (9) BissET, K. A. (1951) Cold Spring Harbor Symposia. 16. 373. (10) BissET, K. A., Grace, J. B. and Morris, E. O. (1951) Exp. Cell. Res. 3. 388. (11) Braun, a. C. and Elrod, R. P. (1946) J. Bact. 52. 695. (12) DiENES, L. (1946) Cold Spring Harbor Symposia. 11. 51. (13) Flewett, T. H. (1948) J. Gen. Microbiol. 2. 325. (14) Grace, J. B. (1951)]. Gen. Microbiol. 5. 519. (15) Klieneberger-Nobel, E. (1945) J. Hyg., Camb. 44. 99. (16) Klieneberger-Nobel, E. (1947a) J. Gen. Microbiol, i. 33. (17) Klieneberger-Nobel, E. (1947b) ibid. i. 22. (18) Krzemieniewska, H. (1930) Act. Soc. Bot. Pol. 7. 507. (19) Krzemieniewski, H. and S. (1928) ibid. 5. 46. (20) Lederberg, J. (1948) Heredity. 2. 145. (21) LiNDEGREN, C. C. and Mellon, R. R. (1932) J. Bact. 2s. 47. (22) LiNDEGREN, C. C. and Mellon, R. R. (1933) Proc. Soc. Exp. Biol. Med. 30. no. (23) Mellon, R. R. (1925)]. Bact. 10. 579. (24) Morris, E. O. (1951a) J. Hyg., Camb. 49. 46. SEXUALITY IN BACTERIA II7 (25) Morris, E. O. (19s ib) ibid. 49, 17s. (26) PuLVERTAFT, R.J. Y. (1950)]. (k'li. Microbiol. 4. xiv. (27) Prevot, a. R. (1953) Symp. Actino. Rome. vi. Int. Cong. Microbiol. (28) ScHAUDiNN, F. (1902) Arch. Proristcnk. 1. 306. (29) ScHAUDiNN, F. (1903) ibid. 2. 421. (30) Stanier, R. Y. (1942) Bact. Rev. 6. 143. (31) Stoughton, R. H. (1929) Proc. Roy. Soc. B. 105. 469. (32) Stoughton, R. H. (1932) ibid. in. 46. (33) TuFFERY, A. A. (1954) J. Gen. Microbiol. 10. 342. CHAPTER VII Life-Cycles in Bacteria A. GENERAL THE lifc-cyclcs of many bacteria arc simple and direct. A cell in the resting stage is transplanted upon a new source of food. It germinates into the vegetative form and multiplies by simple, asexual fission, as well as by more complex methods (Chapter V). The nucleus of the vegetative cell may adopt a variety of appearances. Often it is semi-permanently in the active condition, without a nuclear membrane, and with the chromatinic material in the form of chromosomes or chromosome complexes (Chapter IV). When the food supply begins to be exhausted, and when the waste-products of the culture have accumulated to such an extent as to interfere with metabolic activity, a new generation of resting forms is produced, by a sexual process (Chapter VI). The resting cells may be contained in elaborate fruiting bodies, or may be free. They may or may not be especially resistant. The nucleus is central and vesicular, often staining with an eccentric, chromatinic granule (Chapter IV). In almost every case the diploid phase is short, fusion being almost immediately followed by a reduction division (Chapters VI and X). B: THE LIFE-CYCLE IN MYXOBACTERIA (2, 3, 16, 29, 34, 35) The type of life-cycle described in the previous section is found in its most advanced form in myxobacteria. The unit is the swarm. When a ripe fruiting body, usually whidborne, 118 LIFE-CYCLES IN BACTERIA Iig falls upon a suitable substrate, it releases the thousands of niicrocysts which it contains, and each ot these germinates to form a vegetative bacterium, the whole constituting the swarm. The swarm moves out over the substrate, feeding and multiplying as it goes. From time to time fruiting bodies are produced, under the influence of a specific factor elaborated by the vegetative cells. These are formed bv the aggregation of vegetative bacteria, some of which arc transformed into niicrocysts, and some of which are sacrificed to assist in the formation of the stalk and wall of the fruiting body. The mature fruiting body, in some genera, is of great complexity, and may be borne upon a long stalk, in others it is sessile and simple in form. When ripe, the fruiting body is released from its stem and blows away in the wind. If it alights upon a suitable substrate it germinates and releases the swarm to repeat the cycle. This type of multicellularity is not pecuhar to myxobacteria. It occurs also in an interesting group of organisms, the Acrasiecu. In this case the unit of the swarm is an amoeboid cell instead of a bacterium, but the cycle is in every other way similar. Myxobacteria and Acrasiea' arc probably not in any way related, but have merely adopted a similar mode of life. They also resemble one another in being predatory upon saprophytic bacteria, in the soil, although pathogenic myxe^bacteria do exist. C: THE LIFE-CYCLE IN EUBACTERIA (1,^,5,7,28,37,38 ) The condition in eubacteria and myxobacteria is not dissimilar, although superficially it may appear to be so. Bacteriologists working with pure cultures upon otherwise sterile media mav derive a false impression of the extent to which the distribution of bacteria is achieved by single cells, whether spores or otherwise, inaugurating new growth. Such conditions do not obtain in nature. The soil, which is the natural habitat of most bacteria, 'swarms with micro-organisms of every kind, and competition must often 120 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA Resting Stages Swarm Vegetative Reproduction THE LIFE-CYCLE IN EUBACTERIA AND MYXOBACTERIA be too keen to permit of the immediate success of a single transplant, of this kind. New substrates are more frequently introduced to micro-organisms already present, than vice versa, and are thus inoculated with a considerable number of spores or resting cells, derived from the last period of growth in the same area. In effect, therefore, the unit of growth and reproduction, in eubacteria as in myxobacteria, is the swarm or bacterial culture. The culture grows and ages, physically and physiologically, exactly Hke a multicellular organism, and differs from one mainly in the lack of speciaHsation of its component cells, hi myxobacteria some degree of speciahsation has been achieved in the formation of the various parts of the fruiting body, but in eubacteria all cells have an equal reproductive potential. The similarity between the cytology of the reproductive processes of the two groups makes it apparent that their life-cycles are basically ahke, and this similarity is increased by the existence o( the cytophagas which appear to be intermediate between them in many respects ; resembhng eubacteria in their saprophytic habit and absence o{ fruiting bodies, myxobacteria in the large microcyst and the morphology of the bacterium. LIFE-CYCLES IN BACTERIA 121 What is known of the cycle in Mycohacterintn and Nocardia is entirely com- parable in most respects (Chapters V and VI), but these organisms have some characters in common with the higher bacteria, to which they are probably related. D; THE LIFE-CYCLE IN HIGHER BACTERIA (32, 37) hi streptomyces a truly multicellular organism is formed, and thus the problem of distribution entails the hberation of free, reproductive units, small airborne spores. Streptomyces resemble moulds in their general form, and this resemblance extends to their mode of reproduction. The spore alights upon a suitable substrate, and germinates to form the primary mycehum. Sexual branches arise upon the primary mycelium and initiate the aerial hyphae which bear the spores. During its hfetime the organism or colony produces spores continuously, while conditions are suitable, hi eubacteria, upon the other hand, the re- productive cells arise only in an ageing culture, when almost all may be thus transformed. The true Actinomyces hovis, which is a parasite and microaerophilic also has a complete life-cycle with a primary and secondary mycelium. But the spores, which cannot be expected to benefit by aerial distribution, are borne singly and in relatively small numbers. This may represent a degenerate condition. E: THE LIFE-CYCLE IN C HLAMYDOBAC TE RI A AND CAULOBACTERIA (15, 20, 21, 28) In chlamydobacteria, such as the filamentous, iron bacterium Sphaerotilus discophoms, as in the case of streptomyces, the organism is essentially sessile and multicellular, so that it must be provided with a distributive mechanism CYTOLOGY LIFE-HISTORY BACTERIA Miss PInlhs E Pease) Figs, .'ui and .")4- THE LIFE-CYCLE OF CAULOBACTER The life-cycle of the true, stalked caulobacteria provides an example of the alternation of sessile and motile generations, such as is commonly found in other biological groups. In Fig. 53 the entire cycle is seen foreshortened. The stalked cell is producing a flagellate daughter cell. Fig. 54 shows the complete life-cycle. The stalked generation (1) is shown in process of division. In (2) the flagellate daughter cell is shown. In (3) are two flagellate cells in the process of becoming sessile. In the upper example the stalk has already developed, but the flagellum (slightly outhned for clarity) is retained. In the lower example the stalk is in an early stage of development. Electron micrographs, gold-shadowed. Fig. 53 and Fig. 54 (1) X 20,000; Fig. 54 (2), (3) ■ 15,000. for reproductive purposes. Chlamydobacteria are aquatic and instead o£ aerial spores produce motile swariner cells, which swim actively by means of flagclla. The vegetative cells grow as a long filament, surrounded by a sheath o£ colloidal iron. At the ends of the sheath the cells may be transformed into flagellated swarmers, which swim oft' to found new filamentous colonies. LIFE-CYCLIiS IN BACTliRIA 123 Fig 54 (See Legend on page 122) 124 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA It is unfortunate that the cytology of this interesting process has never been properly described. The vegetative cells are quite similar to those of eubacteria, but the minute structure of the swarmcrs has never been described. Certain caulobacteria possess a short cycle which is rather similar. The sessile, stalked bacterium produces a succession of motile, flagellated daughter cells, which swim actively until they fmd a suitable attachment upon which to form a stalk, and produce a new, sessile generation. There is no apparent difference between the daughter cell which, having the terminal stalk, remains attached to the substrate, and that which, being free, swims away, but the effect is exactly as though a sessile, asexual generation were producing a series o( motile buds. If this type of organism possesses a sexual stage, it is in the motile cells that it must be sought. F: GONIDIA AS A STAGE IN THE BACTERTiL LIFE-CYCLE (i, 6, 7, 17, iX, 22, 36, 40, 42, 43) This question, so far as it relates to the tubercle bacillus, has already been discussed (Chapter V). Of the many claims that bacteria of other groups may reproduce by the liberation of tiny, fdterable granules or gonidia, few have been accompanied by sufficiently detailed, cytological information. By far the most perfect examples of gonidial reproduction are provided by the nitrogen-fixing bacteria Azotohactcr and Rhizohium. The details in both cases have long been known, but have been described in a rather un- convincing manner, and, in the case of Rhizohium, not entirely accurately. It was supposed that large " barred " bacilli fragmented to form small coccoid swarmers ; each dark bar representing the genesis of a single swarmer. In fact, the process is alike in both genera. The tiny, polar-flagellated gonidia form within the lumen of large mother cells, and are released by rupture of the cell wall. The mother cells in the case of Rhizohium are divided by basophilic septa, probably secretory in function, and these septa are the " bars " of the earlier account. A second, larger type of gonidium, resembling a small vegetative cell, may be produced by Azotohactcr. LIFE-CYCLES IN HACTERIA 125 Many accounts of gonidial reproduction refer to spore-bearing bacilli, and of these the paper published by Allen c[ al. (1939) is the most detailed and convmcnig. These workers described the occurrence of small, rcfractilc granules in the cytoplasm of the bacillus. These granules appeared to reproduce by fission, and were liberated from the cell and transformed into small rods which grew up into normal bacilh. The granules were capable of passing through a Berkefeld filter. The process is probably typical of many which have been described, rather less convincingly, by other authors. It is possible that the G forms of Hadley (1927 ct scq.), which have been reported as tiny cells, forming correspondingly tiny colonies among the normal colonies of several different bacterial species, may be of a similar nature, but in this case also, the evidence is more voluminous than enlightening. The existence of a granular reproductive phase in spirochaetes has long been a subject of controversy. Recent studies have confirmed, however, that members of this group may reproduce both by transverse fission and by the formation of large cysts, usually at the end, but occasionally in the middle of the organism. These cysts contain several small spirochaetes, sometimes in a granular form. In the experience of the author the cysts stain more deeply than the vegetative spirochaetes with basic dyes, and thus probably have an increased nucleic acid content. There is evidence that the granules or gonidia are more resistant than the vegetative spirochaetes, and the latter may survive adverse conditions in this form. The details ot the processes ot formation and germination have not been described, nor is it known whether a sexual process is involved. G; L-ORGANISMS (8, 9, 10, II, 12, 13, 14, 23, 24, 25, 26, 27, 30, 3T, 33, 39, 41) In several papers, Klieneberger (1935 ct seq.) described the occurrence, in cultures of Streptohacillus monilijormis, of small colonies resembling those of the organism of bovine pleuropneumonia. She considered these colonies, 126 THE CYTOLOGY AND LIFE- HISTORY OF BACTERIA t¥. tic?, ^-^ %. -^^S^ {Flwlomicrogniphs hy Dr. H. KH.'nfhfrfitrXofhl. K,f^ro,liiu;i fmin Ihf Journal of Hygifiu). Fig 55 LIFE-CYCLES IN BACTERIA 127 and the occasional swelling of the bacterial filaments which accompanied their presence, to be evidence of a parasitic or symbiotic condition between the L-organism, as it was termed, and the bacterium. Diencs (1939 ct sc(].), upon the other hand, believed the L-organism to be a gonidial stage in the life-cycle of the bacterium. He claimed that the minute colonies were liberated from the swellings upon the bacterial threads, and that the tiny, component organisms grew up into Str. nuniilifonnis. Klicncberger stated that the L-organism could be subcultured upon artificial medium for several years, without reverting to a bacterial condition. Dienes also claimed that the swollen cells, which hberated the L-gonidia, were the result of sexual fusions between filaments, and that similar swellings occurred at the point of contact of Proteus swarms (Chapter VI). Similar swollen filaments have been observed to occur in a variety of bacterial genera, and L-organisms have been found in cultures of Neisseria {^ofiorrlioeae and Fusijormis iiecrophorus. Studies with the electron microscope support the life-cycle hypothesis, and there is little doubt that the L-organisms are in fact a gonidial stage in the life-cycle of these bacteria. Klieneberger has herself adopted this view in a recent paper (1949), and claims that the reproductive bodies arise from a sexual process. The main difference between this phenomenon and gonidial reproduction in many other bacterial groups (Section F, above) is the apparent capacity of the L-stage to reproduce itself for many generations without returning to the bacterial condition. Fig. 55 THE L-STAGE IN THE BACTERIAL LIFE-CYCLE (I), (2) Colonies of the L-form of Fusiformis necrophorusx 200. This gonidial stage reproduces for some time without reverting to the bacterial form. (3) A colony of the L-form, fixed and stained, in sifux 2000. (4) Fusiformis necrophorus, the bacterial form, showing reproductive swellings. Giemsa )k 3500. {Reproduced from the Journal of C.cnenil Microbiology). 1 Fig r,H LIFE-CYCLES IN BACTERIA 129 H: SUMMARY The type of life-cycle which is seen in its most perfect form in myxobactcria is also common to most other bacterial groups. The resting stage, a group of microcystic cells, is transplanted upon a fresh medium, and germinates to produce the vegetative culture or swarm, which is the reproductive condition. When the substrate is exhausted the vegetative cells undergo a sexual process to produce the resting stage, which remains in that condition until again transplanted, or until the food supj ly is renewed. The resting stage may be a resistant spore, or may not be markedly resistant, except to inanition, hi the case of myxobactcria the microcysts are contained in elaborate fruiting bodies. Sessile bacteria, the mycelium-forming streptomyces, filamentous chlamy- dobacteria and stalked caulobacteria can only be distributed by the agency of free, reproductive units. Streptomyces produce aerial spores in large numbers, and the aquatic chlamydobacteria and caulobacteria produce motile swarm cells, which swim away and found new colonies. Many bacteria may produce very small gonidia from which typical bacteria are regenerated, but little is known of their nature, or the circumstances under which they are formed. Such gonidia are found in their most perfect form in the hfe-cycles of Rliizohiiini, Azotobactcr and certain spiral bacteria. BACTERIAL GONIDIA The production of gonidia is seen in its most perfect form in the root-nodule symbiotic bacterium Rhizobium. Small, spherical gonidia are released by the rupture of the cell wall in specialised, large, septate mother-cells. The gonidia have single or occasionally double fiagella and show appearances suggestive of conjugation. They rapidly grow up into small baciUi. (1,2) Production of gonidia. (Acid-Giemsa, •, 3000). (3, 4) The same. (Tannic-acid-violet). (5) Electron micrograph of gonidium. (Gold-shadowed, ,< 20,000). (6, 7) Electron micrographs of gonidia showing appearances suggestive of conjugation. It will be noticed that both members of a pair have well-developed fiagella, which would not normally appear in cases of division, where one daughter cell would have short fiagella or none. (X 16,000). ( (8) Small, motile bacterium, replacing gonidia in some strains of Rhizobium. (x 16,000). ( (1) iind (2) reproduced from the Ji iiiiiainder by courtesy of Miss Phyllis Pease). LIFE-CYCLES IN BACTERIA 131 Fig. 58 LIFE-CYCLE OF AZOTOBACTER An extraordinary degree of complexity is found in the life-cycle of the nitrogen-fixing bacterium Azotobacter. Not only does this organism produce spore-like cysts (not illustrated here), but two distinctly different types of gonidia. The vegetativ'e cell [A] becomes packed with tiny replicas of itself (C), or with motile gonidia (G, H). In both cases, the cycle is initiated by the production, within the mother-cell of an undifferentiated mass of Gram-positive material [B] ; traces of Gram-positivity may be retained to a later stage of gonidium production. The large gonidia (C, D, E) grow up directly into typical vegetative cells, and are retained within the remains of the mother-cell wall. The small, motile gonidia (G, H, I) may reproduce for several generations as small. Gram-negative bacteria. Fig. 57 BACTERIAL GONIDIA Bacterial gonidia are also well seen in A:otobactey, which produces more than one kind, and in some spiral organisms. (1) Azotobacter mother cells showing large and small gonidia (compare Fig. 58). The large types grow directly into vegetative cells, the small gonidia may reproduce as such for several generations. (Gram's stain, x 3000). (2) Small gonidium of Azotobacter. It resembles those of Rhizobium but has more fiagella and is less nearly spherical. (Electron micrograph, gold-shadowed, X 16,000). (3) Electron micrograph of Spirilliiin sp. showing attached cysts, from which the gonidia are produced. ( ■ 6000). (4) ^Mature cyst with fiagella still attached. ( ■ 12,000). (5) Developing gonidia, of Spirillum, each with a single polar flagellum. Note the liU^pharoplasts. ( ■ 12,000). 132 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA BIBLIOGRAPHY Allen, L. A., Appleby, J. C. and Wolf, J. (1939) Zbl. f. Bakt. II. 100. 3. Anscombe, F.J. and Singh, B. N. (1948) Nature, Lond. 161. 140. Beebe,J. M. (i94i)J. Bact. 42. 193. BissET, K. A. (1949) J. Hyg., Camb. 47. 182. Bisset, K. a. (1950) J. Gen. Microbiol. 4. i. BissET, K. A. and Hale, C. M. F. (195 i) J. Gen. Microbiol. 5. S9-- Bisset, K. A. and Hale, C. M. F. (1953) ibid. 8. 442. DiENES, L. (1939) J. Inf. Dis. 65. 24. DiENES, L. (1940) Proc. Soc. Exp. Biol. Med. 44. 470. DiENES, L. (1942)). Bact. 44. 37. DiENES, L. (1943) Proc. Soc. Exp. Biol. Med. 53. 84. DiENES, L. (1946) Cold Spring Harbor Symposia. 11. 51. DiENES, L. (1947)]. Bact. 54. 231. DuGUiD, J. P. (1948)]. Path. Bact. 60. 265. Fischer, A. (1897) Vorlesnngcn nber Bakterien, Leipzig. Garnjobst, L. (1945) J. Bact. 49. 113. Hadley, p. (1927) J. Inf. Dis. 40. i. Hadley, p. (1937) ibid. 60. 129. Hadley, P. (1939) ibid. 65. 267. Henrici, a. T. and Johnson, D. E. (1935) J. Bact. 30. 61. Houwinck, a. L. (1949) Address to Soc. Gen. Microbiol. Jones, D. H. (1920) J. Bact. 5. 325. Klieneberger, E. (1935) J. Path. Bact. 40. 93. Klieneberger, E. (1936) ibid. 42. 587. Klieneberger, E. (1938) J. Hyg., Camb. 38. 458. Klieneberger, E. (1940) ibid. 40. 204. Klieneberger, E. (1942) ibid. 42. 485. Klieneberger-Nobel, E. (1945) ihid. 44. 99. Klieneberger-Nobel, E. (1947a) J. Cien. Microbiol, i. 33. Klieneberger-Nobel, E. (1947b) J. Hyg., Camb. 45. 407 133 LIFE-CYCLES IN BACTERIA (31) Klieneberger-Nobel, E. (1947c) ibid. 45. 410. (32) Klieneberger-Nobel, E. (i947d) J. Gen. Microbiol, i. 22. (33) Klieneberger-Nobel, E. (1949) ibid. 3. 434. (34) Krzemieniewski, H. and S. (1926), Act. Soc. Dot. Pol. 4. i. (35) Lev, M. (1954) Nature, Lond. 173. 501. (36) LoHNis, F. (1921) Mem. Nar. Acad. Sci. 16. i. (37) Morris, E. O. (1951a) J. Hyg., Camb. 49. 46. (38) Morris, E. O. (1951b) ibid. 49. 175. (39) PoKROwsKAjA, M. (1930) Zbl. f. Bakt. I. 119. 353. (40) Shrewsbury, J. F. D. and Barson, G. J. (1949) The Life History of Snprospira Pers. Comni. (41) Smith, W. E., Mudd, S. and Hillier, J. (1948) J. Bact. 56. 603. (42) WoRATZ, H. (1954) Zbl. f Bakt. i. 160, 613. (43) Wyckoff, R. W. G., Hampp, E. G. and Scott, D. B. (1948);. Bact. 56. 755. CHAPTER VIII Macro for Illations A: THE MYXOBACTERIAL FRUITING BODY (ll, 12, 14) THE most perfect and elaborate multicellular structures formed by bacteria are the fruiting bodies of myxobacteria. Other macroscopic formations, however elaborate, are little but the result ot the reaction between the growth potential of the organism and the physical restraint ot the environment, and slight variations in the latter may aftect the result to an apparently disproportionate extent. The fruiting body, however, although it may be prevented, by unsuitable conditions, from forming at all, is otherwise independent, in its form, of small environmental changes, and is characteristic of the species. The co-ordination ot cellular activity which initiates the formation of the fruiting body is stimulated by a specific substance, analogous to a hormone which diffuses from the vegetative cells. The great majority of these cells are transtormcd into typical microcysts and thus survive. Some are embodied in the stalk and envelope, and are sacrificed. Little is known of the mode ot formation oi these structures. It has been stated that the cells which take part in their formation are cemented together by dried mucus, but this appears to be mere supposition. It is known that the physical properties of the envelope vary considerably in dirterent species, and it may even be entirely absent. The envelope varies especially in its physical strength and in its resistance to water. The fruiting bodies of some species of myxobacteria burst open as soon as they are wetted. Others remain intact. This characteristic has been considered to be of taxonomic value by some botanists, but there is no reason to believe that it indicates biological relationship. Whether the variation is due to differences in structure and composition of the envelope is not known. 134 MACRO FORMATIONS I35 {After Krzemifniewski. Drawn /torn the photomicrographs.) Fig. 59 MYXOBACTERIAL FRUITING BODIES B: THE MYXOBACTERIAL SWARM (11, 12, 16) 111 the vegetative stage also, myxobacteria give the impression of a degree of organisation far beyond that of other bacteria. The swarm moves outwards from the centre of germination in a regular fashion, following the hues of physical stress in the substrate. It concentrates in chosen areas, converging towards the incipient fruiting body, and piling up, the bacteria crawling over each other, to encyst in an elevated mass. The appearance of ordered purpose is most remarkable in so lowly an organism. C; THE SWARM OF PROTEUS (9, 12, 13, 15) Some degree of cell-specialisation and organisation in the swarm of Proteus is indicated by Klieneberger-Nobel (1947) and other workers. It is suggested that the swarm commences its activity when an initial generation of large cells has produced a sufficient concentration oi metabolites to provide the energy for swarming. The swarm cells are filamentous. They move out rapidly over the substrate until their reserve oi energy is exhausted, and then f rest until it becomes possible to repeat the process. Although this phenomenon 136 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA is most advantageously seen upon the surface of an agar plate, there is no reason to believe that it does not occur in nature. Proteus may accordingly be regarded as having achieved a minor degree of cellular specialisation. It is also a temporary specialisation, because the swarm filaments, which are distributive in function, are the descendants as well as the parents of the " somatic " cells which accumulate the energy for the swarm. In multi- cellular animals and plants, although not always in fungi, the continuity o( the germ plasm is very close, and the somatic cells proper have no reproductive function, when once the body of the organism is elaborated. Even in the case of myxobacteria, the microcysts may survive to germinate, but the cells which form the stalk and envelope of the fruiting body have no descendants. D; CHLAMYDOBACTERIAL AGGREGATES (5.7) As far as is known, the condition in chlamydobactcria, as typified by the filamentous, iron bacteria, resembles that in Proteus. The aggregates of gelatinous sheaths, the by-product of metabolism, arc devoid of structural specialisation, except in so far as some are dead casts left behind by the cells which were responsible for their formation, whereas others are inhabited, and still increasing in size. The hold-fasts by which the ends of some types of filament are attached to the substrate are a possible exception to this rule. Those bacteria in which the character is most strongly developed are often classed as caulobacteria for precisely this reason. After a period of vegetative growth, motile swarm cells are produced which swim away to form new colonics of filaments. There is no evidence that all the cells comprising the sessile filaments may not be equally capable of transformation into swarmcrs ; and indeed, what little is known of their cytology suggests a close resemblance to eubacteria, in this and other respects. MACROFORMATIONS 1 37 E: THE MEDUSA-HEAD COLONY (1.2,3,6) In most bacteria the colony, however complex its structure, is an accidental growth, each cell of which is equivalent to all the others. The colony docs not alternate with the swarm, but is itself the swarm, or vegetative mass. Where conditions are such that motile bacteria may exercise their motility, no colony is formed at any stage ot culture. Colonies are not confined to conditions of artificial culture, although there they appear in their most perfect form. The type of bacterial colony whose structure was first recognised, although not understood, was the so-called " medusa-head " colony of the anthrax bacillus. This form of colony, which consists of long, coiled, bacillary threads, is common to all bacteria of what we have termed rough morphology. In the case of the anthrax bacillus, and similar, large, spore-bearing bacilli, it is easily seen with a hand lens, whereas the much smaller size of, tor instance, lactobacilh, renders it less obvious, so that its presence, except in these large genera, was long unsuspected. Although it has long been known that the virulent anthrax bacillus possesses this type of colony, whereas the smooth colonies of the avirulent anthrax vaccine were composed of individual bacilh, the obvious corollary that the respectively virulent and avirulent smooth and rough colonies ot BcKk'nacca' might possess the same type of structure in each case, escaped attention until much later. This was partly due to the studies which several different workers made upon the mode of cell division in smooth and rough strains ot bacteria (Chapter V), which, by crediting rough bacteria with a " snapping " mode of division, erroneously claimed that the rough colony was composed of zig-zag chains of bacilli. This error, which has persisted tor a quarter of a century, and is included in nearly all text-books of bacteriology, would never have arisen if these observations had been checked by examination of a rough colony, in situ, for it can readily be seen that the structure of a rough colony of a typhoid bacterium, and the " medusa-head " colony ot the anthrax 'bacillus are identical. 138 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA < i' ABC F G {Reproduced from the Journal of Pathology and Bacteriology.) Fig. 60 STAGES IN THE GROWTH OF A MEDUSA-HEAD COLONY A. Original bacillus. B, C. Elongation and looping. D. Primary coil. E. Further growth, against the friction of the medium cau.ses infolding of the coil. F. Secondary infolding of the coil. G. Continued growth of all parts of the colonj' causes complex folding and convolution. M ACROFORM ATIONS I39 The structural complexity, even beauty, of this type of colony has tended to produce the impression that there exists an intrinsic tendency towards the formation of the structure, as in the case of a true, multicellular organism. This is not so. A rough bacillus growing upon a frictionless surface, would produce a straight, or slightly spiral thread of indefinite length. Upon an agar plate, however, after the thread has grown a short way across the surface, its rigidity is not sufficiently great to permit it to extend further in a straight line. It therefore kinks, and because of its slightly spiral growth, tends to form flat coils upon the surface of the plate. All portions of tliis primary coil are growing simultaneously, and to accommodate this growth it produces (Reproduced from the Journal of Pathology and Bacteriology.) Fig. 61 GROWTH OF A ROUGH COLONY Primary coil of a rough colony. Shigella flexneri, impression preparation < 7(»(). secondary loops and coils, and upon them still further and more complex convolutions. The outer portions of the colony lie flat upon the medium, and the internal coils ovcrhe one another to a small extent. The appearance and complexity of the colony vary with the rigidity of the bacterial thread, and the resistance of the surface of the medium. Colonies upon lactose-taurocholatc agar may be noticeably flat and widespread, presumably in consequence of the high concentration of electrolytes and low I surface-tension of such media. 140 THE CYTOLOGY AND LIFE-HISTORY BACTERIA Bacteria of rough morphology vary considerably in rigidity between such extremes as Bacillus mycoidcs, which is so rigid that it seldom produces any structure more complex than a primary coil, and grows out, as long threads across the agar, and, upon the other hand those rough Bacteriacecv whose colonies are almost indistinguishable from smooth variants. Rough colonies are not confined to the surface of artificial culture medium. They may torm wherever a flat surface is presented to growth, provided that the substrate is not too fluid. The structure of these colonics may be entirely disguised by the production of mucoid capsular material. The anthrax bacillus may produce its polypeptide capsule under suitable cultural conditions, o( which COo tension is one of the most important. The phenomenon of smooth-trough variation in pneu- mococci is entirely concerned with the production or loss of polysaccharide capsular material, concealing or revealing the rough appearance ot the colony (Section G). >5^:w-; v^r^: {Ri-proclucitl fiiim the Joiinuil af Pathology and Bacteriology.) A B Fig. 62 SMOOTH AND ROUGH COLONIES A. Smooth colony, Bact. coli, impression preparation x 500. H. RouKh colony arising from the perimeter of a smooth colony, Shigella flcxneri, ■ .sod. ■ ' ' MACROFORMATIONS I4I F: SMOOTH COLONIES (1.2) There is little to be said ot the structure and mode of formation of the colonies of smooth bacterial variants. The constituent bacteria separate completely after cell division (Chapter III), and produce a structureless colony. It tends to be less flat than the medusa-head colony, because its edges lack the cohesion which is necessary to force their way out over the medium. Con- sequently it is a less efficient colony, because so great a proportion of the component cells lose contact with the substrate. Such colonies are often far from truly smooth in appearance, as surface concentrations of hydrophobe lipo-proteins or insoluble polysaccharides may give them a dusty or even a rocky appearance. Alternatively, like rough colonies, they may be enveloped in a mass of mucoid capsular material. G; ROUGH AND SMOOTH COLONIES OF STREPTOCOCCI Streptococci and pneumococci, like rod-shaped bacteria, may grow in the form of threads and chains, or may separate completely after cell division. Accordingly, the former produce a modified medusa-head colony, and the latter a relatively structureless colony. The phenomenon of smooth^rough variation in the pneumococcus, as usually described, is not, however, con- cerned with this change, but with the loss o{ the capsule, which exposes the rough structure of the long-chained colony (Section E above). No distinction is drawn, by most bacteriologists, between S->R variation, as it is termed, in bacteria and in pneumococci. Both are associated with an antigenic change in the surface material, and both with an alteration in the appearance of the colony, as seen by the naked eye ; but the reason for this difference is not the same in each case, and the employment of the same expression to describe two phenomena which are analogous without being homologous, is un- fortunate and has given rise to a certain amount oi confusion. 142 THE CYTOLOGY AND LIFE-HISTORY OF ACTERIA x^A^ <^».v>$ [Reproducfd from the Journal of General Microhiolosy.) Fig. 63 COLONIES OF STREPTOCOCCI 1. Long-chained " rough " colony. 2. Short-chained " smooth " colony. Impression preparations ■ 300. H: COLONIES OF STREPTOMYCES (4, lo) The colonics ot strcptoniyccs may logically be considered as single, multi- cellular organisms, in which the functions of the various types of component cell are almost completely specialised. The colony consists of a vegetative, haploid, primary mycehum, upon which arises a reproductive, diploid, secondary mycelium. The condition is snnilar to that which obtains in higher tungi. The spores are specialised distributive cells, and a single spore may give rise to the complete colony. Although fragments of mycelium will grow if they are transplanted, the colony is essentially a unit which remains fixed upon its substrate, so long as the food supply permits, and it reproduces solely by the release of spores. The cells of the mycelium perish when the food supply is exhausted or when conditions become unsuitable for growth. Thus, in bacteria, the type of multicellularity represented by the swarm finds its greatest perfection in myxobacteria, the sessile colony in the strep- tomyces. In each the mode of reproduction and distribution is admirably designed to the case. MACROFORMATIONS U3 /: COLON ins OF CAULOBACri-RIA (7,8) The aggregates formed by stalked caulobacteria may consist of small clumps attached to a single point upon the substrate, or of free colonies, in the form of rosettes or ribbons, composed of numerous bacteria joined together bv their stalks. Fig. 64 CHLAMYDOBACTERIAL AGGREGATES A group of filaments of the chlamydobacterium Sphaerotilus discophorus. The sheath is composed of colloidal ferric hydroxide. The older portions are the thickest and may be vacant. The growing filaments protrude from the sheath at the thinner end. (Unstained, • 1000). Fig. 65 CAULOBACTERIAL AGGREGATES A colony of Caulobacter attached to filaments of Sphaerotilus. Each cell has an independent vS OV BACTI-RIA I53 The flagellation of typical Bactcriaccac represents an intermediate sta^e of evolution in this respect. If any weight whatsoever can be placed upon the theory of recapitulation, the fact that the flagella in germinating microcysts of Bactcriaccac appear first in the polar position must be regarded as of some interest. The problem of movement on land has been solved in a different manner by myxobacteria, which have adapted the cell wall to a crawling action, as in the Myxophyccae. The evidence at present available does not enable it to be determined whether this character is indicative of a relationship with the Myxophyceac or, through the more flexible spirilla and trichobacteria, with the typical bacteria, but cytological evidence is strongly in favour of the latter. E: AERIAL DISTRIBUTION The most highly evolved bacteria appear to be those which have not only colonised the terrestrial habitat successfully, but have also developed adaptations for the aerial distribution of their reproductive elements. Of these the best example is provided by the streptomyces, which (presumably by evolutionary convergence) resemble minute moulds with aerial conidia, although their cytological structure and behaviour reveals their affinities with other bacteria. Here also, a complete series of forms of varying degrees of complexity serves to connect them, through Nocardia TmA Mycobacteria, with the Gram-positive eubacteria, among which other devices for aerial distribution may be found. Consideration of this tendency provides a possible explanation ot the nature oi the bacterial endospore. Although less efficient in this respect than those of the streptomyces, the spores of Gram-positive bacilli are capable of being distributed in air and dust, and it is worthy of consideration whether it may be this factor, rather than their remarkable powers of resistance, which has conferred a genetical advantage upon their possessors. Although the exact differences between the physico-chemical constitution of spores and vegetative cells of the same species remain obscure, it has repeatedly been claimed that the water-content oi the former is reduced. 154 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA It is readily demonstrable that the spore-nucleus is in a condition of turgour, and these pieces of evidence, taken in conjunction with the exceedingly small relative size of most spores, and the resistance to denaturation of their proteins, suggest very strongly that the spore is reduced in size and weight by reduction of water-content, as an adaptation to aerial distribution, and that its resistance to heat and antiseptics follows as a consequence, indeed as a by-product of this process. Previous concepts of the nature of spores have been based upon information, much of which is incomplete, and some definitely misleading. In the latter category, the theory that sporulation takes place in response to the stimulus of an unfavourable environment is so completely exploded as to require no further comment. But a less obvious misconception, arising from a failure to realise that more-or-less spore-like resting cells are of universal occurrence among bacteria, is the belief that spores are especially resistant to the processes of inanition, that is to say, of oxidation, which is the main cause of death in resting cells. Obviously spores arc more resistant than the corresponding vegetative cells, but there is no evidence at all to suggest that they are more so than the resting cells of non-sporing bacteria ; and this is the true com- parison. It is a commonplace of practical bacteriology that bacteria dried iti vacuo will survive indefmitely. This is a crucial point in the argument. If spores were more resistant than microcysts to inanition, then their possession would confer a most decided advantage, and no other explanation of their existence would be required. But the agencies to which spores are, in fact, especially resistant are most unlikely to be encountered under natural conditions, and, it they were, this resistance would confer little or no genetic advantage since the spore can neither metabolise nor reproduce. It may survive indefmitely in a hot spring, but evolve it cannot. Previous theories have attempted to account for the spore in terms of its importance, not to the bacillus but to the bacteriologist, and the comparisons which have been made have not been true ones. Whereas the aerobic genus Bacillus comprises a wide variety of specific types, retaining the relatively tiny spore, and capable of profiting by aerial distribution ; in the much smaller genus Clostridiiiiii, which by the genetic THE EVOLUTIONARY RELATIONSHIPS OF BACTERIA I55 accident of anacrobiosis (perhaps associated with the habit of parasitism in the animal gut) has lost this opportunity, the spore itself is much larger. This is readily explicable as a retrogressive step in evolution, the loss of a character which has ceased to be of service, for which innumerable parallels could be cited from other fields. An analogous difference exists between Streptomyccs and Micromonospora. The latter has adopted a thermophilic life in manure heaps and a semi-aquatic life ni lake muds, and its spore is much less freely airborne than in the, probably less degenerate, Streptomyccs. True ActinoniYCcs and their sporogenous relatives may well be still more degenerate, parasitic descendants of freely-sporing ancestors. The Gram-positive cocci resemble the sporing bacilli not onlv m their septate structure, but also in a high degree of adaptation to a terrestrial environ- ment. The individual coccus is much reduced in size, and is not only capable of drifting in the dust like a spore, but, possibly because of a similar if less extreme concentration of the proteins, is among the most resistant of vegetative bacteria. A parallel, but apparently independent line of evolution is found in the myxobacteria, of which the most highly evolved types are completely adapted to a terrestrial life. As already described, they are able to crawl upon moist surfaces, and their distribution is achieved by the enclosure of entire swarms in fruiting-bodies, often borne upon stalks to catch the air currents, which dry up and blow away in the wind. F: RELATIONSHIPS OF AUTOTROPHIC BACTERIA (3,4) Although inappropriate to detailed discussion in this book, it is an interesting confirmation of the validity of an evolutionary scheme for the classification of bacteria, such as has been outhned in this chapter, that the conclusions derivable from the parallel concept of progressive loss of synthetic power in the course of evolution are in excellent accordance with those based upon purely morphological reasoning. This is well seen if the systematic relationships of the autotrophic bacteria 156 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA are considered. Leaving aside diosc such as Bc<^iatoa, which arc ahnost certainly not true bacteria, almost all autotrophs are either spirilla, vibrios, pseudomonads or those colonial pseudomonads some oi which are classed as chlamydobacteria. They must therefore be regarded as primitive, aquatic forms ; which might reasonably be expected in any sound evolutionary system. G: SUMMARY There exists morphological evidence which suggests that the bacteria have evolved, in parallel with other groups of living organisms, from an aquatic to a terrestrial mode ot lite. The most primitive bacteria arc the spirilla, which have characters inter- mediate between those of typical bacteria and flagellates. The most highly evolved bacteria are terrestrial and have special mechanisms for the aerial distribution o£ their resting stages. By the same criteria, the autotrophic bacteria are relatively primitive in respect of their morphology, as they appear to be in their physiology. The evolutionary significance of different types of flagellation and of the bacterial endospore are discussed. BIBLIOGRAPHY (i) Beroi-y's Manual of Dctcriii'uiativc Bacteriolo^iy. 6th cd. (1948) London : Balliere, Tindal & Cox. (2) BissET, K. A. (1950) Nature. 166. 431. (3) BissET, K. A. (1952) Bacteria. Edinburgh : Livingstone. (4) BissET, K. A. and Grace, J. B. (1954) Symposium : " Autotrophic Micro- organisms." Cambridge University Press. {5) Grace, J. B. (1954) J. Gen. Microbiol. 9. 325. (6) Kluyver, A. J. and van Niel, C. B. (1936) Zbl. f Bakt. IL 94. 369. (7) PijPER, A. (1946)]. Path. Bact. 58. 325. (8) Prevot, a. R. (1933) Ann. Sci. Nat. Bot., Ser. 10. 15. 23. CHAPTER X The Genetics of Bacteria A: GENETICAL CONI- 1 R M A TION OF CYTOLOGY (6. S. II, 12, 13, 14) m NY organism which Hvcs and rcprcHluccs its kind must possess a L\ mcclianism of inheritance, and there is no reason to heheve that -^ -^ this mechanism in bacteria is different from that which is found in other Hving cells. The necessity for a linear arrangement of genes was emphasised by Lmdegren in 1935, and shortly afterwards the short, chromosome-like body, whose existence he postulated, was described, or redescribed with great clarity by Stille (1937), Piekarski (1937), and Robinow (1942), so that its existence and nature have become generally recognised. After a period in which multiple or even branched chromosomes were postulated, for which there is no acceptable cytological evidence, the single chromost^me has been genetically vindicated (Jinks, 1954). Granted the existence ot a chromosome-like body, it must be presumed that the genes which it carries are susceptible to the same hazards and chances of alteration or injury as the genes of other cells. Thus true, genetic mutation must be responsible, in bacteria as in other living organisms, for the appearance ot permanent, heritable variation. Such genetic evidence as is available supports the picture of the nuclear structure and sexual behaviour o( bacteria which has been drawn in the previous pages of this book. In a review, Lederberg (1948) has stated : " The evidence suggests that this bacterium [Bact. coli) has a life-cycle com- parable to Zy^^osacclumnnyces : the vegetative cells are haploid (but not necessarily uninucleate) ; fertihsation is homothallic or unrestricted genetically ; the putative zygote undergoes immediate reduction without any intervening mitosis." To which it need only be added that this state ot affairs is not pecuHar to Bact. coli, but is common to most other bacteria for which evidence is available. 157 158 THE CYTOLOGY AND LIFE-HISTORY OF More recently Lederberg and his collaborators have produced evidence that vegetative diploid strains of Bact. coli may exist. The cells are longer than the haploids and have a larger number of similar nuclear elements. i%f^ ^- •*-'*a: \cs^ »^ ^ '%% (Keproducfd from Cold SpnnK Ha/hor Symposia). Fig. 68 DIPLOID OR POLYPLOID FORMS OF NUCLEUS (1) Cells of Bacterium coli known to be genetically diploid. {By permission of Dr. J. Lederberg) . (2) Presumed diploid or polyploid nuclei in post-fusion stages of a .sporing bacillus. (3) Secondary mycelium of Streptomyces. In the sections deahng with the nature of the chromosome it has already been argued that the behaviour of the nuclear material at cell division indicates that all the nuclear units contained in the vegetative cell must be of equal value, that is, the cell is haploid but multinucleate. In sexual conjugation the two partners appear to be similar in status, and in bacteria of this type reduction THE GENETICS OF BACTERIA 1 59 occurs ininiediately after fusion. The status of the vegetative nucleus is also confirmed by the work of Witkin (1951), which is described in Chapter IV, Section G. Some bacteria, such as sporing bacilli, streptomyces and the anaerobic actinomyces appear to have a prolonged diploid phase, and an examination of the genetical behaviour ot such forms would greatly assist in the interpreta- tion of their cytological appearances. This correlation of conclusions between strictly genetical and strictly cytological studies gives grounds for increased confidence in both, and its importance cannot be too greatly stressed. B: SOURCES OF GENETICAL EllDENCE (i, 2, 3, 4, 5, 6, 7, y, 10, 15, 16) The evidence upon which these genetical conclusions are based is mainly concerned with the nutritional requirements of the bacteria, that is to say, with their synthetic abilities. Problems of variation in resistance to dis- infectants and to the bacteriophage have also been studied. Gene recombination has been detected by the appearance of nutritionally unexacting " wild-type " strains of bacteria in mixed populations of nutrition- ally exacting mutants, and has been confirmed by a variety of ancillary experiments. The segregation and recombination of mutant characters shows interactions which may be regarded as indicative of linkage, and some evidence of the hncar arrangement of genes has been obtained. The comparative rarity of the recombination process gives reason for believing that conjugation must normally be autogamous, and only occasion- ally sexual. Here also the genetic and the cytological evidence are in agreement. With bacteria, as with plants, animals and fungi, an increased mutation rate is induced by irradiation with ultra-violet light or X-rays. The mutation rate is proportional to the dosage, but is not related to the lethal effect. These mutations may become evident immediately after irradiation, or may appear only in the offspring of the treated cells ; presumably after conjugation. The suggestion that hereditary factors in higher organisms may reproduce automatically in the cytoplasm o( vegetative cells, independent of nuclear control, has its bacteriological parallel in the transformation of antigenic l60 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA types of capsular material, in the pneuniococcus. If a non-capsulated strain is brought into contact with a deoxyribose nucleic acid derived from a strain of different antigenic structure, it will commence to produce the capsular material appropriate to the strain from which the nucleic acid is derived, irrespective of the antigenic type of the capsule which it originally possessed. The genetic material, in this case, may be nuclear in origin, or may be the surface, secretory material of the cell, which is capable, as enzyme systems in other cells have been presumed to be capable, of reproducing itself without reference to the genetic constitution oi the nucleus. Mutations in the power of resistance to antibacterial agents, and in bacteriophage sensitivity, have both given evidence of a genetic system resembling that of higher organisms. There is, in fact, little to suggest that the entire, complex pattern of bacterial dissociation may not be explicable in terms o( multiple and interrelated gene changes. C; SUMMARY The cytological evidence, presented in this work, is confirmed by genetical evidence. BIBLIOGRAPHY (i) Avery, O. T. cr ai. (1944) J. Exp. Med. 79. 137. (2) Beale, G. H. (1948) J. Gen. Microbiol. 2. 131. (3) Braun, W. (1953) Bacterial Genetics. Philadelphia. Saunders. (4) Bunting, M. I. (1946) Cold Spring Harbor Symposia. 11. 25. (5) Demerec, M. and Latarjet, R. (1946) ibid. it. 38. (6) Jinks, J. L. (1954) Bacterial Genetics. Pers. Comm. (7) Lederberg, J. (1947) Genetics. 32. 505. (8) Lederberg, J. (1948) Heredity. 2. 145. (9) Lederberg, J. and Tatum, E. L. (1946) Cold Spring Harbor Symposia. 11, 113. (10) Lederberg, J., Lederberg, E. M., Zinder, N. D., and Lively, E. R. (1951) Cold Spring Harbor Symposia. 16. 413. (11) Lindegren, C. C. (1935) Zbl. f Bakt. IL 92. 40. (12) LuRiA, S. E. (1947) Bact. Rev. 11. i. (13) Morris, E. O. (1951)). Hyg., Camb. 49. 46. (14) McCarty, M., Taylor, H. E. and Avery, O. T. (1946) Cold Spring Harbor Symposia. 11. 177. (15) Piekarski, G. (1937) Arch. f. Mikrobiol. 8. 428. (16) Robinou', C. F. (1942) Proc. Roy. Soc. B. 130. 299. (17) Schrodinger, E. (1944) What is Life ; Cambridge. (18) Stille, B. (1937) Arch, f Mikrobiol. 8. 125. (19) Tatum, E. L. and Lederberg, J. (1947) J. Bact. 53. 673. (20) Taylor, H. E. (1949) Symposium : "The Nature of the Bacterial Surbce," Soc. Gen. Microbiol. 61. (21) Witkin, E. M. (195 1 ) Cold Spring Harbor Symposia. 16. 357. Ind Acid-C;iemsa, 0, 7, S, 12, 52, 53, 55, 57, 66, 75. 7f>. ^o. 9~, 100, i2g Acid hydrolysis, 51, 59 Actinomyces, 81, g^,g6, log, 112, 113, 114, 155 Actomyosin, 44 Aerial distribution, iig, 121, 153-155 Agar, 89, 139, 140 Agglutination, 33 Agriculture, i Alcohol, 7 Aldehyde, 6, 7 Alga '{see also Blue-green-algae), 44, 55, 99, 147, 149, 150, 153 Alkah, 7 Allen, et al., 53, 102, 125 Amino-acid, 31 Animal, 88, 136, 147, 150, 159 Antibody, 10, 17, 18, 22 Antigen, 18, 31, 33, 44, 47, 159, 160 Antigenic structure, 160 Artefact, 2, 5, 7, 8, 9, 10, 60, 97, 98 Autotrophic bacteria. 155, 156 Autogamy, 81, 82, 83, 100, 103, 109, 159 Azotobacter. 13, 55, 66, 67, 74, 75, 124, 129, 131 Bacillacees, 11, 14, 18, 31, 36, 59, 79, 8r, 105, 106, no, 149 Bacillus anthracis, 28, 137 Bacillus cereus, 20, 27, 31, 38, 41, 55 Bacillus megaterium, 31, 41, 55 Bacillus mycoides, 140 Bacillus subtilis, 59 Bacteriaceee, 31, 55, 57, 78, 79, 81, 91, 95, 105, III, 137, 140, 149, 151, 153 Bacteriologists, 4, 5, 119, 141, 154 Bacteriophage, 159, 160 Bacterium coli, 38, 47, 57, 62, 65, 71, 76, 77, 80, 106, 140, 157, 158 Bacterium malvacearum, 77, in Badian, 53 Basophilia, 32-38, 47, 55, 59, 125 Beggiatoa, 156 Bergersen, 38 Bergey's Manual, 146, 147 Berkefeld filter, 125 Bipolar staining, 33 Biochemistry, 3 Biology, 4 Birefringence, 32 Bisset, 32, 147, 149 Blepharoplast, 38, 43, 44, 47, 48, 150 Blood-agar, 21 Blue-green algae, 44, 147, 149, 150, 153 Borax carmine, 10 Botanist, 134 Bouin's fixative, 12, 21 Bovine pleuropneumonia, organism of, 125 Branching, 38, 95-98, 104, 113 Brieger and Fell, 96 Brownian movement, 4 * Budding, 95-98, 124 Capsule, f4. ^ 3/ 43, 47, 141, 160 cx Carbohydrate, 29 Carbon-dioxide, 140 Caryophanon latum, 27, 28, 55, 62 Cationic detergent, 1 2 Caulobacteria, 38, 87, 121, 122, 123, 124, 129, 136, 143, 144 Cell, I, 6, 7, 10, 12, 62, ()4, 65, 68, 91, 92, 94. 102, animal, 10, 62, 72, 83, 102 division, 27, 29, 31, 35, 36, 37, 64, 65, 66, 68, 73, 75, 80, 88, 92, 96, 102, 109, 137, 142 envelopes, 7, 12, 14, 16, 31, 35, 36, 38, 55 fusion, 67-71, 90, 91, 99, 100 initial, 113, 1 14 membrane, 5, 6, 7, 12, 20, 22, 25-48, 87 mother, 36, in, 124, 131 plant, 10, 62, 72, 83, 102 precursor, 103 resting, 57, 64, 65, 112, 118, 120, 154 somatic, 88 trinucleate, 68, 69, 103 vegetative, 47, 64, 71, 86, 100, 109, in, 119, 124, 134 wall, 20, 22, 25-48, 6S, 71, 87, 92, 96, 97, 124, 150 Chapman, 20 Chlamydobacteria, 62, 71, 91, 121, 122, 129, 136, 143, 144, 156 Chloramphenicol, 38 Chondrococcus exiguus, 109 Chromatin, 33, 62, 66, 70 Chromosome, 41, 59, 62, 64, 66, 67, 69, 70, 72, 73. 74. 77. 78, 81, 82, 83, 103, 112, 113, 118, 157, 158 Chromosome complex, 66, 67, 69, 71, 74, 75, 82, 90, 103, log, 1 18 Cilium, 40 Classification, i, 4 Clostridium, 59, 100 Clostridium tetani, 79, 80 Clostridium welchii, 59, 100 Coccus {see also generic names), 13, 27, 28, 29, 37. 55, 59. 60, 65, 66, 67, 86, 87, 149, 155 Coil, primary, 139 Colony, 21, 22, 74, 89, 97, 121, 122, 125, 134- 144 Complex vegetative reproduction, 67-72, 79, 85-98 Conjugation {see Sexuality) Constriction {see Cell division) Corynebacteria, 5, 12, 55, 65, 66, 86, 88, 89, 91, g2, g6, g8 Corynebacterinm diphtheria;, 5, 2g, 57, gi, 92, 95. 97 Coverslip, 7, g, 14, 89 Cross-wall, 5, 20, 27, 2S, 31, },2, 36, 37, 38, 43, 62, i4g Crystal violet {see also Tannic-acid-violet), 8, II. 59 Culture, 65, 67, 77, 78, Si, 87, 89, 92, 97, 109, 118, 119, 120, 129, 137 Cyst, 67, 125 1 62 Cytochemistry, 4-22, 29, 31 Cytoplasm, 5-9, 18, 22, ^z, 35, 41, 37, 59, 78, 87, 102, 125, 159 Cvtophaga, 59, 65, 81, 105, 106, 109, 120 Dark-ground illumination, 16 Dehydration, 14 DeLamater and Mudd, 53 Dienes, 127 Diploid [see also Nucleus), 102, 105, 113, 142, 158, 159 Distortion, 2, 6, 14, 62 Drying, 2, 5, 0, 9, 32 Dyes, 4, 5, 8, 9, 10, i 35. r-25 Eisenberg, 32 Elasticotaxis, 40 Electrolyte, 139 Electron microscopy, 3, 16, 17, 20, 32, a. 35. 37. 38, 41. 4^. 47. 53. 59, 62, 122, 129 Embedding, 17, 20, 02 Endospore (sff Spore) Enzyme, 9, 17, 18, 22, 160 Eosin, 8 Eubacteria, 51, 62, 65, 67, 82, 86, 91, 95, 99, 103, 104, 105, 106, 109, 113, 115, 119. 120, 121, 124, 136, 144, 153 Evolution, 146-160 Fat [see Lipid) Fell, 96 Feulgen reaction, 3, 6, 7, 31, 57, 75, no Fibre, 40 Filament, 37, 62, 67-72. 80, 86, 95, 96, 103, 113, 122, 127, 129, 136 Film, 4, 8, 9 Filterable stages [see also Gonidia), 94 Fission, 65, 85-92, 94, 98, 104, 118 Fixation [see also Heat-fixation), 7, 12, 17, 35. 62, 91 Fixative, 7, 12 Flagellum, 3, 14, 16, 21. 37, 38, 40, 42, 43, 44, 47, 48, 122, 124, 131, 144, 149-15^ Flexion, 25, 40 Formalin, 35 Fragmentation, 69-73, 92, 94, 97 Fruiting body, 106, 118, 119, 120, 134, 135, 136, 144,' 155 Fuchsm, 9, 53, 77 Fungi, 55, 93. 96, 99. 103, 136, 147. 15°, i59 Fusiformis ywcrophoriis, iij G-forms, 125 Gamete, 82, 91, 109, 113 Gene, 64, 72, 157, 160 Genetics, 3, 64, 72, 73, 99, 109, 154, 157-160 Germination, 41, 44, 47, 48, 64, 65, 67, 86, 97, 104, 109, 118, 129 Germination pore, 59 Giant bacteria, 28 Giemsa (see also acid-Giemsa), 6, 7, iS, 51, 33, 59. I "9 Glass, 14 Gonidia, 1 24-131 Grace, 147 Graham-Smith, 88 Gram stain, 2, 4, ; 147, 149, 133, 155 Granule, 10, 33, 37, 74, 9 Growth cycle, 85, 86, 97 Growing point, 33, 37, 38, 44, 47, 48 S-36, 47, 60, 71, 79, 97, 2, 124, 125 Hadley, 123 Haematoxylin, 10, 31 Hale, 27, 31, 32, 149 Haploid {see also Nucleus), 102, 113, 142, 150. 157. 158 Heat-fixation, 2-3, 9, 21, 31, 57, gz, 97 Henry, 33, 33 Heterozygote, 90 Hillier, 20 Hiss's stain, 12 Hofmann, 35 Holdfast, 136 Holdsworth, 29 Homothallic conjugation, 157 Hormone, 134 Hydrolysis, 6, 7, 22, 35, 41, 53, 37, 59, 60 Hypha, 79, 113. i^i Immersion oil, 4 Impression preparations, 21, 22, 97 Inanition, 81, 134 Indian ink, 12 Industry, i Infolding of colonies, 137-140 Iron bacteria, 121, 122, 136 Irradiation, 74, 159 Jinks, 157 Klieneberger, 125, 127 Klieneberger-Nobel, 32, 64, 102, 113, 135 Kluyver and van Niel, 147, 149 L-organisms, 125, 127 Lactobacillus, 70, 71, 78, 137 Lag-phase, 65, 85, 86, 97 Lamina, 43 Lederberg, 157, 158 Lens, 4, 7 Lewis, 78 Life-cycle, i, 3, 104, 118-131, 144 Light,' 14 Light, ultra-violet, 53, 62, 75, 159 Lindegren, 91, 94, in, 157 Lipid, 7, 10, 12, 22, 27, 29, 31. 35. 47. 78 Lipoprotein, 11, 141 Logarithmic phase, 65, 85, 86, 97 Lysozyme, 10, 18, 35, 43 Macroformation, 134-144 Masking effect, 5, (> Maturation of resting cells, 79, 81, 100, 104- 112 Medicine, i Medium, lo, 35, 85, 86, 119, 129, 140 Medusa-head colony {see also Colony), 36, 141 Meiosis, 74, 102, 103, 112 Mellon, Qi, 94, 1 1 1 Metabolism, 1 18 Metal, 17 Metachromatic granules, 5, 10, 37, 91, 92 Methacrylate, 1 7 Methylene blue, 8, 9 Methylene-blue-eosin, 8, 53, 57, ()o Methyl green (see also Hale), 12 Methyl violet, 60 Micrococcus cryophilus, 55 Microcyst, 37," 44, 47, 48, 57, 64, 65, 77, 78, 81, 82, 83, 86, 97, 104, 105-109, III, 119, 120, 129, 134, 136, 144 M icromonospora, 155 Microscope, 3, 7, 9, 12, 62, 65 Microscopy {see also Electron, Phase-contrast, Ultra-violet), 3, 14, 33, 40 Microtome, 17 Mitchell and Moyle, 35 Mitochondria, 10, 32, 33, 38, 40 Mitosis {see also Nucleus), 53, 55, 66, 72, 73, 74 Mordant, 12, 32 Morris, 32, 104, 112, 113, 114 Motility, 4, 40, 43, 48, 137 Mucus, 134 Muir's stain, 1 1 Multicellularity, 6, 25-38, 55, 64, 66, 67, 86, 88, 92, 94. 99. 119, 121. 134. 136. 139. 142, 149 Murray, 31, 36 Mycelium, 79, 82, 95, 96, 98, 105, 113, 121, 129, 142, 158 Mycobacteria, 32, 65, 81, 86, 92, 94, 95, 96, 98. Ill, 153 Mycobacterium tuberculosis, 6, 57, 66, 75, Sz, 91, 94. 96. 97. 102, III, 112, 124 Mycobacterium lacticola, 60 Mycobacterium phlei, 92, 94 Mycologist, 2 Myosin, 44 Myxobacteria, 2, 3, 7, 8, 40, 48, 51, 59, 62, 65, 70, 71, 77, 81, 82, 90, 91, 103, 104, 105, 106, 109, 115, 118, 119, 120, 129, 134. 135. 136. 142. 144. 155 Myxococcus fulvus, 109 Myxococcus virescens, 109 Myxophyceae {see Blue-green algae) Naphthol dyes, 10 Neisseria gonorrhcece, 127 Nigrosin, 8, 12, 60 Nitrogenous material, 29 Nocardia, 27, 60, 104, 106, 109, 121, 153 Nuclear membrane, 41, 74, 118 Nucleic acid, 35, 51, 65, 160 deoxyribose, 6, 9, 35, 36, 160 ribose, 6, 7, 9, 35, 77 Nucleoprotein, 3, 6, 7, 21. 33, id<, 47, 55 deoxyribose, 6, 7, 15 ribose, 6, 7, 22, 35, 77, 85, 86, 97 EX 163 Nucleotide, 85, 86, 97 Nucleus, 3, 6-12, 18, 20, Z2. 32, 36, 41, 51-88, 87, 90, 97, 100, 103, 105, 106, 109, 113, 118, 147, 154 central, 57, 82, 1 18 chromosomal, 59, 60, 75 crescentic, 57, 59, 60 cycle, 75, 99-1 16 diploid, 118, 158 extra-cytoplasmic, 60 fusion, 68, 69, 71, 90, 102, io<> primary, 70, 76 resting! 9, 53-61, 65, 66, 67, 75-82, 86, 97, 1 02, 1 05- 1 16 rod-shaped, 65, 78-82, 87, 100, 102, 106, 1 1 3 158 secondary, 75-79, 106-111 spherical vegetative, 55, 60, 65, 66, 75 spiral, 78 vegetative, 9, 55, 60, 62, 65, 70, 76, 150 vesicular, 59, 60, 64, 65, 67, 74, 75, 81, 82, 112, 118 Nutrition, 10, 86 Objective, 14 Oil, immersion, 4 Oligosaccharide, 29 Oogenesis, 102 Oscillospira, 28, 52, 62, 106 Osmium tetroxide (osmic acid), 7, 10, 20, 38, 62 Oxidation, 38, 154 Paillot, 53 Parasitism, 121, 127, 147, 155 Pentose, 7, 35 Perchloric acid, 7 Periodic acid, 29 Peristalsis, 40 pH, 10, 18 Phase-contrast, 3, 10, 12, 16-20, 73, no Phosphoric ester, 35 Phosphomolybdic acid {see also Hale), 12, 22, 32 Photomicrography, 2, 14, 71, 103 Physiology, i Phytomonas tumefaciens, no Piekarski, 51, 75, 76, 157 Pijper, 33, 151 Plant, 29, 37, 81, 136, 147, 150. 159 Plant pathogen, 77 Plant seed, 57 Plasmolysis, ^2 Pleuropneumonia bovine, 125 Pneumococcus, 140, 141, 160 Polar body, 102, 109 Polarity, 87 Polypeptide, 2d<. 43, 47, 140 Polyploid, 158 Polysaccharide, 10, 18, 29, ^2, 35, 36, 40, 41, '43, 47, 140, 141 Polystyrene, 17 Post-fission-movements, 88-99 1 64 , 2.2, 31, 32, 35, Prevot, 115 Primary coil, 139 Protein, 3, 6, 7, 10, 12, 17 40, 44, 47, 48, 59, 154 Protein precipitants, 17, 32 Proteus, 43, 44, 47, 57, 71, 115, 127, 135. 136, 144, 15^ Protista, 44, 149 Protoplasm, 6, 35 Protoplast, 5, 42, 88 Protozoa, 99, 103 Pseiidomonas, 27, 38, 43, 47, 151 Pus, 4 Radiation, 72, 74 Reagent, 7, 14 Reduction, nuclear, 65, 69, 73, 75, 99, io-2, no, 112, 115, 118, 150 Reflecting microscope, 3 Reproduction, 68-72, 92 Reproduction, complex vegetative, 67-72, 79, 85-98 Reticulocyte, 33 Rhizobiiim, 36, no, 124, 129 Ribonuclease, 9 Ribonucleic acid {see Nucleic acid) Robinow, 27, 31, 32, 36, 41, 51, 64, 157 Romanowsky stains, 8 Rooyen, van, 97 Rough morphology, 28, 33, 36, 64, 65, 67, 70, 82, 83, 86, 89, 99, 100, 102, 106, 109, 137. 139. 140 Rubin, 72 Salmonella typhi, 47, 57, 77, 15^ Sap vacuole, 32 Sarcina, 60, 149 Schaudinn, 36, 79, 102 Schiff's reagent, 6 Schizomycetes, 147 Seed, 81 Segregation, 74 Sectioning, 16, 17, 20, 21, iz, 27, 32, 33, 37, 41, 62 Septum, 12, 18, 21, 27, 29, 32-38, 47, 86, 87, 96, 124 Sexuality, 73, 77-83, 88, 90, 91, 97, 99-116, 118, 125, 129, 158 Shigella flexneri, 55, 68, 71, 139, 140 Shigella schmitzii, 57 Shrinkage, 5, 6, 14 Slide, 4, 7, 9, 14 Slime layer, 12, 41 Smooth morphology, 28, 33, 36, 37, 64, 67, 70, 78, 82, 83, 86, 88, 89, 90, 91, 100, 102, 106, 141 Soil, 119 Spherophoriis, 115 Spiral morphology, 25, 67, 146-156 Spirillum, 27, 35, 38, 42, 44, 129, 131, 150-156 Sph(srotilus discophorus, 121 Sphcsrotilus natans, gi Spirochaetes, 125 Sporangium, 100, 102, 115 Spore, 41, 44, 52, 57, 59, 64, 65, 67, 71, 76, 79, 80, 81, 82, 96, 99, 100, 102, no, 113, 115, 120, 121, 129, 142, 144, 149, 153, 154 Spore coat, 41, 48, 57, 59, 113 Sporulation, 27, 76, 83, 100, 102, 106, 158 Stacey, 35 Stalk, 87, 119, 124, 134, 136, 144 Staphylococcus, 29, 87 Stille, 51, 157 Stoughton, 53, 75, 77, 109, III Streptobacillus moniliformis, 97, 115, 125, 127 Streptococcus, 27, 28, 29, 69, 70, 141, 142 Streptococcus fcscalis, 68, 71 Streptomyces, 62, 79, 80, 81, 82, 95, 96, 102, 113, 115, 121, 129, 142, 144, 153, 155, 158 Sudan dyes, 12 Sulphydryl, 44 Surface tension, 25, 139 Swarm, swarm cells, 25, 36, 71, 72, 90, 115, 118, 119, 120, 124, 129, 135, 136, 137, 139. 144 Symbiosis, 127 Symplasm, 90, 103 Systematics, 146-156 Tannic acid, 12, 22, 28, 31, 32, 59 Tannic-acid-violet, 12, ^2, 92, 95, 129 Teece, 35 Temperature, 6, 7, 85 Text-book, 14, 88, 137 Tomcsik, 17, 20, 28, 35, 37, 43 Trichloracetic acid, 7, 55 Trichobacteria, 153 Trypsin, 10, 18 Tyrosine, 35 Ultraviolet light, 53, 62, 75, 159 Vaseline, 14 Vibrio, 43, 44, 151, 15^, 156 Victoria blue, 12, 31 Volutin, 10 Water-immersion lens, 7 Wax, 14 WeibuU, 35 Witkin, 74, 159 Wyckoff and Smithburn, 94 X-rays, 159 Yeast, 96 Ziehl-Neelsen stain, 33, 91 Zygosaccharomyces, 157 Zygospore, 72 Zygote, 75, 102, 157 Printed m Great Britain MtLAfiAN & Ci MMiNi, Ltd., Kdi>