|UFDC Home||myUFDC Home | Help|
|Table of Contents|
|2. Symptomatology of viral diseases...|
|3. Transmission of plant viruses...|
|4. Identification of plant...|
|5. Strains, mutation, acquired...|
|6. The transmission of plant viruses...|
|7. Aphid transmission of stylet-borne...|
|8. Virus-vector relationships:...|
|9. Local-lesion assay of plant...|
|11. Serology: Techniques used in...|
|12. Electron microscopy: Principles...|
|13. Structure and function of regular...|
|14. Structural biochemistry of...|
|15. The biochemistry of virus...|
|16. Control of plant virus...|
|17. Speculations on the origins...|
|18. Viruses and molecular...|
|Appendix 1. Form and function:...|
|Appendix 2. Virus diseases...|
|Appendix 3. An introduction to...|
|PRIVATE ITEM Digitization of this item is currently in progress.|
|Table of Contents|
Table of Contents
2. Symptomatology of viral diseases in plants
3. Transmission of plant viruses by grafting, dodder, seed, and mechanical inoculation
4. Identification of plant viruses
5. Strains, mutation, acquired immunity, and interference
6. The transmission of plant viruses in soil
7. Aphid transmission of stylet-borne viruses
8. Virus-vector relationships: Vectors of circulative and propagative viruses
9. Local-lesion assay of plant viruses
11. Serology: Techniques used in plant virus research
12. Electron microscopy: Principles and application to virus research
13. Structure and function of regular virus particles
14. Structural biochemistry of plant viruses
15. The biochemistry of virus infection
16. Control of plant virus diseases
17. Speculations on the origins and nature of viruses
18. Viruses and molecular taxonomy
Appendix 1. Form and function: A problem in virology
Appendix 2. Virus diseases of arthropods
Appendix 3. An introduction to the tumor viruses
Ti .+.+*+++++ +.+.+..+. + .++ .+
Lj A ,A A
DR. L. 0. KUNKEL-1884-1960
*11 *k+A *k. A Ak A Ik .k AL .k AL 9k. A 9@* A*;@ eA e AL eAe
M. K. CORBETT
H. D. SISLER
University of Florida Press
A AI& A AIL A A& A A& A Ak+#" A A& A A& A A& A A& A
low YIWYIWYIW YWYW
A University of Florida Press Book
PUBLISHED WITH ASSISTANCE FROM THE NATIONAL SCIENCE FOUNDATION
COPYRIGHT @ 1964 BY THE BOARD OF COMMISSIONERS OF STATE INSTITUTIONS OF FLORIDA
All rights reserved. No part of this book may be reproduced in any form without permission in
writing from the publisher, except by a reviewer who wishes to quote brief passages in con-
nection with a review written for inclusion in magazines or newspapers.
Library of Congress Catalog Card Number: 64-23327
PRINTED BY ROSE PRINTING COMPANY, INC., TALLAHASSEE, FLORIDA
BOUND BY UNIVERSAL DIXIE BINDERY, INC., JACKSONVILLE, FLORIDA
,, L I d+ I &I & ,
,-.-- +-.- .t Editors' Preface
THIS BOOK has been compiled from a series
of lectures and laboratory exercises pre-
sented during a Southern Regional Gradu-
ate Summer Session in Plant Virology,
which was held at the University of Mary-
land, June 24 to August 2, 1963. The cur-
riculum for the course was designed to
provide students with the fundamentals
and recent advances in the field of plant
virology. The science of plant virology had
its beginning near the end of the nineteenth
century when Beijerinck described the fil-
terable infectious extract from mosaic dis-
eased tobacco plants as a virus or a
contagium vivum fluidum. Today plant
virology encompasses many disciplines as
exemplified by the table of contents of this
book. The subject matter ranges over a
wide variety of topics including symptoma-
tology, vector relationships, chemical na-
ture, electron microscopy, and structure
and substructure of viruses, which requires
a student who wishes to master the science
of virology to have a wide knowledge in the
biological and physical sciences. It is dif-
ficult for one individual to accumulate suf-
ficient knowledge to be an authority in all
phases of virology. Thus, to cover each of
these topics, specialists in the various areas
were invited to participate in the course
for periods of 1-2 weeks and present lec-
tures and laboratory exercises in their field
of specialization. Each lecturer prepared a
review of his subject matter so that it would
be available to a much larger audience than
the 32 students enrolled from the 15 par-
ticipating southern institutions. These lec-
tures have been published in anticipation
that they will stimulate further research
and development in the science of virology.
The need for a textbook of plant virology
has long been evident by the number of
review articles that have appeared in the
last few years in many scientific journals.
It is recognized that this text will not make
a virologist out of a student but it is hoped
that it will suffice as an introduction to the
subject of plant virology and provide a
stimulus for students to continue in the
search for knowledge rather than to accept
The success of this course depended
greatly upon the cooperation of the fol-
lowing to whom special recognition and
acknowledgment is hereby given: to the
Regional Committee of the Southern Re-
gional Education Board for its work in the
initial planning stages; to Drs. S. J. P. Chil-
ton and J. P. Fulton, members of the
Executive Committee, for assistance in
preparing the curriculum, selection of lec-
turers, and preparation of the grant pro-
posal; to Dr. William L. Bowden, As-
sociate Director for Regional Programs,
Southern Regional Education Board, who
worked with the Regional and Executive
Committees throughout the course of the
project; to Dr. Ronald Bamford, Dean of
-A A AA L AkA*LAAkA k AAkA k
the Graduate School and former Head
of the Department of Botany, University of
Maryland, who served as representative of
the Regional Advisory Council on Graduate
Education in the Agricultural Sciences and
made many valuable contributions at all
stages of the program; to the University of
Maryland, for so generously providing the
necessary facilities for conducting the
course; to the staff of the Botany Depart-
ment, University of Maryland, for their
assistance and cooperation; to the lecturers
and students who made the course so suc-
cessful; and finally to the National Science
Foundation for providing the grant which
made the course possible.
Appreciation is expressed to President
J. Wayne Reitz and Provost for Agriculture
E. T. York, Jr., of the University of Florida
for their cooperation in arranging for pub-
lication of this volume by the University of
Acknowledgment is made to Mrs. J. C.
McCollum, Mrs. N. E. Link, and Mrs. A. A.
Bell for typing manuscripts, and to Mr.
R. K. Jones for assistance in proofreading.
M. K. COBBETT
Chairman Executive Committee and
H. D. SISTER
ELLEN M. BALL
PLANT PATHOLOGY DEPARTMENT
UNIVERSITY OF NEBRASKA
F. C. BAWDEN
ROTHAMSTED EXPERIMENTAL STATION
HARPENDEN, HERTFORDSHIRE, ENGLAND
R. H. E. BRADLEY
DIVISION OF PLANT PATHOLOGY
AGRICULTURAL RESEARCH STATION
FREDERICTON, NEW BRUNSWICK, CANADA
DIVISION OF PLANT PATHOLOGY
GLASSHOUSE CORPS RESEARCH INSTITUTE
LITTLEHAMPTON, SUSSEX, ENGLAND
D. L. D. CASPAR
THE CHILDREN'S CANCER RESEARCH
THE CHILDREN'S MEDICAL CENTER
THE HARVARD MEDICAL SCHOOL
M. K. CORBETT
PLANT VIRUS LABORATORY
UNIVERSITY OF FLORIDA
ROBERT W. FULTON
DEPARTMENT OF PLANT PATHOLOGY
COLLEGE OF AGRICULTURE
UNIVERSITY OF WISCONSIN
C. E. HALL
DEPARTMENT OF BIOLOGY
MASSACHUSETTS INSTITUTE OF TECHNOL-
B. D. HARRISON
ROTHAMSTED EXPERIMENTAL STATION
HARPENDEN, HERTFORDSHIRE, ENGLAND
F. 0. HOLMES
THE ROCKEFELLER INSTITUTE
NEW YORK, NEW YORK
C. A. KNIGHT
UNIVERSITY OF CALIFORNIA
DEPARTMENT OF MICROBIOLOGY
MAX A. LAUFFER
DEPARTMENT OF BIOPHYSICS
UNIVERSITY OF PITTSBURGH
ROBERT A. MANAGER
LABORATORY OF VIRAL ONCOLOGY
NATIONAL CANCER INSTITUTE
THE BOYCE THOMPSON INSTITUTE FOR
PLANT RESEARCH, INC.
YONKERS, NEW YORK
W. C. PRICE
PLANT VIRUS LABORATORY
UNIVERSITY OF FLORIDA
D. A. ROBERTS
DEPARTMENT OF PLANT
UNIVERSITY OF FLORIDA
A. F. ROSS
DEPARTMENT OF PLANT PATHOLOGY
ITHACA, NEW YORK
H. D. SISLER
UNIVERSITY OF MARYLAND
COLLEGE PARK, MARYLAND
KENNETH M. SMITH
DEPARTMENT OF BIOPHYSICS
UNIVERSITY OF PITTSBURGH
R. L. STEERE
PLANT VIROLOGY LABORATORY
CROPS RESEARCH DIVISION
WILLIAM N. TAKAHASHI
DEPARTMENT OF PLANT PATHOLOGY
UNIVERSITY OF CALIFORNIA
*AL A&, AL* Contributors
K. NM. SlITH
C. E. HALL
R. H. E. BRADLEY
H. D. SISLER
R. L. STEERE
D. L. D. CASPAR
ELLEN NM. BALL
W. N. TAKAHASHI
F. C. BAWVDEN
B. D. HARRISON
F. 0. HOLMES
M. K. CORBETT
R. WV. FULTON
D. A. ROBERTS
M. A. LAUFFER
A. F. Ross
W. C. PRICE
C. A. KNIGHT
1-M. K. CORBETT
1-1. DEVELOPMENT OF THE SCIENCE
OF VIROLOGY . . .
1-2. NOMENCLATURE AND CLASSIFI-
1-3. ECONOMIC IMPORTANCE
1-4. LrrERATURE CTrED . . .
2-F. 0. HOLMES
Symptomatology of viral diseases in
2-1. INTRODUCTION .
2-2. FOLIAGE SYMPTOMS .
A. ABNORMALITIES OF LEAF COLOR
B. ABNORMALITIES OF LEAF NUMBER
C. ABNORMALITIES OF LEAF SHAPE
D. ABNORMALITIES OF LEAF SIZE
E. ABNORMALITIES OF LEAF TEXTURE
F. ABNORMALITIES OF LEAF LUSTER
G. ABNORMALITIES OF LEAF POSITION
H. ABNORMALITIES OF LEAF HISTOLOGY
I. ABNORMALITIES OF LEAF CYTOLOGY~
J. ABNORMALITIES OF MISCELLANEOUS
PHYSIOLOGICAL PROCESSES IN LEAVES
K. ABNORMALITIES COMPRISING DEATH
OF WHOLE LEAVES OR OF MACRO-
SCOPIC PARTS OF LEAVES
2-3. FLOWER SYMPTOMS
A. ABNORMALITIES OF FLOWER COLOR
B. ABNORMALITIES OF FLOWER SIZE
C. ABNORMALITIES OF FLOWER NUM-
D. ABNORMALITIES OF FLOWER SHAPE
E. ABNORMALITIES OF FLOWER POSITION
F. ABNORMALITIES OF POLLEN AND EGG
VIABILITY IN FLOWERS
G. ABNORMALITIES IN TIME OF BLOS-
H. ABNORMALITIES OF FLOWER PERSIST-
2-4. FRUIT SYMPTOMS .
A. ABNORMALITIES OF FRUIT NUMBER
B. ABNORMALITIES OF FRUIT SIZE
C. ABNORMALITIES OF FRUIT SHAPE
D. ABNORMALITIES OF FRUrr HABIT
E. ABNORMALITIES OF FRUIT COLOR
F. ABNORMALITIES OF FRUIT TEXTURE
G. ABNORMALITIES OF FRUIT MATURA-
H. ABNORMALITIES OF FRUIT FLAVOR
I. ABNORMALITIES OF SEED CONTENT
OF FRUITS 31
J. ABNORMALITIES OF PERSISTENCE OF
K. ABNORMALITIES OF YIELD OF FRUITS 31
2 L. DEATH OF PARTS OF FRUITS 31
2-5. STEM SYMPTOMS 32
A. ABNORMALITIES OF GROWTH HABIT
10 OF STEMS 32
15 B. ABNORMALITIES OF INTERNODE
LENGTH OF STEMS 32
C. ABNORMALITIES OF NUMBER OF
D. ABNORMALITIES OF STEM SHAPE 32
E. ABNORMALITIES OF BARK OF STEMS 33
F. ABNORMALITIES OF STEM COLOR 33
17 G. ABNORMALITIES OF STEM RIGIDITY 33
19 H. ABNORMALITIES OF STEM HISTOLOGY 33
9 I. ABNORMALITIES OF STEM PHYSIOL-
2 OGY 34
2 J. ABNORMALITIES OF CHEMICAL CON-
4 STITUTION OF STEMS 34
4 K. ABNORMALITIES OF GUM PRODUCTION
4 IN STEMS 34
4 L. DEATH OF STEMS OR PARTS OF
5 STEMS 34
2-6. BUD SYMPTOMS .
A. ABNORMALITIES OF DORMANCY OF
B. ABNORMALITIES OF STRUCTURE OF
C. ABNORMALITIES OF VIABILITY OF
2-7. ROOT SYMPTOMS .
A. ABNORMALITIES OF ROOT NUMBER
B. ABNORMALITIES OF ROOT SIZE
C. ABNORMALITIES OF ROOT SHAPE
D. ABNORMALITIES OF ROOT TEXTURE
E. ABNORMALITIES OF ROOT COLOR
F. ABNORMALITIES OF CHEMICAL
COMPOSITION OF ROOTS
G. DEATH OF ROOTS OR PARTS OF
2-8. SYMPTOMS EXPRESSED BY THE
PLANT AS A WHOLE .
OF LENGTH OF LIFE
OF GROWTH RATE
OF SIZE OF PLANT
OF CHEMICAL COM-
H. ABNORMALITIES OF HISTOLOGY
ALA AkAAdLAAi&AALjLA& A ALA AL4
,W'Vl Nr IV IMF V, W'Vl W'Vl w lrwv NV v
I. ABNORMALITIES OF LATEX FLOW
J. ABNORMALITIES OF GROWTH HABIT
2-9. MASKING OF SYMPTOMS .
2-10. LITERATURE CITED
3-R. W. FULTON
Transmission of plant viruses by grafting,
dodder, seed, and mechanical
3-1. INTRODUCTION .
3-2. GRAFT TRANSMISSION .
3-3. DODDER TRANSMISSION OF PLANT
A. THE MECHANISM OF DODDER TRANS-
B. PRACTICAL ASPECTS OF DODDER
C. METHODS OF USING DODDER
3-4. SEED TRANSMISSION OF PLANT
A. THE RATE OF VIRUS TRANSMISSION
B. POLLEN TRANSMISSION
C. ELIMINATION OF VIRUS FROM SEED
D. LATENT VIRUS IN SEED
E. CORRELATION OF SEED TRANSMISSION
WITH OTHER VIRUS CHARACTERISTICS
F. MECHANISMS RESTRICTING SEED
3-5. MECHANICAL INOCULATION .
B. METHODS OF APPLYING INOCULUM
C. SPRAYING METHODS OF APPLYING
D. PRICKING AND INJECTION APPLICATION
E. HOST SUSCEPTIBILITY; THE USE OF
F. PHYSIOLOGICAL FACTORS OF HOST
G. EFFECT OF ILLUMINATION
H. EFFECT OF HOST NUTRITION ON
I. EFFECT OF HEAT TREATMENTS ON
J. OTHER FACTORS AFFECTING SUSCEPTI-
K. INFECTIVITY OF INOCULUM; THE
L. INHIBITORS IN PLANT EXTRACTS
M. POLYPHENOLS AND ANTIOXIDANTS
N. SOURCES OF INOCULUM
0. THE MECHANISM OF INFECTION AND
FACTORS AFFECTING IT
P. RIBONUCLEASE AND VIRUS TRANS-
3-6. CONCLUSIONS .
3-7. TABLES .
3-8. LrrERATURE CITED .
4-A. F. ROSS
Identification of plant viruses-68
4-1. INTRODUCTION . . ...
4-2. ESTABLISHING THAT THE "UN-
KNOWN IS A VIRUS . .
4-3. ESTABLISHING THAT ONLY ONE
VIRUS IS PRESENT . ..
4-4. GENERAL APPROACH TO IDENTI-
FICATION . . .
4-5. TECHNIQUES AND TESTS USED IN
IDENTIFICATION . . .
A. SYMPTOMATOLOGY AND HOST RANGE
B. PROPERTIES IN CRUDE JUICE
C. TRANSMISSION CHARACTERISTICS
D. INTERACTION WITH OTHER VIRUSES
E. RESPONSE TO PRESENCE OF SPECIFIC
GENES IN THE HOST
F. SEROLOGICAL TESTS
G. VIRUS PARTICLE CHARACTERISTICS
4-6. CHOICE OF CRITERIA TO USE IN
ESTABLISHING THE IDENTITY OF
AN UNKNOWN . . .
4-7. VIRUS DESCRIPTIONS AND NAMES .
4-8. ROUTINE DIAGNOSES . .
4-9. LITERATURE CITED . .
46 5-W. C. PRICE
47 Strains, mutation, acquired immunity, and
. . 63
EVIDENCE FOR MUTATION
PERMANENCE OF MUTANTS
RATE OF MUTATION
CHARACTERISTICS ALTERED BY
E. ARTIFICIAL INDUCTION OF MUTA'
5-4. STRAINS .
A. SIMILARITIES AND DIFFERENCES
B. IDENTIFICATION OF STRAINS
C. NATURALLY OCCURRING STRAINS
D. USEFUL STRAINS
5-5. ACQUmIED IMMUNITY 103
A. ACQUIRED IMMUNITY FOLLOWING
RECOVERY FROM DISEASE 104
B. CROSS PROTECTION 108
5-6. INTERFERENCE 110
A. INTERFERENCE RESULTING FROM
USE OF MIXED INOCULA 111
B. INTERFERENCE WITHIN AN INOCU-
LATED LEAF 111
C. LOCALIZED INTERFERENCE IN
HYPERSENSITIVE HOSTS 112
D. SYSTEMIC INTERFERENCE IN HYPER-
SENSITIVE HOSTS 112
E. INTERFERENCE BY INACTIVE VIRUS 114
5-7. LrrERATURE CT 114
6-B. D. HARRISON
The transmission of plant viruses in
6-1. INTRODUCTION 118
6-2. ECONOMIC IMPORTANCE OF
SOIL-BORNE VIRUSES .. 119
6-3. GENERAL PROPERTIES OF THE
A. NEMATODE-TRANSMITTED VIRUSES
WITH POLYHEDRAL PARTICLES
B. NEMATODE-TRANSMITTED VIRUSES
WITH TUBULAR PARTICLES (NETU-
C. TOBACCO NECROSIS AND ALLIED
D. LETTUCE BIG-VEIN AND ALLIED
E. WHEAT MOSAIC AND ALLIED VI-
F. TOBACCO MOSAIC VIRUS 126
6-4. METHODS OF TRANSMISSION .
A. TRANSMISSION BY NEMATODES
B. FUNGUS-ASSISTED TRANSMISSION
C. METHOD OF TRANSMISSION UN-
6-5. VIRUS ECOLOGY . .
A. METHODS OF SURVIVAL
B. DISTRmIBUTION AND SPREAD
C. INCIDENCE OF INFECTION
6-6. METHODS OF CONTROL
A. PREVENTION OF SPREAD TO NEW
B. CONTROL MEASURES ON INFESTED
6-7. CONCLUDING REMARKS .
6-8. LITERATURE CITED . .
7-R. H. E. BRADLEY
Aphid transmission of stylet-borne
7-1. INTRODUCTION .. 148
7-2. FRAMEWORK .. 149
A. APHIDS 151
B. VIRUSES 153
C. PLANTS 154
7-3. UPTAKE .. 155
A. DURING SUPERFICIAL PROBES 155
B. WHERE ON THE SOURCE? 156
C. HOW IS THE LABIUM PLACED IN RE-
LATION TO THE EPIDERMAL CELLS? 157
D. Is UPTAKE FROM WITHIN OR BE-
TWEEN EPIDERMAL CELLS?
F. UPTAKE OTHER THAN DURING SU-
G. WHY DOES UPTAKE MAINLY OC-
CUR DURING SUPERFICIAL PROBING?
H. EFFECT OF KEEPING APHIDS OFF
PLANTS PRIOR TO UPTAKE
7-4. CARRY OVER .
A. ON PLANTS
B. OFF PLANTS
C. STYLET TREATMENTS
7-5. INOCULATION .
7-6. LITEATURE TED .
Virus-vector relationships: Vectors of
circulative and propagative
A. CIRCULATIVE TRANSMISSION
B. PROPAGATIVE TRANSMISSION
C. VIRUS INTERRELATIONSHIPS AND
D. LOSS OF TRANSMISSIBILITY
E. Vmus ACQUISITION AND TRANS-
F. VIRUS INTERRELATIONSHIPS
8-8. EFFECTS OF PLANT VIRUSES 01N
VECTORS AND NONVECTORS
A. ASTER YELLOWS VIRUS
B. RICE VIRUSES
8-9. PLANT VIRUSES PATHOGENIC TO
THEIR VECTORS . . .. 187
A. EUROPEAN WHEAT STRIATE MOSAIC
B. RICE DWARF VIRUS 188
8-10. BENEFICIAL VIRUS EFFECTS . 189
8-11. CONCLUSIONS . .. 190
8-12. LrrERATURE CTED .. 191
9-D. A. ROBERTS
Local-lesion assay of plant viruses-194
9-1. INTRODUCTION . . .. 194
9-2. METHODS FOR ASSAYING PLANT
VIRUSES . . .. 194
B. CHEMICAL 196
C. SEROLOGICAL 196
D. BIOLOGICAL 197
9-3. THE LOCAL-LESION ASSAY 198
A. THE DILUTION CURVE 198
B. ASSESSING RELATIVE INFECTIVITY 200
C. FACTORS INFLUENCING THE LOCAL-
LESION ASSAY 205
9-4. DiscussioN 208
9-5. LITERATURE CITED 209
10-R. L. STEERE
10-1. INTRODUCTION . .
10-2. HOST SELECTION . .
A. ASSAY HOST
B. PRODUCTION HOST
10-3. DETERMINATION OF INITIAL
BUFFER . . .
10-4. MACERATION OF INFECTED TIS-
SUES AND JUICE EXTRACTION .
10-5. REMOVAL OF CONTAMINANTS .
A. CLARIFICATION OF THE EXTRACT
B. FURTHER PURIFICATION OF CLARI-
10-6. CONCENTRATION PROCEDURES .
A. REMOVAL OF SOLVENT FROM THE
B. REMOVAL OF VIRUS FROM SUS-
10-7. STORAGE OF PURIFIED VIRUS .
10-8. DETECTION OF IMPURITIES .
B. ELECTRON MICROSCOPY
D. DENSITY-GRADIENT CENTRIFUGATION
F. AGAR-GEL FILTRATION
10-9. SAMPLE PROCEDURE FOR PURI-
FYING VIRUS PARTICLES AND
KEEPING THEM IN SUSPENSION
THROUGHOUT PURIFICATION AND
10-10. LITERATURE CITED .
D. ADJUVANTS OTHER THAN FREUND'S
E. ACQUISITION OF BLOOD SAMPLES
11-3. SEROLOGICAL TECHNIQUES .
A. DILUTION END-POINT PROCEDURES
B. AGAR DIFFUSION PROCEDURES
C. CONJUGATED ANTIBODY PROCEDURES
11-4. APPLICATION . . .. 248
11-5. L RATURE CITrrED . .. 250
11-6. SUPPLEMENTARY READING . 252
12-C. E. HALL
Electron microscopy: Principles and
application to virus research-253
211 12-1. INTRODUCTION
211 12-2. THE ELECTRON MICROSCOPE
12-3. IMAGE CHARACTERISTICS
212 A. Focus
212 C. GENERAL CONSIDERATIONS
213 12-4. TECHNIQUES . .
214 A. SHADOW CASTING
217 C. OTHER TECHNIQUES
228 12-5. MICROSCOPY OF PLANT VII
A. TOBACCO MOSAIC VIRUS
OOo B. SPHERICAL VIRUSES
12-6. LrrERATURE CITED .
11-ELLEN M. BALL
Serology: Techniques used in plant virus
11-1. INTRODUCTION .. 235
11-2. IMMUNIZATION TECHNIQUES 237
A. INTRAVENOUS INJECTION 237
B. INTRAMUSCULAR INJECTION IN
FREUND'S ADJUVANT 238
C. COMPARATIVE STUDIES AND COM-
BINATIONS OF ROUTES OF INJECTION 238
13-D. L. D. CASPAR
Structure and function of regular virus
13-1. INTRODUCTION . . .
13-2. DESIGN PRINCIPLES AND SELF-
13-3. STRUCTURAL CLASSIFICATION AND
INTERRELATIONS OF VIRUSES .
13-4. VIRUS SUBSTRUCTURE AND THE
X-RAY DIFFRACTION METHOD
13-5. SYMMETRY AND MOLECULAR
MORPHOLOGY OF TOBACCO
MOSAIC VIRUS . . 275
A. HELICAL ARRANGEMENT OF SUB-
B. INTERNAL STRUCTURE 277
C. DEFORMED HELICAL STRUCTURES 279
D. SYMMETRY AND STRUCTURE OF
POLYMERIZED PROTEIN 280
13-6. ASSEMBLY OF THE TOBACCO
MOSAIC VIRUS PARTICLE . 282
13-7. DESIGN OF ICOSAHEDRAL VIRUS
PARTICLES . . 285
13-8. LITERATURE CITED . . 290
14-C. A. KNIGHT
Structural biochemistry of plant
14-1. INTRODUCTION .
14-2. COMPOSITION OF VIRUSES
A. GENERAL FEATURES AND RELATIONS
AMONG THE MAJOR CLASSES OF
B. PLANT VIRUSES
14-3. STRUCTURE OF VIRUSES .
A. SUBDIVISIONS OF PROTEIN STRUCTURE
B. PREPARATION OF VIRAL PROTEINS
C. SOME PROPERTIES OF VIRAL PROTEINS
D. PRIMARY STRUCTURAL CHARACTER-
ISTICS OF SOME VIRAL PROTEINS
E. SECONDARY, TERTIARY, AND QUA-
F. METHODS FOR PREPARING PLANT
VIRUS NUCLEIC ACIDS
G. SUBDIVISIONS OF NUCLEIC ACID
H. RECONSTITUTION OF VIRUS FROM ITS
14-4. VIRUS MUTANTS ....
A. SPONTANEOUS MUTANTS
B. CHEMICALLY INDUCED MUTANTS
C. THE CODING RELATIONSHIP
14-5. LITrrERATURE CITED .
15-W. N. TAKAHASHI
The biochemistry of virus infection-
15-1. INTRODUCTION .
15-2. CONCEPTS OF NATURE OF PLANT
VIRUS MULTIPLICATION .
A. OLD CONCEPT
B. NEW CONCEPT
C. BIPARTITE SYNTHESIS OF VIRUS
15-3. INFECTION PROCESS .
A. ADSORPTION PHASE
B. THE INFECTIOUS ENTITY
C. SHEDDING OF PROTEIN COAT
D. HYPERACTIVITY OF THE NUCLEUS
15-4. VIRUS PROTEIN .
A. MACROMOLECULAR NONINFECTIOUS
B. TOP COMPONENT OF TURNIP YELLOW
MOSAIC VIRUS PREPARATION
C. ANOMALOUS PROTEIN OF TMV
D. SOME PROPERTIES OF THE VIRUS
E. THE ROLE OF THIS PROTEIN IN
THE BIOSYNTHESIS OF VIRUS
15-5. PROTEIN SYNTHESIS .
A. STRUCTURE OF PROTEIN
B. SYNTHESIS IN NORMAL CELLS
MECHANISM OF PROTEIN SYNTHESIS
PROTEIN SYNTHESIS WITH ISOLATED
PROTEIN SYNTHESIS WITH ISOLATED
RIBOSOMES OF E. coli
SYNTHESIS OF TMV PROTEIN USING
TMV-RNA AS MESSENGER
FAILURE TO FORM TMV PROTEIN
WITH RIBOSOMES OF PLANT ORIGIN
15-6. INFECTIOUS VIRUS NUCLEIC ACID 323
A. INTEGRITY OF INFECTIOUS RNA 323
B. TMV-RNA 324
C. REPLICATION OF RNA 324
D. RNA POLYMERASES 325
E. EXPERIMENTS REPORTING SYNTHESIS
OF INFECTIOUS RNA 325
300 15-7. DISCUSSION
ele 15-8. LITERATURE CITED
Control of plant virus diseases-330
307 16-1. EPIDEMIOLOGY .
309 A. IDENTITY OF VIRUS
311 B. SOURCE OF VIRUS
3 C. HOST SUSCEPTIBILITY
313 D. IDENTITY AND EFFICIENCY OF VEC-
E. ACTIVITY OF VECTORS
16-2. EXCLUSION AND PROTECTION-
I. INTRODUCTION OF VIRUSES INTO
CROPS BY MAN, MAMMALS, AND
A. VIRUSES SPREAD BY CONTACT
B. PROTECTED CROPPING
C. HYGIENE MEASURES
D. VIRUS INHIBITORS AND INACTIVATORS
E. CULTURAL MODIFICATIONS
F. INFECTION WITH MILD VIRUS
G. VIRUSES INTRODUCED IN VEGETA-
TIVELY PROPAGATED PLANT PARTS
H. SEED INFECTION
16-3. EXCLUSION AND PROTECTION-
11. INTRODUCTION OF VIRUSES IN-
TO CROPS BY INSECTS AND MITES
BARRIER AND COVER CROPS
ELIMINATION OF EXTERNAL SOURCES
OF VIRUS OR VECTOR
BREAKING IN CROP CYCLES
AVOIDANCE OF VECTORS
INFECTION WITH MILD STRAINS OF
J. KILLING VECTORS BEFORE THEY EN-
TER THE CROP
K. KILLING VECTORS AFTER ARRIVAL
16-4. ERADICATION OF VIRUS .
A. ERADICATION FROM SOIL
B. ERADICATION FROM SEEDS
C. ERADICATION FROM VEGETATIVELY
PROPAGATED MATERIAL BY HEAT
D. ERADICATION BY TIP CULTURE
E. ERADICATION BY ROGUING
F. ERADICATION BY CHEMICAL TREAT-
16-5. ERADICATION OF VECTORS .
A. ERADICATION OF VECTORS WITHIN
CROPS BY INSECTICIDES
B. ERADICATION BY PREDATORS AND
C. ERADICATION OF VECTORS IN SOIL BY
C. INFORMATION TRANSFER: REPLICA-
TION, TRANSCRIPTION, TRANSLATION 390
D. MOLECULAR BASIS OF GENETIC HO-
E. THE NIRENBERG-OCHOA CODES 393
18-3. GENERAL TAXONOMIC CONSIDER-
52 A. WHAT IS TAXONOMY?
53 B. WHAT IS A CHARACTER?
53 C. POWER OF THE MOLECULAR AP-
PROACH THROUGH NUCLEIC ACIDS
55 D. HOMOLOGY CRITERIA BASED ON NU-
355 CLEIC ACIDS
E. OTHER MOLECULAR APPROACHES:
55 GENETICS AND PROTEINS
18-4. BACTERIA AND THE MUTUAL
16-6. PLANT IMMUNITY AND TOLER-
ANCE . . . . 35
A. USE OF IMMUNE OR TOLERANT VA-
RIETIES OF PLANTS 357
B. PLANT AGE AT THE TIME OF INFEC-
C. CONTROL BY ALTERING NUTRITION 359
16-7. CONCLUSION . . . 35
16-8. LrrERATURE CITED . 3
17-F. C. BAWDEN
Speculations on the origins and nature of
17-1. INTRODUCTION . .. 365
17-2. CHANGING IDEAS ABOUT VIRUSES 366
17-3. THE ORIGINS OF VIRUSES . 368
A. THE IRRELEVANCE OF POTATO PARA-
B. CHANGES IN THE BEHAVIOUR OF
NUCLEIC ACIDS 369
17-4. INFECTION AND MULTIPLICA-
TION . . .. 374
A. ESTABLISHING INFECTION 377
B. SYNTHESIS AND ASSEMBLY 379
17-5. GENETIC VARIABILITY . 381
17-6. LrrERATURE CITED . 384
Viruses and molecular taxonomy-386
18-1. INTRODUCTION .. 386
18-2. CONCEPTUAL BACKGROUND OF
MOLECULAR TAXONOMY . 387
A. ORGANIZATION OF GENETIC INFOR-
MATION: CODING RELATIONS 387
B. HEREDITARY VARIATION: EFFECTS
OF MUTATION 389
VALIDATION OF CLASSICAL AND
MOLECULAR TAXONOMY . 405
18-5. VIRAL TAXONOMY . . 408
A. A BRIEF ORIENTATION 408
B. SOME RECENT MOLECULAR-TAXO-
NOMIC STUDIES 410
C. A NEOCLASSICAL SYSTEM OF VIRUSES 415
18-6. PROSPECTS AND SPECULATIONS 417
A. TAXONOMIC DISCREPANCIES AND VI-
RAL EVOLUTION 417
B. DIRECTION OF MOLECULAR EVOLU-
18-7. SUMMARY . . 421
18-8. LITERATURE CITED . 423
Al-M. A. LAUFFER
Form and function: A problem in
Al-1. INTRODUCTION . . .
Al-2. THE OPERATIONAL APPROACH .
Al-3. METHODS FOR IDENTIFYING FORM
WITH FUNCTION . . .
A. CORRELATION OF INFECTIOUSNESS
WITH CHARACTERISTIC PARTICLE
C. DIFFUSION COEFFICIENT
D. DESTRUCTION OF INFECTIOUSNESS
G. ELECTROPHORETIC MOBILITY
Al-4. SOUTHERN BEAN MOSAIC VIRUS .
Al-5. TOBACCO MOSAIC VIRUS RNA .
Al-6. LITERATURE CITED . .
A2-K. M. SMITH
Virus diseases of arthropods-439
A2-1. INTRODUCTION ..... 439
A2-2. TYPES OF ARTHROPOD VIRUSES 439
A. NUCLEAR POLYHEDROSES 440 B.
B. CHEMICAL COMPOSITION OF THE
NUCLEAR POLYHEDROSIS VIRUSES C.
AND THEIR POLYHEDRA 440 D.
C. CYTOPLASMIC POLYHEDROSES 441 E.
D. THE GRANULOSES 443 F.
E. CHEMICAL COMPOSITION 443 G.
F. THE FREE VIRUSES 444 H.
A2-3. ORDERS IN WHICH VIRUSES HAVE 1.
BEEN DESCRIBED .. .444 K.
A2-4. DESCRIPTION OF SOME VIRUS L.
DISEASES OF ARTHROPODS 445
A. CHEMICAL COMPOSITION 446 M.
A2-5. ISOLATION AND PURIFICATION 447 N.
A2-6. MORPHOLOGY AND ULTRASTRUC- A3-4
A2-7. METHODS OF SPREAD
A2-8. LATENT VIRUS INFECTIONS .
A2-9. TRANSMISSION BETWEEN DIFFER-
ENT SPECIES AND ORDERS .
A2-10. VIRUSES MULTIPLYING IN
PLANTS AND INSECTS .
A2-11. LrrERATURE CITED . .
A3-R. A. MANAGER
An introduction to the tumor viruses-457
A3-1. INTRODUCTION . .. 457
A3-2. THE AVIAN-TUMOR VIRUSES . 457
A. FOWL LEUCOSIS 457
B. AVIAN SARCOMA VIRUSES 459
A3-3. THE MARINE GROUP OF TUMOR
VIRUSES . .. 469
A. INTRODUCTION 469
THE MOUSE MAMMARY TUMOR
MURINE LEUKEMIA VIRUSES
THE GRoss VIRUS
THE GRAFFI VIRUS
THE FRIEND VIRUS
THE MOLONEY VIRUS
THE SCHOOLMAN-SCHWARTZ AGENT
THE RAUSCHER VIRUS
THE C-60 VIRUS
OTHER LEUKEMOGENIC AGENTS
LEUKEMOGENIC AGENTS IN TISSUE
LEUKEMIA VIRUS MORPHOLOGY
POLYOMA VIRUS TUMORS IN MICE
1. OTHER VIRUSES THAT INDUCE
TUMORS IN RODENTS. . 485
SIMIAN VIRUS 40 485
A3-5. RABBIT TUMOR VIRUSES .
A. SHOPE RABBIT PAPILLOMA VIRUS
B. THE RABBIT FIBROMA VIRUS
A3-6. YABA VIRUS . .
A3-7. PLANT TUMOR VIRUSES . 491
A. THE WOUND-TUMOR VIRUS 491
A3-8. OTHER TUMOR VIRUSES . 492
A3-9. VIRUSES ASSOCIATED WITH HU-
MAN LEUKEMIA . 492
A3-10. LITERATURE CITED .
e'Vg..F **IMF vO+I+*"+*++4O
M. K. CORBETT
THE WORD virus was described in the
Phillips dictionary of 1720 as "a poison,
venom, also a rammish smell as of the arm-
pits; also a kind of watery matter, whitish,
yellowish, and greenish at the same time,
which issues out of ulcers and stinks very
much; being indued with eating and malig-
nant qualities." The modern concept of a
virus as a pathogenic agent did not develop
immediately, but required years of pains-
taking observation and research. By 1950,
a virus was described by Bawden (1950)
as an obligatelyy parasitic pathogen with
dimensions of less than 200 mM." Holmes
(1948) proposed that viruses are "etiologi-
cal agents of disease, typically of small size
and capable of passing filters that retain
bacteria, increasing only in the presence of
living cells, giving rise to new strains by
mutation, not arising de novo." Luria
(1959) defined viruses as submicroscopicc
entities, capable of being introduced into
specific living cells and reproducing inside
such cells only." Various other definitions
have been proposed from time to time, but
these few will suffice to demonstrate how
the connotation of the word virus has been
altered as knowledge has accumulated. In
the following pages, I wish to show, by
presenting some of the important findings,
how this knowledge developed and how
these discoveries have contributed to the
present meaning of the word virus and to
the science of virology.
Undoubtedly, many of the plant diseases
which we now know are caused by viruses
were recognized many years ago, but few
of them attracted attention before the turn
of the nineteenth century. Perhaps one of
the oldest such plant diseases is that of
tulip mosaic, or "breaking" as it was com-
monly called. The condition has been
termed beneficial by some investigators be-
cause it induced color changes in the tulip
flowers that enhance their beauty. They
were so sought after by tulip fanciers dur-
ing the sixteenth and seventeenth centuries
that speculators gambling on the tulip
market created the craze "tulipomania." Al-
though the cause of the breaking was un-
known at that time, planters knew how to
transmit the condition by bulb grafts, and
Blagrave in 1675 gave exact procedures
and details for grafting such bulbs (McKay
and Warner, 1933). The actual association
of "breaking" with a virus had to wait until
1926 when it was shown that the virus
could be transmitted mechanically and by
aphids (McKay and Warner, 1933). An-
other early recognized plant virus disease
was that of Jessamine (Jasminum) mottle.
In a letter published in the philosophical
transactions in 1720 Mr. Henry Cane re-
ported that in 1692 he had transmitted the
mottle condition to a common white Jas-
minum by grafting to it a yellow-striped
Jasminum. In the words of Mr. Cane: "I
have tried several other sorts of variegated
Chapter 1-M. K. Corbett
plants but do not find any of them trans-
mute as that Jessamine will do." A similar
situation was noted by Vibert in 1863 to
occur in trees. He reported that apple trees
budded with buds from aucubaa" plants
produced variegated leaves on the stock
the following spring. In some cases the
buds failed, but the tree still produced
variegated leaves. He concluded that the
scion and stock need only be together long
enough to allow sap to pass from the scion
to the stock. He also reported a similar
condition to occur in grafts between plants
of rose and dog-rose.
1-1. DEVELOPMENT OF THE
SCIENCE OF VIROLOGY
Virus diseases undoubtedly damaged
many crop plants and ornamentals at this
early date, but little was done to find their
cause until the middle of the nineteenth
century. Mayer (1886), an agricultural
chemist working at Wageningen, the Neth-
erlands, investigated a mosaic disease of
tobacco, which had been termed "bunt,"
"rust," or "smut" by growers. To prevent
confusion, Mayer suggested the interna-
tional name of "mosaic disease of tobacco."
The causal agent of the disease was un-
known, though many theories had been
proposed. Mayer attempted many experi-
ments on the etiology of the mosaic disease
and found that the causal agent was trans-
missible to healthy plants in juice extracts.
He postulated that the disease was caused
by an unorganized or organized ferment
and that an unorganized ferment like an
enzyme, capable of self-reproduction was
unheard of. He found that continual heating
at 600C did not alter the infectivity but
that it became weaker at 65-75C and
was lost after several hours heating at 80C.
He was unable to retain infectivity after
clarification and precipitation with weak
alcohol. Thus, he concluded that the in-
fectious agent was subject to the living
conditions of organized ferments such as
bacteria and fungi. Fungi were ruled out
as the causal agent because they were too
large to go through the filter paper. He
concluded that the causal agent of the mo-
saic disease of tobacco was a bacterium
about which little was known concerning
its mode of life or infectious form. Peach
yellows disease, according to Smith, may
have been recognized as early as 1750, but
nothing was known about the nature of the
disease until 1891 when he showed that
it was contagious, had a long incubation
period, and was bud transmitted.
In 1890, Ivanowski (1892) noted two
diseases of tobacco in the Crimea. One was
a pox-disease, and the other a mosaic dis-
ease similar to that reported by Mayer. He
believed that the two diseases were inde-
pendent rather than different stages of one
disease. He verified Mayer's results on
transmission, thermostability, and the ab-
sence of fungi and other parasites. He did
not agree with Mayer, however, with re-
spect to his statement on filtration through
double filter paper. Ivanowski's preparation
was still infectious after filtering through
double filter paper and he knew that such
filtration would not retain bacteria. Fur-
thermore, he found that his preparation
was still infectious even after filtering
through a Chamberland filter-candle that
would retain bacteria. Ivanowski thought
his results were due either to a toxin se-
creted by the bacteria or to the penetration
of the bacteria through the pores of the
filter. The cause of these unknown diseases
was still thought to be corpuscular, but
they were now known to be caused by
agents transmissible mechanically and by
grafting. The first departure from the cor-
puscular theory for the etiology of the to-
bacco mosaic disease came with the work
of Beijerinck (1898), who concluded after
experiments on agar diffusion that the dis-
ease was not caused by microbes but by a
contagium vivum fluidum. He also con-
cluded from serial inoculation that the con-
tagium reproduced itself in the living plant.
It is also in Beijerinck's work that the word
virus is used to describe the contagium. He
found that the virus would infect and in-
vade young tissue more rapidly than ma-
ture tissue, moved in both the xylem and
the phloem, was graft transmissible and
able to infect plants by means of the roots.
He also reported that the virus was still
infectious after drying for two years in dis-
eased leaves, would survive the winter in
soil, and was inactivated after exposure to
formalin or boiling.
According to Fukushi the importance of
insects as vectors of viruses was first shown
by Hashimoto, who proved the relationship
of a leafhopper to the dwarf disease of rice
in 1894-95. It was thought at that time that
the disease was caused by a leafhopper,
and it was not until 1900 that it was found
that only one species of leafhopper, Nepho-
tettix apicalis var. cincticeps, was capable
of producing the disease. In 1906-8 it was
recognized that leafhoppers from certain
areas of Japan induced the disease whereas
leafhoppers from other areas did not and
that these noninfective leafhoppers became
able to induce the disease if they fed for
about five days on diseased plants. Thus it
became evident that the leafhopper N.
apicalis did not cause the disease but was
only transmitting its causal agent (1934).
It was thought by some that cutting to-
bacco plants back caused the mosaic dis-
ease, and Woods (1902) proposed that the
trouble could not be due to parasites but
must be attributed to a disturbance of a
normal physiological activity of the cells in
question. These disturbances, he proposed,
were due to the enzymes peroxidase and
Baur (1904), working with the variega-
tion of Abutilon that he termed infectious
chlorosis chlorosiss infectiosa), stated that
the infectious substance could not be a
living organism because in Europe varie-
gated and nonvariegated plants were grow-
ing side by side without any evidence of
transmission. Such a limited capacity for
movement, according to Baur, was incon-
sistent with parasitic organisms. He thought
that it was important to recognize that
there were infectious diseases for which
living organisms could not be considered
the cause. He stated: "For a further insight
into the etiology of these diseases the old
dogma of the unconditionally parasitic na-
ture of all infectious diseases seems to me
to be only an obstruction."
From 1904 to 1935 many discoveries were
made concerning the nature of virus dis-
eases and the nature of the disease-causing
entity. The published contributions during
this period were numerous. Some of the
outstanding discoveries that have affected
our interpretation of the word virus and
contributed to the characterization of vi-
ruses will be discussed. Allard in 1914
showed that of the various plants tested
only those from the family Solanaceae were
susceptible to tobacco mosaic virus. He also
reported that pokeweed mosaic and to-
bacco mosaic were distinct diseases. He
confirmed much of the earlier work and
showed that clear filtrates obtained by the
passage of sap from diseased plants through
Berkfeld filters were as infectious as the
unfiltered sap. In 1915 Allard showed that
the disease could be caused by very small
amounts of the virus. He found that the
virus could withstand dilutions of 1:10,000
and in some cases 1:1,000,000. He argued
that this small amount of material needed
to cause infection was inconsistent with the
enzymatic theory of Woods. He proposed
that there was something in the virus quite
extraneous to the protoplasmic constitution
of healthy plants and that once introduced
into healthy plants it rapidly increased and
was parasitic in nature. Working on the
properties of tobacco mosaic virus, Allard
(1916) partially purified the virus by pre-
cipitation with aluminum sulfate. The pre-
cipitate was still infectious whereas the
clear supernatant was not, even though it
had good peroxidase activity. He con-
cluded that there was no correlation be-
tween peroxidase or catalase activity and
the infectious agent. He believed that to-
bacco mosaic was caused by an ultramicro-
scopic parasite. Although the necessity of
wounds for infection was recognized earlier,
Allard (1917) showed that tobacco plants
were readily infected through the trichomes
by rubbing leaves or stem with juice from
diseased plants. He also found that washing
with soap and water was a practical and
efficient means of removing the virus from
Mild and severe forms of the sugar-beet
curly-top virus were recognized by Carsner
(1925a). He passed a severe form of the
virus through plants of Chenopodium mu-
rale, Rumex crispus, or Suseda moquini and
found that it then caused a mild form of
the curly-top disease in beets. Similar mild
forms were found to occur in beet leaf-
Chapter 1-M. K. Corbett
hoppers collected in their natural breeding
areas. Carsner proposed that the virus was
attenuated, but he was most likely sorting
out strains of the virus. Strains of viruses
were also noted by McKinney (1929) to
occur in isolates of a virus from Nicotiana
glauca collected in the Canary Islands.
Most plants had a light green or mild dark
green mosaic. One plant had a pure yellow
mosaic that produced yellow spots on sub-
sequent inoculation. Subinoculation from
these yellow spots produced a pure yellow
mosaic type. McKinney proposed that such
plants may be infected by mixtures of vi-
ruses or that the virus may become altered
in the plant, producing mutations.
After extensive study on the aster yellows
virus, Kunkel (1926) showed that both
nymphs and adult leafhoppers were able to
transmit the disease only after a 10-day
incubation period. He proposed that this
period of time was necessary for the causa-
tive agent to develop or multiply in the
tissues of the leafhopper. He found that
the virus was not transmitted through the
eggs of the insect carrier or through the
seeds of the aster, nor was it mechanically
Studies on the nature of tobacco mosaic
virus (TMV) led Mulvania (1926) to pro-
pose that the virus behaves as a typical
colloid with regards to isoelectric point and
mobility. Vinson (1927) showed that the
virus could be precipitated out of solution
by means of acetone, absolute alcohol, or
ammonium salts. The precipitate in all cases
was infectious whereas the supernatant was
The antigenic nature of plant viruses was
discovered when Dvorak (1927) produced
antisera in rabbits to sap from healthy po-
tatoes and to sap from potatoes infected
with a mosaic inducing virus. She was able
to show that both sera had antibodies in
common but that they exhibited a higher
titer for the homologous than the heter-
ologous reaction. In the preceding year
Mulvania (1926) had injected a TMV
preparation into the marginal ear veins of
rabbits and had attempted to reisolate the
virus from blood of the opposite ear. He
made no attempt to produce a specific
antiserum. More extensive studies were
conducted on the antigenic nature of vi-
ruses by Purdy (1928, 1929). She found
that antiserum contained a highly specific
antibody for virus sap and that the specific
antiserum would neutralize infectivity.
One of the most significant discoveries in
the field of plant virology was that local
lesions could be used for quantitative
studies with TMV (Holmes, 1929). Until
then, very little attention had been given
to the reaction of viruses at the site of
inoculation. Holmes found that several
species of Nicotiana reacted by producing
necrotic lesions at the site of infection with
TMV and that the numbers of such lesions
correlated with virus concentration in the
inoculum when an efficient method of inoc-
ulation, such as rubbing with a gauze pad,
was used. The reaction to TMV of plants
of N. glutinosa made this host-virus com-
bination a very desirable one for quantita-
tive work. Holmes pointed out that the
method could be compared with the
poured-plate method for counting bacteria.
Another interesting phenomenon noted
to occur in virus-infected plants was that of
recovery from severe necrotizing diseases.
Wingard (1928) was apparently the first
to notice that leaves in the final or recovery
stage of tobacco ringspot disease, although
containing active virus, would not show
necrotic spots as the result of reinoculation.
Wingard suggested that the plants were
immune. A type of interference phenomena
was also suggested from the work of Thung
(1931), who found that tobacco plants in-
oculated with mixtures of yellow and com-
mon strains of TMV developed symptoms
of both diseases and that the viruses could
be recovered separately. Thus, Thung con-
cluded that only one type of virus could
occupy any given cell. He also found that
a tobacco plant affected with the yellow
mosaic did not develop additional symp-
toms when it was inoculated with the
Little was known about the morphology
of virus particles, except that they were
very small in size, until Takahashi and
Rawlins (1933) ingeniously utilized the
technique of stream double refraction and
showed that the virus of tobacco mosaic
was composed of rod-shaped particles.
During the 30-year period 1904-35, many
virus diseases were described on the basis
of symptomatology and methods of trans-
mission. The virus as a disease-causing
agent during this period was shown to dif-
fer from bacteria in more ways than size
alone. It was shown to exist and to increase
only in living cells, to have host specificity,
in some cases to behave like a proteinaceous
colloid of rod-shaped particles, to mutate
to different strains, to be antigenic, to be
transmitted by insects in which it may mul-
tiply, and, in some instances, to induce in
some plants a type of acquired immunity.
In the next period, from 1935 to date, many
changes occurred in our thinking concern-
ing the nature of viruses. The first major
contribution came in 1935, when Stanley
obtained crystals of TMV. He found that
sap from infected tobacco plants could be
fractionated with ammonium sulfate until
a crystalline protein was obtained. He was
able to associate infectivity with the crys-
talline protein and proposed that TMV
could be regarded as an autocatalytic pro-
tein which required the presence of living
cells for multiplication. The next year
Bawden et al. (1936) showed that the
liquid crystalline preparations of TMV con-
sist of a protein and nucleic acid of the
ribose type and that the virus particles
were composed of subunits in a rod shape
which exhibited the phenomenon of ani-
sotropy of flow. Immediately following 1936
numerous papers characterizing this virus
by physical and chemical methods were
published. The early X-ray studies of Ber-
nal and Fankuchen (1941) demonstrated
that the virus particles are built up of sub-
units arranged in a regular way. The elec-
tron microscope, although developed ear-
lier, was used in 1939 by Kausche et al. to
demonstrate that virus particles associated
with tobacco mosaic are slender rods. Al-
though the quality of the electron micro-
graphs was poor as judged by today's stand-
ards, the morphology of other viruses was
studied by Stanley and Anderson (1941).
Some were found to be rods whereas others
were spherical. The use of the electron
microscope to study virus particle morphol-
ogy received a new stimulus in 1945 when
Williams and Wyckoff demonstrated the
use of shadow casting to improve contrast.
The visualization of viruses within their
host cells was soon accomplished when
techniques of embedding and sectioning
were improved (Porter et al., 1945). Al-
though multiplication of a plant virus in its
insect vector was suggested by earlier work-
ers, the first direct evidence was presented
by Maramorosch (1952) for multiplication
of aster-yellows virus in the leafhopper
Macrosteles divisus. The chemical composi-
tion of tobacco mosaic virus has been stud-
ied in detail, and the importance of the
nucleic acid portion of the virus particle to
infectivity was pointed out by Epstein
(1953), when he showed that the radiosen-
sitive portion of the virus was the nucleic
acid and not the protein. Treatment of the
virus with phenol by Gierer and Schramm
(1956) showed that the protein could be
removed and that the ribonucleic acid por-
tion of the virus particle was infectious.
Other procedures (Fraenkel-Conrat, 1956)
have been used for deproteinization studies
of viruses, and some of the rod-shaped vi-
ruses have been reconstituted. Thus, most
viruses are now considered to consist of an
infectious nucleic acid surrounded by a
large number of protein subunits all of
which may be identical. The use of nega-
tive staining techniques for electron micros-
copy (Hall, 1955; Brenner and Home,
1959) has greatly enhanced the use of the
electron microscope for studies on the
morphology and structure of viruses. These
studies, coupled with modem X-ray crystal-
lography and physiochemical data, have
given us much information on the nature
of viruses, the details of which will be
discussed in later chapters. The fortuitous
discovery of tobacco mosaic virus and its
unique characteristics which permit it to be
studied so easily by so many procedures
immediately becomes apparent, and from
this brief review of some of the outstand-
ing contributions to our knowledge on the
nature of viruses one realizes how much we
know and how much there is left to know
before the questions, what is a virus, is it
animate or inanimate, and how did it arise,
may be answered in full. The fallacy of at-
tempting to describe all viruses under one
definition when the information is not
Chapter 1-M. K. Corbett
complete is readily recognized from the
changes made by recent studies in virology.
Today we may define a virus as an infec-
tious nucleoprotein or an infectious nucleic
acid only to have our definition drastically
changed in the near future to a group of
bases or a single base or to a single com-
pound. Who knows what wonders await us?
1-2. NOMENCLATURE AND
Nomenclature and classification of plant
viruses have been in a state of chaos and
confusion ever since the first virus was
named. This confusion arose in part from
research workers giving different names
and descriptions to the same virus. Most of
these investigators received their training
in a biological discipline where the use
of Latinized names, keys, and methods of
classification were a very important part
of their formal education. The carry over of
this training to the field of plant virology
becomes very evident to the reader who,
after attempting to unscramble the pro-
posed systems of nomenclature and classi-
fication of plant viruses, is left completely
At first, plant virus diseases were named
for the most conspicuous symptom in the
host, and the cause of the disease received
the same name with the word virus at-
tached. That this system is merely based
upon symptoms is immediately evident, and
the pitfalls of such a system are recognized
when consideration is given to the variabil-
ity of symptoms. The investigators who
describe new virus diseases and propose
new systems of classification should be the
first to recognize this variability, and should
rely upon a more complete study of the
disease and the causal agent before pro-
posing new names and systems of classifica-
tion. The point is well put by Bawden
(1950): "Virus literature contains many de-
tailed descriptions of symptoms, usually
first accounts of 'new' viruses, in which dif-
ferences from clinical pictures previously
recorded are stressed as evidence of new-
ness. How valueless such comparisons can
be is evident from the number of times
potato viruses X and Y, cucumber mosaic
and tobacco mosaic viruses, have been
identified as new viruses and given new
names. Nevertheless, virus workers seem re-
luctant to appreciate the fact that variabil-
ity is the normal, and they continue to
describe with a wealth of picturesque de-
tail symptoms that may never again be
Regardless of the confusion that has been
created, work on the various systems of
classification has contributed to our under-
standing of virology and has stimulated
much research. It is easy for us today to
look at the problem of classification and
nomenclature and to think how worthless
all the controversy over what you should
call a particular virus and how you should
classify it, when it is easy just to distinguish
them by common names. This was not so
true in 1927 when Johnson proposed a sys-
tem for classifying viruses. In 1927, little
information concerning the causal agent of
mosaic was available, beyond the fact that
it was filterable, invisible, and infectious.
Investigators were just realizing that there
is considerable variability in the number
and type of hosts that are susceptible and
that there is variability in the symptoms
produced. When this was realized, the need
for a system of classification seemed impor-
tant so that research workers would have a
means of identifying the virus with which
they were working. Johnson (1927) found
that the diagnostic features of some symp-
toms, when properly interpreted from
comparative studies, had some value in
classification of the viruses of tobacco, but
that it was difficult to give a descriptive
name to all the viruses that occur in one
host. That he recognized the value of such
symptom variability is noted from the fol-
lowing: "Nowhere in the realm of plant
pathology are symptoms of less value in
description than in plant virus diseases, be-
cause of the remarkable influence of en-
vironmental factors, and the possible co-
existence of two or more viruses in a
single plant." He proposed that a system
be established of naming viruses based
upon the host in which each virus was first
discovered together with a number to indi-
cate the specific virus. Under this system,
TMV would become tobacco virus 1 and
the other viruses of tobacco would receive
a number in their order of discovery. In
1927, this system had its merits for the
number of known viruses affecting tobacco
was small, and it was possible for a research
worker to remember and associate certain
important features of a virus with a num-
ber. Today, such a system would be im-
possibly abstruse, for there is nothing
characteristic or meaningful about a num-
ber that will associate it with a particular
virus or disease. It also classes together
viruses that have nothing in common ex-
cept that they infect the same host. The
system also raises the difficulty of what
becomes of the number if a specific virus is
removed from the system after further re-
search shows that it does not belong to the
particular group in which it was classified.
Johnson (1929) further attempted to clas-
sify the virus diseases of potatoes that had
been reported from the United States and
Europe. He used several diagnostic fea-
tures, such as symptoms in the Bliss Tri-
umph potato variety, physical properties
(thermal-inactivation, longevity in vitro,
and dilution end point), influence of chem-
icals, methods of transmission, incubation
period, host range, variation in cytological
and histological detail, and filterability. He
concluded that such a study was more
meaningful than symptomatology alone and
that rugosee mosaic" was the same as
Smith (1931), working with a composite
potato virus disease of the mosaic group,
designated the letters X and Y for the two
components. This use of letters and com-
mon names has persisted and is perpetu-
ated even today for the naming of the virus
diseases of potatoes.
Use of the relationship of viruses to their
vectors as means of classification was first
proposed by Elze (1931). He proposed
three classes: (1) viruses not spread by
insects; (2) viruses spread by different in-
sects; and (3) viruses adapted to particular
insects. The third class was subdivided into
(a) those viruses which, in addition, are
mechanically transmitted, and (b) those
which are adapted for a short incubation
period and those with a long incubation
period. Storey (1931) proposed that the
species of insect vector is usually charac-
teristic of the virus and that the virus is
better characterized by its insect vector
than by the host or the symptoms induced
in the host. He cautioned that the speci-
ficity of the virus to its vector may not be
absolute and that strains of the virus and
races of the insect vector may determine
the possibility of transmission.
Another departure from classification of
viruses by symptomatology was proposed
by Chester (1935), who demonstrated the
usefulness of specific serological reactions
for determining relationships among viruses.
He showed that the viruses could be di-
vided into serological groups such as the
TMV group including the virus of tomato
aucuba mosaic, Johnson's yellow tobacco
mosaic 1, Holmes' symptomless tobacco
mosaic, and Jensen's brilliant yellow, white,
and slow-moving types. Although very
diverse in symptomatology, these viruses
were serologically related. Similarly, latent-
mosaic of potato virus, X virus, healthy-
potato virus, potato ringspot virus, and
British Queen streak virus all belonged
to the same group. The system of serology,
undoubtedly one of the most reliable means
of grouping viruses, is however not infal-
lible. Chester (1935) obtained reactions
suggesting that the vein-banding virus of
potato (potato virus Y) was serologically
related to cucumber mosaic virus and that
they might be strains of one virus. These
two viruses have since been shown to be
distinct entities (Ross, 1950), and Chester's
results may be explained on the basis that
the viruses he was working with were not
separate isolates of Y or cucumber mosaic
virus or that his virus isolates were con-
Smith (1937) proposed a variation of
Johnson's system of classification for about
144 viruses, excluding strains, in his book
A Textbook of Plant Virus Diseases. He
adopted Johnson's use of numbers but used
the Latin generic name of the host instead
of the common name. This has the same
disadvantages as Johnson's system, but did
make it possible to be consistent with the
naming of the host, especially in those cases
where no generally accepted popular name
of the host is available. Ainsworth (1939)
Chapter 1-M. K. Corbett
recognized the difficulty of obtaining inter-
national conformity to any detailed list that
included all strains. He proposed that a
convenient method of distinguishing strains
would be to add a number (in roman type)
to the italicized name of the virus with any
particular designation given by the author
who described the variant. Thus, TMV
would be designated Nicotiana tabacum
virus 1 and strains would be distinguished
as follows: Nicotiana tabacum virus 1,
masked strain, Holmes 1934. Ainsworth's
recommendation may have some merit with
regard to distinguishing strains, but it has
the same faults as the already proposed
On behalf of the Committee for Virus
Nomenclature appointed by the Council of
the American Phytopathological Society,
Bennett (1939) selected five characteristics
of viruses that he believed were most im-
portant. These were (1) type of symptoms
produced on different species and varie-
ties of susceptible plants; (2) morphologi-
cal and cytological disturbance produced;
(3) relation of insect vector to virus trans-
mission; (4) antigenic reactions in animals
and plants; and (5) the chemical and
physical properties of the viruses them-
selves. He thought, at the time, that the
host plant in all probability was the most
logical basis for generic distinction of vi-
ruses but that purification and crystalliza-
tion of the virus protein may lead to a more
logical and natural basis for grouping them.
Of the various possibilities, three were sug-
gested. (1) Viruses might be considered as
organisms and given binomial designations
following the practices already used for
recognized organisms. (2) They might be
considered chemical compounds. (3) Arbi-
trary systems having no reference to the
nature of the entities might be used as a
basis for names. He proposed that, due to
the difficulties of the systems in use, names
should replace numbers. Thus, tobacco vi-
rus 1 of Johnson might become tobacco vi-
rus altathermus or Nicotiana virus altather-
mus. If, on subsequent investigations,
viruses were shown to be organisms, then
they might be given proper generic binom-
ial names such as paracrystalis altathermus.
If, on the other hand, they proved to be
chemical compounds, the suffix "vir" might
be used to signify virus in the same way as
the termination "ase" designates enzymes.
Thus, tobacco virus 1 would become
In the same volume, same number, of
Phytopathology, Holmes (1939a) proposed
an extension of the binomial system of no-
menclature to include viruses. He pointed
out that such a system had the advantages
of grouping viruses on fundamental simi-
larities such as serological and immunologi-
cal tests, type of disease, and all other ac-
cumulated data. The system has the further
advantage of allowing names to be re-
moved or added without affecting the con-
tinuity of the system. He proposed a king-
dom Vira with two divisions, Phytophagi
for plant viruses and Zoophagi for animal
viruses. Each division consisted of classes,
families, and genera with Latinized bino-
mials for the individual viruses. Holmes
(1939b) later expanded the system in his
Handbook of Phytopathogenic Viruses to
include descriptions of 129 individual vi-
ruses. As with all new proposals, as soon as
the binomial system of nomenclature was
published it had its critics, its supporters,
and those who wanted to modify it. Soon
after Holmes proposed the Latinized bi-
nomial-trinomial system, Valleau (1940)
classified the viruses causing diseases of to-
bacco. He was dissatisfied with Holmes'
genus Marmor because it was too hetero-
geneous and ill-defined. Thus, he suggested
that the genus be used as a catchall. In
place of the Marmor genus, he substituted
a new genus, Musivum, based on Holmes'
Marmor tabaci var. vulgare as a type spe-
cies. He set up several new genera and re-
defined some of Holmes' original genera.
He did not consider any groupings above
the genus level.
Fawcett (1940) proposed a system of
nomenclature which, in his judgment, com-
bined the best of the systems proposed
by Johnson (1927), Smith (1937), and
Holmes (1939a). He described it briefly as
a simplification of Smith's system without
the confusion of Johnson's numbers, and
Holmes' generic difficulties. He proposed
using the binomials as suggested by Bennett
(1939), obtaining the genus from the host
plant in the way proposed by Johnson
(1927) and Smith (1937), and using spe-
cific names as does Holmes, instead of num-
bers. Fawcett's rule of obtaining generic
names was simply to apply the stem "vir" to
the Latin genitive of the name of the genus
of the host in which the virus was first re-
ported. Thus, the viruses of peach rosette,
sugar-beet curly-top, and potato yellow-
dwarf become Prunivir rosettae, Betavir
eutetticola and Solanivir vastans, respec-
tively. Fawcett also recognized that such a
system would create names that would be
difficult to pronounce, such as Chrysanthe-
mivir. Using his system, Fawcett proposed
names for citrus viruses. Here again is a
system of nomenclature and classification
based mainly on the host and symptoms.
Thornberry (1941) proposed that all in-
tracellular infectious agents that do not in-
crease in a cell-free medium be assigned to
one order, Biovirales, an adjunct to the
bacteria (class Schizomyceta phylum Thal-
lophyta of the Plant Kingdom). The plant
viruses were to be assigned to the genus
Phytovirus of the family Phytoviraceae, and
the classification of the genus Phytovirus
was to be based upon host range, methods
of transmission, and symptoms in standard
hosts. The species portion of the binomial
was to be based upon the first or more syl-
lables of the generic names of an important
host prefixed to a Latin word implying one
or more of the characteristic symptoms in a
standard host. Thus tobacco mosaic virus
would presumably become Phytovirus nico-
mosaicum variety vulgare. To master such
a system, one would have to be a Latin
scholar and etymologist in addition to being
a plant pathologist.
Bawden (1941) opposed the binomial
system and suggested that nomenclature of
viruses should be based on a system of clas-
sification so that the virus names would in-
dicate some characteristic property of the
virus or their interrelationships and that any
permanent virus classification should be on
the virus and not on the host plant. He sug-
gested that it might be possible to classify
those viruses about which serological, mor-
phological, and chemical information was
known and that all others be put in a group
with a name to indicate that they were
there because of ignorance and not because
of our knowledge. He proposed that a list
of approved common names would be use-
ful for the virus worker or plant pathologist
to distinguish or make reference to the
virus in question. Similar to all other pro-
posals, Bawden's suggestions precipitated
three more papers on the subject in volume
7 of Chronica Botanica (Fawcett, 1942;
Johnson, 1942; Valleau, 1942).
Ainsworth (1943) suggested that the
work of a plant pathologist would be
greatly facilitated by a list of standardized
English names of virus diseases from which
the common name of the virus could be de-
rived, similar to that compiled for the fungi
in 1939 (common names of British plant
diseases) and that a distinction should be
made between obligate synonyms (those
based on the same type) and facultative
synonyms (based on different types).
McKinney (1944) proposed a kingdom
Phyta with a division Viriphyta for the
plant viruses. He described type species for
12 genera, again based on a binomial sys-
tem. Holmes (1948) expanded his earlier
classification to include all viruses, and he
grouped 67 species in the genus Marmor,
which contain the viruses of the mosaic
group. Some of these he admitted were not
described in sufficient detail and needed
additional investigation. But he argued that
they could be reclassified, as relationships
with other groups were determined. To
date, this has been his most ambitious treat-
ment of the subject, and recently he has
stated that he would most likely not revise
it (personal communication).
The whole subject of virus classification
and nomenclature was discussed in a series
of papers presented at a conference on "Vi-
rus and Rickettsial Classification and No-
menclature," held in New York City,
January 1952, by the Section of Biology of
the New York Academy of Science, and
edited by R. W. Miner (Vol. 56, 381-622,
1953). The authors of the previously dis-
cussed papers again presented their views,
objections, and opinions on the subject
without reaching any conclusive accepted
procedure for plant virus nomenclature and
Another binomial, Latinized system of
Chapter 1-M. K. Corbett
virus nomenclature, in which the generic
name is composed of symbols representing
three independent characteristics of the vi-
rus, was proposed by Hansen (1960). The
symbols refer to direct transmission, to vec-
tor transmission, and to particle type. He
proposed that a combination of these three
characteristics would define the virus in
question; thus potato virus X would be
Minflexus solani (M = mechanical trans-
mission; in = virus without specific arthro-
pod vector; flexus = flexible threadlike
particle). In this case the specific epithet
(solani) is derived from the main host.
Recently, work on classifying viruses in
relation to their morphological and serolog-
ical relationships is being attempted on the
viruses with elongated particles (Brandes
and Wetter, 1959). Thus far, they have
classified 46 plant viruses on the basis of
particle morphology, normal length of the
particle, and serological reactions. This is
really the first attempt to classify viruses on
natural relationships rather than on host re-
actions, and it should meet with approval of
the critics of earlier classification. Brandes
and Wetter have continued to use common
names derived from the host reactions for
the viruses. This system allows for expan-
sion to include new viruses as they are
shown to have virus particles of similar
morphology and serological reactions and
to remove viruses if they should be more
closely related to another group. It also al-
lows for the use of common names (Review
of Applied Mycology, 1957) by which vi-
ruses are so frequently named and called,
especially by the plant pathologists.
An extension of this natural classification
was proposed recently by Lwoff et al. in
1962. They suggested that the system in-
clude only those entities that exhibit in
their life cycle an infectious particle con-
taining only one type of nucleic acid. The
system is based mainly on morphology,
structure, and symmetry of the virus parti-
cle as determined by electron microscopy,
negative staining, and X-ray diffraction. In
their system the mature virus is the virion,
which is composed of a capsid built of
structure units. The structure units can as-
sociate into symmetric units termed the
capsomeres, and the capsid (protein) plus
nucleic acid is the nucleocapsid. Thus, vi-
rions were divided into 2 groups on the
basis of their nucleic acid: group D, deoxy-
ribonucleic acid (DNA), and group R, ri-
bonucleic acid (RNA). These groups were
subdivided on the basis of symmetry, heli-
cal (H) and cubic (C). All virions belong
to one of 2 categories: some having a naked
capsid (N), some an enveloped capsid (E).
The groups were further subdivided on the
number of capsomeres for the cubic capsids
and on the diameter of the helical capsids.
Thus, TMV would be an RHN, 170-200A
virus, and turnip yellow mosaic virus would
belong to the REO group designated RCN,
32c. Such a system, although based on nat-
ural characteristics of the virus, would re-
quire that every laboratory have X-ray
equipment and a crystallographer to clas-
sify a virus. Eventually such a system may
be useful, but at present I believe it is long
before its time in the field of plant virology.
1-3. ECONOMIC IMPORTANCE
Since 1898, more than 400 plant viruses
have been described and named. The total
number of diseases caused by these viruses
has not been calculated. Some viruses, like
cucumber mosaic virus, affect many kinds
of crop plants. Others, like southern bean
mosaic virus, affect only a few kinds. Al-
most all known forms of plant life suffer
from the effects of plant viruses. In the
plant world only the Gymnosperms and
Pteridophyta still appear to have escaped
their effects, but until recently the fungi
and algae were also included in this group,
and now even the fungi and algae have
been reported to have virus diseases
(Gandy and Hollings, 1962; Safferman and
Morris, 1963). Why virus diseases have not
been discovered in gymnosperms and ferns
is unknown. It is probable that these groups
are affected by viruses which will be
revealed as they become economically
important and extensively studied.
Plants of economic importance all suffer
from losses due to viruses. The extent of the
losses will vary greatly, depending upon
the value of the crop and the type of dam-
age. Losses may be either quantitative or
qualitative. Quantitative losses are those
which are directly associated with yield re-
duction, such as fewer or smaller potato
tubers per plant, or fruit per tree, whereas
qualitative losses are usually involved in
lower market values, which result when
such symptoms as internal necrosis of po-
tatoes, color breaks in ornamentals, or size
reduction in flowers are present. The diffi-
culty in obtaining a monetary evaluation of
any crop loss due to virus diseases is imme-
diately recognized when consideration is
given to the complexity of the market situa-
tion. In some areas, a particular grower or
group of growers may lose an entire crop to
one virus disease, while at the same time
another grower or group of growers in a
different location will receive an increased
evaluation on their harvest because of the
law of supply and demand. Thus, all crop
losses are only estimates on an individual
basis, and extrapolation of such losses to a
world market is meaningless.
Both annual and perennial plants are af-
fected. Sometimes, the annual crops are
more dramatically affected as judged from
symptoms and immediate losses. Crops
such as potatoes, citrus, bananas, strawber-
ries, raspberries, fruit trees, and sugar cane
which are vegetatively propagated, will
collect various viruses and eventually suffer
severe losses. Not only does the grower lose
the immediate yield in the case of peren-
nial crops but the cost of replacement is
much greater than that of annual crops.
Numerous reports of losses due to virus dis-
eases are available in the literature, but few
contain actual market values. Some specific
examples will be given in both annual and
perennial crops to illustrate the extent to
which virus diseases may occur and their
drastic effects upon yield. Roque and
Adsuar (1941) reported that a mosaic dis-
ease of chili peppers (Capsicum frutes-
cens) occurred in epiphytotic proportions
at the Agricultural Experiment Station at
Isabela and later spread to other parts of
Puerto Rico. The estimated crop losses due
to the disease was 50-80%. The "big-bud"
virus of tomatoes has been reported
to cause losses in all parts of Australia
(Samuel et al., 1933). Usually, only a small
percentage of plants is affected, scattered
at random over the field, but infections of
50-100% have been reported in New South
Virus diseases of crucifers such as
swedes, rape, turnip, and cauliflower have
been reported from various parts of the
world. It was recognized in New Zealand
in 1932 (Chamberlain, 1936) that losses in
rape equivalent to 25% occurred during
1934-35. Ling and Yang (1940) reported
that a mosaic of rape destroyed more than
30% of the crop in China and that the re-
duction in seed yields ranged from 37 to
86% of the samples examined. Commercial
plantings of cauliflower in the coastal areas
of central California are severely affected
by a virus disease that causes 20-30% loss
in some fields (Tompkins, 1934). A similar
virus disease was reported to be wide-
spread in Devon and Cornwall and to have
affected as many as 75% of the plants in a
field; entire crops were rendered unmarket-
able (Caldwell and Prentice, 1942). Strains
of turnip mosaic virus have been reported
to damage commercial plantings of horse-
radish in Wisconsin, Illinois, Missouri, and
Washington. In some areas 100% of the
crop was infected (Pound, 1948).
In certain areas of Western Australia,
farmers were not aware of the insidious na-
ture of bean mosaic virus, which caused
widespread damage in bean crops ranging
from 20 to 50% (Cass Smith, 1945). This
yield reduction is seldom noticed and rec-
ords are often not available. Van Hoof
(1956) determined from small plots of rhu-
barb in the Heemstkerk and Zwolle district
of the Netherlands that the diseased plants
were not appreciably less productive than
healthy plants in the first year but that they
were 23.6% less productive in the second
A yellow mosaic virus has been reported
to cause severe damage to grasses in the
New Delhi area, where 50-80% of the
plants were affected (Vasudeva et al.,
1948). In 1935 a survey of the spotted wilt
disease of lettuce in the Salinas Valley of
California indicated that the virus has be-
come progressively more destructive. Fields
which formerly were unaffected showed a
slight loss in the early summer but, as the
season advanced, the loss increased until at
the beginning of autumn it amounted to
Chapter 1-M. K. Corbett
more than 3 of the crop in some plantings
(Harris, 1939). In the lettuce-growing
areas of southeastern North Carolina, the
virus diseases big vein and mosaic contrib-
uted significantly to the low yields and
poor quality of the spring crop of lettuce,
and seed from mosaic-free lettuce yielded
14,448 lbs per acre as compared to 12,432
lbs from fields planted with regular seed
(Aycock and Winstead, 1955). Chupp and
Paddock (1949) reported an epidemic of
lettuce big vein virus in Long Island, New
York, in 1949. In some counties the losses
ranged from 10 to nearly 100% of the crop.
Big vein virus in 10 commercial lettuce
fields in the Salinas Valley was found by
Zink and Grogan in 1954 to affect the yield
of marketable lettuce without necessarily
causing any economic loss. Profits obtained
from fields almost 100% infected occasion-
ally surpassed those obtained from pre-
dominantly healthy fields. The profits were
apparently directly related to the consumer
demand at the time of harvest: if demand
was great, poor quality lettuce was ac-
cepted. Cucumber mosaic virus has been
shown to cause severe losses to cucumbers
both in the field and in the greenhouse.
Doolittle (1924) reported that losses from
this virus had been on the increase for five
years in the greenhouse industry and that
losses in a single locality in 1922 were esti-
mated at $75,000. Losses were so continu-
ous that in some areas cucumbers were be-
ing replaced by other crops. Tobacco
ringspot virus has been reported by Pound
(1949) to cause losses of 10% to water-
melons in Wisconsin. The same virus has
been reported to have caused up to 50%
losses to watermelons in Texas (Rosberg,
Cereal crops are affected by at least 20
viruses causing various estimated losses.
Sill et al. (1955) estimated that the soil-
borne wheat mosaic virus caused a loss of
$3,000,000 to farmers in Kansas in 1953-54.
Losses caused by wheat streak mosaic virus
were estimated at $14,000,000. The disease
was very widespread in most parts of the
state. Barley false stripe virus present in
most states of the upper Mississippi Valley
has been reported by Timian and Sisler
(1955) to occur in 93% of the 214 barley
fields examined in North Dakota. Infection
ranged from a trace to 15%. The yield re-
duction depended upon the variety but
ranged from 17 to 24%. A mean loss in
yield of 10.8% was recorded for winter
wheat varieties grown in soil infested with
soil-borne wheat mosaic virus (Bever and
Oswald and Houston (1951) reported a
new outbreak of barley yellow dwarf virus
in barley in California in 1951 and esti-
mated that it caused a 10% loss in the bar-
ley crop (Oswald and Houston, 1953). The
virus is widespread around the world
and infects oats, barley, wheat, many for-
age grasses, and numerous wild grasses.
Rochow (1961) concluded that the virus is
as destructive as any cereal virus yet dis-
covered. Mosaic of winter wheat has been
reported by Zazhurilo and Sitnikova (1939)
to occur each year in the Voronczh Prov-
ince in Russia where, under favorable con-
ditions, up to 15-20% of the plants are af-
fected; infected plants yield less than
Lupines have been used in various parts
of the world as a forage, cover, or green-
manure crop, and during the past 15 years
they have been rather extensively grown
in southeastern United States. In 1950
Weimer reported that two virus diseases
occurred in epiphytotic proportions in the
state of Georgia. Corbett (1955) reported
that the United States Department of Agri-
culture, Bureau of Agricultural Economics,
showed that the acreage of lupines grown
for seed production in Florida decreased
between 1950 and 1954. In 1950, 130,000
acres were planted, of which 16,000 acres
were harvested for seed with a production
value of $510,000. In 1954, 120,000 acres
were planted, of which only 7000 were
harvested with a production value of
$172,000. The reduction in crop value was
attributed mainly to infection by bean
yellow mosaic virus.
The motley virus disease of carrots
has been reported by Stubbs (1948) to be
widespread in Australia and to cause se-
vere losses to all carrot varieties. He re-
ported the same disease from California
(Stubbs, 1956) but noted that losses were
correlated with vector population. Aster
yellows virus, in addition to causing losses
in crops such as lettuce, onions, and spin-
ach, has been reported by R. D. Watson
(1945) to have affected 80-90% of the
carrots in the Lower Rio Grande Valley
Sugar beets are grown throughout the
world. In the United States they are
grown in 22 states and contribute approxi-
mately 20% of the national sugar produc-
tion. The by-products of the sugar-beet
industry, consisting of tops, molasses, and
pulp, are used for feeding livestock. Un-
fortunately, sugar-beet production is seri-
ously affected by virus diseases. Carsner
(1925a) reported that only about 25% of
the normal sugar-beet crop was harvested
from large acreages in 1924 in the Yakima
Valley of the state of Washington. The
rapid expansion of the sugar-beet industry
was abruptly checked in Idaho in 1919 after
the first very severe outbreak of curly-top
virus (Murphy, 1946). Bennett et al.
(1954) inoculated sugar beet with yellows
virus by aphids and found that the average
reduction due to virus infection was 39.9%
in root weight and 35.8% in sucrose con-
tent. The potential damage of the yellows
disease to sugar production was pointed out
by M. A. Watson (1942) as being under-
estimated in Great Britain. She reported
that the yields of roots and sugar were con-
siderably reduced. Early infection of late-
sown beets caused a loss of 67% of the root
and 71% of the sugar yield. Watson et al.
(1946) reported that the average yield of
sugar beets for Great Britain during the
years of severe outbreaks of beet yellows
virus was 1.6 tons per acre less than during
the other years. It appeared that the yield
reduction of sugar was proportional to the
percentage of diseased plants, which was
estimated to be 5%. In Great Britain, Hull
(1954) calculated a loss of 0.39 tons per
acre of sugar-beet roots due to infection by
sugar-beet yellows. Total losses from sugar-
beet yellows in Great Britain in 1957 (Hull,
1958) were about 1,000,000 tons. In Den-
mark, where the disease has been reported
to occur sporadically, losses of up to 3% of
the sugar content have occurred (Gram,
1942). In Western Germany (Heiling,
1953), where severe outbreaks of beet yel-
lows virus occur, the root yield is approxi-
mately 50-60% of the normal, and sugar
yield is 60-70% of normal. In addition, the
leaves wilted and decayed more rapidly
than healthy ones, and virus-infected plants
contained 30-50% less protein. Losses of
from 29 to 35% of the sugar were reported
for late infection of sugar beets in Sweden
(Bjorling, 1949). In addition, the quality of
the root is poor and the weight of seed
per plant, percentage germination, and
number of seed balls per ounce are
reduced (Brewbaker, 1942).
Losses are very great in crops such as
sugar cane and potatoes that are planted
from vegetative parts when healthy propa-
gative material is not used. In addition to
causing losses in the individual plant, the
virus in the infected "seed" provides a
source of inoculum for within-field spread.
Sugar cane is affected by at least five major
virus diseases; mosaic, streak, sereh, Fiji,
and ratoon stunting. Field experiments in
Puerto Rico (Landrau and Adsuar, 1953)
showed that cane grown from noninfected
seed pieces produced significantly higher
yields (42 tons per acre) than infected
stock (34 tons per acre). Ratoon stunting
has been reported to cause losses of from 11
to 37% in the Q-28 variety in Queensland
(Hughes, 1955). In India, Chona (1944)
reported that losses due to the sugar-cane
mosaic in the Co. 213 variety, even though
100% infected, was only 10-12% reduction
in yield of sugar; there was no loss in ex-
traction or quality of juice. He concluded
that mosaic virus does little damage to the
Co. canes in India, but that even losses of
10% would amount to 3,300,000 rupees
The effects of viruses upon the yields of
potatoes were probably noted in the late
eighteenth century in Europe. Since then,
about 18 viruses have been reported as
causing diseases of potatoes. Each has its
various effects upon yield, and without cer-
tification it would most likely be impossible
to grow potatoes profitably. Scott (1941)
reported that the yield of potatoes may be
reduced 14-25% by a mild mottle caused
by mild strains of potato virus X or A. Mild
mosaic caused by virus X reduced the yield
by 30-40%, whereas borderline severe
Chapter 1-M. K. Corbett
mosaic, generally due to a combination of
viruses A and X, and severe strains of virus
X reduced the yield by 50-60%. Various
combinations of these viruses caused yield
reductions of 65-85%, and the leaf-roll vi-
rus reduced the yield by 50-95%. He also
found that, in addition to reducing the
yield, virus infections tended to induce
early ripening of the potatoes. Bonde et al.
(1943) reported results similar to Scott for
the rate of virus spread and effects upon
the yield of potatoes in Maine and in Long
Island, New York. They found that the
spread of potato viruses in Maine during
a 19-year period varied from season to sea-
son. Mild mosaic varied from 4 to 97%,
leaf-roll from 2 to 100%, and spindle tuber
from 1 to 61% of plants infected. The
spread was most extensive in seasons most
favorable for the aphid vectors common in
Maine. They reported that two potato vari-
eties, Green Mountain and Bliss Triumph,
when completely infected with leaf-roll,
spindle tuber, or different types of mosaic
yielded significantly less than when disease
free. The reduction ranged between 18 and
56% in Maine and between 26 and 64% in
Long Island. Schultz and Bonde (1944)
found that potato virus X was harbored by
potatoes more generally than any other vi-
rus and that it was responsible for losses of
9-22% in yield, representing annual losses
of millions of bushels. Bald and Norris
(1941) assumed that more than 90% of po-
tatoes grown in Australia were infected
with potato virus X. They estimated that
losses from this virus were as heavy as
those from all other viruses combined and
probably amounted to a loss of $1,750,000
Peach trees in the eastern temperate re-
gion of North America have been suffering
from the effects of peach yellows virus for
about 150 years. In 1891, Erwin F. Smith
wrote, "In July, 1891, I saw hundreds of
bushels of this worthless fruit in upper
Maryland and Delaware, and the entire
loss thereby in 1891 certainly exceeded half
a million dollars." Peaches are susceptible
also to phony peach, little peach, X or yel-
low-red disease, and peach mosaic, all of
which cause losses. Stout (1989) reported
that during 1938 approximately 18,000 new
infections of peach mosaic were found in
southern California, and this figure amounts
to only one-half that of 1937. Approxi-
mately 81,000 diseased trees were found of
which 63% were removed; 200,000 trees
Similarly, citrus is greatly affected by
virus diseases, but actual losses to the citrus
industry are difficult to determine. Psorosis
has been reported in 8% (15,100) of the
mature trees in 220 orange orchards in
southern and central California (Moore et
al., 1955). The reduction in yield varied
with the stage of the disease. The yield in
stage one was not significantly different
from that in the controls, whereas trees in
stage three yielded 28% less than those
in stage one. Tristeza was first observed in
Argentina in 1930-31. It probably occurs
in all the major citrus-producing areas of
Brazil, Argentina, and Uruguay. Bennett
and Costa (1949) reported that in about 12
years the virus spread to all citrus-produc-
ing areas of the state of Saio Paulo, infecting
and destroying upwards of 6,000,000 trees,
or about 75% of the orange trees in the
The Manila hemp plant, widely used for
the production of rope fibers, was reported
by Ocfemia (1949) to be infected by a
mosaic virus that caused losses ranging
from 90% in some plantings to almost 100%
in others in the eastern part of Mindanao
of the Philippine Islands. One of the most
devastating viruses on record is that of ca-
cao swollen shoot. Posnette (1945) re-
ported that in the eastern province of the
Gold Coast more and more cacao farms
were being completely destroyed by the
disease. On one farm over a five-year pe-
riod, the virus killed 74% of the 30- to 40-
year-old trees and 43% of the 20- to 30-year-
old trees. The total production decreased
from 30 tons per annum in 1926-29 to 20
tons in 1936-39, and to only 6 tons in
1943-44. In 1947, Posnette reported that
over 1,000,000 cacao trees had been de-
stroyed in the eastern province, and pro-
duction was reduced from 116,000 tons in
1936 to 64,000 tons of cacao in 1945. One
of the more dramatic diseases of recent
times that is causing extensive losses is that
of cadang-cadang (yellow mottle decline)
of coconut palms in the Philippines. Al-
though the disease is still of unknown etiol-
ogy, it is thought to be caused by a virus.
De Leon (1952) reported that over
1,500,000 trees were dying from the disease
in the Bicol area, with yearly losses
amounting to $1,800,000. Price (1958) re-
ported that approximately 7,927,000 dis-
eased trees, or about 47% of all trees,
were destroyed or diseased in the Bicol
region, representing a direct loss to the
farmers of 33,293,000 pesos in 1956.
From these few cases it becomes evident
that viruses extract an annual loss from
most commercial crops. With an ever-in-
creasing world population, this loss of food
is very important to peace and the well-
being of mankind. With increased knowl-
edge of the nature of viruses, it is antici-
pated that these losses will be reduced if
This work was supported in part by
Research Grant AI-03148 from the Na-
tional Institute of Allergy and Infectious
Diseases, National Institutes of Health,
U.S. Public Health Service.
1-4. LITERATURE CITED
Ainsworth, G. C. (1939). Chronica Botan. 5, 193.
Ainsworth, G. C. (1943). Ann. Appl. Biol. 30, 187.
Allard, H. A. (1914) U.S. Dept. Agr. Bull. 40.
Allard, H. A. (1915). J. Agr. Res. U.S. 3, 295.
Allard, H. A. (1916). J. Agr. Res. U.S. 6, 649.
Allard, H. A. (1917). J. Agr. Res. U.S. 10, 615.
Aycock, R., and Winstead, N. N. (1955). Res.
Farming N. Carolina 14, 7.
Bald, J. G., and Norris, D. 0. (1941). J. Council
Sci. Ind. Research Aust. 14, 187.
Baur, E. (1904). Ber. Deut. Botan. Ges. 22, 453.
(Phytopathological Classic 7, 55. 1942).
Bawden, F. C. (1941). Chronica Botan. 6, 385.
Bawden, F. C. (1950). Plant Viruses and Virus
Diseases. Chronica Bot. Co., Waltham, Mass.
Bawden, F. C., Pirie, N. W., Bernal, J. D., and
Fankuchen, I. (1936). Nature 138, 1051.
Beijerinck, M. W. (1898). Verhandel. Koninkl.
Ned. Akad. Wetenschap. (Amsterdam) 65, 3.
(Phytopathological Classic 7, 33).
Bennett, C. W. (1939). Phytopathology 29, 422.
Bennett, C. W., and Costa, A. S. (1949). J. Agr.
Res. U.S. 78, 207.
Bennett, C. W., Price, C., and Gillespie, G. E.
(1954). Proc. Am. Soc. Sugar Beet Technolo-
gists 8, 236.
Bernal, J. D., and Fankuchen, I. (1941). J. Gen.
Physiol. 25, 111.
Bever, W. M., and Pendleton, J. W. (1954). Plant
Disease Reptr. 38, 366.
Bjorling, K. (1949). Socker Handl. 5, 119.
Bonde, R., Schultz, E. S., and Raleigh, W. P.
(1943). Maine Agr. Exp. Sta. Bull. 421, 128.
Brandes, J., and Wetter, C. (1959). Virology 8,
Brenner, S., and Home, R. W. (1959). Biochim.
et Biophys. Acta 34, 103.
Brewbaker, H. E. (1942). Proc. Am. Soc. Sugar
Beet Technologists 3, 381.
Caldwell, J., and Prentice, I. W. (1942). Ann.
Appl. Biol. 29, 366.
Cane, Henry. (1720). Roy. Soc. [London] Phil.
Trans. B. 31 (366), 102.
Carsner, E. (1925a). Phytopathology 15, 745.
Carsner, E. (1925b). U.S. Dept. Agr. Official Rec.
Cass Smith, W. P. (1945). J. Dep. Agr. W. Aus-
tralia 22, 20.
Chamberlain, E. E. (1936). New Zealand J. Agr.
Chester, K. S. (1935). Phytopathology 25, 686.
Chona, B. L. (1944). Indian Farming 5, 178.
Chupp, C. H., and Paddock, W. C. (1949). Plant
Disease Reptr. 33, 280.
Corbett, M. K. (1955). Soil Sci. Soc. Florida Proc.
De Leon, D. (1952). Foreign Agr. 16, 103.
Doolittle, S. P. (1924). U.S. Dept. Agr. Circ. 321.
Dvorak, M. (1927). J. Infectious Diseases 41, 215.
Elze, D. L. (1931). Phytopathology 21, 675.
Epstein, H. T. (1953). Nature 171, 394.
Fawcett, H. S. (1940). Science 92, 559.
Fawcett, H. S. (1942). Chronica Botan. 7, 7.
Fraenkel-Conrat, H. (1956). J. Am. Chem. Soc.
Fukushi, T. (1934). J. Fac. Agr. Hokkaido Imp.
Univ. 37, 41.
Gandy, D. G., and Hollings, M. (1962). Rep.
Glasshouse Crops. Res. Inst. 1961, 103.
Gierer, A., and Schramm, A. (1956). Nature 177,
Gram, E. (1942). Tidsskr. Planteavl. 47, 338.
Hall, C. E. (1955). J. Biophys. Biochem. Cytol. 1, 1.
Hansen, H. P. (1960). Am. Potato J. 37, 187.
Harris, M. R. (1939). Calif. Dept. Agr. Bull. 28,
Heiling, A. (1953). Zucker. 6, 27.
Holmes, F. 0. (1929). Botan. Gaz. 87, 39.
Chapter 1-M. K. Corbett
Holmes, F. 0. (1939a). Phytopathology 29, 431.
Holmes, F. 0. (1939b). Handbook of Phytopatho-
genic Viruses. Burgess, Minneapolis.
Holmes, F. 0. (1948). The Filterable Viruses, in
Bergey's Manual of Determinative Bacteriology.
Suppl. 2, 1127. Williams and Wilkins, Balti-
Hughes, C. G. (1955). Cane Growers Quart. Bull.
Hull, R. (1954). Plant Pathol. 3, 130.
Hull, R. (1958). Agriculture (London) 65, 62.
Ivanowski, D. (1892). St. Petersb. Acad. Imp. Sci.
Bull. 35, 67. (Phytopathological Classic 7, 27).
Johnson, J. (1927). Wisconsin Univ. Agr. Expt.
Sta. Research Bull. 76.
Johnson, J. (1929). Wisconsin Univ. Agr. Expt.
Sta. Research Bull. 87.
Johnson, J. (1942). Chronica Botan. 7, 65.
Kausche, G. A., Pfankuch, E., and Ruska, H.
(1939). Naturwissenschaften 27, 292.
Kunkel, L. 0. (1926). Am. J. Botany 13, 646.
Landrau, P., and Adsuar, J. (1953). J. Agr. Univ.
Puerto Rico 37, 19.
Ling, L., and Yang, J. Y. (1940). Phytopathology
Luria, S. E. (1959). General Virology. John Wiley
and Sons, Inc., N.Y.
Lwoff, A., Home, R., and Tournier, P. (1962).
Cold Spring Harbor Symp. Quant. Biol. 27, 51.
McKay, M. B., and Warner, M. F. (1933). Nat.
Hort. Mag. 12, 179.
McKinney, H. H. (1929). J. Agr. Res. U.S. 39,
McKinney, H. H. (1944). J. Wash. Acad. Sci. 34,
Maramorosch, K. (1952). Phytopathology 42, 59.
Mayer, A. (1886). Landwirtsch. Vers. Sta. 32, 451.
(Phytopathological Classic 7, 11).
Moore, P. W., Nauer, E., and Yendol, W. (1955).
Calif. Citrograph 40, 82.
Mulvania, M. (1926). Phytopathology 16, 853.
Murphy, A. M. (1946). Proc. Am. Soc. Sugar Beet
Technologists 4, 408.
Ocfemia, G. 0. (1949). Philippine Agriculturist 33,
Oswald, J. W., and Houston, B. R. (1951). Plant
Disease Reptr. 35, 471.
Oswald, J. W., and Houston, B. R. (1953). Phyto-
pathology 43, 128.
Phillips, Edward (1720). New World of Words or
Universal English Dictionary. 7th ed. Kings Arm
in St. Paul's Church Yard, London.
Porter, K. R., Claude, A., and Fullam, E. F.
(1945). J. Exptl. Med. 81, 233.
Posnette, A. F. (1945). Cacao Res. Conf. 1945,
Posnette, A. F. (1947). Ann. Appl. Biol. 34, 388.
Pound, G. S. (1948). J. Agr. Res. U.S. 77,97.
Pound, G. S. (1949). J. Agr. Res. U.S. 76, 647.
Price, W. C. (1958). Food and Agr. Organization,
U.N. Report 850.
Purdy, H. A. (1928). Proc. Soc. Expt. Biol. Med.
Purdy, H. A. (1929). J. Exp. Med. 49, 919.
Rochow, W. F. (1961). Advances in Agronomy 13,
Roque, A., and Adsuar, J. (1941). J. Agr. Univ.
Puerto Rico 25,40.
Rosberg, D. W. (1953). Plant Disease Reptr. 37,
Ross, A. F. (1950). Phytopathology 40, 445.
Safferman, R. S., and Morris, Mary-Ellen (1963).
Science 140, 679.
Samuel, G., Bald, J. G., and Eardley, C. M.
(1933). Phytopathology 23, 641.
Schultz, E. S., and Bonde, R. (1944). Am. Potato
J. 21, 278.
Scott, R. J. (1941). Scot. J. Agr. 23, 258.
Sill, W. H., Fellows, H., and King, C. L. (1955).
Plant Disease Reptr. 39, 29.
Smith, E. F. (1891). U.S. Dept. Agr. Div. Veg.
Path. Bull. 1.
Smith, K. M. (1931). Nature 127, 852.
Smith, K. M. (1937). A Textbook of Plant Virus
Diseases. P. Blakiston's Son and Co., Inc., Phila-
Stanley, W. M. (1935). Science 81,644.
Stanley, W. M., and Anderson, T. F. (1941). J.
Biol. Chem. 139, 325.
Storey, H. H. (1931). 2nd Intern. Congr. Path.
Paris II Compt. rend. Comm. 471.
Stout, G. L. (1939). Calif. Dept. Agr. Bull. 28,
Stubbs, L. L. (1948). Australian J. Sci. 1, 303.
Stubbs, L. L. (1956). Plant Disease Reptr. 40, 763.
Takahashi, W. N., and Rawlins, T. E. (1933). Sci-
ence 77, 26.
Thornberry, H. H. (1941). Phytopathology 31, 23.
Thung, T. H. (1931). Handel. Ned. Ind. Natuur.
Cong. 6'e, 450.
Timian, R. G., and Sisler, W. W. (1955). Plant
Disease Reptr. 39, 550.
Tompkins, C. M. (1934). Phytopathology 24, 1136.
Valleau, W. D. (1940). Phytopathology 30, 820.
Valleau, W. D. (1942). Chronica Botan. 7,152.
Van Hoof, H. A. (1956). Mededeel. Proefsta.
Groenteteelt Ned. 3, 761.
Vasudeva, R. S., Raychaudhuri, S. P., and Patha-
nian, P. S. (1948). Current Sci. (India) 17, 244.
Vibert, M. (1863). J. Soc. Imp. Central. Hort. 9,
Vinson, C. G. (1927). Science 66, 357.
Watson, M. A. (1942). Ann. Appl. Biol. 29, 358.
Watson, M. A., Watson, D. J., and Hull, R.
(1946). J. Agr. Sci. 36, 151.
Watson, R. D. (1945). Plant Disease Reptr. 29,
Weimer, J. L. (1950). Plant Disease Reptr. 34,
Williams, R. C., and Wyckoff, R. W. G. (1945).
Proc. Soc. Exp. Biol. Med. 58, 265.
Wingard, S. A. (1928). J. Agr. Res. U.S. 37, 127.
Woods, A. F. (1902). U.S. Dept. Agr. Bur. Plant
Ind. Bull. 18.
Zazhurilo, V. K., and Sitnikova, G. M. (1939).
C. R. Acad. Sci. U.S.S.R., N. S. 25, 798.
Zink, F. W. and Grogan, R. G. (1954). Plant Dis-
ease Reptr. 38, 844.
F. 0. HOLMES
Symptomatology of viral diseases in plants
AN IMPRESSIVE array of symptoms can be
recognized today as expressions of viral dis-
eases in plants. This is largely a result of
the rapid growth of scientific literature on
this subject in relatively recent years. New
techniques have enriched observations. Our
present data are far more varied in charac-
ter than the initial findings of investigators
three or four decades ago, when relatively
few viral diseases of plants had been
Pioneers in virology confined their atten-
tion almost entirely to foliage abnormalities,
failing to note many of the changes that
occurred in other parts of infected plants.
Various modifications of chlorotic effects in
leaves were the principal phenomena that
led to recognition of the fact that viruses
could cause diseases in plants just as fungi
and bacteria had long been known to do.
An all-inclusive category, known under
the name of "infectious chloroses," seemed
adequate at first to contain all the viral dis-
eases of plants. No sharp distinction was
made between constituent types. Some var-
iations were noted from the beginning, of
course, but were not explicitly defined or
differentiated until a considerable number
of viral diseases had been studied with
Gradually it came to be recognized that
one large group of infectious chloroses, the
mosaic diseases, could be differentiated
from another large group, the true yellows
diseases. Both were found to be transmissi-
ble through graft unions, but the mosaic
diseases were seen to be typically chlorotic-
spotting maladies whereas the true yellows
diseases were observed never to show any
spotting patterns at all.
Further studies disclosed that the mosaic-
type diseases could in many cases, although
not invariably, be transmitted mechanically
and that they usually had aphid vectors in
nature. They proved generally resistant to
brief heat treatments and characteristically
did not terminate flowering or affect dor-
mancy of buds. In these respects they dif-
fered sharply from the yellows-type dis-
The true yellows diseases showed a ten-
dency to produce virescent flowers and
eventual cessation of blossoming, to break
dormancy of all axillary buds, and to be
leafhopper-transmitted. They seemed in
general to be susceptible of cure by brief
exposures at relatively high temperatures,
such as 500-55C.
The ringspot diseases were recognized
later as a third group. They resembled the
mosaic diseases in having spotlike primary
and secondary lesions, but had a strong
tendency to disappearance of symptoms
soon after onset and to development of
what has been called "nonsterile" immunity
to reinfection (that is, an immunity when
Chapter 2-F. 0. Holmes
reinoculated, insofar as reinfection would
be shown by recurrence of symptoms of on-
set after conditions of chronic disease were
once established). They have proved not to
be transmitted by aphids or by leafhoppers,
but some of them are now known to be
transmitted by nematodes (as summarized
by Raski and Hewitt, 1963).
Additional major and minor groupings of
viral diseases have been suggested in the
recent literature, in some cases partly be-
cause of symptom resemblances (Bawden
and Pirie, 1942; Hollings, 1957; Bos and van
der Want, 1962); but in other cases largely
on the basis of characteristics such as physi-
cal properties of the etiological agents
(Broadbent and Heathcote, 1958; Cadman
and Harrison, 1959; Harrison and Nixon,
1960), vector relationships (Freitag et al.,
1952; Harrison et al., 1961; Brunt and Ken-
ten, 1962), serological overlaps (Bagnall
et al., 1958; Bercks and Brandes, 1961),
and tendencies to mutual suppression in
plant hosts (Bawden and Kassanis, 1941,
1945; Valenta, 1959).
Symptomatology may be expected to ex-
tend far beyond our current conceptions
eventually, for the presently known symp-
toms are largely such gross abnormalities as
are discernible without recourse to careful
laboratory-based studies. Virology has de-
veloped vigorously in recent years but ob-
viously has not yet reached its culmination
Viruses and their innumerable strains
appear able to upset in almost every con-
ceivable way the processes by which plants
grow and maintain themselves. Hence
symptom expressions are of many kinds.
Such discrete symptoms as have been re-
ported in the virological literature form a
discontinuous spectrum of abnormalities
covering nearly the whole range of plant
It should not be concluded, however,
that reported symptoms represent the only
levels of abnormality that exist in nature.
They seem to represent only a large num-
ber of deviations observed at random along
a partially explored and fundamentally con-
tinuous spectrum, just as the inch markings
and their subdivisions represent arbitrary
points of reference along a foot rule.
Primary lesions illustrate this to advan-
tage. They have been variously described
in such terms as necrotic spot lesions, dis-
tinctly yellow primary lesions, chlorotic or
faintly chlorotic spot lesions, ringlike le-
sions, chlorophyll-retaining lesions, and
masked lesions that can be disclosed only
by appropriate stains for starch content.
Actually it is possible to produce lesions of
intermediate character over this whole
range of symptoms, by choosing sometimes
more virulent and sometimes less virulent
strains of each virus or by changing host
plants to attain slightly higher or slightly
lower levels of tolerance. Other variations
in symptoms can be induced by utilizing
the subtle effects of changes in light inten-
sity and temperature that superpose on the
virus-host system a great range of physio-
logical modifications. Hence the specific
symptoms that have been reported in the
literature must be regarded as mere sam-
ples from a potentially unlimited range of
symptoms of each type.
In the future we may expect to see addi-
tions to our present knowledge in the form
of altogether unanticipated classes of symp-
toms. There will be, of course, a great
number of intensity variants of these new
kinds as well as new intensity variants of
previously recognized types.
In reviewing the symptoms that have
been reported thus far, it will not be feasi-
ble to refer to all of the viral diseases that
can display each one. It seems important,
however, to mention at least one or a few
diseases in connection with each, because
the cited diseases can then be looked up
in textbooks or library records to locate the
original literature dealing with the specific
symptoms. It is not feasible to give specific
literature citations for the very large num-
ber of symptoms specified here nor is it
possible to guarantee that the diseases men-
tioned as examples are the first reported,
the most conspicuous, or the most appropri-
ate ones that might have been selected to
exemplify each symptom. When a disease
name includes the name of a host plant, the
symptom under discussion will be under-
stood to occur in the host thus implied
unless some other host is explicitly named.
The following account represents as ac-
curately as possible our present accumula-
tion of factual data in respect to the
symptomatology of viral diseases in plants.
The symptoms are treated here by dealing
first with the extensive list of those that are
expressed in diseased foliage and then with
the relatively few, although still numerous,
symptoms that are concerned with other
parts of the plant hosts, such as flowers,
fruits, stems, buds, and roots, or that in-
volve the plant as a whole. Masked symp-
toms are discussed in a final section.
2-2. FOLIAGE SYMPTOMS
Foliage symptoms include abnormalities
in leaves with respect to color, number,
shape, size, texture, luster, and position, as
well as abnormalities of histology, cytology,
and basic physiological processes; a variety
of effects also involve premature death of
whole leaves or of macroscopic parts of
A. ABNORMALITIES OF LEAF COLOR
Leaves affected by viral diseases show a
great number of color abnormalities. These
may be separated into the following cate-
gories: Abnormalities of over-all color, vein
patterns, mottling patterns, spotting pat-
terns, ring patterns, and line patterns.
1. Abnormalities of over-all color in
leaves.-Foliage is usually less green than
normal in plants affected by viral diseases,
but a few diseases are characterized by
darker than normal green foliage.
Foliage is reported to be abnormally
green in peach phony disease, dark green
in Fiji disease of sugar cane, potato spindle
tuber, raspberry alpha leaf curl, and black-
raspberry streak. It is described as dark
bluish green in blue dwarf of oats. Leaves
are normally green or abnormally dark
green in potato green dwarf (potato curly
top), and darkened at first in tobacco
On the other hand, chlorosis appears in
all newly developing foliage in aster yel-
lows and many other yellows-type diseases.
Leaves are lighter green than normal in
strawberry witches' broom, light in color in
potato infected by potato virus S, and pale
in sugar-cane dwarf.
A yellow cast of foliage appears in potato
plants carrying the yellow mottle strain of
potato X virus. Flavescence of leaflets is
seen in potato witches' broom. Leaves are
yellowed in later stages of potato yellow
dwarf. Young leaves are yellowed in tomato
bushy stunt. Extreme chlorosis of foliage,
somewhat resembling the effect of mag-
nesium deficiency, is shown by soybean
after infection by the virus of alfalfa mo-
saic. Brightly yellowed foliage in the early
season is characteristic of yellow mosaic in
grapevines. Outer leaves are yellowed in
A diffuse bleaching of uninoculated tri-
foliate leaves has been reported in cowpea
systemically affected by pea wilt. Leaves
of sweet orange trees on sour orange roots
acquire a dull ashen color in quick decline
(tristeza) disease. Silvery upper leaf sur-
faces are said to be characteristic of cherry
vein-clearing disease as seen by reflected
light. Bronzed leaves that are slightly
glistening are reported for raspberry alpha
leaf curl. Leaves become uniformly golden
bronze green to olive brown as early as
June in cherry albino disease. Young leaves
are usually purplish underneath in tomato
A diffused and blended yellow and red
discoloration of foliage has been noted in
peach yellow-red virosis. This same disease
in another host plant produces even more
striking color changes, a brilliant red to yel-
low coloration of foliage six to seven weeks
after growth starts in spring, intensifying
with the season, in chokecherry in its first
year of manifesting yellow-red virosis;
there is a dulling of reddened foliage in the
second and third years of this disease in
Yellowing of tips of leaves in summer
with later premature reddening of leaf
margins occurs in blueberry stunt. Foliage
is reddened or orange-yellow in carrot af-
fected by motley dwarf disease. A bronzy
red color is acquired early in ocean-spray
witches' broom. Red to purple discoloration
appears in leaves affected by the Mesa Cen-
tral type of corn stunt disease, but only a
tinge of red appears in leaves affected by
the Rio Grande type of this disease. Leaves
are at first dull green but later prematurely
Chapter 2-F. 0. Holmes
bright red in strawberry green petal dis-
ease. On the other hand, leaves may re-
main green after normal foliage takes on
autumn coloration in Muir peach dwarf
Old leaves become chlorotic near their
tips in wheat streak mosaic. The margins of
leaves are yellow in strawberry yellow
edge. Areas of leaves appear rusty in cherry
2. Vein patterns.-The movement of vi-
ruses along veins is more rapid than through
tissues of the leaf lamina in general. This
causes a variety of vein-oriented symptoms.
A transitory chlorotic vein-clearing is the
first observable evidence of systemic dis-
ease in aster yellows and tobacco mosaic,
as well as in many other yellows-type and
mosaic-type diseases. It has been reported
also in cacao swollen shoot disease. A pro-
nounced yellow clearing of veins is charac-
teristic of the onset of turnip yellow mosaic.
Veins show clearing and thickening in
peach rosette. Flecks of vein-clearing in
lime foliage are hailed as the first evidences
of the presence of tristeza virus after ex-
perimental transmission by grafting.
Veins appear greener and hence darker
than normal in tobacco leaf curl. Dark
green veins are reported also in Pierce's dis-
ease of grape, caused by the virus of
alfalfa dwarf. Veins of young leaves are
blackened in broad bean vascular wilt.
Veinbanding mosaic, in which green
bands of tissue border the veins in other-
wise mottled leaves, is characteristic of
infections by potato Y virus in tobacco, as
well as of mild mosaic in potato. Brown
veinbanding is reported in some rose varie-
ties affected by rose streak.
Persistent yellowing occurs along veins
in Euonymus mosaic. Yellowing along veins
has been reported in red clover vein mo-
saic; the causative virus of this disease
induces the appearance of whitish bands
along veins in infected plants of Vicia faba
L. Yellowing of veins that tends to be
masked in summer is said to be characteris-
tic of alfalfa vein yellowing disease. Foliage
is yellow veined in yellow vein disease of
grapevines. Bright yellow chlorosis of veins
and veinlets accounts for the name of
sugar-beet yellow net disease. Yellow chlo-
rosis of veins and veinlets is observed in
yellow net disease of tomato. A network of
yellow veins encloses islands of green tis-
sue in yellow vein-mosaic of okra. Yellow-
ing along the margins of veins is found in
peach golden-net disease. Yellow spots and
blotches on the larger veins, extending into
the adjacent parenchyma a distance of one
millimeter or more, characterize sugar-beet
yellow vein disease.
Chlorotic starlike lesions occur along
veins in peach asteroid spot disease. Star-
like spots, consisting of coalescent pellucid
veins, are found in asteroid mosaic of grape.
Chlorotic flecking, mostly along veins of
dwarfed and puckered leaves, occurs in
apple leaf-pucker disease. Chlorotic spots
appear on veins in strawberry veinbanding.
Interveinal reddish browning on both
leaf surfaces, extending from margins in-
ward, has been reported as a summer symp-
tom in crimson clover infected by the virus
of potato yellow dwarf. Interveinal areas
become reddish in filaree red leaf disease.
Interveinal chlorosis occurs in barley yellow
dwarf and hop yellow net. Fine brown lines
parallel to main veins, followed by death
and dropping out of included areas, are
recorded for tatter leaf disease in sweet
3. Mottling in leaves.-Chlorotic mot-
tling, apparently caused by the nonuniform
establishment of secondary lesions in de-
veloping leaves, is typical of mosaic dis-
eases in general. If the degree of chlorosis
is slight, the condition is sometimes de-
scribed as green mottling; when extreme,
it may be referred to as yellow mottling or
even white mottling.
Green mottling is attributed to typical
cucurbit mosaic and tobacco mosaic. Mild
dark green mosaic mottling is caused by an
isolate of so-called mild dark green mosaic
virus from Nicotiana glauca L. when this is
introduced into plants of cultivated to-
bacco. Yellow mottling is mentioned as
occurring in Hyoscyamus henbanee) mo-
saic, turnip yellow mosaic, and lettuce yel-
low mosaic. Yellow mottle is seen in potato
aucuba mosaic and dandelion yellow mo-
saic. White mottling has been reported in
connection with a mosaic caused in tobacco
by cucumber mosaic virus of Price's strains
2 and 4. Gray mottling is said to be found
with yellow mottling in Ornithogalum mo-
Systemic chlorotic variegation occurs in
Abutilon mosaic. The chlorotic spots in this
disease have a tendency to be limited by
veins more than is the case in such diseases
as tobacco mosaic or cucumber mosaic.
Feathery chlorotic patterns, consisting of
finely branched yellow mottle along small
veins with or without many small diffusely
margined spots, account for the name
of sweetpotato feathery mottle disease.
Feather-like mosaic mottling is said to oc-
cur also in cacao swollen shoot disease. Vein
feathering is mentioned in connection with
yellow bud mosaic in almond.
4. Spotting patterns in leaves.-Spotting
of leaves sometimes represents the local
damage at sites of infection in foliage (pri-
mary lesions) but often is caused in sys-
temic disease by the formation of relatively
few and hence widely separated and dis-
crete secondary lesions.
Yellowish-green primary lesions are oc-
casionally seen in tobacco mosaic, although
they tend to be so faintly contrasted with
the color of the healthy leaf surface as to be
overlooked. Yellowish primary lesions in
Nicotiana rustica L. infected by typical
potato yellow dwarf virus are used for
quantitative estimation of viral concentra-
tions in inocula. Raised yellow primary
lesions have been described in pineapple
infected by tomato spotted wilt virus.
Primary lesions are disclosed by chloro-
phyll retention in older leaves of tomato
affected by bushy stunt disease. Chloro-
phyll is retained in lesions as leaves turn
yellow in peach asteroid spot disease.
Pale spotting of leaf bases is reported in
maize streak. Leaves show a fine chlorotic
stippling in achaparramiento, or corn stunt,
disease. Small pale-yellow to paper-white
spots aggregated near tips of leaves are
reported in plum white spot disease. Pale
spots between veins near tips of older
leaves are found in maize leaf fleck. Pale
spots appear in leaves in cotton leaf curl.
Petioles are white-streaked or spotted in
western celery mosaic. Yellow spotting oc-
curs in interveinal areas of leaves in red
clover vein mosaic. Small bright yellow dots
appear on leaves in bean yellow dot dis-
ease. Small yellow spots in irregular patches
on leaves are characteristic of bean yellow
stipple. Small, circular, stellate, or dendritic
yellowish spots occur in spring in Pelargo-
nium leaf curl; the centers of these spots
usually become necrotic and leaves become
twisted by unequal growth of unaffected
and affected tissues.
Orange to reddish blotches appear on
leaves in red leaf of oats. Purple or purple-
brown spotting on young leaves occurs in
pea streak. A red pigmentation in spots that
later become dry and drop out has been
reported for vine mosaic. Chlorosis at leaf
base extends upward in plum blotch; the
chlorotic blotches in developing foliage be-
come reddish brown and develop shot
holes. A blotching type of chlorosis has
been reported in sugar-beet mosaic. A red-
dish discoloration of pulvini of leaves and
leaflets occurs in bean red node.
5. Ring patterns in leaves.-Whitish ne-
crotic rings and concentric markings are
typical of tobacco ringspot and allied dis-
eases. Incomplete chlorotic rings develop
interveinally on newly formed leaves in
Bergerac ringspot of tobacco. Rings consist-
ing of nonnecrotic areas alternating with
necrotic patches are characteristic of to-
bacco broken ringspot. Bronze ringlike sec-
ondary lesions are characteristic features of
tomato spotted wilt. Ringlike necrotic pri-
mary lesions are found in tomato bushy
stunt. Small black necrotic rings or spots
appear in cabbage black ringspot. Both
local and systemic lesions appear as small,
black, necrotic rings in tomato black-ring
disease; if the rings coalesce, affected tis-
sues often collapse and die.
6. Line patterns in leaves.-Line patterns
are especially common in the viral diseases
of grasses. In these they may represent dis-
tortions of spotting patterns or fused spot-
ting patterns. In other plants, line patterns
are doubtless related often to extensive
Chlorotic streaks near the bases of young
leaves, later fusing to form continuous
stripes, have been observed in maize stripe.
Fine chlorotic striae followed by yellowing
of leaves occur in wheat striate mosaic.
Broken chlorotic streaks on tips of young
Chapter 2-F. 0. Holmes
leaves characterize wheat streak mosaic.
Yellow bands or irregular chlorotic rings
on mature leaves have been noted in del-
phinium ringspot; yellow streaks at the
bases of developing leaves, in onion yellow
dwarf; and green and yellow zigzag bands,
in celery calico.
Long, narrow, longitudinal streaks of
creamy or white tissues have been reported
in sugar-cane chlorotic streak disease; light
colored line patterns, in Prunus line pattern
disease; and pale stripes from midrib to
margin of leaf, in marble disease of
Necrotic or yellowish concentric lines on
affected leaves have been seen in Odonto-
glossum orchid ringspot; reddish brown to
blackish pitting and sunken streaks in
leaves and pseudo-bulbs, in Cattleya or-
chid virosis; chlorotic elongate areas and
later necrotic spots and streaks, in black-
streak disease of Cymbidium orchids.
Oak-leaf patterns, extending farther
along main veins than between these, are
occasionally noted in many and perhaps
all mosaic-type and ringspot-type diseases.
B. ABNORMALITIES OF LEAF NUMBER
Tobacco plants infected by severe etch
virus show an increased rate of leaf produc-
tion as well as a delay of flowering; hence
more leaves are produced by infected than
by normal plants.
Fewer than normal leaves appear in
sweet cherry necrotic rusty mottle because
some buds and leaf spurs die in this disease.
Leaves are few in pear stony pit.
In general, leaves are more numerous
than is normal in viral diseases that break
the dormancy of buds and less numerous
than normal in those that induce premature
C. ABNORMALITIES OF LEAF SHAPE
A well-nigh universal expression of viral
diseases in plants is distortion of leaf con-
tours. This distortion varies greatly in
degree. There are almost imperceptible
deviations from normal shape in leaves of
so-called "recovered" (i.e., chronically dis-
eased as opposed to acutely diseased)
plants affected by ringspot diseases. Better
recognized are the moderately pinched
leaf tips and intermediate irregularities of
leaf margins displayed in many mosaic dis-
eases. Occasionally one may see extreme
conditions of fern-leaf and shoe-string man-
ifestations, in which the leaf lamina be-
tween veins is poorly developed or hardly
developed at all, as is sometimes the case
in tomato plants affected by cucumber mo-
saic, by special strains of tobacco mosaic, or
by complexes including one or both of these
as constituents. Leaves are distorted and
sometimes filamentous in the eggplant mo-
saic of India. Filiform leaves characterize
late stages of teasel mosaic.
Leaves are almost round and lack teeth
in their margins in spoon leaf of red cur-
rant, caused by a strain of raspberry ring-
spot virus. In contrast to this, leaves become
lobed because of varying degrees of
imperfect growth of lamina in tobacco
Definite nations, which are leafy out-
growths of variable size, appear conspicu-
ously in sweet cherry rasp leaf and pea
enation mosaic. In the latter disease, al-
though the nations are usually blister-like
or ridgelike pseudo-enations or true lamina-
like nations on the undersides of pea leaf-
lets and stipules, they have been observed
also as imbricated, prominently veined,
leaflike tissue with a short stalklike attach-
ment to the main stem of the pea plant.
Conspicuous nations occur in aspermy dis-
ease in Nicotiana glutinosa L., generally
as leafy outgrowths from the lower surface
of the leaf but occasionally as outgrowths
from the upper surface of the midrib,
producing a structure resembling a small
Lamina-like nations seem to be the ex-
treme members of a series of abnormalities
that would include veins greened and
thickened, as in tobacco leaf curl and cot-
ton leaf curl (which sometimes also show
fully developed nationss, veins swollen in
sugar-beet leaf curl, and unevenly thick-
ened veins that are depressed below the
upper surface of the leaf (also sometimes
with full nations) in crimson clover big
vein disease (caused by wound tumor vi-
rus). In Fiji disease of sugar cane, galls
appear on the lower surfaces of leaves,
formed on vascular bundles through prolif-
eration of phloem and nearby cells. Sharp
protuberances on lower surfaces of veins as
well as knotlike swellings resembling galls
on distorted veins occur in sugar-beet curly
top. Nations develop on veins within 18 to
33 days after transmission to lime of citrus
vein-enation virus by the aphid Myzus
Leaves are thickened in tomato big bud,
sugar-beet curly top, potato yellow dwarf,
potato leaf roll, and sugar-beet yellows.
They are thinner than normal, in their
yellow areas, in sugar-beet mosaic and
mosaic diseases in general.
Leaves are described as shortened and
crumpled in Fiji disease of sugar cane, short
in sugar-beet yellows, dwarfed in peach
rosette, and dwarfed and curled downward
in beet savoy.
Leaves are blistered in cucurbit mosaic,
cotton leaf curl, and citrus infectious mot-
tling, sometimes savoyed in clover club
leaf, and crinkled in tobacco leaf curl,
raspberry leaf curl, and bean mosaic.
Leaves are said to be rolled in sugar-beet
curly top and potato leaf roll, rolled up-
ward in cherry leaf roll, rolled downward
in grape leaf roll, rugose, rolling down-
ward, thickened, and later savoyed in to-
bacco yellow dwarf. They are cupped at
tips of terminal shoots in autumn in apple
star-crack disease, cupped inward in grape
fan leaf, and cupped outward in filaree red
leaf. They are cupped upward in potato
green dwarf, but are said to resemble in-
verted spoons, with their margins curled
downward, in tobacco rattle disease.
Leaves are described as curled in black-
raspberry streak, curled upward in hop
nettlehead, curled lengthwise and upward
in quick decline of citrus, and arched
downward in flowering cherry rough bark
Leaves are tattered in peach yellow-red
virosis (X disease). They are perforated by
the dropping out of spots in vine mosaic
and tatter leaf of sweet cherry. Late in the
season, usually after harvest, affected areas
fall out of leaves, producing a conspicuous
shot-hole effect, in sweet cherry necrotic
Leaves are described as narrow and flat
in black currant reversion disease; short,
narrow, and stiff in abact bunchy top; short
and rigid in blue dwarf of oats. They are
simple rather than normally compound in
shoots from potato tubers formed in potato
witches' broom. Leaves become sickle-
shaped as a result of unsymmetrical chloro-
sis in olive sickle-leaf disease. Leaf blades
are often bilaterally unequal in cherry
Leaves of onion, which are normally
more or less circular in cross section, are
described as flattened in onion yellow
dwarf. Leaves of raspberry are fluted along
veins in raspberry decline. Leaves are said
to be folded in strawberry stunt. They have
a ribbed appearance in tobacco yellow
dwarf. They are folded along their mid-
veins in cherry vein-clearing.
Leaves are warped in citrus psorosis and
twisted in rape savoy. The leaf spindle is
twisted in sugar-cane dwarf.
Leaflets are described as recurved in rose
wilt; narrow, cupped, and twisted in west-
ern celery mosaic; and slightly cupped
convexly in cowpea mosaic.
Midribs are said to be curved in cauli-
flower mosaic, tortuous in strawberry stunt,
split and cracked underneath in flowering
cherry rough bark disease, and bent down-
ward in cherry twisted leaf. Midveins of
leaflets arch downward in strawberry
Veins are reduced in number in black
currant reversion disease. They are short-
ened in raspberry leaf curl. Veins are
crooked in tobacco infected by tobacco
Petioles are undulating in crimson clover
affected by big vein (wound tumor) dis-
ease, thickened in anemone alloiophylly,
spindly in strawberry witches' broom, and
slender in potato spindle tuber. They curve
outward in filaree red leaf disease. In carrot
motley dwarf, petioles are twisted and
sometimes S-shaped or bent backward so
that undersurfaces of leaves are exposed to
view. Stipules are enlarged in cherry
Western X disease.
Leaf margins turn down in strawberry
stunt. They are puckered in cotton leaf
curl, rolled down in bean mosaic, curling
down in raspberry leaf curl, curled down-
ward in a spoonlike fashion in red currant
spoon leaf (caused by a strain of raspberry
Chapter 2-F. 0. Holmes
ringspot virus), and curved downward in
strawberry witches' broom. Fissures are
formed at leaf edges in rape savoy.
D. ABNORMALITIES OF LEAF SIZE
In respect to leaf size, there seem to be
almost no reports of gigantism in leaves,
despite greatly increased size of calyx lobes
in some yellows-type diseases. Leaves that
have been reduced in width often seem
longer than normal and may even be so
without this having been specifically noted
in published literature. Attention should be
given in the future to clarification of this
point. Already there have been some hints.
Leaves on the Drake almond tree when
affected by bud failure disease are reported
to be possibly somewhat larger than those
on nonaffected trees. Some leaves on to-
bacco plants, produced after infection by
severe etch virus, have proved both longer
and wider than comparable leaves on
healthy control plants (their thicknesses
and weights are not reported).
Reduction of leaf size, on the other hand,
is exceedingly common and perhaps in one
degree or another nearly universal. In some
cases leaves become so small that a host
species becomes almost unidentifiable. In
potato witches' broom, black locust witches'
broom, and eggplant little leaf, successively
formed leaves are progressively smaller un-
til even a trained botanist would find diffi-
culty in recognizing the host plants as
potato, black locust, and eggplant, respec-
tively. Except among the yellows-type dis-
eases, however, such uniform and extreme
reduction of leaf size seems not to occur,
yet there is very commonly an appreciable
degree of reduction, as in alfalfa dwarf
where leaves are small. Individual leaves
are occasionally greatly reduced in size, as
in tomato affected by cucumber mosaic, in
which there may be formed no more than
a miniature petiole and midvein to repre-
sent a particular leaf. Absence of the leaf
lamina or narrowing of its apical portion
greatly distorts tobacco leaves affected by
the Brazilian disease, necrose branca, or
Shortened young petioles are described as
occurring in western celery mosaic, and
this is the more important because petioles
are the commercially desirable parts of
celery leaves. Petioles are short in straw-
berry stunt and potato yellow dwarf also.
E. ABNORMALTmES OF LEAF TExTURE
Leaves are described as less rigid than
normal, when young, in corn infected by
the mottle strain of maize streak virus.
They may droop and appear somewhat
wilted in cherry vein-clearing. Leaves ap-
pear wilted but are stiff in cherry leaf roll.
Leaves are described as stiff in sugar-
cane dwarf, stiff and small in potato yellow
dwarf, brittle in rose wilt, rigid in Phenom-
enal berry after infection by the virus of
loganberry dwarf, leathery in little peach
disease, pea leaf roll, and potato leaf roll,
but unusually tender in citrus psorosis.
An interesting feature of strawberry stunt
is that leaves give a papery rattle when
they are brushed by hand. In leaf roll of
cherry, also, leaves rattle when shaken. In
sugar-beet yellows, in which leaves are
thickened and brittle, they are said to
crackle or rustle when brushed. Tobacco
leaves rattle when disturbed in tobacco
F. ABNORMALITIES OF LEAF LusTER
Leaves are glazed in prune dwarf. They
show a glazed-appearing upper surface in
tobacco infected by tomato big bud virus.
Foliage appears dry in raspberry leaf
curl. It is dull in strawberry stunt. It is
sometimes less glossy than normal in peach
line pattern disease.
G. ABNORMAImES OF LEAF PosrnoN
Leaves are abnormally erect in most yel-
lows-type diseases. They are rigidly upright
in Drake almond bud failure disease in-
stead of bending away from the twig as in
normal trees. They are unusually close to-
gether, forming rosettes on terminals, in
yellow-red virosis of chokecherry. Short
stems bear leaves in rosettes in peach ro-
sette disease. The leaves are bunched in
tobacco yellow dwarf. Yellowed or red-
dened leaves curl downward close to the
ground on dwarfed plants in lily rosette, or
yellow flat, disease. Rosetting occurs in
barley and other cereal plants affected by
enanismo, or dwarf disease, in Colombia.
In black-raspberry streak, leaves are ab-
normally close together on canes and often
are twisted so as to be upside down. Leaves
are mostly restricted to the tips of branches,
giving a "lion's tail effect," in citrus xylo-
porosis. Leaflets are crowded in rose wilt.
Leaves may be prostrate in onion yellow
dwarf. They hang close to the stem in to-
bacco after infection by tomato big bud
virus. They droop at the pulvini in southern
bean mosaic. Petioles of older leaves be-
come horizontal in western celery mosaic.
Small incurved new leaves form a com-
pact head in sugar-beet leaf curl. Lettuce
mosaic, on the other hand, is characterized
by defective heading of plants.
H. ABNORmALITIES OF LEAF HISTOLOGY
Vascular bundles in leaves are enlarged,
irregular in shape, and fused in sugar-cane
dwarf. They are said to be more numerous
than usual in anemone alloiophylly. Vascu-
lar tissues are described as darkened in
Phloem degeneration is followed by for-
mation of supernumerary sieve tubes in
sugar-beet curly top. Necroses occur in
phloem and adjoining sheath tissues in lily
rosette. Phloem necrosis has been noted in
winter wheat mosaic.
Palisade cells are reported as abnormally
short in anemone alloiophylly; chloroplasts
are smaller and fewer in these cells. Winter
wheat mosaic is also reported to show
chloroplasts that are few and small in af-
fected tissues, as do many other infectious
chloroses, such as sugar-cane streak.
I. ABNoRmALITIES OF LEAF CYTOLOGY
Intracellular inclusions of kinds not
found in healthy plants characterize many
viral diseases. Some of these inclusions are
found only in the cytoplasm but others are
intranuclear or intranucleolar.
Abnormal intracellular inclusions are
spherical or oval in Fiji disease of sugar
cane. Vacuolate intracellular bodies have
been described in connection with sandal
spike, rice dwarf, tobacco mosaic, Hippea-
strum mosaic, and potato mottle.
Amoeboid intracytoplasmic bodies are
abundant in subveinal epidermis in turnip
Intracytoplasmic crystalline inclusions oc-
cur commonly in tobacco mosaic, but never
in the closely mimicking mosaic induced by
infection with cucumber mosaic virus. In-
tracytoplasmic amorphous inclusions that
tend to crystallize, forming needle-shaped
birefringent bodies, are reported in tobacco
etch. Crystalline inclusions in red clover
vein mosaic are said to contain ribonucleic
acid but not deoxyribonucleic acid, starch,
or fats; amorphous inclusions are reported
in this disease also. Protein crystals are
found in affected cells in oat pupation dis-
ease. Cigar-shaped inclusion bodies are
found in cactus virosis.
Isometric crystals occur in host cell nu-
clei and cytoplasm of Vicia faba L. affected
by pea mosaic. Intranuclear crystalline in-
clusions, in the form of thin rectangular
plates, occur in tobacco etch. Spherical,
sometimes multiple, inclusion bodies in
hypertrophic nuclei of phloem parenchyma
cells have been reported in cotton leaf
Nuclei are enlarged and contain extra
nucleoli in winter wheat mosaic. Intranu-
cleolar inclusions of proteinaceous type are
recorded for bean yellow mosaic in broad
J. ABNO ALIrrmES OF MISCELLANEOUS
PHYSIOLOGICAL PROCESSES IN LEAVES
A wide variety of disturbances of physi-
ological processes not obviously associated
with abnormalities of structure have been
noted as symptoms of viral diseases.
Early unfurling of new leaves is charac-
teristic of abaca bunchy top. In contrast to
this, young leaves are said to be slow to un-
fold in crimson clover affected by clover
club leaf. Delay in foliage development
has been reported for red raspberry mosaic,
Pierce's disease of grape, star-crack disease
of apple, and weak peach disease.
Leaf tissues are described as having
excessive starch content, as well as an in-
creased concentration of reducing sugars,
in potato leaf roll; serine and aspartic acid
are less, but y-aminobutyric acid more,
concentrated than normal in this disease.
In sugar-beet yellows, old leaves contain an
abnormally large content of carbohydrate,
and the greater part of the additional car-
Chapter 2-F. 0. Holmes
bohydrate consists of reducing sugars; also,
two substances are formed that fluoresce
and appear as white spots in chromato-
grams. Sugar-beet mild yellows is not char-
acterized by these fluorescent substances
but produces at least six yellow or orange
pigments in addition to those in healthy
Starch is slow to accumulate in primary
and secondary lesions of tobacco mosaic in
young tobacco leaves; once accumulated,
it is slow to disappear from these same
lesions during subsequent periods of dark-
ness; thus, at appropriate times it is pos-
sible to detect starch-deficient and starch-
retaining lesions by the use of iodine stains.
Starch is said to be removed abnormally
slowly from darkened leaves in citrus
Ribonuclease activity is said to increase
twice as much in leaves of cowpea inocu-
lated with cucumber mosaic virus as in
leaves abraded without inoculum.
It has been noted that there is a slightly
decreased respiration just before, and a
strongly increased respiration immediately
after, the appearance of primary necrotic
lesions in leaves of Nicotiana glutinosa L.
as a result of infection by tobacco mosaic
virus (TMV). Infection of tobacco leaves
by tobacco etch virus does not change
respiration rates until leaves show external
symptoms, but respiration then rises
sharply and remains high throughout the
life of the leaves.
An excess of proline in midsummer dis-
tinguishes peach Western-X-diseased
leaves from normal; abnormal retention of
pipecolic acid after springtime has also
been noticed and introduction of D,L-pi-
pecolic acid into leaves of healthy trees
produces reddening of veins and other
effects resembling the symptoms of the
disease in nature.
The attraction of vascular bundles for
dodder as a parasite is said to be lowered
in sugar-beet curly top. The attractiveness
of peach trees to Japanese beetles is in-
creased by peach yellows.
Leaves absciss prematurely in sour
cherry yellows, peach yellow-red virosis (X
disease), cacao swollen shoot, and cherry
rusty mottle. All infected leaves absciss
promptly after the appearance of necrotic
primary lesions in Tabasco pepper plants
inoculated with tobacco mosaic virus, thus
saving the plants from systemic infection.
Pea plants tend to be defoliated when in-
fected by alsike clover mosaic virus 2.
Inoculated leaves wilt but remain at-
tached in pea wilt. Leaves yellow and die
but remain attached when dry in olive
die-back. Leaves drop or dry on tree in
quick decline of citrus (tristeza disease).
A clear viscid exudate, which may later
turn black and sticky, appears on petioles,
midribs, or veins on the lower surface of
leaves affected by sugar-beet curly top.
K. ABNORMALITIES COMPRISING DEATH
OF WHOLE LEAVES OR OF MACROSCOPIC
PARTS OF LEAVES
The usefulness of so-called local-lesion
hosts for measuring relative concentrations
of viruses in samples of inoculum has led to
conscious search for plants capable of re-
sponding sharply to initial infections and
specifically for plants in which tissues tend
to die promptly in the immediate vicinity
of the first infected cells, thus producing
conspicuous necrotic primary lesions. Ex-
perience has shown that most mosaic-type
and most ringspot-type diseases have such
hosts among potentially susceptible but not
usually infected herbaceous species, but
that no such local-lesion hosts can be found
for yellows-type diseases.
Necrotic primary lesions are described in
the literature as brown in leaves of Tabasco
pepper infected by tobacco mosaic virus,
dull red in cowpea infected by cucumber
mosaic virus, reddish in cowpea infected by
tomato bushy stunt virus, white in tobacco
infected by tobacco ringspot, and black in
tomato infected by alfalfa mosaic virus.
They may be simple dead spots as in cu-
cumber infected by tomato bushy stunt
virus or zonate lesions as in pea wilt and
most ringspots. Leaves may show brown
necrotic primary lesions with gray centers
as in Nicotiana rustica L. infected by an
especially severe strain of potato yellow
dwarf virus. A light yellow nonnecrotic halo
may surround a dark brown necrotic pe-
ripheral ring which in turn surrounds a light
brown center, as in primary lesions in cu-
cumber infected by the virus of pea streak.
Sometimes indistinct yellowish primary le-
sions, resembling the haloes mentioned
above, become slowly necrotic, as in to-
bacco broad ringspot. Necrotic primary
lesions may be surrounded by chlorotic
haloes in Physalis peruviana L. after inocu-
lation with tobacco etch virus. Similar
necrotic lesions usually lack conspicuous
chlorotic haloes in Nicotiana glutinosa L.
after inoculation with TMV. In hybrid de-
rivatives of Nicotiana rustica L. inoculated
with TMV, individual plants characteristi-
cally display such necrotic lesions sur-
rounded by conspicuous chlorotic haloes if
a recessive gene is present but the necrotic
lesions lack haloes and appear more
promptly after inoculation if a dominant
allele replaces this.
Secondary necrotic lesions in young
leaves may be isolated spots or rings, or
may be large numbers of spots so close to-
gether as to cause collapse of whole leaves.
Sometimes whitish secondary necrotic le-
sions may appear not as solid spots but as
indistinct flecks or arcs and rings, as in
tobacco infected by the virus of alfalfa
mosaic. Necrotic flecking is reported in
chlorotic areas of leaves affected by cauli-
flower mosaic. Necrotic etching along veins
and peripherally around chlorotic primary
and secondary lesions occurs in tobacco
Necrotic lesions produced by tobacco
necrosis virus as a result of inoculation of
leaves of French bean are smaller, and
sometimes more numerous, if the so-called
satellite virus is also present; this satellite
virus has been found to multiply detectably
only in the presence of tobacco necrosis
virus, from which, however, it is serologi-
When necrosis occurs within the creamy
or white streaks of sugar-cane chlorotic
streak disease, it begins in mesophyll and
involves vascular bundles late.
Necrotic spots of irregular shape and of
blackish-brown color occur in foliage of
doubly-infected tomato plants wherever to-
bacco mosaic virus and potato X virus
invade the same individual cells. Necrotic
flecks in leaves of Easter lily are distinctive
features of the double-virus infection that
involves lily-symptomless virus plus cucum-
ber mosaic virus.
Outer leaves are reported to die from
their tips back in spinach affected by sugar-
beet mosaic. Leaf edges are scorched in
some varieties of lettuce affected by lettuce
mosaic. Foliage shows burning and scald-
ing, with death of leaf margins, in Pierce's
disease of grape. Inoculated leaves are
killed in pea and bean plants infected by
alsike clover mosaic virus. They are re-
ported to collapse in Nicotiana rustica L.
infected heavily by potato yellow dwarf.
Veinal necrosis of older leaves appears
in tobacco that has been infected by tomato
big bud virus. Young leaves show a bronze
veinal necrosis in lettuce infected by dan-
delion yellow mosaic virus. Interveinal
necrosis is described as a feature of sugar-
beet yellows disease in spinach.
2-3. FLOWER SYMPTOMS
Flower symptoms include abnormalities
of flower color, size, number, shape, posi-
tion, pollen and egg viability, time of blos-
soming, and persistence.
A. ABNORMALTrES OF FLOWER COLOR
"Broken" or variegated flowers have long
been recognized as features of some viral
diseases. In the case of tulip mosaic the
resulting patterns were once valued as or-
namental and still are sometimes admired
despite the relatively poor growth of af-
fected plants. Streaks on petals may be a
result of decreased pigmentation, as is
usual in tulip mosaic, or a result of in-
creased pigmentation, as in tulip color-ad-
ding disease. Streaks as a result of local
intensification of flower pigments charac-
terize cranberry false blossom.
Transparency of flowers has been noted
in black currant reversion disease. Opaque
frostlike streaks on perianth and cup of
flower are reported for narcissus mosaic.
White boat-shaped spots and streaks, some-
times with pink centers, are seen in pink-
flowered peach affected by peach mosaic.
Vinca flowers that normally are white with
red centers show progressive suppression of
red pigment after infection by the virus of
corn stunt. Blotches of light and dark green
Chapter 2-F. 0. Holmes
on flower stalks occur in Ornithogalum
One of the commonest abnormalities of
flower color is virescence, or greening of
petals, a nearly universal characteristic of
the true yellows diseases, especially note-
worthy in aster yellows, tomato big bud,
and strawberry green petal.
B. ABNORMALITIES OF FLOWER SIZE
Flowers are dwarfed to a greater or less
extent in many viral diseases, as for ex-
ample in chrysanthemum flower distortion
disease. Occasionally parts of flowers be-
come abnormally large. A bladder-like
structure may result from fusion of en-
larged sepals, as in tomato big bud. Gigan-
tic though often still discrete calyx lobes
may be produced, as in tomato experimen-
tally infected by cranberry false blossom
virus. In strawberry green petal disease,
also, sepals are enlarged. Enlarged sepals
on male flowers but not on female flowers
are reported in cucumber witches' broom.
C. ABNORMALrES OF FLOWER NUMBER
Abnormally profuse blooming has been
noted as a feature of sandal spike and
cherry vein-clearing diseases. Flower buds
are more often geminate (with two com-
plete flower buds on a common receptacle,
enclosed by a single set of bud scales) in
peach mosaic than in healthy peach trees.
Fewer than the normal number of
flowers develop in each inflorescence in
chirke disease of cardamom. Blossoming
ceases in pigeon-pea sterility disease. Flo-
rets are blasted in oat yellow leaf disease.
First-formed floral heads sometimes are
killed in lettuce mosaic, but side shoots
later produce seeds.
Yellows-type diseases usually show an
unfruitfulness of flowers that is often a prel-
ude to complete suppression of flowering,
as in alfalfa witches' broom. There is grad-
ual suppression of flowering in Vinca in-
fected by chrysanthemum flower distortion
D. ABNORMALITIrES OF FLOWER SHAPE
Petals are narrow and distorted in prune
dwarf. The corolla tends to split and petals
are curiously "apiculated," showing thread-
like tips, in the Brazilian disease necrose
branca, or white necrosis, which is probably
the same as the North American tobacco
streak. The calyx and corolla may be
twisted and distorted in sour cherry ne-
Diseases that produce virescent flowers
tend to produce a modified floral axis bear-
ing small leaves (phyllody) or proliferating
flower buds, as in aster yellows in some
host species, sunn-hemp phyllody, and clo-
E. ABNORALrrTIES OF FLOWER POSITION
In yellows-type diseases flowers as well
as branches are often abnormally erect.
This is conspicuous in cranberry false blos-
som disease, because cranberry flowers are
pendant on healthy plants. Flower stalks
are sometimes bent in onion yellow dwarf.
Flower pedicels are short in cherry leaf
F. ABNORMALITIES OF POLLEN AND
EGG VIABmLTY IN FLOWERS
In tomato aspermy disease, many pollen
cells and egg cells fail to function. Abortive
pollen grains are reported to be numerous
in tobacco ringspot. Pollen from cherry
trees affected by sour cherry yellows in-
duces poorer sets of fruit than normal when
applied to healthy or diseased cherry trees.
Ovules are said to become leaflets in kok-
G. ABNORMALTES IN TIME OF BLOSSOMING
Blossoming may occur prematurely or, in
contrast to this, it may be delayed as a
result of viral infection.
Flowers are sometimes formed prema-
turely, i.e., during the first year of growth,
in normally biennial species, as in carrot
bolting disease. Premature flowering in
citrus is characteristic of quick decline
(tristeza disease). New Zealand flax blos-
soms prematurely when affected by Phor-
mium yellow leaf. Flowers are reported to
appear early in chrysanthemum stunt.
Flowering in spring may be slightly de-
layed in Muir peach dwarf. Retarded blos-
soming has been reported for sweet cherry
necrotic rusty mottle. Affected trees blos-
som late in spring in sour cherry pink fruit
disease. Tobacco plants affected by severe
etch disease are said to be delayed in
H. ABNORMALITIES OF FLOWER PERSISTENCE
Flowers may abort, as in cowpea mosaic.
Pistils of flowers are described as aborted
in prune dwarf. Flowers drop off from to-
mato plants after infection by sugar-beet
2-4. FRUIT SYMPTOMS
Fruits express symptoms as abnormalities
of number, size, shape, habit, color, texture,
ripening, flavor, seed content, persistence,
and total yield; some viral diseases also
cause death of parts of fruits.
A. ABNORrmALITIES OF FRUIT NUMBER
Fruit production is often seriously de-
creased in viral diseases but rarely if ever
Complete sterility occurs eventually in
most yellows-type diseases, usually as a
result of gross flower abnormalities or fail-
ure of plants to continue the production of
blossoms. Complete unfruitfulness without
damage to pollen has been reported in
connection with female sterility disease of
tobacco. Fruiting is slight in tomato
aspermy disease because pollen cells and
ovules are largely nonfunctional.
Mature fruits are few, although blossoms
are numerous, in prune dwarf. The crop of
fruits is reduced or wanting, despite abnor-
mally abundant blossoms, in cherry vein-
clearing disease. Fruits are fewer than
normal in peach phony disease, Drake al-
mond bud failure, and cucumber mosaic.
Cone production is greatly reduced in hop
nettlehead. Only occasional fruits are left
on trees in peach X-disease, because of
abnormal fruit drop. The crop is progres-
sively reduced in successive years in Muir
peach dwarf and in advanced cases there
may be no more than a dozen fruits on
each tree. Lambert cherry trees set only a
light crop when affected by Lambert mot-
tle. Fruit is lacking or scanty in grape
yellow vein disease.
B. ABNORMALITIES OF FRUrr SIZE
Reduction of fruit size is a common but
not invariable symptom of viral diseases.
Fruits are much reduced in size in little
peach disease, which is caused by infection
with a mild strain of peach yellows virus.
Despite good foliage to support their
growth, cherry fruits remain small because
of deficient cell division in otherwise nor-
mal-appearing trees affected by little
cherry disease. Kernels of maize are re-
duced in size by corn stunt but remain via-
ble and edible. Fruits are small, pale, and
conical in cherry buckskin disease. They
are small and turn white in cherry albino
disease. Infected seeds tend to be small in
barley false stripe.
Cotton bolls are reduced in size by cotton
leaf crumple disease. Cantaloupe fruits are
small, poorly netted, and low in sugar on
plants infected by sugar-beet curly-top vi-
Fruits are often larger than normal and
of excellent quality, but few, in prune
dwarf disease. Fruits are larger than nor-
mal and are said to be of good quality in
sour cherry yellows, but the number of
fruiting spurs is reduced.
C. ABNORALITIES OF FRUrr SHAPE
Fruits, like leaves, are likely to be
changed to some extent in shape, and
sometimes are radically distorted, by viral
Cucumbers are conspicuously misshapen
as well as small and mottled in cucumber
mosaic. Navel orange fruits are misshapen,
with rind abnormally thick near the stem
end but thin near the stylar end, in stub-
born disease of citrus. Fruits become
spineless in Datura distortion mosaic.
Dwarfed, ribbed, or lobed fruits have
been reported in apple decline. Flattened
patches of green skin distort fruits in dapple
apple disease. Irregular fruits are formed in
cranberry false blossom. Fruits are de-
formed in apple false sting disease and
lumpy in apple green crinkle. They are se-
verely deformed and pitted in stony pit of
pear. Symmetry of fruits is destroyed in
side rot (spotted wilt disease) of pine-
apple, which causes fruits to bend toward
Chapter 2-F. 0. Holmes
the affected side. The suture sides of fruits
often bulge in peach red suture disease.
Fruits may be blistered, welted, or warty in
peach wart disease. In ring pox of apricot,
about two weeks before ripening, fruits de-
velop protuberances that tend to disappear
during the ripening process and to be sup-
planted by discolored blotches within
which cracks may develop.
Fruits may show small depressions in
which the skin is abnormally pigmented,
sometimes with superficial russet ring pat-
terns, in leaf-pucker disease of apple.
Coarse depressed rings, sometimes concen-
tric, have been noted on fruits in psorosis
B, more rarely in psorosis A, of citrus
(grapefruit). Depressed light-yellowish
longitudinal streaks appear on green fruits
of avocado affected by graft-transmissible
sun blotch disease. Water-soaked blemishes,
later appearing as scars on fruits, charac-
terize apple scar-skin. Star cracks on fruits
characterize star-crack disease of apple.
Shallow skin depressions or furrows are
underlaid by brown corky flesh in trans-
missible corky pit of the Flemish Beauty
D. ABNORMALITIES OF FRUrr HABIT
Cranberry fruits are small and erect in
cranberry false blossom disease. Fruit pedi-
cels may be short and curved in Lambert
mottle of sweet cherry.
E. ABNORMALITIES OF FRUIT COLOR
Fruit color is often defective in viral dis-
eases but occasionally is intensified or
radically changed instead.
Fruits are usually more highly colored in
peach trees affected by phony disease than
in normal trees of the same varieties. In
peach varieties that normally develop some
red color in the skin and around the pit, the
skin may become abnormally highly
colored and spotted with red and purple in
peach yellows disease; at the same time the
flesh of the fruit may become streaked with
crimson, with a pronounced red coloring
around the pit.
On the other hand, fruits often fail to
reach full color, as in tomato spotted wilt,
which induces production of yellow mot-
tling or concentric-ring patterns, sometimes
involving subepidermal necrosis that ap-
pears externally as concentric brown bands.
Light or rusty spots are reported on figs
affected by fig mosaic. Circular patches of
green skin fail to color properly at ripening
time in dapple apple disease. Apple fruits
affected by ringspot in New Zealand dis-
play patches of russeted tissue edged with
smooth dark brown bands and occasional
dark brown concentric rings. Cherry varie-
ties that normally have red or pink fruits
develop little or no red pigment in cherry
Reddish concentric-ring patterns occur
on pods in red node of bean. Pods may be
blotched with dark green, shiny, short, mal-
formed, and sometimes curled in southern
bean mosaic. They are purple or marked
with purple in pea streak. They are darker
green than normal and severely mottled as
well as short, curled, twisted, and warty in
bean pod mottle.
Green or dark brown rings occur on yel-
lowing fruits in Puerto Rican papaya mo-
saic. Chlorotic or yellow rings conspicu-
ously distinguish ripening fruits affected by
the Hawaiian papaya-"ringspot," which is
in fact an aphid-transmitted mosaic-type
F. ABNORMAU=ES OF FRUIT TEXTURE
Fruits may be woody, as in tomato
big bud. Woody fruits, with thickened peri-
carp and deficient pulp as well as
deepened color, are reported for passion-
fruit woodiness disease.
Watermelon fruits are warty with ne-
crotic spots that exude drops of a viscous
liquid when infected by tobacco ringspot
virus. Warty tissues in fruits may be hard
and bony or just leathery and tougher than
usual in peach wart disease; they are su-
perficial, with underlying tissues coarse
and containing gum pockets.
Fruits are described as dry in mild streak
of black raspberry. Fruits show drying
when foliage wilts in Pierce's disease of
grape. Hard, seedy, small fruits may be
produced in strawberry stunt.
Concentric hemispheres or spheres with
corky or watersoaked boundaries have been
reported in winter squash infected by to-
mato ringspot virus. Punky areas are re-
ported to occur in fruits in virus gummosis
of apricot. Fruits may show corky spots in
In red suture disease, peach fruits may
show coarse and stringy flesh that is ex-
ceptionally watery. Crumbly fruits, which
are also abnormally globose and irregular,
have been noted in raspberry decline.
G. ABNORMALITIES OF FRUIT
Fruits that do not shrivel and fall may
ripen prematurely in peach yellow-red vi-
rosis (X disease). Uneven ripening is re-
ported in loganberry dwarf and peach red
suture disease. Fruits seldom ripen in
strawberry green petal disease. Peaches
ripen several days to three weeks later than
usual in little peach disease.
H. ABNORMALITIES OF FRUTr FLAVOR
Cherries are said to be insipid as a result
of cherry rusty mottle disease. Peach fruits
generally have an insipid flavor in little
peach disease. Lack of flavor and of sweet-
ness characterizes fruits in little cherry dis-
ease. Fruits are poor in flavor in mild streak
of black raspberry. Tomato fruits affected
by internal browning disease are unsatis-
factory in flavor. Ripe berries have an un-
pleasant flavor in blueberry stunt disease.
Cherries at picking time are somewhat
bitter in small bitter cherry disease; they
are delayed in ripening but eventually lose
their bitterness and then taste somewhat
fermented. Fruits are bitter and unpalata-
ble, and seeds do not develop within the
pits, in peach yellow-red virosis.
I. ABNORMALITIES OF SEED CONTENT
Seed number is reduced to some extent
in virtually all viral diseases; it is rarely if
ever increased, even temporarily. Set of
seeds is almost completely suppressed in
tomato aspermy disease, because pollen
viability and egg cell viability are both
greatly lowered. Grain is rarely formed in
spring wheat, barley, and oats infected by
the virus of winter wheat mosaic. Partial
sterility of heads of infected plants is re-
ported in wheat streak mosaic. Seed forma-
tion is inhibited in peanut rosette. Pods
often contain aborted ovules in bean pod
mottle. Fruits are often seedless in grape
yellow vein disease.
J. ABNORMALITIES OF PERSISTENCE OF
The number of mature fruits may be re-
duced by abscission of fruits when young,
as in black currant reversion disease. Many
fruits shrivel and fall soon after the appear-
ance of foliage symptoms in peach yellow-
red virosis. Fruits drop prematurely in fig
mosaic. On the other hand, ripe berries
stick abnormally tightly to the stem in
blueberry stunt disease.
K. ABNORMALITIES OF YIELD
Reduction in yield of fruits is nearly uni-
versal in viral diseases, usually being slight
when foliage symptoms are mild and pro-
gressively more serious when these are
increasingly severe. It has been studied
with care in tomato infected by a variety of
strains of TMV introduced at various stages
in the growth of the host plant. Reduction
of crop is often due to a combination of re-
duced fruit size and reduced fruit number.
Retardation of tree growth reduces the
crop in cherry rasp leaf.
Reduction of yield in cowpea mosaic has
been reported as greater in some varieties
of cowpea than in others. Yield of seeds is
said to be reduced in onion yellow dwarf.
No reduction in yield of fruits has
been noted in almond calico, despite well-
defined leaf symptoms (chlorotic blotches).
L. DEATH OF PARTS OF FRUITS
Fruits may contain dead embryos in
chokecherry affected by yellow-red virosis.
Necrotic lesions are found in seeds of pea
plants that have been infected by tomato
spotted wilt virus. In early browning dis-
ease of pea, pods show flecklike and ring-
like purplish brown necrotic patterns, and
seeds in affected pods become chlorotic
and wrinkled. Necrotic spots, rings, or
streaks appear on pods in bean stipple
streak. Necrosis occurs in pineapple fruits
affected by side rot. Spots and flecks of
necrosis in internal tissues of tomato fruits
characterize tomato internal browning dis-
Chapter 2-F. 0. Holmes
ease. Necrosis of the fruit pedicel occurs in
cherry twisted leaf disease.
2-5. STEM SYMPTOMS
Stem symptoms involve abnormalities in
growth habit, internode length, number,
shape, bark characteristics, color, rigidity,
histology, physiology, chemical constitution,
and gum production, as well as premature
death of parts.
A. ABNORMALITIES OF GROWTH
HABrr OF STEMS
Yellows-type diseases tend to change the
angle of growth of branches, making new
shoots abnormally erect. After cure by heat
the normal angle of growth is resumed,
only to be lost again if reinfection occurs, as
has been demonstrated in peach yellows.
Aerial tubers are formed in potato
witches' broom, and underground tubers
are abnormally small in this disease. In pur-
ple top wilt, potato tubers sometimes are
formed in leaf axils; underground tubers,
which are often soft and spongy, may fail
to germinate, form weak sprouts, or de-
velop secondary tubers without formation
of shoots and foliage.
B. ABNORMALrES OF INTERNODE
LENGTH OF STEMS
Internode length may be much modified.
It is reduced moderately in many diseases,
as in tobacco yellow dwarf and rice dwarf.
Rosette diseases, such as peach rosette,
show extreme reduction of internode
length. Runners of strawberry are conspicu-
ously short in strawberry witches' broom.
Soon after infection newly developing
internodes are abnormally short in tomato
bunchy top, but subsequently formed inter-
nodes again lengthen to produce a spindly
type of growth. Unusually long internodes
and enlarged nodes are reported as occur-
ring in potato witches' broom.
C. ABNORMALITmES OF NUMBER OF
All yellows-type diseases tend to break
the dormancy of axillary buds and hence to
increase the number of branches in their
host plants. Abnormally numerous laterals
with extra branches are shown in ocean-
spray witches' broom. Canes are weak
and more numerous than usual in Rubus
stunt. Profuse branching is reported to
occur in winter wheat mosaic, profuse shoot
development in oat pupation disease.
Lateral branches are fewer than normal
in flowering cherry rough bark disease.
D. ABNORMALITIES OF STEM SHAPE
Upward bending of runner tip is caused
in cucurbitaceous plants by infection with
sugar-beet curly-top virus; affected stems
in this disease may be deep green. Lateral
bending of the stem tip occurs in broad
bean vascular wilt. A tendency of stems to
zigzag at the nodes has been noted in grape
fan leaf and tea phloem necrosis.
Spines are suppressed in Ceiba by infec-
tion with cacao swollen shoot virus. Thorns
are attenuated or absent in thorny varieties
of eggplant affected by little leaf disease.
Branchlets are thin and wiry in ocean-
spray witches' broom. Axillary and basal
branches are spindly in potato witches'
broom. Twigs become slender and pendu-
lous in peach willow twig disease. Tubers
are cylindrical, tapered, long, smooth, soft,
and have tender skins and conspicuous eyes
in potato spindle tuber. Loganberry dwarf
disease produces short and spindly new
Stubby stems are recorded in connection
with peach stubby twig disease. Twigs are
short and thick in Muir peach dwarf; these
twigs tend to be pale green instead of the
normal reddish brown color, presumably
because of heavy shade from abnormally
Stems are swollen or die back in areas of
recent ligneous growth in cacao swollen
shoot disease. Slightly swollen parts appear
in twigs of Napoleon sweet cherry affected
by black canker of cherry; later rough black
cankers occur at the same sites. Apical
swelling in flax occurs as a result of double
infection by aster yellows virus and the
virus of flax crinkle (oat blue dwarf virus),
but not in singly-infected plants.
Woody galls develop near the infection
site after grafting, or elsewhere later, in cit-
rus vein-enation disease, especially appear-
ing in wounded tissues. Galls develop on
stems of sweetclover plants infected by
wound-tumor virus (clover big vein virus),
especially at the sites of natural or artificial
E. ABNORMALITIES OF BARK OF STEMS
Bark is rough and split in flowering
cherry rough bark disease; it may also be
unusually deep brown. Bark is cracked in
pear stony pit. Bark splitting occurs in
apple bark-splitting disease.
Bark may be increased in thickness in
stubborn disease of citrus. It becomes
thickened and roughened, often in a dia-
mond-shaped area around a wound, in
prune diamond canker disease. Blisters and
bark splitting on the main stem characterize
blister canker disease of the Williams pear;
severe phloem necrosis in this disease may
kill susceptible scions. Blisters and concen-
tric ring cankers on bark surface produce
extremely rough bark with marked ridges
and flutes in pear bark-measles disease.
Bark is roughened by blister-like lesions
and cankers in sweet cherry necrotic rusty
Scaling of bark is characteristic of citrus
exocortis. Outer bark scales away in citrus
psorosis; gum may exude before the bark
scales away, with necrosis after scaling,
producing concavities on trunks and large
limbs or troughlike pockets in bark or wood;
the wood is sometimes gray to red in this
The smooth interface between the bark
and wood of trees is sometimes deformed
by death of parts and compensatory growth
of adjacent tissues. Minute holes on the
inner surface of bark of sour orange occur
just below the bud union in citrus tristeza,
producing the macroscopic symptom called
honeycombing; minute pegs from the
woody cylinder fit into the holes. Pegs of
wood fit into pits in thickened bark in stem-
pitting of coffee. In contrast to this, pegs of
bark fit into pits in wood in rootstocks of
citrus affected by xyloporosis. Wood-pitting
occurs in Virginia crabapple trees affected
by stem-pitting disease. Mild stem-pitting
has been noted sometimes in apple chlo-
rotic leaf spot disease. Pitting in wood and
cambial face of bark is reported for citrus
F. ABNORMALTrrES OF STEM COLOR
Bluish violet dots, spots, or stripes appear
near the bases of new canes that are af-
fected by black-raspberry streak. Brownish
or greenish ring patterns, often appearing
watersoaked, are found on canes in rose
streak. Light yellowish sunken streaks occur
on green stems in avocado sun blotch.
Browning at nodes and brownish or pur-
plish necrotic streaks in petioles and stem
are characteristic of Wisconsin pea streak.
A reddish discoloration of nodes accounts
for the name of red node of bean.
G. ABNORMAITmES OF STEM RGIDrrITY
Brittle stems characterize pea streak dis-
ease. Trunks and branches are especially
subject to breakage in windstorms in sour
Flexible stems occur in rubbery wood of
apple as a result of inhibition of lignifica-
tion in many xylem fibers and vessels.
Stems are less woody than normal in
black currant reversion disease. Affected
branches are more flexible than normal
branches of the same diameter in peach
willow twig disease.
H. ABNORMALITIES OF STEM HISTOLOGY
Vascular bundles are reddened by a col-
ored gum in sugar-cane sereh disease and
are darkened in sugar-beet mosaic. They
are larger than usual and also abnormally
numerous in anemone alloiophylly. They
show a reddish-yellow discoloration near
nodes in sugar-cane ratoon stunting disease;
a feature of this disease is failure to show
a blue-green color at the leaf scar level in
parenchyma around fibrovascular bundles
when longitudinal sections of basal nodes
of stems are flooded with 3% hydrogen
peroxide for 10-15 seconds, then blotted
dry, and flooded with concentrated hydro-
Inner phloem is hypertrophied in tomato
big bud. Stems of tomato are hollow when
affected by potato witches" broom disease.
Phloem necrosis occurs in mahaleb root-
stock carrying scions of sweet cherry af-
fected by buckskin disease. Phloem necro-
sis is contained within circular hyperplastic
areas in the primary phloem of stems in
potato rosette of Tasmania.
Chapter 2-F. 0. Holmes
I. ABNORMALITIES OF STEM PHYSIOLOGY
Raspberry canes are easily winterkilled
when affected by raspberry leaf curl. Shoots
are reddish and late to develop in spring as
a result of raspberry-decline disease.
J. ABNORMALITIES OF CHEMICAL
CONSTITUTION OF STEMS
Ray cells stain red in phloroglucinol-HCl
in exocortis disease of Poncirus trifoliata
Raf. A fluorescent substance, characteristic
of carnation mosaic, is found in extracts of
infected terminal shoots; butanol is added
to the extract, which is shaken, allowed to
settle, and made slightly alkaline; on expo-
sure to ultraviolet light fluorescence is pink.
Glutamine and glutamic acid are reported
to occur in abnormal amounts in eyes of
K. ABNORMALITIES OF GUM PRODUCTION
Gum exudes profusely through fissures in
bark in leaf roll of cherry. Gumming is pro-
fuse on branches and trunk in apricot gum-
mosis. Gum is deposited in phloem and in
rings of xylem in citrus cachexia. Gummosis
of conductive elements is said to be com-
mon in sugar-cane chlorotic streak. Gum-
mosis has been observed in the xylem, ac-
companied by excessive formation of
tyloses, in Pierce's disease of grapevine.
L. DEATH OF STEMS OR PARTS OF
Progressive die-back of branches occurs
in rose wilt. Terminal die-back is found in
apple chlorotic leaf spot disease. Die-back
of year-old shoots is reported for apple star-
crack disease. New shoots die back in
apricot gummosis. Branches die back in
cherry leaf roll.
Death of growing points permits subse-
quent growth of axillary buds in tomato
bushy stunt. Top necrosis in some field-
resistant varieties of potato is caused by
infection with potato mottle virus. Death of
growing points induces new growth from
stem buds of Vicia faba L. infected by red
clover vein mosaic. Death of terminal
shoots is caused by the tip-blight strain of
tomato spotted-wilt virus. Stem tips grow
unilaterally, curve toward the affected
sides, and eventually die, become brittle,
and break off in bud blight of soybean
caused by infection with tobacco ringspot
Stem streak occurs in southern bean
mosaic. Streaking of blue lupine stems by
pea mosaic causes bending of the growing
point toward the injured side. Stem streaks
and crookneck have been reported in to-
bacco affected by tomato spotted wilt. Dark
brown or black necrotic streaks occur on
stems, petioles, and leaf veins in bean stip-
ple streak. Stem necrosis is seen in potato
yellow dwarf. Corky, raised necrotic streaks
are found in stems and petioles of gerani-
ums affected by Pelargonium leaf curl.
Stem cankers occur occasionally in sugar-
cane mosaic. Cankers are reported around
buds in star-crack disease of apple.
Certain hybrid tea roses show necrotic
primary lesions, sometimes girdling the
canes, at the places where buds from plants
affected by rose streak are inserted; ne-
crotic secondary lesions may appear on
young lateral branches below the inserted
buds. Pea stems are sometimes girdled by
necrotic lesions in Wisconsin pea streak.
Phloem destruction below bud unions gir-
dles trees in quick decline (tristeza) of
Necrosis of cortex and pith in potato
tubers has been noted in potato aucuba
mosaic. Phloem necrosis occurs in tubers of
some potato varieties as a result of potato
leaf roll; the tubers affected by leaf roll are
inclined to be small, crisp, and few; sprouts
from them are often spindly. Tuber necro-
sis is reported in the Red La Soda potato
after infection by a strain of alfalfa mosaic.
In potato yellow dwarf, seed tubers that
germinate may show dying shoots, but
tubers often remain hard, glassy, and un-
rotted in the ground instead of germinating;
newly formed tubers are close to the stem,
few, small, and often cracked, with flesh
discolored by scattered brown flecks.
2-6. BUD SYMPTOMS
Symptoms shown by buds include ab-
normalities of dormancy, of structure, and
A. ABNORMALITIES OF DORMANCY
Yellows-type diseases stimulate axillary
buds of stems to grow prematurely; this
produces the witches'-broom type of
growth of affected plants. Tuber buds share
in this lack of dormancy in potato witches'
In green dwarf of potato, the small tubers
that are formed are slow to germinate. Lat-
eral buds fail to grow in pear stony pit.
Buds fail to grow in peach and almond
after graft transmission of bud failure dis-
ease of Drake almond.
B. ABNORMALITIES OF STRUCTURE OF
Flower buds are enlarged in tomato big
bud, and their sepals fail to separate as
C. ABNORMALITIES OF VIABILITY OF BUDS
Lateral leaf and fruit buds die and are
shed in winter in peach willow twig dis-
ease. Buds grow to a length of a few milli-
meters and then appear yellow and remain
at a standstill for several weeks in peach
yellow bud mosaic; later some of the ab-
normal buds die and others produce
rosettes of small distorted leaves, with or
without mottling. Upper buds die in late
spring in Lambert mottle of sweet cherry;
the development of lower leaf buds and
flower buds is late and irregular. In necrotic
rusty mottle of sweet cherry, part of the
buds are killed; this produces bare, rangy
2-7. ROOT SYMPTOMS
Symptoms expressed by roots include ab-
normalities of number, size, shape, texture,
color, and chemical composition; in some
cases there may also be premature death
A. ABNORMALITIES OF ROOT NUMwBE
An increased number of rootlets has been
noted in sugar-beet curly-top. Excessive
numbers of adventitious roots are reported
in connection with sugar-cane sereh disease.
Root development in cuttings of alfalfa is
reduced in alfalfa mosaic.
B. ABNORMALITIES OF ROOT SIZE
Fiji disease of sugar cane is described as
characterized by small roots that appear
bunchy. Carrot roots are reduced in growth
in carrot motley dwarf disease. Roots are
severely stunted in corn stunt disease.
C. ABNORMALITIES OF ROOT SHAPE
Numerous galls develop on roots of
Rumex and sweetclover plants after infec-
tion by wound-tumor virus, especially at
the sites of natural or artificial wounds.
Roots are reported to be abnormally swol-
len in cacao swollen shoot disease. The tap
root may be reduced to a shrivelled vascu-
lar cylinder in broad bean vascular wilt.
D. ABNOmAmUTIES OF ROOT TExTURE
Roots are said to be brittle in black locust
witches' broom disease; in this disease, also,
rootlets have been reported as branching
excessively. Roots snap instead of bending
in horseradish brittle root disease, caused
by infection with sugar-beet curly-top virus;
these brittle roots show brown to almost
black streaks when sliced longitudinally.
Dark, hard, corky spots in the flesh of
sweet potatoes appear during storage, most
conspicuously at 70-85F, in internal cork
E. ABNORMALITIES OF ROOT COLOR
Discolored wood in roots is reported in
alfalfa dwarf. Slight external browning and
phloem browning have been recorded in
abnormally small root systems that are
characteristic of tobacco yellow dwarf.
F. ABNORMATIES OF CHEnmICAL
COMPOSITION OF ROOTS
Roots of peach trees that are affected by
phony disease show flecked wood, and root
sections show red to purple spots after
treatment with methyl alcohol and hydro-
chloric acid. Root sections stain yellow by
the phloroglucinol-nitrophenolic acid tech-
nique in peach mosaic, but pink in normal
A yellow to brown discoloration of
phloem tissues, accompanied by a faint
odor of wintergreen, occurs in roots of elm
trees after infection by elm phloem necro-
Chapter 2-F. 0. Holmes
sis; affected roots soon die, causing death
G. DEATH OF ROOTS OR PARTS OF
Root decay occurs in tomato affected by
sugar-beet curly-top. Roots die in citrus
tristeza disease. Death of roots causing sub-
sequent death of plants has been reported
in Phormium yellow leaf disease. Phloem
necrosis occurs spottily in roots as well as
in stems and leaves in tea phloem necrosis.
Phloem necrosis in roots is reported also for
Our knowledge of symptoms in roots is
still scanty because of the relative paucity
of root records. This is especially true for
root systems under field conditions.
2-8. SYMPTOMS EXPRESSED BY
THE PLANT AS A WHOLE
It is difficult to separate the discussion of
symptoms that are expressed by the plant
as a whole from the discussion of those that
concern only parts of the plant. Yet there
are some features of viral diseases that logi-
cally fall under the more inclusive category.
A. ABNORMALITIES OF YIELD
Some degree of reduction in total yield
has been found in nearly all viral diseases
of plants. It may be slight in masked in-
fections, such as paracrinkle in King
Edward potato, or devastating, as in potato
yellow dwarf, potato leaf roll, and cotton
leaf curl. A reduced rate of increase has
been noted for planting stock of bulbous
iris affected by iris mosaic. Onion yellow
dwarf is characterized by production of
abnormally small bulbs. Dry matter yield is
much reduced in alfalfa mosaic. A slight
increase in tuber yield of potatoes claimed
to occur after infection by mild strains of
potato X virus requires confirmation, as
does a claim of increased height and weight
of Cleome spinosa L. after infection of roots
by tobacco necrosis virus.
B. ABNORMALITIES OF TURGIDITY
Wilting is rather a rare symptom of viral
diseases, but occasionally it is striking.
When it occurs, it often is a forerunner of
death of the entire plant.
Wilting of Tabasco pepper, but not of
most varieties of the garden pepper, is
characteristic of infection by tobacco-etch
virus. Wilting at midday is recorded for
sugar-cane chlorotic streak. Individual
leaves or whole stems wilt in bean stipple
Wilting followed by death in potato is
said to be a result of infection by parastol-
bur virus. Elm phloem necrosis is charac-
terized by wilting and sudden death in the
summer of the second year of disease, or by
gradual decline over a period of 12-18
months before death. Wilting and death
occur in peach rosette and in sudden-death
disease of the clove tree. Wilting is fol-
lowed by death in blue lupine affected by
pea mosaic. Wilting and death may occur
in yellow wilt disease of sugar beet in Ar-
gentina. Loss of turgidity throughout the
whole plant may be followed by death in
broad bean vascular wilt.
C. ABNORMALITIES OF LENGTH OF LIFE
Death ensues within a few years after
infection of cherry by leaf roll. Apricot
trees affected by virus gummosis character-
istically show gumming and necrosis in
small branches during three or four suc-
cessive years and then die. Fuller's teasel
is soon killed by teasel mosaic.
Systemic necrosis may occur in, but be
limited to, a few uninoculated parts of
plants that tend to localize mosaic-type in-
fections, and in this case length of life may
not be appreciably affected; but if necrotic
secondary lesions become very numerous,
as in very young Black Beauty eggplants
infected by TMV, the plants may collapse
and die within a few days.
Peach trees affected by peach yellows do
not stop growing as cold weather ap-
proaches but continue vegetative growth
until the tender tips of branches are frozen
and killed. Winter hardiness of Ladino clo-
ver is reduced by infection with alfalfa
Premature death of oat plants is a feature
of blue dwarf disease. On the other hand,
prolongation of life has been reported in
oats affected by red leaf disease.
China aster plants remain alive later
than normal when affected by aster yel-
lows, perhaps because this disease reduces
and finally stops seed production. Potato
plants grown in greenhouses ordinarily ma-
ture their foliage, produce tubers, and die;
if infected by witches' broom disease, how-
ever, they produce tubers but continue
vegetative growth throughout successive
D. ABNORMALITIES OF GROW RATE
Growth rate is markedly reduced in
sugar-cane chlorotic streak; this disease was
first discovered by observation of the com-
parative growth rates of heated and un-
heated cuttings, cure of the disease occur-
ring regularly in cuttings immersed in a
water bath for 20 minutes at 52C. Many
other viral diseases tend to share this char-
acteristic of producing a more or less con-
spicuously reduced growth rate. Almost
complete cessation of upward growth has
been reported in tomato bushy stunt. On
the other hand, diseased vines are said to
be sometimes abnormally vigorous vegeta-
tively in yellow vein of grape, perhaps
because their fruits are few and often
E. ABNORMALITIES OF SIZE OF PLANT
Dwarfing of the whole plant is a very
common attribute of viral diseases. It has
been noted, for example, as a characteristic
of tomato big bud, Pierce's disease of grape,
blueberry stunt, potato yellow dwarf, barley
yellow dwarf, cucurbit mosaic, and potato
leaf roll. A slight dwarfing, associated with
premature flowering, is sometimes the only
obvious symptom of chrysanthemum stunt.
Plants that grow all new parts above
ground each year actually become smaller
progressively, as has been noted for alfalfa
dwarf and for black-raspberry streak.
Dwarfing of trees occurs in flowering
cherry rough bark disease. Severe dwarfing
of the affected side of the plant, and later
of the whole plant, is characteristic of
sugar-beet yellow vein.
F. ABNOMALITIE OF TRANSPIRATION
Decreased transpiration probably occurs
in viral diseases much more often than has
been reported. It has been noted in alfalfa
dwarf. Some lowering of transpiration has
been found also in connection with the
sudden-death disease of clove trees.
Increased transpiration seems to have
been observed rarely. Tomato plants after
infection by tobacco mosaic virus are re-
ported to show a sharp drop in transpi-
ration at the time of first appearance of
symptoms; after this the transpiration rate
increases gradually; in long-infected plants,
mosaic leaves transpire at a substantially
faster rate than the healthy leaves of un-
infected control plants.
G. ABNORMALITIES OF CHEMICAL
Marked reduction of ribose nucleic acid
content has been found in Prunus mahaleb
L. seedlings doubly-infected with prune
dwarf virus and Prunus recurrent-ringspot
virus. Catalase activity is low but peroxi-
dase activity is high in barley yellow dwarf.
H. ABNORMALITIES OF HISTOLOGY
Degeneration of sieve tubes and hyper-
trophy of ray and cortical parenchymas
have been noted in rapid decline of apple.
Breakdown of tissue in and around vascular
bundles of affected foliage is reported for
artichoke curly dwarf; this structural
change is associated with decline of vigor,
sometimes followed by death of the plant.
Phloem degeneration, involving death of
phloem parenchyma and companion cells
with resultant collapse of sieve elements,
is recorded in connection with barley yel-
low dwarf. Phloem cells proliferate and the
supernumerary cells assume the character-
istics of sieve elements in sugar-beet curly-
top and aster yellows.
I. ABNORMALmTIES OF LATEX FLOW
Latex flow on wounding ceases in papaya
bunchy top; removal of the affected tops of
trees effects cure.
J. ABNORMALITIES OF GROWTH HABIT
Growth is erect, stiff, and spindly in po-
tato spindle tuber. Plants as a whole are
abnormally erect in strawberry stunt. On
the other hand, plants appear flat in straw-
berry yellow edge and western celery mo-
saic. The top of the plant has a flattened
appearance in broad bean vascular wilt and
Chapter 2-F. 0. Holmes
in lily yellow flat disease. Sugar-cane stools
are bushy in Fiji disease.
2-9. MASKING OF SYMPTOMS
All symptoms are masked at high envir-
onmental temperatures in red raspberry
mosaic. Many viral diseases share this mask-
ing effect of high temperature to a greater
or less degree at some stage of the disease
or at some season of the year. A conspicu-
ous example is pelargonium leaf curl, which
shows no symptoms in leaves produced in
summer and autumn, but reappears as a
leaf spotting disease in late winter and
spring each year unless cured by consist-
ently maintained high temperatures.
Several so-called latent viruses appear to
produce no obvious symptoms at any time
in their natural hosts. Thus carnation latent
virus is said to produce no symptoms in car-
nation and Sweet William plants, dodder
latent virus is not known to produce defi-
nite symptoms in dodder, and strawberry
latent virus is regarded as producing no re-
liable symptoms, but perhaps a slight re-
duction of vigor, in Fragaria vesca L., when
no other viruses are present. King Edward
potatoes were regarded as healthy in ap-
pearance when only stocks infected by
paracrinkle virus were available for study,
but after the variety had been freed from
virus by meristem culture it became evi-
dent that virus-free plants produced
slightly more leaves, larger leaves, and
greater dry weight than the supposedly
symptomless infected plants. Presumably
nearly all, if not all, symptomless carriers
are slightly less vigorous than their virus-
The symptoms of viral diseases are not
limited to effects that can be seen at a hasty
glance or detected by touch. If they were
so, most of the symptoms that we now con-
sider important would be of little value for
an investigator with poor eyesight, yet all
these symptoms and what we now regard
as masked symptoms would be present and
awaiting the coming of a more gifted or a
more resourceful investigator.
The study of masked infection has shown
that leaf tissues may be strikingly affected
chemically even when leaf shape and color
are not noticeably modified. In affected but
outwardly symptomless leaves, patterns of
starch distribution may be as revealing as
our more frequently recognized patterns of
chlorophyll damage. Such patterns have
been sought principally when more obvious
expressions of disease processes have not
been present to appease the curiosity of the
In diseases in which symptoms are con-
spicuous, we may predict that unnoticed
but important changes are about as fre-
quently present as in masked infections.
When these less obvious, but perhaps
highly significant, symptoms come to be
fully described in the scientific literature,
we shall have a more complete picture on
which to base our attempts to understand
viral diseases and the ever changing rela-
tionships of invading viruses to their plant
2-10. LITERATURE CITED
Bagnall, R. H., Wetter, C., and Larson, R. H.
(1958). Phytopathology 48, 391. (Abstract.)
Bawden, F. C., and Kassanis, B. (1941). Ann.
Appl. Biol. 28, 107.
Bawden, F. C., and Kassanis, B. (1945). Ann.
Appl. Biol. 32, 52.
Bawden, F. C., and Pirie, N. W. (1942). Brit. J.
Exptl. Pathol. 23, 314.
Bercks, R., and Brandes, J. (1961). Phytopathol.
Z. 42, 45.
Bos, L., and van der Want, J. P. H. (1962). Tijd-
schr. Plantenziekten 68, 368.
Broadbent, L., and Heathcote, G. D. (1958). Ann.
Appl. Biol. 46, 585.
Brunt, A. A., and Kenten, R. H. (1962). Virology
Cadman, C. H., and Harrison, B. D. (1959). Ann.
Appl. Biol. 47, 542.
Freitag, J. H., Frazier, N. W., and Flock, R. A.
(1952). Phytopathology 42, 533.
Harrison, B. D., Mowat, W. P., and Taylor, C. E.
(1961). Virology 14, 480.
Harrison, B. D., and Nixon, H. L. (1960). Virol-
ogy 12, 104.
Hollings, M. (1957). Ann. Appl. Biol. 45, 44.
Raski, D. J., and Hewitt, Wm. B. (1963). Phyto-
pathology 53, 39.
Valenta, V. (1959). Acta Virol. (Prague, English
Edition) 3, 145.
ROBERT W. FULTON
Transmission of plant viruses by grafting, dodder,
seed, and mechanical inoculation
THE PROPERTY of transmissibility is a funda-
mental characteristic of viruses as it is of
other biological agents that cause disease.
In the development of knowledge of plant
viruses, however, transmission has played a
more fundamental role than with micro-
scopically visible pathogens. For years the
transmission of a virus provided the only
experimental evidence of its existence as an
The pathologist is concerned with virus
transmissibility both from the practical
standpoint of trying to prevent or circum-
vent natural transmission and from an ex-
perimental standpoint in that most research
depends on his ability to transmit virus at
will under controlled conditions. This chap-
ter is concerned mainly with experimental
3-2. GRAFT TRANSMISSION
Grafting is the most nearly universally
applicable method of virus transmission, re-
quiring only that the virus become systemic
in plants that can be joined by grafting.
Grafting is an ancient horticultural practice
(Roberts, 1949), and it might be pertinent
here to point out that the word "inoculate"
originally referred to the insertion of an
"eye" (i.e., bud) into another plant. Baur
(1904), in one of the early classical papers
on graft transmission, points out that horti-
culturists had known for 200 years that
certain types of leaf variegation could be
transferred to healthy plants through a
graft union. Dubos (1958), in describing
"tulipomania," points out that grafting was
employed in the early fifteenth century to
induce the highly desirable "breaking" or
white streaking of solid-colored petals of
tulip. As Baur (1904) points out, the dis-
tinction of transmissibility (as opposed to
mere perpetuation in the part grafted on)
is a fundamental one, indicating an infec-
tious disease. He goes further and points
out that although his infectious variegation
appears to have no other means of spread,
it would not exist naturally unless it did
have some other means of perpetuating
One of the early and very extensive in-
vestigations of a virus disease transmitted
by grafting was that of Smith (1888) on
peach yellows. He pointed out that practi-
cal growers had known for years that
yellows was transmitted by budding.
Numerous examples could be cited of
the transmission a century and more ago of
rather poorly defined conditions by graft-
ing. The distinguishing feature of the work
of Smith and of Baur, however, was their
recognition that they were dealing with
an unusual agent of some sort. Speculation
about the nature of these agents, and to-
bacco mosaic virus, presented problems for
which no solution was then apparent. The
Chapter 3-R. W. Fulton
apparently anomalous differences between
virus and living, microscopic pathogens
have stimulated virology ever since.
Before inoculation by leaf rubbing with
Carborundum was introduced, mechanical
inoculations were often uncertain or impos-
sible. Thus, much of the early work in
virology involved grafting. This is parti-
cularly true with potato virus diseases.
Quanjer et al. (1916) pointed out that leaf
roll of potatoes could be transmitted by
stem and tuber grafts. These were used
very commonly in early work (Schultz and
Folsom, 1920, 1921; Quanjer, 1920; Atana-
soff, 1924). In 1926, Murphy and McKay
(1926) and Goss (1926) described core
grafting of potato tubers in which a plug of
tuber tissue removed with a cork borer is
inserted in a hole made with a slightly smal-
ler cork borer in another tuber. This was
more rapid and more efficient than bind-
ing the cut surfaces of two tubers together
as had been done previously.
Other types and modifications of grafts
are too numerous to describe here. Basi-
cally, a shoot (scion) is severed from its
stem and the cut is closely matched by a
reverse cut in a rooted plant (stock). The
scion is then bound to the stock so that the
cambium of the two is as closely matched
as possible. As Muzik and LaRue (1954)
point out, the presence of meristematic cells
at the graft union is of importance, rather
than cambium per se.
The scion will be subject to damaging
dehydration unless it is dormant or without
leaves. All but bud leaves are usually re-
moved from actively growing scions to pre-
vent water loss. Propagating chambers sup-
plied with intermittent mist can be used
for grafted plants until the graft union has
healed. Enclosing individual scions in
polyethylene bags is convenient and effec-
Slanting cuts of the same length in stock
and scion are often used. Increasing the cut
surface by an additional vertical cut, to
form a tongue of exposed tissue, increases
the surface for tissue union and helps pre-
vent slippage when wrapping the graft. For
virus transmission it is often convenient to
trim a shoot tip, or the nodal portion of a
young stem bearing a bud, to a wedge
shape and insert this into a cut slanting
inward in the stem of the stock plant.
Approach grafting or inarching is done
by binding cut surfaces of two plant stems
together without severing either from its
own roots. Thus, both plants are supplied
water through their own roots until the
graft union heals.
To hold tissues together firmly until they
unite and to prevent water loss, grafts must
be wrapped. For this, strips of rubber are
convenient because they disintegrate out-
of-doors in a few weeks. Grafts wrapped
with raffia, string, or cloth, and then cov-
ered with wax must be cut away to pre-
vent girdling when the stem enlarges.
Budding is a specialized form of grafting
in which the scion consists of a bud, usually
dormant, and the underlying bark. This is
inserted beneath the bark flaps resulting
from a T-shaped cut in the bark of a young
stem, and bound in place.
Finally, as these methods of mechanical
inoculation became more widely used, it
was found that although some diseases
could be transmitted in this way, others
could not, and viruses were grouped on the
basis of how they could be transmitted.
Eventually it became evident that this dis-
tinction reflected deficiencies in methods
of mechanical transmission required to
transmit less infectious viruses.
Grafting is successful in transmitting
some viruses, where other methods fail,
either because a sufficient concentration of
infective virus cannot be obtained in ex-
tracts, or because mechanical inoculation
methods do not introduce virus into tissue
in which it can multiply. Leafhopper-trans-
mitted viruses are apparently confined to
vascular tissue, or are in higher concentra-
tion in this tissue than in parenchymatous
tissue. The transmission of these viruses
thus seems to require a period sufficiently
long to allow some development of vascular
tissue at the graft union. Kunkel (1938)
pointed out that peach yellows and rosette
viruses moved across a graft union only
after 8-14 days, whereas peach mosaic virus
passed the union in 2-3 days. None of these
viruses has yet been transmitted mechani-
cally; the 2-8-day period required for trans-
mission of peach mosaic suggests that some
tissue interaction was involved rather than
simple mechanical transmission. Bennett
(1943) found similar differences in compar-
ing sugar-beet curly-top virus with tobacco
ringspot and cucumber mosaic viruses. The
curly-top virus passed a graft union in to-
bacco only after 6 days or more; the
2 mechanically transmissible viruses passed
in 2 or 3 days.
In the experiments reported by Bennett
(1943), strains of tobacco ringspot and cu-
cumber mosaic viruses were used that were
transmissible mechanically with some dif-
ficulty. These 2 viruses did not pass graft
unions in contact for 24 hours. As with
peach mosaic, certain tissue changes, per-
haps the development of plasmodesmata
between parenchymatous cells of the stock
and scion, may have occurred after 48 hours
that permitted virus transmission.
Evidence also indicates that with at least
some viruses in certain hosts a response that
might be described as an unsuccessful graft
will still result in virus transmission. The
insertion of pieces of infected leaves under
flaps of bark of healthy trees has been a
technique used successfully rather com-
monly (Sreenivasaya, 1930; Cochran and
Rue, 1944; Wallace, 1947). Wallace dem-
onstrated tissue union between the leaf
tissue and the stock. Usually this occurred
at the cut edges of veins; transmission was
improved by increasing the amount of
cut surface on the piece of leaf. J. D.
Moore (personal communication) trans-
mitted cherry necrotic ringspot virus by
placing under bark flaps of healthy cherry,
pieces of leaves, sepals, pistils, anthers, and
filaments of diseased trees. Transmission
was not obtained, however, from pollen or
Virus may also be transmitted by insert-
ing other types of tissue under bark flaps
or into stem slits. Jorgensen (1957) de-
scribed transmission of strawberry viruses
by inserting pieces of petioles with the
edges trimmed to expose vascular bundles
into slits in the petioles of healthy straw-
berries. Such transmission is not unexpected
when tissue of the same species is inserted;
a degree of tissue fusion may occur even
though no permanent union is evident and
the inserted tissue lacks the capability of
continued growth. Transmission of virus by
inserting in healthy plants tissue that would
not appear to make a union in any accepted
sense suggests that a continuous protoplas-
mic connection did not exist, and that a
form of "mechanical" transmission of virus
occurred. Blattn5r and Limberk (1956) have
described transmission of stolbur virus
(leafhopper-transmitted) by implanting
Convolvulus tissue into stems of tomato and
tobacco plants. Boyle et al. (1949) reported
transmitting necrotic ringspot virus by im-
planting infected cucumber tissue beneath
cherry bark. The author has confirmed this,
but as Boyle et al. pointed out, the rate of
transmission was so low that reliability of
the results was uncertain. Kegler (1959)
reported transmission of apple mosaic virus
without the graft partners fusing.
Disadvantages of grafting in transmitting
virus are numerous. It is time consuming,
and is often limited by compatibilities of
stock and scion. Unless woody plants are
beginning a stage of active terminal growth,
symptom expression may be delayed.
Kunkel (1930) pointed out that peach yel-
lows symptoms appear earlier if grafts are
made near the tips of rapidly growing
shoots than if made low. Also, many woody
species that have stopped growing within
the previous month or so may be stimulated
to produce a new flush of leaves by defoliat-
ing. Hildebrand (1941, 1942) pointed out
advantages of stimulating a flush of new
growth by pruning graft-inoculated plants.
Quite commonly it is pointed out that
graft transmission applies to viruses affect-
ing dicotyledons, because monocotyledon-
ous plants cannot be grafted. One can only
wonder if this statement has not been re-
peated more often than graft transmission
has been tried with monocotyledonous tis-
sue. Muzik and LaRue (1954) have de-
scribed grafts with grasses. While the lack
of a ring of cambial tissue might prohibit
permanent tissue union as it occurs in the
dicotyledons, the fact that a structurally
sound union is not always necessary for vi-
rus transmission in the dicotyledons should
encourage trials with monocotyledons.
An idea seems more or less prevalent
that graft transmissions are highly efficient
in transmitting virus. Published data indi-
Chapter 3-R. W. Fulton
cate that this is not the case. Reasons for
failure to transmit are not always clear.
Probably one of the commonest reasons is
that the virus is not completely systemic in
the plant supplying scions, and that scions
are chosen that carry no virus. Rozendaal
and van der Want (1948) point out that
potato stem mottle virus, known not to in-
vade potato completely, was not trans-
mitted by tuber grafts, and not by all stem
grafts. Davidson (1955) reported up to 48%
transmission of potato leaf roll virus by
tuber grafts. The highest percentage trans-
mission occurred when the diseased core
was within one-fourth inch of a healthy eye,
suggesting that the virus did not move
readily or rapidly from the diseased core,
or that vascular tissue was necessary for vi-
rus movement, and that this did not always
develop. Valenta (1962) has pointed out
that the rate of graft transmission of stolbur
and potato witches' broom viruses varied
with the hosts grafted, and that failure to
transmit did not necessarily indicate im-
munity of the stock plant. J. P. Fulton
(1957) and Cropley (1958) have recorded
erratic transmission of strawberry viruses
with the leaf grafting technique of Bring-
hurst and Voth (1956). Even though
tissue union occurred, some viruses were
transmitted more often than others.
Natural transmission of viruses by graft-
ing is apparently quite rare because of the
relatively rare occurrence of natural grafts.
Hunter et al. (1958) described the trans-
mission of apple mosaic virus through
naturally occurring root grafts. Thomas and
Baker (1952) described transmission of
carnation mosaic virus through root grafts
when plants were growing in the same pot
but otherwise prevented from touching.
Sheffield (1952) described transmission of
the sudden-death disease of cloves through
natural root grafts.
3-3. DODDER TRANSMISSION
OF PLANT VIRUSES
One of the drawbacks of virus transmis-
sion by grafting is the difficulty or impossi-
bility of obtaining an organic union be-
tween unrelated plants sufficient to permit
passage of virus. In part, this difficulty may
be circumvented by dodder transmission.
Kunkel (1943a) pointed out that it is sel-
dom that one host is suitable for all types of
experimental work with plant viruses. Dod-
der is one method of transferring "difficult"
viruses to new hosts.
Dodder (Cuscuta spp.) is a parasitic
vine, lacking chlorophyll and leaves, of the
family Convolvulaceae. It parasitizes higher
plants by twining around them and forming
haustoria that ultimately connect with the
vascular tissue of the host. Lackey (1953)
has pointed out that both the beet leaf-
hopper and dodder are apparently at-
tracted to the vascular bundles of sugar
beet. Dodder haustoria formed on beet
petioles opposite vascular bundles, not be-
tween them. Phloem degeneration in curly-
top-infected beet adversely affected the
accuracy of the dodder in locating the
The species of dodder commonly used in
virus transmission work will parasitize a
rather wide range of host plants. A given
species of dodder may flourish on some
hosts but be unable to maintain growth on
another host (Lackey, 1949). Different
species of dodder will parasitize each other;
Lackey (1946) has described the parasitism
of each other by C. californica and C.
A. THE MECHANISM OF DODDER
Bennett (1940a) and F. Johnson (1941a)
described the transmission of a number of
plant viruses by C. subinclusa and C.
campestris, the two species subsequently
used most widely. Many species of dodder
will transmit one or more viruses, but there
is some specificity (Table 3-1). Bennett
was able to separate cucumber mosaic virus
from tobacco mosaic virus (TMV) because
it persisted in the dodder when the dodder
was grown on hosts immune to both viruses.
TMV did not persist in the dodder, and
there was thus no evidence that it multi-
plied in the dodder. Its transmission is
evidently a passive movement from the
vascular system of one host plant to the
other. Costa (1944a) found a few cases in
which TMV was transmitted by dodder
removed from infected plants, but usually
this did not occur, and there was never
transmission after one or two transfers on
plants immune from TMV.
G. W. Cochran (1946) presented further
evidence that TMV is carried passively in
the food stream of dodder when he showed
that pruning the dodder (to remove grow-
ing points) and shading the healthy "recip-
ient" plant gave 75% transmission; un-
shaded plants did not become infected.
Cochran (1947) also described a "dodder
graft" by which a detached stem tip was
attached to an intact plant by winding
growing tips of dodder around each. The
detached stem tips were kept moist and
sprayed with indole butyric acid.
The evidence is good (Costa, 1944b;
Bennett, 1944a; Kunkel, 1943b, 1944, 1945)
that many viruses multiply in the dodder
species that will transmit them; stems of
such dodder growing on infected plants will
transmit the virus when detached and
transferred to another host plant. Bennett
(1944b) pointed out that in addition to
making direct tracheal connections with
the xylem and phloem of the host, there
were also indications of plasmodesmatal
connections between parasite and host.
It seems probable that certain species of
dodder do not transmit certain viruses be-
cause they do not infect and multiply in the
dodder. The evidence seems clear that
TMV is merely carried in the dodder food
stream, and is transmitted without infecting
the dodder. Its transmission may be aided
by the high concentration reached in many
hosts, a concentration not closely ap-
proached by many other viruses that are
not transmitted by dodder.
There are peculiarities of dodder trans-
mission that remain unexplained. Peach X-
disease virus can be transmitted by C.
campestris from peach to periwinkle, carrot,
and other hosts. It has not, however, been
transmitted from the herbaceous hosts back
to peach. Possibly this is due to some inter-
action, or lack of it, between the peach and
the dodder, since Kunkel (1945) trans-
mitted cranberry false-blossom virus from
tomato back to cranberry, a rather woody
host. Hildebrand (1945) was unable to
transmit cucumber mosaic virus to peach
with dodder, although peach has been
found infected by this virus. Canova (1955)
reported that Cuscuta epithymum failed to
transmit cucumber mosaic virus to beet but
that aphids acquired the virus by feeding
on dodder growing on diseased beets. This
would indicate that the dodder had ac-
quired the virus, but that it was unable to
A number of insects feed readily on
dodder, and Giddings (1947) pointed out
that the beet leafhopper will feed more
readily on dodder and will transmit curly-
top virus more readily from the dodder
than from the species on which the dodder
A number of mechanically transmitted
viruses have been recovered from the
pressed juice of dodder, indicating that
the viruses occur in appreciable amounts in
the dodder. A number of species of dodder,
however, contain substances inhibiting vi-
rus infection when mixed with the virus
in vitro (Miyakawa and Yoshii, 1951;
Schmelzer, 1956a). There is no evidence
that such inhibitors are involved in the
specificity of virus transmission by dodder.
B. PRACTICAL ASPECTS OF DODDER
Dodder is probably seldom, if ever, in-
volved in virus transmission in nature to
economically important plants. In annual
crops its occurrence is usually rare, except
when seed is grossly contaminated with
dodder seed. The possibility of dodder seed
carrying virus over from one annual crop to
the next has not often been considered.
Bennett (1944a) described dodder latent
virus, which causes a severe disease on
some hosts, and which is transmitted
through the seed of C. californica. Cran-
berry false blossom was not transmitted
through seed of C. campestris (Kunkel,
C. METHODS OF USING DODDER
For experimental work dodder can be
started conveniently from seed. Sowing
dodder seed with the seed of a congenial
host will result in the parasitism of the
young green seedlings after germination. A
somewhat more rapid method of obtaining
a quantity of vigorous dodder is to germi-
Chapter 3-R. W. Fulton
nate the dodder seed on filter paper. When
the dodder seedlings are an inch or two
long the rootlike "peg" can be immersed in
a very small tube of water and attached
near the top of a stem of a host large
enough and growing vigorously enough to
support rapid and abundant growth of
dodder. Growth movements of the dodder
in making initial contacts with its host are
facilitated if the "peg" is immobilized.
Dodder will flower rather abundantly
when it has produced considerable vegeta-
tive growth. Once flowering starts, vegeta-
tive growth slows markedly. Transfers from
the dodder plant after flowering starts are
likely to continue flowering without making
much vegetative growth, thus, fairly fre-
quent transfers are needed to keep dodder
vegetative. Removal of flowers as soon as
they form also may help keep dodder
3-4. SEED TRANSMISSION OF
The possibility that virus diseases might
be transmitted through the seed of infected
plants was not neglected in the earliest
work. Mayer (1886) mentions that there
was no more mosaic in plants grown from
seed of diseased tobacco than in plants
grown from seed of healthy plants. Later,
when the highly infectious nature and sys-
temic distribution of the virus were better
understood, the puzzling question was, why
are so few plant viruses transmitted through
the seeds? Although 30 or more viruses are
now known to be seed transmitted (Table
3-2), the problem of why there are not
more is still not entirely solved.
McClintock (1916, 1917) presented some
of the first evidence that plant viruses are
transmitted through seed in the case of
cucumber mosaic and lima bean mosaic
viruses. The evidence, however, was more
suggestive than conclusive. Reddick and
Stewart (1919) and Doolittle and Gilbert
(1919) presented uncontestable evidence
of seed transmission of bean mosaic and
cucumber mosaic in 1919. Seed transmis-
sion of a number of other viruses was
demonstrated in the next few years.
A. THE RATE OF Vmus TRANSMISSION
The frequency with which a seed-trans-
mitted virus appears in the seed from an
infected plant varies greatly with the virus
and with the species infected. Up to 100%
transmission has been found in seed of in-
dividual soybean plants infected with to-
bacco ringspot virus (Athow and Bancroft,
1959); lettuce plants infected with lettuce
mosaic may produce 3-10% of the seed
carrying the virus (Newhall, 1923; Couch,
1955). Seed transmission of virus at very
low rates might be overlooked or ignored
as being due to contamination from some
other source. There are, however, relatively
few reports of very low rates, suggesting
that when seed transmission does occur, the
percentage of seed carrying virus is usually
high enough to be detectable without ex-
tensive testing or elaborate precautions
Where rates of seed transmission from
individual plants have been compared,
rather wide differences have been found.
Differences in rates of transmission of bar-
ley stripe mosaic virus were associated with
different varieties (McNeal and Afanasiev,
1955; Singh et al., 1960). Varietal resistance
factors may be involved in this variation.
Clinch and Loughnane (1948) found sugar-
beet yellows seed transmitted in 47.5% of
the seed of one breeding line of beets, but
not in other lines. Nikolic (1956), however,
reported a low incidence of seed transmis-
sion of this virus in sugar-beet varieties.
Grogan and Bardin (1950) found that
lettuce mosaic virus was consistently trans-
mitted through the seed in higher percent-
ages in some varieties than in others. This
virus apparently is not seed transmitted in
the variety Cheshunt Early Giant (Kas-
sanis, 1947; Couch, 1955). Fajardo (1930)
found the percentage of transmission of
bean mosaic virus higher with late varieties
than with early varieties. Couch (1955)
found considerable variation between dif-
ferent flower heads in their rates of seed
transmission of lettuce mosaic virus. Fajardo
(1930) found that the location in the pod
of bean seed carrying bean mosaic virus
was random. Nelson (1932) found that
plants infected during the growing season
transmitted virus to fewer seeds than plants
infected the entire season.
Variation in rate of seed transmission may
also be due to the stage at which the plant
became infected. A high percentage of
transmission resulting from early infection
has been reported for barley stripe mosaic
virus (Singh et al., 1960), tobacco ringspot
virus in soybean (Athow and Bancroft,
1959), lettuce mosaic virus in lettuce
(Couch, 1955), and with several viruses by
Crowley (1959). On the other hand, Eslick
and Afanasiev (1955) reported the highest
percentage of infection with barley stripe
mosaic in barley when plants were inocu-
lated ten days before heading than from
earlier or later infections. Harrison (1935a)
reported that, with bean mosaic virus, pods
set early contained more seeds with virus
than pods set late. Some reports indicate
that plants grown from infected seed pro-
duce a higher percentage of infected seeds
than plants that contract infection during
the growing season. Other reports indicate
that this is not always the case; the differ-
ence may lie in differences in times of "cur-
rent season infection." Crowley (1959)
found that southern bean mosaic, tobacco
ringspot, and barley stripe mosaic viruses
infected the gametes of their hosts. Em-
bryos were infected in early stages of their
development, but in lower percentages, and
were not infected in later stages.
B. POLLEN TRANSMISSION
Pollen transmission of bean mosaic virus
was reported by Reddick (1931). Nelson
and Down (1933) found that the virus was
seed-transmitted in about the same per-
centage when flowers on infected plants
were pollinated with pollen from healthy
plants as when flowers on healthy plants
were pollinated with pollen from infected
plants. This was confirmed with bean mo-
saic virus by Medina and Grogan (1961)
and with elm mosaic virus by Callahan
(1957). As far as tested, most seed-trans-
mitted viruses appear to be pollen-trans-
mitted also (Way and Gilmer, 1958; Gold
et al., 1954). Thus, self-pollination will re-
sult in more infected seed than pollinations
of infected plants by healthy, or vice versa.
The presence of virus in pollen immedi-
ately raises the question of virus transmis-
sion to healthy plants by pollination. Red-
dick (1931) suggested this might occur,
and Das and Milbrath (1961) described
the infection of healthy plants (squash) by
pollination with pollen from infected
squash. Crowley (1959) pointed out earlier,
however, the notable lack of evidence of
this type of virus transmission. Thus far, the
evidence indicates that if it does occur
with seed-transmitted viruses, it must be
rare or occur with only a few of the viruses.
Elm mosaic virus, for example, is trans-
mitted by pollen to seed in fairly high per-
centages. Since elm is wind-pollinated and
sets seed copiously even a low rate of
transmission of virus from pollen to a ma-
ture tree would mean a high incidence of
C. ELIMINATION OF VIRUS FROM SEED
With most seed-transmitted viruses there
is good evidence that the virus is carried
internally. Attempts to treat the seed in
ways that should inactivate external virus
have not reduced the percentage contain-
ing demonstrable virus (Harrison, 1935b;
Athow and Bancroft, 1959). Heat treatment
of infected seed has not eliminated the vi-
rus. Muskmelon mosaic (squash mosaic?)
and bean mosaic viruses are not highly
tolerant of heat in vitro, but temperatures
far above their thermal inactivation points
have failed to eliminate the virus from seed
(Reddick and Stewart, 1919; Rader et al.,
1947; Harrison, 1935b). Well-dried seed is
surprisingly resistant to high temperature;
virus in such seed appears to tolerate as
much heat as the seed tolerates, suggesting
that it, too, is less hydrated than in sap and
thus has acquired greater resistance to heat
Most attempts to eliminate virus from
seed by heat have been done with rather
high temperatures for relatively short peri-
ods. The apparent disappearance of virus
from seed stored for a number of years
suggests that elimination of virus might be
hastened by moderate heat for longer
Valleau (1939) found a lower percentage
of tobacco seed carrying tobacco ringspot
Chapter 3-R. W. Fulton
virus after 5%-years storage. He pointed
out, however, that a differential loss of
viability might account for the results. R.
W. Fulton (unpublished) found that the
percentage of Prunus pensylvanica seed
carrying cherry necrotic ringspot virus
(60-70%) remained relatively constant the
first 4 years of storage at 20C, but by the
6th year less than 5% of the seed produced
infected seedlings. Loss of viability was
minor, indicating a true loss of virus from
the seed. Similar marked decreases in per-
centage of seed carrying muskmelon mosaic
virus after 3-years storage were reported by
Rader et al. (1947). Presumably this virus
is the same or closely related to squash
mosaic virus, which Middleton (1944)
found did not disappear from squash seed
stored 3 years. Athow and Bancroft (1959)
found no decrease in percentage of soybean
seed infected with tobacco ringspot virus
after 10-months' storage.
The chief exception to the statement that
most virus is carried internally in seed is
TMV on tomato seed. Early workers dis-
agreed on whether the virus was seed-trans-
mitted (Dickson, 1922; Gardner and Ken-
drick, 1922; Bewley and Corbett, 1930;
Doolittle and Beecher, 1937). Differences
in results may have been due to differences
in methods of extracting and cleaning seed
from infected fruit. Chamberlain and Fry
(1950) and Taylor et al. (1961) found that
most of the virus present with tomato seed
is in and on the seed coat. This could be
eliminated by acid extraction or trisodium
phosphate treatment, but not by thorough
washing in detergent solutions. A small
number of seeds had virus in the endo-
sperm; this was not affected by acid or
phosphate treatment, but was inactivated
slowly during storage. Howles (1961) re-
ported that virus associated with tomato
seed could be reduced by heat treatment.
The infection of tomato seedlings by
TMV carried with the seed occurs during
and after germination (Taylor et al., 1961),
and may occur, partly at least, because vi-
rus from the seed coat is transferred
mechanically to the seedlings when they
are handled. Taylor (1962) found that
germinating seed remained healthy when
accompanied by trash containing TMV.
Evidently the transmission depends on
fairly intimate association of virus with
parts of the seed other than the embryo.
McKinney (1952) reported transmission of
TMV in pepper seed, but it is not known
whether the virus was carried internally or
externally or whether the strain of virus or
line of pepper were determining factors.
D. LATENT VmIus IN SEED
Of considerable practical importance is
the fact that a number of seed-transmitted
viruses are of the ringspot-type; infected
seedlings are in the "recovered phase" from
the time the seed germinates, and may not
be readily detected. Valleau (1939), how-
ever, noticed that tobacco seedlings carry-
ing yellow tobacco ringspot are more
chlorotic than normal seedlings. Seedlings
of a number of species of Prunus carrying
necrotic ringspot virus may show no detect-
able symptoms. Wallace and Drake (1962)
reported that some avocado trees, symp-
tomless themselves, produce seedlings that
carry the sunblotch virus without symp-
toms. McKinney (1953) described a barley
line with a high rate of seed transmission of
barley stripe mosaic in nearly symptomless
E. CORRELATION OF SEED TRANSMISSION
wrrI OTHER Vinus CHARACTERISTICS
There has been some tendency to de-
scribe certain plant families, Leguminosae
for example, as particularly prone to seed
transmission of infecting viruses. As more
examples of seed transmission have been
investigated, however, it seems more likely
that seed transmission is a characteristic of
the virus rather than a plant family. Lister
(1960) has pointed out the general tend-
ency for soil-borne viruses also to be seed-
transmitted. Nitzany and Gerechter (1962)
found that seed transmission of barley
stripe mosaic virus occurred rather gen-
erally in many grasses.
F. MECHANISMS RESTRICTING SEED
The problem of explaining the lack of
seed infection by many plant viruses has
stimulated investigations of mechanisms
and tissues involved in seed transmission.
Duggar (1930) suggested that TMV might
be inactivated in the seed. Kausche (1940)
described virus inactivation by aqueous
extracts of tobacco seed. As Crowley (1955)
pointed out, however, virus inactivators oc-
cur commonly in many plant extracts, and
there is no evidence that any of them affect
virus in vivo. Crowley (1957a) could find
no evidence of any diffusible substance as-
sociated with developing tomato embryos
on White's medium that was capable of
reducing the infectivity of TMV in the cul-
Cheo's (1955) results with southern bean
mosaic virus, on the other hand, indicated
that this virus was present in embryos of
immature seeds, but disappeared very
rapidly as maturing seed became dehy-
drated. This loss of virus did not occur
when seed coats or leaves were dried.
Crowley (1957b) found that bean yellow
mosaic, tomato spotted wilt, and tobacco
mosaic viruses (not seed-transmitted) did
not occur in the embryos.
Thus, two mechanisms may operate to
prevent seed transmission: (1) elimination
of virus from embryos of maturing seed,
and (2) exclusion of virus from developing
seed. Too few viruses not seed-transmitted
have been investigated to know whether
they are consistently absent from embryos
and pollen. It might be assumed that both
exclusion and elimination must be operative
to prevent seed transmission in plants in-
fected before reproductive organs are
formed. There is some doubt, however, that
many viruses completely invade meriste-
matic tissue (Valleau, 1935; Bennett, 1940b;
Limasset and Cornuet, 1949; Crowley and
Hanson, 1960). Thus, virus might not reach
reproductive cells until they were suffi-
ciently differentiated to be "insulated" from
the rest of the infected plant. Sheffield
(1941) found inclusion bodies of severe
etch virus in the testa, but not in other
parts of tobacco seeds. Bennett (1940b)
pointed out that tobacco mosaic and beet
mosaic viruses, not seed-transmitted, but
easily transmitted mechanically could not
be transmitted mechanically from pollen.
It is easy enough to see, as pointed out
by Bennett and Esau (1936), how the ab-
sence of direct phloem connections between
the developing seed and the maternal par-
ent could prevent a phloem-limited virus
from entering the seed. Bennett (1940b)
suggested that the absence of plasmodes-
matal connections between the seed and
the maternal plant might prevent virus
from entering the seed. The extent to
which viruses depend on plasmodesmata to
pass from cell to cell is uncertain. Kassanis
et al. (1958) found no plasmodesmata in
cultured tobacco tumor tissue, but TMV
spread through such tissue at about the
same rate as through leaf tissue, where cells
were connected by plasmodesmata. There
must be some barrier, however, usually pre-
venting virus movement between develop-
ing seed and maternal plant. Not only do
nonseed-transmitted viruses fail to enter
developing seeds, but many viruses that are
seed-transmitted fail to pass from seed to
maternal plant when introduced into the
seed by pollen.
Bennett (1940b) suggested that plant
viruses that are not seed-transmitted are
unable to maintain themselves in gameto-
phytic tissue. Crowley (1957b) concluded
that this explanation was the only one fit-
ting all the data. The general failure of cit-
rus viruses to enter nucellar seed (Weath-
ers and Calavan, 1959) would seem to
indicate that the meristematic condition
might be the critical factor rather than the
haploid nuclear condition. Caldwell (1962)
has suggested that a smaller amount of high
energy phosphate compounds in embryos
than in leaves may be concerned with in-
ability of virus to maintain itself in embryos.
3-5. MECHANICAL INOCULATION
Such a large part of our knowledge of
plant viruses has depended on mechanical
inoculation that its importance as an experi-
mental technique cannot be overestimated.
This was emphasized in a review by Yar-
Investigations of virus outside the host
are all dependent, in the final analysis, on
the ability to demonstrate and measure the
infectiousness of the material. The methods
of doing this have improved greatly in the
past 50 years, and improvements in effi-
ciency are still being made.
Chapter 3-R. W. Fulton
Mechanical inoculation techniques have
been described as highly inefficient in terms
of numbers of infections obtained with the
numbers of virus particles applied (Steere,
1955). The nature of the plant cell, with its
cellulose cell wall, and the necessity for
introducing virus into such cells make it
seem probable that inoculation efficiency of
plant viruses will never approach the
highly efficient bacteriophage, equipped
with its own mechanisms for penetrating
the bacterial cell. Improvements in effi-
ciency of plant virus inoculation in recent
years, however, have resulted in mechani-
cal transmission of many viruses once
designated "not mechanically transmissi-
ble." A more realistic goal may be the
mechanical transmission of all plant viruses
rather than the ability to cause infection
with each virus particle.
The earliest investigators of TMV used a
variety of inoculation methods. Mayer
(1886) sucked juice of diseased plants into
glass capillaries and inserted these into
midribs of healthy plants. Beijerinck (1898)
injected juice with a syringe and inserted
pieces of dry leaf in wounds in the stem.
Both Clinton (1915) and Allard (1917)
mention that a high percentage of infection
was obtained by lightly rubbing tobacco
leaves with diseased plant extract on hands
or a brush. Schultz and Folsom (1920) in-
oculated potatoes by bruising leaves with
their fingers and then applying juice of
virus diseased potato leaves. They later
(1923) characterized the "pin-prick"
method of inoculation as useless. Fromme
et al. (1927) and McKinney (1928) pointed
out the apparent efficiency of wiping or
swabbing leaves with juice or extracts of
Despite this evidence of the efficiency of
lightly rubbing leaves with juice of an in-
fected plant, until about 1930 inoculations
were commonly made by pricking through
drops of juice into healthy leaves with fine
pins. In 1929, Holmes (1929a) published
results that greatly stimulated the study of
viruses. He demonstrated that the numbers
of necrotic lesions on certain Nicotiana
species were proportional to the dilution of
the inoculum. Such lesions had been ob-
served previously, but their quantitative
significance apparently had not been ap-
preciated. This provided a method, long
needed for quantitative measurement of vi-
rus. It also demonstrated unequivocally the
superiority of wiping inoculum on the sur-
face of leaves. The relative efficiency of
various inoculation methods could be com-
pared quickly and far more accurately than
by the use of large numbers of plants
reacting with systemic symptoms. Appar-
ently the previous evidence of the efficiency
of the leaf-wiping method had not been
widely accepted because of skepticism of
the data on its reliability. There also may
have been some tendency to equate plant
virus inoculation methods with the injec-
tion or scarification methods used in animal
virology; it did not seem quite logical that
much virus was going to be introduced
into a plant by lightly rubbing the leaves.
B. METHODS OF APPLYING INOCULUM
There have been many refinements of the
basic technique of applying inoculum to
the surface of leaves gently enough to avoid
lethal damage to the cells. Some of these
have been introduced as matters of con-
venience, some have increased the effi-
ciency or uniformity of application. Com-
monly, inoculum is applied with a pad of
cheesecloth or gauze periodically dipped
in the inoculum. This, and the necessity of
supporting the leaf during inoculation,
means that the operator's hands will be
contaminated with virus. This can be
avoided by using a fairly stiff brush as an
applicator (Allard, 1917; McKinney, 1928;
Takahashi, 1947; and Yarwood, 1952b) with
the leaf supported with some impervious
material such as waxed paper. Spatulas
made by flattening one end of a glass rod
and roughening the bearing surface with
abrasive are often used (Samuel, 1931).
These are convenient in conserving inocu-
lum or when making inoculations with
single lesions (McWhorter, 1951). Single
lesions can be ground between two spatulas,
which are then used to transfer inoculum to
plants. MacClement (1937) described an
apparatus involving a ground glass disc
mechanically rotated against the leaf sur-
face. This permitted application of a meas-
ured amount of inoculum with uniform
pressure provided by a tension spring.
Many other types of applicators have
been used. The operator's forefinger is quite
convenient, but necessitates thorough de-
contamination. McKinney (1948) noted
the relative ease with which grass leaves
can be inoculated by drawing them be-
tween the thumb and forefinger, compared
with rubbing with a cloth pad. The end of
the pestle used for grinding inoculum can
be used, mainly as a matter of convenience
rather than efficiency. Bent and twisted
pipe cleaners (Krietlow, 1961) make cheap
disposable applicators, as do "Q-tips."
The optimum amount of rubbing and
pressure to apply probably vary with the
host being inoculated. Usually, more than
one or two passes over a leaf area will
result in successively fewer lesions, prob-
ably because of increasing injury to the
cells. Ross (1953) found that continued use
of a cloth pad without redipping in inocu-
lum resulted in a rapid progressive reduc-
tion in lesion counts of potato virus Y on
The relative efficiency of rubbing meth-
ods of inoculating suggests at once that leaf
hairs are involved in virus transmission.
Boyle and McKinney (1937) found, how-
ever, that trichomes were not unusually
susceptible. When only the trichomes of a
leaf were damaged in the presence of virus,
there was considerably less infection than
when the whole leaf surface was rubbed.
Benda (1956) also found that direct
wounding of individual hair cells did not
result in a high rate of infection. Boyle and
McKinney (1937) and Yarwood (1962b)
have pointed out that some plants with few
or no trichomes are nevertheless quite sus-
ceptible to virus infection. Kontaxis and
Schlegel (1962), using CX4-labeled TMV
found that virus might be deposited in basal
septa of broken trichomes and enter unin-
jured cells via plasmodesmata.
C. SPRAYING METHODS OF APPLYING
One drawback to applying inoculum to
leaves by wiping or rubbing is that it is
somewhat laborious if large numbers of
plants are involved. A technique that has
been used with some success on large num-
bers of plants is pressure spraying of inocu-
lum. Richards and Munger (1944) used
this method for field inoculation of large
numbers of plants with bean mosaic and
cucumber mosaic viruses. Subsequently,
McKinney and Fellows (1951), Timian et
al. (1955), and Dean (1960) found the
method useful in inoculating wheat streak-
mosaic, potato virus X, and sugar-cane mo-
saic viruses to large populations of plants to
detect resistance. With the type of agricul-
tural spray apparatus used the method was
not highly effective in terms of complete
infection of every susceptible plant. It did,
however, provide a quick method for elimi-
nating a large part of the susceptible popu-
Lindner and Kirkpatrick (1959) refined
the spraying technique by using an artist's
air brush, a tool conveniently handled for
inoculating individual plants and one lend-
ing itself to close standardization of pres-
sure, fineness of spray, and distance from
leaf surface. They reported a 10-20-fold in-
crease in efficiency over the conventional
rubbing method of inoculation of tobacco
mosaic virus on cucumber cotyledons.
Optimum conditions for infection of cu-
cumber cotyledons by tobacco mosaic virus
were: 60 psi spraying pressure, a distance
of 1 cm from the spray nozzle to the
cotyledon, an inoculum delivery rate of
10 ml/min, a spraying time of 4 sec to
cover a cotyledon, and the presence of 1%
(w/v) 600-mesh Carborundum in the in-
oculum. The spraying distance was closely
regulated by fitting over the spray nozzle
a cork with a funnel-shaped opening. Dur-
ing spraying the cork was kept from con-
tacting the inoculated surface by a cushion
of air from the nozzle. Toler (1962) has
reported better infection with oat mosaic
virus using an air brush than by rubbing.
D. PICKING AND INJECTION
APPLICATION OF INOCULUM
In spite of the general convenience and
efficiency of applying inoculum by rubbing
there are a few diseases for which pin
pricking seems to be a better, or the only
method, of mechanical transmission. There
Chapter 3-R. W. Fulton
is much evidence that sugar-beet curly-top
virus is closely associated with, or limited
to, the phloem. Apparently, the virus is un-
able to multiply in epidermal tissues or is
unable to move from the epidermis. Severin
(1924) infected sugar beets with curly top
by pricking through drops of juice into the
crowns. Dana (1932) infected healthy
plants by pricks through young infected
leaves. Others have confirmed this (Ben-
nett, 1934). Fulton (1955) infected tobacco
with curly-top virus by pricking inoculum
into axillary buds. Brakke et al. (1954)
transmitted wound-tumor virus to a low
percentage of Trifolium incarnatum plants
by pricking drops of tumor extract into the
Sugar-cane mosaic virus has been trans-
mitted with various degrees of success by
hypodermic injection or by needle pricking
through drops of extract into the base of
young leaves (Brandes, 1920; Matz, 1933;
Bain, 1944; Liu, 1949). Bruner (1922)
transmitted the virus by pricking through
young infected leaves into the bases of
young healthy leaves. The virus can also be
transmitted by leaf rubbing (Matz, 1933),
but the percentage of infection is low. Bain
(1944) and Costa et al. (1952) obtained
better infection by leaf rubbing with Car-
borundum than by leaf pricking, but Bain
infected only about half the plants inocu-
lated. Liu (1949) reported 95% infection
by pin pricking. Bird (1961) obtained
higher infection with an air brush operated
at 75 psi than by pin pricking.
Hypodermic injection of infectious ex-
tracts has been used sporadically with plant
viruses, usually without marked success.
Harpaz (1959), however, reported 25%
transmission of a maize virus by hypoder-
mic injection of stalks, where leaf rubbing
with abrasives failed. Pfaeltzer (1960) in-
fected a higher percentage of cherry seed-
lings with cherry Pfeffinger virus by inject-
ing sap into stems than by rubbing leaves.
Symptoms appeared in one month after
leaf rubbing, however, compared with one
year for injection.
E. HOST SUSCEPTIBILITY; THE USE
The leaf-rubbing and local-lesion tech-
niques immediately revealed differences in
susceptibility among leaves on one plant,
and among plants (Holmes, 1929a). It
seems obvious that differences in the thick-
ness or toughness of the cuticle and the
external cell wall of the epidermis would
affect the efficiency of the rubbing method
of inoculation. Probably the most important
improvement of the leaf-rubbing method
was the introduction of Carborundum (sili-
con carbide) as an abrasive by Rawlins
and Tompkins (1934, 1936). This appar-
ently provided a means of making the
small, nonlethal wounds needed for the in-
troduction of virus. The numbers of primary
lesions with a number of viruses are in-
creased 10 to 100 times when virus is
applied to leaves lightly dusted with Car-
borundum. Rawlins and Tompkins (1936)
reported much higher percentages of infec-
tion by tomato spotted wilt, celery mosaic,
broad bean mosaic, cauliflower mosaic, and
sugar-beet mosaic viruses. These were all
difficult to transmit without Carborundum.
The use of abrasives was not new;
Harrison (1935b), Fajardo (1930), and
Fajardo and Maranon (1932) had used
fine sand in transmitting bean mosaic and
sincamas mosaic viruses. Vinson and Petre
(1931) had noted increased infection with
tobacco mosaic in the presence of charcoal.
These abrasives, however, had been used so
as to produce visible injury to the epider-
mis, in an approximation of needle scratch-
ing or pricking, and quantitative methods
for estimating increased infections were not
available. A combination of Carborundum
and gentle wiping of virus on a local-lesion
host left no doubt of the efficiency of non-
lethal wounds for inoculation.
There have been many modifications in
the use of abrasives. Kalmus and Kassanis
(1945) pointed out that Celite (diatoma-
ceous earth), charcoal, aluminum oxide
corundumm), and kaolin (aluminum sili-
cate) were all effective. Celite has an ad-
vantage in that it is light and does not
settle out as rapidly when mixed with
inoculum as Carborundum. Different
grades of charcoal had various effects, pos-
sibly due to varying adsorptive capacities.
The use of abrasives has become routine
in mechanical inoculation. Thus, the num-
ber of viruses mechanically transmissible
with abrasive, but not transmissible without
abrasive is not known. Kassanis (1949)
credits mechanical transmissibility of sugar-
beet yellows to the use of Celite or Car-
borundum, and Bawden et al. (1950) were
able to transmit mechanically potato para-
crinkle virus (long an enigma) using
abrasives. Probably many other viruses be-
long in this category.
It seems obvious that finely divided
abrasive will wound or penetrate epidermal
cells. That this is its main effect seems
probable from the fact that different abra-
sives are all relatively effective (Costa,
1944b; Kalmus and Kassanis, 1945; Beraha
and Thornberry, 1952). Particle fineness of
various grades of abrasive is expressed as
the spacing of screen mesh per inch that
will permit passage. Various reports on the
effect of abrasive particle size have indi-
cated greater efficiency for one size or an-
other in the 150-800 mesh range. As
Lindner et al. (1959) pointed out, however,
the size range of particles varies in different
lots of the same grade. Beraha et al. (1955)
presented evidence indicating that effi-
ciency was related to the numbers of par-
ticles present rather than their size. Lindner
et al. (1959) thought that infectivity was
not directly proportional to the amount of
Carborundum in the inoculum, but was
proportional to the calculated number of
particles/epidermal cell up to one par-
ticle/cell. The inference here is that one
wound/cell is optimum for infection and
that greater wounding may kill cells or
damage them so greatly that they will not
support virus multiplication.
The question of whether the effect of
Carborundum was only in wounding or
whether it carried adsorbed virus into the
cell was considered by Costa (1944b) and
Beraha et al. (1955), who found no evi-
dence that virus was adsorbed to the Car-
F. PHYSIOLOGICAL FACTORS OF
Although by breaching physical barriers
of epidermal cells to virus penetration
abrasives decrease variations in suscepti-
bility among leaves and plants, the remain-
ing variation is still great. The basis of
much of this variation is obscure; to say it
is due to physiological factors covers much
Certain species of plants seem to have an
inherent susceptibility to many different
viruses. Phaseolus vulgaris, Nicotiana taba-
cum, Cucumis sativus, Chenopodium ama-
ranticolor are representative of plants used
widely in virus research, partly because of
their general susceptibility. Many plants are
susceptible to fewer viruses or are more
difficult to infect. Tomato, for example, is
relatively difficult to infect mechanically
with tobacco ringspot virus. Undoubtedly,
many plants have an absolute immunity to
many viruses. To what extent this category
is determined by the efficiency of inocula-
tion methods, however, is not clear. Most
investigators refrain from reporting a spe-
cies "immune"; "not infected" is a much
safer category in view of the wider and
wider host ranges found as improved meth-
ods of transmission are developed.
Inherent variation in susceptibility
within species was described by Holmes
(1939) in Lycopersicon esculentum. Cer-
tain lines tended to escape infection by
TMV, and this tendency was inherited
(Holmes, 1943, 1954). The same charac-
teristic in tobacco was reported by
Schwartz and Cuzin (1951). Troutman and
Fulton (1958) showed that this characteris-
tic in tobacco was relatively nonspecific
against a number of different viruses.
G. EFFECT OF ILLUMINATION
Bawden and Roberts (1947, 1948) re-
ported that in summer reduced light inten-
sity increased susceptibility of plants to
virus. The virus content of plants was
higher under reduced light intensity. This
effect seemed not related to leaf morphol-
ogy because it could be obtained by dark-
ening the plants for 24 or 48 hours.
The effect of darkness on susceptibility
indicates that there should be diurnal vari-
ations in susceptibility. These have been
found frequently, but high susceptibility
has not always been associated with a
previous dark period. Matthews (1953a)
reported maximum susceptibility of beans
to TMV in the afternoon. This was between
Chapter 3-R. W. Fulton
March and May in New Zealand. In Madi-
son, Wisconsin, during the short days of
winter, tobacco is most susceptible to TMV
late in the afternoon. Matthews (1953b)
reported that the daily variation in suscep-
tibility persisted even when plants were
kept in a constant environment. Yarwood
(1956a, 1962a) found beans were more
susceptible to TMV in the early afternoon
than in the early morning. He also reported
that other factors interacted with light in
With viruses difficult to transmit, darken-
ing plants before inoculation may permit
transmission otherwise impossible. Costa
and Bennett (1955) and Mundry and
Rohmer (1958) have reported that pre-
inoculation darkening increased susceptibil-
ity of sugar beets to beet yellows virus.
The basis of the dark-induced suscepti-
bility is not known. The evidence indicates
that certain amount of light or length of
day is optimum for susceptibility. Darken-
ing may counteract superoptimal light.
Humphries and Kassanis (1955) attempted
to correlate dark-induced susceptibility
with changes in chemical composition of
the leaves. An increase in nitrate nitrogen
was correlated with increased susceptibil-
ity, but they concluded that the relation-
ship was indirect. Troutman and Fulton
(1958) found that the susceptibility of a
variety somewhat difficult to infect was in-
creased much more by darkening than the
susceptibility of a rather susceptible vari-
ety, indicating inherent differences among
plants in the response of susceptibility to
changes in light.
H. EFFECT OF HOST NUTRITION ON
Because the multiplication of viruses is
so dependent on the biochemical apparatus
of the plant, it is not surprising that it is
nearly an axiom that young, well-nourished,
vigorous plants are most susceptible.
Spencer (1935a) pointed out that the
amount of nitrogen supplied to beans and
tobacco affected susceptibility to TMV.
Susceptibility was highest when growth
was somewhat retarded by excess nitrogen,
indicating that factors other than host
growth were involved. Moderately high po-
tassium decreased susceptibility without
affecting plant growth (Spencer, 1935b).
With variations in phosphorus supply, sus-
ceptibility was correlated directly with
growth of the plants. Bawden and Kassanis
(1950) found that susceptibility of N. glu-
tinosa to TMV was decreased by a de-
ficiency or an excess of either nitrogen
or phosphorus. Chessin and Scott (1955)
reported that calcium deficient N. glutinosa
were less susceptible to TMV than normal
I. EFFECT OF HEAT TREATMENTS ON
Kassanis (1952) found with a number of
viruses that keeping plants at 36C for
some time before inoculation increased
their susceptibility. This was not simply an
effect of temperature stimulating virus
reproduction, since with those viruses mul-
tiplying at 36C, plants kept at 36C after
inoculation developed fewer lesions than
those kept at lower temperatures. Tobacco
necrosis, tomato bushy stunt, and cucumber
mosaic viruses apparently did not multiply
at 36C, and keeping inoculated plants at
this temperature a day or more after
inoculation prevented lesion formation al-
Yarwood (1956b) has investigated heat-
induced susceptibility with a number of
viruses. Beans were treated by dipping
them in water at various temperatures (40-
55C) for periods up to 30 sec. Greatest
increases in susceptibility were obtained
with leaves that had become somewhat
resistant to infection due to age. The in-
creased susceptibility of bean leaves was
observed for up to 3 days after treatment.
The effect of heat in increasing susceptibil-
ity was observed only with bean. Yarwood
et al. (1962) later found that the effect of
heat was not a local phenomenon. Heating
one of a pair of primary bean leaves in-
creased the susceptibility of the opposite
In some cases (Yarwood, 1958) the ap-
pearance of lesions was delayed by heating
leaves 20 sec at 50C between 6 hr and 3
days after inoculation. More and larger
lesions developed, however, and lesions ap-
peared on plants inoculated with peach
yellow bud mosaic virus, which ordinarily
did not induce lesions. The increase in
lesion numbers by treatment after inocula-
tion evidently resulted from the activation
of virus latent under ordinary conditions.
Increases in susceptibility may result from
the destruction of some material, or the
temporary inhibition of some normal proc-
ess of the plant that tends to oppose
J. OTHER FACTORS AFFECTING
There are a number of chemical and
physical factors that affect the susceptibility
of leaves to viruses. Kalmus and Kassanis
(1944) found that exposing plants to 30-
60% CO2 reduced susceptibility. Yarwood
(1954) found that immersion of bean
leaves 10 min in low concentrations of zinc
sulphate increased their susceptibility, as
did soaking them in water (Yarwood,
1959). Rubbing the lower surface of bean
leaves increased the susceptibility to virus
applied on the upper surface (Yarwood
et al., 1962).
Yarwood (1953b) found that application
of pressure to bean leaves just before or
after inoculation increased the number of
lesions produced by TMV, tobacco ringspot
virus, and tobacco necrosis virus. Panzer
(1959), however, found no effect of atmos-
pheric pressures as high as 20 psi after
inoculation of N. glutinosa with TMV or
bean with alfalfa mosaic virus.
Tinsley (1953) found that tobacco and
N. glutinosa receiving unlimited water be-
fore inoculation were more susceptible to
tomato bushy stunt virus and potato viruses
X and Y than plants receiving only enough
water to prevent wilting. The differences
decreased when plants were grown under
shade or inoculated with Celite in the
inoculum and may have been due to struc-
tural differences associated with succu-
lence. Panzer (1957) found that lowest
lesion counts were obtained on leaf disks
floated on relatively concentrated sucrose
solutions or on plants with their stems im-
mersed in these solutions. He suggested that
variations in osmotic pressure of the tissues
inoculated may be involved in variations
K. INFECTIVITY OF INOCULUM; THE
In any inoculation, it is essential that
inoculum be infective. There are many fac-
tors that may increase or decrease the
infectivity of plant viruses. The evidence
is good that many of these factors affect
the susceptibility of the host rather than
affecting the virus directly. They are
treated here, however, as inhibitors of in-
fectivity because their net effects are rather
similar to materials directly affecting virus,
and the methods of dealing with such
inhibitors usually involve methods of han-
Mechanical inoculation usually involves
virus suspended in a liquid medium, and
the nature of this medium may greatly
affect the efficiency of inoculation. Thorn-
berry (1935a) reported that dibasic phos-
phate salts in inoculum greatly increased
infectivity of TMV. This has been repeat-
edly confirmed, although not for all viruses
on all hosts (Yarwood, 1952c). The relative
effect of phosphate may be greater on one
host than another. Yarwood (1952c) has
interpreted the effect of phosphate as being
on the host. It seems more probable, how-
ever, that it affects some part of the initial
process of the entry or attachment of virus
to a susceptible site on a wounded cell.
Kahn and Schachtner (1954a,b), analyzed
the effect of phosphate by abrading leaves
dry and then applying solutions, or inocu-
lum in various sequences. Applying water
after abrasion and before applying inocu-
lum resulted in less infection than applying
buffer or applying inoculum. The effect of
phosphate in increasing infection was ap-
parent at the time of inoculation or im-
mediately after, but not when applied be-
fore inoculation. As Kahn and Libby
(1958) point out, however, more lesions
were produced if dry-abraded leaves did
not receive either water or phosphate
before inoculum was applied. Yarwood
(1962a) also found that the maximum ef-
fect of phosphate is obtained when it is
present at the time virus is applied and
the cells are wounded. Application of phos-
phate before or after application of virus
had less effect.
Chapter 3-R. W. Fulton
The hydrogen ion concentration of the
suspending medium also may markedly
affect the amount of infection obtained.
Thornberry (1935a) reported that the op-
timum for infectivity of TMV was between
pH 7 and 8.5. Yarwood (1952c) commonly
uses a solution of K2HPO4; this, unad-
justed would have a pH of about 8.5. In
general, slightly alkaline inocula are more
infectious than slightly acid inocula; infec-
tivity decreases rapidly with additional
increase in acidity.
The concentration of salts dissolved in
inoculum also affects infectivity. Thorn-
berry (1935a) found greatest infectivity
with 0.1 M phosphate, and this concen-
tration is commonly used. Certain viruses,
however, are more infectious in 0.03 M or
even lower concentrations of phosphate
than in 0.1 M (R. W. Fulton, 1957). The
leaves of some plants may be damaged by
0.1 M phosphate (or with any salt at this
concentration) which may account for
fewer lesions developing. The buffering ca-
pacity of 0.03 M phosphate is not great;
there is no assurance that pH is maintained
in inocula containing any quantity of ex-
traneous material. Some other salts also
seem to improve transmission to some ex-
tent; Yarwood (1956a) pointed out that
sodium sulphite had a beneficial effect that
could not be explained on the basis of its
antioxidant properties. Ross (1953) found
that 0.1 M borate buffer was superior to
phosphate when inoculating Physalis florid-
ana with potato virus Y. Inoculated leaves
were severely injured, however, unless they
were rinsed within one minute.
L. INnmrrORS IN PLANT EXTRACTrs
The presence of inhibitors in plant ex-
tracts has been known since Duggar and
Armstrong (1925) described the inhibition
of TMV infectivity by extracts of Phyto-
lacca decandra. These materials either
have little or no effect on the virus in vivo
or they are formed only after the tissue is
ground (Bawden, 1954). The chemical na-
ture of many of these inhibitors is un-
known; undoubtedly they are a heteroge-
neous group. Kassanis and Kleczkowski
(1948) found the inhibitor in Phytolacca
was a mucoprotein; Kuntz and Walker
(1947) described two inhibitors in spinach
juice, one which absorbed to charcoal, the
other, removed by adding CaCl2 and fil-
tering, was probably an oxalate.
Part of the evidence that the action of
one type of inhibitor is on the host rather
than on the virus is that the full effect of a
given amount of inhibitor is obtained im-
mediately on mixing with viruses; incubat-
ing mixtures does not increase the amount
of virus inactivated. Many inhibitors ap-
parently do not permanently inactivate
viruses. Dilution of inhibitor-virus mixtures
usually results in an increase in infectivity
(Stanley, 1934; Black, 1939; Ross, 1941;
Kuntz and Walker, 1947; Bawden, 1954).
The action of these inhibitors is condi-
tioned by the host inoculated. Gendron and
Kassanis (1954) found that several plant
saps inhibited infection when inoculations
were made to other species, but not when
inoculated to the species supplying the sap.
Tannins are of particular interest and
importance in the transmission of plant
viruses because of their wide occurrence
and virus-inhibiting properties. Allard
(1918) and Stanley (1935) demonstrated
inhibition of infectivity of TMV by tannic
acid. Thornberry (1935b) found that the
degree of inhibition depended on the con-
centration of tannic acid and on the time it
was in contact with the virus. Bawden and
Kleczkowski (1945) found that extracts of
strawberry leaves contained material, pre-
sumably tannin, that precipitated proteins
from the extracts.
Thornberry (1935b) and Hirth (1951)
found evidence that tannin formed a loose
complex with TMV, and that this complex
dissociated at pH 8.2-8.5. TMV presumably
is unusually resistant to denaturation by
tannin. Other viruses may be denatured; in
fact, the characterization of many viruses
as "unstable" probably reflects their tend-
ency to become rapidly and permanently
destroyed by tannins or other substances in
Numerous methods have been devised
to avoid the effects of tannins in transmit-
ting virus. Tannins form insoluble precipi-
tates with proteins; Thomrnberry (1935b)
partially restored infectivity of tannic acid-
TMV mixtures by adding gelatin. Alkaloids
form insoluble complexes with tannin.
Thung and van der Want (1951) used
nicotine sulphate precipitation to remove
tannins from extracts of raspberry leaves.
Thresh (1956) pointed out that nicotine
sulphate was only partially effective in pre-
venting precipitation of TMV by tannin in
raspberries and that by itself it reduced
infectivity of the virus. Cadman (1956),
however, has transmitted raspberry viruses
by grinding leaves with 4 ml of 40% nico-
tine sulphate, centrifuging and then dialys-
ing overnight to remove the nicotine sul-
Cadman (1959) later found nicotine base
more effective than the sulphate. He
pointed out that the degree to which in-
fection was inhibited depended on the virus
rather than the host. Inactivation could be
reversed by dilution or raising the pH to 8
with some viruses but not with others. Lis-
ter (1958) used nicotine treatment in pre-
paring antigen from infected strawberry.
Diener and Weaver (1959) have used a
0.5% solution of caffeine as an aid in im-
proving transmissibility of cherry necrotic
ringspot virus from cherry leaves to herba-
Cornuet (1952) reported transmission of
a virus of strawberry using inoculum pre-
pared by freeze-drying infected tissue, then
extracting it with alcohol to remove tan-
nins. Vaughan (1956a) described a modi-
fied Soxhlet apparatus for extracting frozen
dried tissue with alcohol or chloroform so
as to avoid hydration of the tissue or solvent
during the process. The method was effec-
tive in removing tannins, but attempts to
transmit raspberry and strawberry viruses
using the method were unsuccessful
(Vaughan, 1956b). It may be that not all
viruses will withstand the freezing and ex-
Another promising method of removing
tannins is by adsorption. Cadman (1959)
found that alumina effectively removed
inhibitors from raspberry and strawberry
leaves. Lister (1959) used extracts contain-
ing alumina in mechanically transmitting
cassava brown streak virus.
Other inhibitive substances in extracts
also may be removed with resulting in-
crease in infectivity. J. Johnson (1941)
showed with a variety of materials that
diffusing the inhibitor away from the virus
resulted in an increase in infectivity.
The beneficial effect of charcoal on in-
fectivity of TMV was noticed by Vinson
and Petre (1931), Stanley (1935), and Bald
(1937). This effect was also thought due to
the abrasive action of charcoal, but Stanley
found that filtrates after removal of char-
coal were more infectious than untreated
extracts. TMV was adsorbed by charcoal,
particularly at pH 3-5, but adsorbed virus
was still infective. Kalmus and Kassanis
(1945) found that charcoal from various
sources had widely different effects on
TMV; some lots were inhibitory. Presum-
ably inclusion of charcoal in extracts of
citrus infectious variegation virus by Grant
and Corbett (1960, 1961) is effective be-
cause it removes inhibitors. Thung and
Dijkstra (1958) and Thung and Noordam
(1959) removed virus-inhibiting material
from carnation extract by adsorption to
montmorillonite. They pointed out that this
clay mineral removed some virus also.
Less clear is the nature of action of some
otier compounds reported to stabilize virus
in extracts or to improve transmissibility.
The nature and chemical characteristics of
virus-stabilizing compounds, however, pro-
vide a key that eventually may be used in
explaining unusual instability of some vi-
ruses. Englebrecht and Regenmortel (1960)
reported that salicylic acid and "deinhib-
ited rabbit serum" prolonged infectivity of
a stone fruit ringspot virus. Kegler (1961)
stabilized Stecklenburger and Pfeffinger vi-
ruses with hydroxylamine hydrochloride.
He found increased infection when infected
Prunus leaves were homogenized with so-
dium diethyldithiocarbamate (DIECA).
Grant and Corbett (1961) and Desjar-
dins and Wallace (1962) used sucrose as
well as charcoal in the inoculum of citrus
psorosis virus. Sucrose increased infectivity
and was thought to act by preventing rup-
ture and release of enzymes from mitochon-
M. POLYPHENOLS AND ANTIOXIDANTS
A class of compounds related to tannins
and important in interfering with mechani-
cal transmission of viruses are the oxidiza-
Chapter 3-R. W. Fulton
ble phenolic compounds. These apparently
are relatively innocuous in the reduced
state, but oxidize readily when tissue ex-
tracts are exposed to air, and become
highly inhibitory. Perhaps most of the vi-
ruses characterized as highly unstable are
sensitive to this type of inactivator. Bald
and Samuel (1934) investigating tomato
spotted wilt virus found that the rate of
loss of infectivity was increased by aeration
and retarded in an atmosphere of nitrogen.
Chemical oxidizing agents increased inac-
tivation; reducing agents (sodium sulphite)
retarded inactivation. Best and Samuel
(1936) concluded that the effects of oxida-
tion were not primarily on the virus. Best
(1939) found that sodium salts of gluta-
thione, thioglycollic acid, ascorbic acid, and
potassium cyanide all had preservative ef-
fects on the virus in extracts.
Numerous viruses have since been found
that are stabilized in extracts by antioxi-
dants (Ainsworth and Ogilvie, 1939;
R. W. Fulton, 1949; Limasset, 1951; Hou-
gas, 1951; Martin, 1952; Hildebrand,
1956b). Usually the transmission of these
viruses is possible without antioxidants, lhut
such materials increase efficiency. This is
particularly true with concentrated ex-
tracts. Use of undiluted sap in an effort to
inoculate with as high a concentration of
virus as possible may be self-defeating.
R. W. Fulton (1957) showed that undiluted
extracts of cherry necrotic ringspot virus
lost most of their infectivity 90 sec after
grinding the tissue. More dilute extracts
lost infectivity more slowly, and those con-
taining antioxidants lost little or no infec-
tivity for several hours. Since chemical re-
actions are involved it might be expected
that the speed of the reaction would be
greater with more concentrated reagents.
Sulphur-containing compounds have been
used commonly as antioxidants. Other com-
pounds that combine more readily with
oxygen than do the phenols in the extract
are effective if they do not adversely affect
infectivity otherwise. Iron filings, for exam-
ple, partially stabilize tobacco streak virus
(R. W. Fulton, 1949).
Enzymes are involved in the oxidation of
phenolic compounds in plant extracts.
Hampton and Fulton (1961) found that
sodium diethyldithiocarbamate (DIECA)
stabilized infectivity of extracts of cherry
necrotic ringspot virus without the use of
antioxidants. DIECA inhibits polyphenol
oxidase activity by sequestering copper,
necessary for enzymatic activity. Cyanide
also forms complexes with copper ion, and
it is also effective in stabilizing a number
of viruses (Best, 1939; R. W. Fulton,
1949, 1957), although less so than DIECA.
Azide and citrate helped stabilize tobacco
necrosis virus (Bawden and Pirie, 1957).
Other compounds inhibiting polyphenol
oxidase by providing a competitive sub-
strate (p-nitrophenol, benzaldehydeoxime)
also stabilized infectivity of cherry necrotic
ringspot virus (Hampton and Fulton,
The intensity of the phenol-polyphenol
oxidase reaction varies with different tis-
sues and at different seasons of the year.
Hampton and Fulton (1961) found that
prune dwarf virus retained infectivity in
extracts of infected squash leaves longer in
the late spring and early fall than during
the winter. This was apparently due to less
enzyme as well as less substrate because
the addition of either to the extract has-
The polyphenol-polyphenol oxidase sys-
tem is also of interest because it appears to
be much more prominent in virus-infected
than in healthy tissue. Best (1937) found
that an enzyme catalyzing the oxidation of
several phenols was present in tomatoes
infected with tomato spotted wilt virus,
but was not demonstrable in healthy tissue.
It was present in infected tobacco, but not
in Tropaeolum majus.
Martin (1954) found that tyrosinase ac-
tivity was higher in mosaic-infected dahlias
than in healthy plants. He later (1958) re-
ported one maximum in the activity of this
enzyme 24-72 hours after inoculating to-
bacco with several viruses, and another
maximum after 10 days when symptoms
developed. Martin and Morel (1958) found
that chlorogenic acid and its derivatives-
substrates of polyphenol oxidase-were
present in greater amounts in infected tis-
sue. Similar increase in this enzyme-sub-
strate system was found by Hampton and
Fulton (1961) in squash and cucumber
infected with stone fruit viruses. Bawden
and Pirie (1957) found stronger inactiva-
tion of tobacco necrosis virus by material
sedimented from tobacco ringspot virus-
infected plants than from healthy plants.
The virus-inactivating action of oxidized
phenolic compounds apparently results in
irreversible loss of infectivity with some
viruses. Some oxidized phenolic compounds
can be reduced by proper reducing agents.
Once cherry necrotic ringspot virus had
been in contact with o-quinone (oxidized)
loss of infectivity was not reversed by re-
ducing the o-quinone to catechol, the re-
duced form, which was innocuous to the
virus (Hampton and Fulton, 1961). The
"stable" viruses are evidently not sensitive
to oxidized phenolic materials, or require
much higher amounts for their inactivation.
How many of the "unstable" viruses are
permanently inactivated is not known. The
mechanism of inactivation is rather subtle,
since serological specificity is retained
(Hampton and Fulton, 1961).
When a virus inactivator is formed after
tissue is ground, and permanently inacti-
vates virus, its formation must be pre-
vented if inoculations are to be consistently
successful. Antioxidants provide one means
of doing this. Other attempts to avoid
oxidative changes have involved making
inoculations without grinding tissue and
preparing extracts. Yarwood (1953a) de-
scribed a technique involving rubbing the
freshly cut edges of a stack of leaf disks
over the surface of leaves previously dusted
with Carborundum and sprayed with phos-
phate. Hildebrand (1956a) used the
method to transmit sweet-potato internal
cork virus. Berg (1962a, 1962b) has trans-
mitted a virus disease of poplars using a
modification of the method.
N. SOURCES OF INOCULUM
It should be obvious that tissue selected
to provide inoculum should contain a high
concentration of virus. Ordinarily this will
be young tissue, in which the virus concen-
tration is highest within a relatively few
days after infection, whether by virus intro-
duced directly, or virus entering system-
ically from another part of the plant. Tissue
with the highest virus concentration, how-
ever, may not always supply the most infec-
tious inoculum. Results (unpublished)
have indicated maximum infectivity of
cherry necrotic ringspot virus in cucumber
cotyledons 3-4 days after inoculation. After
7-10 days, virus concentration measured
serologically was high, but infectivity of
extracts of the leaves was quite low. Evi-
dently in newly infected leaves virus in-
creased more rapidly than inactivating
substances accumulated, but after a few
additional days the concentration of inacti-
vating material was sufficiently high to
depress infectivity to low levels.
Some methods for transmitting "difficult"
viruses involve selection of tissue with a
low inhibitor content for preparing inocu-
lum. Sill and Walker (1952) found that cu-
cumber corollas contained little or no in-
hibitor of cucumber mosaic virus, in
contrast to the rest of the plant. Milbrath
(1953), McWhorter (1953), and Willison
et al. (1956) have used petals as a source
of readily transmissible virus, or of virus
free of undesirable material present in leaf
extracts. Tomlinson (1955) reported that a
latent virus of cherry was more infectious
and more stable in extracts of cucumber
roots than in extracts of leaves.
Yarwood (1957a) described an inocula-
tion method based on the demonstrated
high concentrations of some viruses in the
epidermis. The surface of infected plants
was stroked with a stiff poster brush, which
was then immediately stroked on a healthy
leaf dusted with Carborundum and
sprayed with phosphate. Transmission from
nonhairy leaves with this method was usu-
Schmelzer (1957a) avoided the effect
of an inhibitor in carnation by first trans-
mitting carnation virus to Stellaria media.
Presumably this species, a member of the
same family as carnation, was less affected
by the inhibitor in carnation than other
hosts. The virus then was readily trans-
missible from S. media to tobacco.
O. TIHE MECHANISM OF INFECTION AND
FACTORS AFFECTING IT
Infection of a susceptible cell by a virus
particle is a process rather than a single
event. A number of factors appear to affect
Chapter 3-R. W. Fulton
this process, some of which will be con-
sidered here. Differentiation of these fac-
tors from those considered previously, how-
ever, is rather arbitrary. Many virus
inhibitors, for example, are more or less
effective if applied to inoculated leaves
soon after inoculation (Bawden, 1954).
Evidently the virus (or the host) remains
sensitive to these effects until a certain
stage is reached in the process of infection.
A factor in the amount of infection ob-
tained by mechanical inoculation is the
washing the leaves received after inocula-
tion. Holmes (1929b), in determining
whether the time of contact between virus
and epidermal cells was important, found
that washing leaves after inoculation in-
creased the amount of infection. This prac-
tice has been widely followed, and it is
easy to understand why it might be effec-
tive, since some plant extracts kill epider-
mal cells if allowed to dry on leaves.
Yarwood (1952a) reported, however, that
leaves to which dry virus was applied de-
veloped fewer lesions if they were washed
after inoculation than when they were
left unwashed. Leaves dried rapidly after
inoculation developed more lesions than
leaves dried slowly. When abrasion and
application of inoculum were done as suc-
cessive steps, abrading the leaves wet, then
applying inoculum resulted in fewer lesions
than dry abrasion before inoculation. In a
later report Yarwood (1955a) found that
washing was not detrimental if the period
of washing was 10 sec or less. The deleteri-
ous effect of washing could be demon-
strated up to 3 hours after inoculation at
31C and 9 hours at 20C. When inoculum
contained phosphate, washing always de-
creased infection. When inoculum was pre-
pared with water, a brief washing often
increased infection, particularly when the
wash water contained phosphate. Kahn and
Schachtner (1954a) found that washing
cowpea leaves after inoculation decreased
infection regardless of the presence or ab-
sence of phosphate in the inoculum.
These results, in general, have been con-
firmed by other workers (Allington and
Laird, 1954b; Dale, 1956). Some workers
have reported negligible effects from wash-
ing (Crowley, 1954); others have found
sharp reduction in infection after washing
(R. W. Fulton, 1957). Yarwood (1955a)
believed that the effect of water was to
dilute or remove from the tissue certain
ions that were necessary for infection.
Statistical evidence (Lauffer and Price,
1945) indicates that each infection by a
virus results from the action of a single
infectious particle. With a few viruses, how-
ever, there is evidence that two or more
particles are required at an infection site
for successful infection (R. W. Fulton,
1962). This characteristic makes the
amount of virus of great importance in
transmission if the concentration in infected
tissue were lower than that necessary to
provide the required number of particles
at an infection site.
Hildebrand (1943) described the extru-
sion of a droplet of cytoplasm from a
wounded cell and its rapid withdrawal
back into the cell. Benda (1956) described
this extrusion and retraction when a Nico-
tiana hair cell was punctured through a
droplet of inoculum. D. J. Rossouw and the
author (unpublished) readily observed the
retraction of the protoplasmic droplet when
a cell in air was punctured. When im-
mersed in liquid, however, protoplasmic
granules were extruded from the wound,
but there was no evidence of a return of
any material into the cell regardless of the
composition or concentration of the sur-
How a cell behaves when wounded by
rubbing with an abrasive in the presence
of liquid is not known. It may be significant,
however, that application of dry inoculum
is more effective than inoculum suspended
in water or buffer (Yarwood, 1952a; Kal-
mus and Kassanis, 1945; Kahn and Libby,
1958). Sheffield (1936) found that it was
not necessary (although it was more effi-
cient) to wound a cell in the presence of
virus to obtain infection. The loss of
susceptibility, however, was rapid after
wounding. Allington and Laird (1954b)
found that after dry abrasion plants that
had received a low potassium nutrient
supply were more susceptible to TMV than
those receiving normal potassium. In fact,
the susceptibility of some potassium-defi-
cient plants increased with time for 60 sec
or more. Later Jedlinski (1956) found that
a number of plants retained susceptibility
after dry abrasions for periods up to 10
min after wounding. As the authors of both
these papers point out, it is apparent that
more is involved in infection than the mere
introduction of virus into susceptible cells.
There seem to be different degrees of prob-
ability of successful infection of a cell,
conditioned by events preceding inocula-
tions as well as after inoculation. While the
rapid retraction of a droplet of cytoplasm
into a wounded cell can be envisioned as
carrying virus with it, it is difficult to re-
concile this view with retention of high
infectability of dry abraded cells for 5-10
The effects of humidity are probably in-
volved indirectly in transmission by deter-
mining the length of time a leaf remains
wet after inoculation. Howles (1947), Yar-
wood (1952a), Panzer (1959), and Lind-
ner et al. (1959) obtained more infection
on plants placed at low humidity after in-
oculation than on plants at high humidi-
ties. Yarwood (1952a) found that some-
what wilted leaves were more susceptible
than fully turgid ones.
P. RIBONUCLEASE AND Vmius
After Schramm et al. (1955) demon-
strated the infectivity of the nucleic acid
portion of TMV there was some basis for
supposing that some plant viruses might
exist only as "naked" nucleic acid, and that
mechanical transmission of these might be
prevented by ribonuclease in leaf extracts.
Babos and Kassanis (1962) described a
variant of tobacco necrosis virus more
readily transmissible in water-saturated
phenol extracts of infected leaves or in 0.5
M borax, than in sap. They suggested that
the variant usually existed largely as nu-
cleic acid. Brandenburg (1962) reported
mechanically transmitting potato leaf roll
virus, using phenol extracts of diseased
leaves (ribonuclease being removed by the
phenol treatment). Govier (1963), how-
ever, was unable to repeat this.
Gordon and Smith (1960) reported in-
fecting Rhoeo discolor by TMV nucleic acid,
whereas the plant could not be infected by
intact virus. It was soon found, however,
that the plant was also susceptible to intact
TMV, but that the virus increased only at
high illumination (Gordon and Smith,
1961). Bawden (1961) pointed out that vi-
rus increase after inoculation with intact
virus was less readily detected than with
nucleic acid inoculation because of residual
virus on the leaves.
There is little evidence in the literature
that many attempts have been made to
transmit "nonmechanically transmissible vi-
ruses" by phenol extracts or extracts pre-
pared with bentonite to check inhibitive ef-
fects of ribonuclease in leaf extracts. While
the hypothesis is attractive, there is very
little evidence indicating that many plant
viruses exist in the plant solely as "naked"
Increasing the efficiency of mechanical
inoculation may be desirable to reduce
variability in quantitative measurement of
infectivity, and to permit transmission of vi-
ruses from hosts in which their study is
difficult to hosts better adapted for experi-
mental work. Viruses present problems in
woody hosts, for example, that are avoided
in herbaceous hosts. Woody plants are sub-
ject to periods of dormancy; cultivated
woody plants are usually clones, which
necessitates clonal propagation of experi-
mental material; most woody plants are not
only difficult to transmit virus from, they
are also difficult to infect by mechanical
inoculation. Annual plants have obvious
advantages as experimental hosts, particu-
larly if the seed is reasonably uniform
genetically or can be made so by selling.
There seems to be no method of predict-
ing where susceptible plants may be found
for unknown viruses. Certain families
(Leguminosae, Cucurbitaceae, Solanaceae,
Chenopodiaceae) contain species that are
susceptible to a wide range of viruses. Pos-
sibly suitable hosts for an unknown virus
are more common in these families than in
others; or these families may only appear
as susceptible to a wide range of viruses
because they have been tested more often.
The author has gained the impression in
Chapter 3-R. W. Fulton
his own work as well as in reviewing the of technique, each relatively minor. It is
literature that no one factor in transmission to be hoped that what appear as relatively
techniques can be credited with permitting minor improvements in technique will be
mechanically transmission of "difficult" vi- documented as far as possible with quanti-
ruses. Success may depend on many factors tative data.
TABLE 3-1. VIRUSES REPORTED AS TRANSMITTED BY VAmOUS SPECIES OF Cuscuta
Species of Diseases, the viruses of which have Diseases, the viruses of which have
Cuscuta been transmitted been reported as not transmitted
Cucumber mosaic (Schmelzer, 1956b)
Ageratum yellow veinbanding, crotalaria
witches' broom (Thung and Hadiwid-
Tobacco etch, sugar-beet curly-top, cu-
cumber mosaic, tomato spotted wilt
(Bennett, 1944b), dodder latent (Ben-
Tobacco mosaic, aster yellows, tomato
bushy stunt, cucurbit mosaic (Johnson,
1941a,b), sugar-beet rosette (Bennett
and Duffus, 1957), lilac witches' broom
(Brierly, 1955), chrysanthemum flower
distortion (Brierly and Smith, 1957),
white clover mosaic (Bos et al., 1959),
clover phyllody (Frazier and Posnette,
1957), potato stem mottle (tobacco rat-
tle) (Schmelzer, 1955), tomato big bud
(Hill and Mandryk, 1954), tomato spot-
ted wilt (Bennett, 1944b), potato
witches' broom (Kunkel, 1943b), peach
X-disease (Kunkel, 1944), cranberry
false blossom (Kunkel, 1945), Vinca
yellows (Maramorosch, 1956), tomato
stolbur (Misiga and Valenta, 1957),
cucumber mosaic (Bennett, 1940b),
alfalfa mosaic (Schmelzer, 1956b), Cen-
trosema mosaic (Van Velsen and Crow-
ley, 1961), Tulare apple mosaic (Yar-
wood, 1955b), dodder latent mosaic
(Bennett, 1944a), sugar-beet yellow
wilt (Bennett and Munck, 1946), Ascle-
pias yellows (Kunkel, 1950)
Potato witches' broom (Fukushi and Shi-
Tomato stolbur (Misiga and Valenta,
1957), alfalfa mosaic (Schmelzer,
Tobacco etch, bean yellow mosaic, po-
tato Y, and potato bouquet (Schmel-
Bean yellow mosaic, potato Y, potato
bouquet, alfalfa mosaic, tobacco etch
(Schmelzer, 1956b), tobacco mosaic,
sugar-beet mosaic, sugar-beet yellow
vein, tomato ringspot, peach mosaic
(Bennett, 1944b), sugar-beet rosette
(Bennett and Duffus, 1957), tobacco
rattle (Schmelzer, 1955), tomato
aspermy (Schmelzer, 1957b)
Tobacco etch, sugar-beet mosaic, sugar-
beet yellow vein, tomato ringspot, cit-
rus psorosis, peach mosaic (Bennett,
1944b), lettuce bigvein (Campbell
and Grogan, 1963), tobacco yellow
dwarf (Hill and Mandryk, 1954), to-
bacco ringspot (Johnson, 1941a), pea
wilt virus (Johnson, 1942), bean yel-
low mosaic, potato Y, potato bouquet
(Schmelzer, 1956b), tomato aspermy
(Schmelzer, 1957b), potato leaf roll
Tobacco etch, bean yellow mosaic, po-
tato Y, potato bouquet (Schmelzer,
Species of Diseases, the viruses of which have Diseases, the viruses of which have
Cuscuta been transmitted been reported as not transmitted
Cucumber mosaic (Schmelzer, 1956b)
Beet yellows, beet yellow net (Ben-
nett, 1944b), alfalfa mosaic, bean yel-
low mosaic, potato Y, potato bouquet,
tobacco rattle, tobacco etch (Schmel-
zer, 1956b), tomato aspermy (Schmel-
Tobacco etch, bean yellow mosaic, po-
tato Y, potato bouquet (Schmelzer,
Lettuce big-vein (Campbell and Gro-
gan, 1963), tobacco etch, bean yellow
mosaic, potato Y, potato bouquet
Bean yellow mosaic, potato Y, potato
bouquet (Schmelzer, 1956b)
C. repens Cucumber mosaic (Hildebrand, 1945)
C. sandwichiana Cucumber mosaic (Sakimura, 1947)
C. subinclusa Cucumber mosaic (Bennett, 1940b), to-
bacco mosaic (Bennett, 1940b), lilac
witches' broom (Brierly, 1955), straw-
berry veinbanding (Frazier, 1955),
strawberry green petal (Frazier and
Posnette, 1956), blueberry stunt (Hutch-
inson et al., 1960), potato stem mottle
(Schmelzer, 1955), alfalfa mosaic
(Schmelzer, 1956b), tomato aspermy
(Schmelzer, 1957b), potato leaf roll
(Williams, 1957), Tulare apple mosaic
(Yarwood, 1955b), dodder latent mo-
saic (Bennett, 1944a)
C. trifolii Tomato stolbur (Misiga and Valenta,
Tomato spotted wilt (Sakimura, 1947)
Tobacco etch, sugar-beet mosaic, citrus
psorosis, tomato ringspot, sugar-beet
yellow vein, peach mosaic (Bennett,
1944b), sugar-beet rosette (Bennett
and Duffus, 1957), bean yellow
mosaic, potato Y, potato bouquet
(Schmelzer, 1956b), sugar-beet yel-
low wilt (Bennett and Munck, 1946)
Cucumber mosaic, alfalfa mosaic, potato
stem mottle (Schmelzer, 1956b), to-
mato aspermy (Schmelzer, 1957b)
Beet yellows (Beiss, 1956), clover phyl-
lody (Chiykowski, 1962), cucumber
mosaic, potato stem mottle (Schmelzer,
Potato witches' broom (Fukushi and Shi-
kata, 1955), tobacco mosaic (Miyakawa
and Yoshii, 1951)
Cucumber mosaic (poorly), alfalfa mo-
saic, potato stem mottle, tobacco etch
Tobacco mosaic (Gualaccini, 1955)
A mosaic of Rubus (Azad and Sehgal,
Chapter 3-R. W. Fulton
TABLE 3-2. SEED TRANSMrrrED VmUSES
Virus of: Host Citation
Barley stripe mosaic
Bean red node
Dandelion yellow mosaic
Datura "Q" virus
Dodder latent mosaic
Peach necrotic leafspot
Peanut marginal chloro-
Prune dwarf virus
Sour cherry necrotic
ringspot (peach ring-
Sour cherry yellows
Southern bean mosaic
Abutilon thompsoni X
Wallace and Drake
McNeal and Afanasiev
Reddick and Stewart
Thomas and Graham
Doolittle and Gilbert
Salmon and Ware
Rader et al. (1947)
Wagnon et al. (1960)
Van Velsen (1961)
Gilmer and Way (1960)
Cochran, L. C. (1946)
Cation (1949, 1952)
Das and Milbrath
Cation (1949, 1952)
Shepherd and Fulton
Bennett and Costa
Virus of: Host Citation
Squash mosaic Cucurbita maxima Middleton (1944)
C. moschata Grogan et al. (1959)
Sugar-beet yellows Beta culgaris Clinch and Loughnane
Tobacco mosaic virus Lycopersicon esculentum Bewley and Corbett
Tobacco ringspot Glycine max Desjardins et al. (1954)
Petuna hybrida Henderson (1931)
Nicotiana tabacum Valleau (1932)
Cucumis melo McLean (1962)
Tomato black ring Glycine max Lister (1960)
Tomato bunchy top Solanum incanum McClean (1948)
Tomato ringspot Glycine max Kahn (1956)
Western bean mosaic Phaseolus vulgaris Skotland and Burke
Xyloporosis Citrus spp. Childs (1956)
3-8. LITERATURE CITED
Ainsworth, G. C., and Ogilvie, L. (1939). Ann.
Appl. Biol. 26, 279.
Allard, H. A. (1917). J. Agr. Res. 10, 615.
Allard, H. A. (1918). J. Agr. Res. 13, 619.
Allington, W. B., and Laird, E. F., Jr. (1954a).
Phytopathology 44, 297.
Allington, W. B., and Laird, E. F., Jr. (1954b).
Phytopathology 44, 546.
Anderson, C. W. (1957). (Abstr.) Phytopathol-
ogy 47, 515.
Atanasoff, D. (1924). Phytopathology 14, 521.
Athow, K. L., and Bancroft, J. B. (1959). Phyto-
pathology 49, 697.
Azad, R. N., and Sehgal, 0. P. (1958). Indian
Phytopathology 11, 159.
Babos, P., and Kassanis, B. (1962). Virology 18,
Bain, D. C. (1944). Phytopathology 34, 844.
Bald, J. G. (1937). Ann. Appl. Biol. 24, 77.
Bald, J. G., and Samuel, G. (1934). Ann. Appl.
Biol. 21, 179.
Baur, E. (1904). Ber. Deutschen Bot. Gesellschaft
Bawden, F. C. (1954). Advan. Virus Res. 2, 31.
Bawden, F. C. (1961). J. Biol. Chem. 236, 2760.
Bawden, F. C., and Kassanis, B. (1950). Ann.
Appl. Biol. 37, 46.
Bawden, F. C., and Kleczkowski, A. (1945). J.
Pomol. Hort. Sci. 21, 2.
Bawden, F. C., and Pirie, N. W. (1957). J. Gen.
Microbiol. 16, 696.
Bawden, F. C., and Roberts, F. M. (1947). Ann.
Appl. Biol. 34, 286.
Bawden, F. C., and Roberts, F. M. (1948). Ann.
Appl. Biol. 35, 418.
Bawden, F. C., Kassanis, B., and Nixon, H. L.
(1950). J. Gen. Microbiol. 4,210.
Beijerinck, M. W. (1898). Verhandel. Koninink.
Ned. Akad. Wettensch. (Amsterdam) 65 (2), 3.
Beiss, U. (1956). Phytopathol. Z. 27, 83.
Benda, G. T. A. (1956). Virology 2, 820.
Bennett, C. W. (1934). J. Agr. Res. 48, 665.
Bennett, C. W. (1940a). (Abstr.) Phytopathology
Bennett, C. W. (1940b). Botan. Review 6, 427.
Bennett, C. W. (1943). Phytopathology 33, 818.
Chapter 3-R. W. Fulton
Bennett, C. W. (1944a). Phytopathology 34, 77.
Bennett, C. W. (1944b). Phytopathology 34, 905.
Bennett, C. W., and Costa, A. S. (1961). Phyto-
pathology 51, 546.
Bennett, C. W., and Duffus, J. E. (1957). Plant
Disease Reptr. 41, 1001.
Bennett, C. W., and Esau, K. (1936). J. Agr. Res.
Bennett, C. W., and Munck, C. (1946). J. Agr.
Res. 73, 45.
Beraha, L., and Thornberry, H. H. (1952).
(Abstr.) Phytopathology 42, 2.
Beraha, L., Varzandeh, M., and Thornberry, H. H.
(1955). Virology 1, 141.
Berg, T. M. (1962a). Nature 194, 1302.
Berg, T. M. (1962b). Tijdschr. Plantenziekten 68,
Best, R. J. (1937). Australian J. Exptl. Biol. Med.
Sci. 15, 191.
Best, R. J. (1939). Australian J. Exptl. Biol. Med.
Sci. 17, 1.
Best, R. J., and Samuel, G. (1936). Ann. Appl.
Biol. 23, 759.
Bewley, W. F., and Corbett, W. (1930'. Ann.
Appl. Biol. 17, 260.
Bird, J. (1961). J. Agr. Univ. Puerto Rico 45, 1.
Black, L. M. (1939). Phytopathology 29,321.
Blakeslee, A. F. (1921). J. Genetics 11, 17.
Blattny, C., and Limberk, J. (1956). Preslia 28,
Bos, L., Delevi6, B., and van der Want, J. P. H.
(1959). Tijdschr. Plantenziekten 65, 89.
Boyle, J. S., Moore, J. D., and Keitt, G. W.
(1949). (Abstr.) Phytopathology 39, 3.
Boyle, L. W., and McKinney, H. H. (1937). Sci-
ence 85, 458.
Brakke, M. K., Vatter, A. E., and Black, L. M.
(1954). Brookhaven Symp. Biol. 6, 137.
Brandenburg, E. (1962). Phytopathol. Z. 43, 420.
Brandes, E. W. (1920). J. Agr. Res. 19, 131.
Brierly, P. (1955). Plant Disease Reptr. 39, 719.
Brierly, P., and Smith, F. F. (1957). Phytopathol-
ogy 47, 448.
Bringhurst, R. S., and Voth, V. (1956). Plant Dis-
ease Reptr. 40, 596.
Bruner, S. C. (1922). Rev. Agr. Com. y Trab.
(Cuba) 5, 11.
Cadman, C. H. (1956). J. Hort. Sci. 31, 111.
Cadman, C. H. (1959). J. Gen. Microbiol. 20, 113.
Caldwell, J. (1962). Nature 193, 457.
Callahan, K. L. (1957). (Abstr.) Phytopathology
Campbell, R. N., and Grogan, R. G. (1963). Phy-
topathology 53, 252.
Canova, A. (1955). Ann. Sper. Agrar. (Rome) 9,
Cation, D. (1949). Phytopathology 39, 37.
Cation, D. (1952). Phytopathology 42,4.
Chamberlain, E. E., and Fry, P. R. (1950). New
Zealand J. Sci. Technol. (A) 32, 19.
Cheo, P. C. (1955). Phytopathology 45, 17.
Chessin, M., and Scott, H. A. (1955). Phytopathol-
Childs, J. F. L. (1956). Plant Disease Reptr. 40,
Chiykowski, L. N. (1962). Can. J. Botany 40, 397.
Clinch, Phyllis E. M., and Loughnane, J. B.
(1948). Sci. Proc. Roy. Dublin Soc. (N.S.) 24,
Clinton, G. P. (1915). Conn. Agr. Expt. Sta. Ann.
Rept. 1914, 357.
Cochran, G. W. (1946). Phytopathology 36, 396.
Cochran, G. W. (1947). Phytopathology 37,5.
Cochran, L. C. (1946). Science 104, 269.
Cochran, L. C., and Rue, J. L. (1944). (Abstr.)
Phytopathology 34, 934.
Cornuet, M. (1952). Compt. Rend. Acad. Sci.
(Paris) 235, 171.
Costa, A. S. (1944a). Phytopathology 34, 151.
Costa, A. S. (1944b). Phytopathology 34, 288.
Costa, A. S., and Bennett, C. W. (1955). Phyto-
pathology 45, 233.
Costa, A. S., de Aguirre, J. M., Segalla, A. L., and
Alvarez, R. (1952). Bragantia 12, 7.
Couch, H. B. (1955). Phytopathology 45, 63.
Cropley, R. (1958). Ann. Rept. East Mailing Res.
Sta. (Kent) 1957, 124.
Crowley, N. C. (1954). Australian J. Biol. Sci. 7,
Crowley, N. C. (1955). Australian J. Biol. Sci. 8,
Crowley, N. C. (1957a). Australian J. Biol. Sci. 10,
Crowley, N. C. (1957b). Australian J. Biol. Sci. 10,
Crowley, N. C. (1959). Virology 8, 116.
Crowley, N. C., and Hanson, J. B. (1960). Virol-
ogy 12, 603.
Dale, J. L. (1956). Dissertation Abstr. 16, 1566.
Dana, B. F. (1932). (Abstr.) Phytopathology 22,
Das, C. R., and Milbrath, J. A. (1961). Phyto-
pathology 51, 489.
Das, C. R., Milbrath, J. A., and Swenson, K. G.
(1961). (Abstr.) Phytopathology 51, 64.
Davidson, T. R. (1955). Can. J. Agr. Sci. 35, 238.
Dean, J. L. (1960). Plant Disease Reptr. 44, 874.
Desjardins, P. R., and Wallace, J. M. (1962). Plant
Disease Reptr. 46, 414.
Desjardins, P. R., Latterell, R. L., and Mitchell,
J. E. (1954). Phytopathology 44, 86.
Dickson, B. T. (1922). Macdonald College J. Ser.
Tech. Bull. No. 2.
Diener, T. 0., and Weaver, M. L. (1959). Phyto-
pathology 49, 321.
Doolittle, S. P. (1920). U. S. Dept. Agr. Bull. 879.
Doolittle, S. P., and Beecher, F. S. (1937). Phyto-
pathology 27, 800.
Doolittle, S. P., and Gilbert, W. W. (1919). Phyto-
pathology 9, 326.
Dubos, R. J. (1958). In M. Pollard (ed.) Perspec-
tives in Virology. 291-99. Wiley and Sons, Phila-
Duggar, B. M. (1930). (Abstr.) Phytopathology
Duggar, B. M., and Armstrong, J. K. (1925). Ann.
Missouri Botan. Garden 12, 359.
Engelbrecht, D. J., and Regenmortel, M. H. V.
(1960). S. African J. Agr. Sci. 3, 617.
Eslick, R. F., and Afanasiev, M. M. (1955). Plant
Disease Reptr. 39, 722.
Fajardo, T. G. (1930). Phytopathology 20, 469.
Fajardo, T. G., and Marafion, J. (1932). Philip-
pine J. Sci. 48, 129.
Frazier, N. W. (1955). Phytopathology 45, 307.
Frazier, N. W., and Posnette, A. F. (1956). Na-
ture 177, 1040.
Frazier, N. W., and Posnette, A. F. (1957). Ann.
Appl. Biol. 45, 580.
Fromme, F. D., Wingard, S. A., and Priode, C. N.
(1927). Phytopathology 17, 321.
Fukushi, T., and Shikata, E. (1955). Mem. Fac.
Agr. Hokkaido Univ. 2, 47.
Fulton, J. P. (1957). (Abstr.) Phytopathology 47,
Fulton, R. W. (1949). Phytopathology 39, 231.
Fulton, R. W. (1955). Plant Disease Reptr. 39,
Fulton, R. W. (1957). Phytopathology 47, 683.
Fulton, R. W. (1962). Virology 18, 477.
Gardner, M. W., and Kendrick, J. B. (1921). J.
Agr. Res. 22, 111.
Gardner, M. W., and Kendrick, J. B. (1922). Bo-
tan. Gaz. 68, 469.
Gehring, F. (1957). Nachrbl. Deut. Pflanzen-
schutzdienst (Braunschweig) Stuttgart 9, 172.
Gendron, Y., and Kassanis, B. (1954). Ann. Appl.
Biol. 41, 183.
Giddings, N. J. (1947). Phytopathology 37, 278.
Gilmer, R. M. (1955). Plant Disease Reptr. 39,
Gilmer, R. M., and Way, R. D. (1960). Phyto-
pathology 50, 624.
Gold, A. H., Suneson, C. A., Houston, B. R., and
Oswald, J. W. (1954). Phytopatholoorv 44, 115.
Gordon, M. P., and Smith, C. (1960). J. Biol.
Chem. 235, PC28.
Gordon, M. P., and Smith, C. (1961). J. Biol.
Chem. 236, 2762.
Coss, R. W. (1926). Phytopathology 16, 233.
Govier, D. A. (1963). Virology 19, 561.
Grant, T. J., and Corbett, M. K. (1960). Nature
Grant, T. J., and Corbett, M. K. (1961). In W. C.
Price (ed.), Proc. Second Conf. Int. Org. Citrus
Virologists. 197. Univ. of Florida Press.
Grogan, R. G., and Bardin, R. (1950). (Abstr.)
Phytopathology 40, 965.
Grogan, R. G., Hall, D. H., and Kimble, K. A.
(1959). Phytopathology 49, 366.
Gualaccini, F. (1955). Boll. Staz. Pat. Veg.
(Rome) 12, 137.
Hampton, R. E., and Fulton, R. W. (1961). Virol-
ogy 13, 44.
Harpaz, I. (1959). Nature 184, B.A. 77-B.A. 78.
Harrison, A. L. (1935a). N.Y. State Agr. Expt. Sta.
(Geneva) Tech. Bull. 235, 48.
Harrison, A. L. (1935b). N.Y. State Agr. Expt. Sta.
Helms, Katie. (1962). Australian J. Biol. Sci. 15,
Henderson, R. G. (1931). Phytopathology 21, 225.
Hildebrand, E. M. (1941). Contrib. Boyce Thomp-
son Inst. 11, 485.
Hildebrand, E. M. (1942). Science 95, 52.
Hildebrand, E. M. (1943). In J. A. Reyniers (ed.),
Micrurgical and Germ-free Techniques. 72-92.
C. C. Thomas, Baltimore.
Hildebrand, E. M. (1945). Plant Disease Reptr.
Hildebrand, E. M. (1956a). Science 123, 506.
Hildebrand, E. M. (1956b). Phytopathology 46,
Hill, A. V., and Mandryk, M. (1954). Australian
J. Agr. Res. 5, 617.
Hirth, L. (1951). Ann. Inst. Pasteur 80, 458.
Hobart, 0. F. (1956). (Abstr.) Iowa State Coll.
J. Sci. 30, 381.
Holmes, F. 0. (1929a). Botan. Gaz. 87, 39.
Holmes, F. 0. (1929b). Botan. Gaz. 87, 56.
Holmes, F. 0. (1939). Phytopathology 29, 215.
Holmes, F. 0. (1943). Phytopathology 33, 691.
Holmes, F. 0. (1954). Phytopathology 44, 640.
Hougas, R. W. (1951). Phytopathology 41, 483.
Howles, R. (1947). Exp. Research Sta. Turner's
Hill, Cheshunt, Herts. Ann. Rept., 33d, 22.
Howles, R. (1961). Plant Pathol. 10, 160.
Humphries, E. C., and Kassanis, B. (1955). Ann.
Appl. Biol. 43, 686.
Hunter, J. A., Chamberlain, E. E., and Atkinson,
J. D. (1958). New Zealand J. Agr. Res. 1, 80.
Hutchinson, M. T., Goheen, A. C., and Varney,
E. H. (1960). Phytopathology 50, 308.
Jedlinski, H. (1956). Phytopathology 46, 673.
Johnson, F. (1941a). (Abstr.) Phytopathology 31,
Johnson, F. (1941b). Phytopathology 31, 649.
Johnson, F. (1942). Phytopathology 32, 103.
Johnson, J. (1941). Phytopathology 31, 679.
Jones, L. K. (1944). Phytopathology 34, 941.
Jorgensen, P. S. (1957). Plant Disease Reptr. 41,
Kahn, R. P. (1956). Phytopathology 46, 295.
Kahn, R. P., and Libby, J. L. (1958). Phytopathol-
ogy 48, 57.
Kahn, R. P., and Schachtner, N. D. (1954a).
(Abstr.) Phytopathology 44, 389.
Kahn, R. P., and Schachtner, N. D. (1954b).
(Abstr.) Phytopathology 44, 494.
Kalmus, H., and Kassanis, B. (1944). Nature 154,
Kalmus, H., and Kassanis, B. (1945). Ann. Appl.
Biol. 32, 230.
Kassanis, B. (1947). Ann. Appl. Biol. 34, 412.
Kassanis, B. (1949). Ann. Appl. Biol. 36, 270.
Kassanis, B. (1952). Ann. Appl. Biol. 39,358.
Kassanis, B., and Kleczkowski, A. (1948). J. Gen.
Microbiol. 2, 143.
Kassanis, B., Tinsley, T. W., and Quak, F. (1958).
Ann. Appl. Biol. 46, 11.
Kausche, G. H. (1940). Biol. Zbl. 60, 423.
Kegler, H. (1959). Phytopathol. Z. 37, 170.
Kegler, H. (1961). Tidsskr. Planteavl. 65, 163.
Keur, J. Y. (1933). (Abstr.) Phytopathology 23,
Kontaxis, D. G., and Schlegel, D. E. (1962). Vi-
rology 16, 244.
Kreitlow, K. W. (1961). Phytopathology 51, 808.
Kunkel, L. O. (1930). Science 71, 516.
Chapter 3-R. W. Fulton
Kunkel, L. 0. (1938). Phytopathology 28, 491.
Kunkel, L. 0. (1943a). In Rockefeller Institute for
Medical Research, Virus Diseases. 63-82. Cor-
nell Univ. Press.
Kunkel, L. 0. (1943b). Proc. Am. Phil. Soc. 86,
Kunkel, L. 0. (1944). (Abstr.) Phytopathology
Kunkel, L. 0. (1945). Phytopathology 35, 805.
Kunkel, L. 0. (1950). (Abstr.) Phytopathology 40,
Kuntz, J. E., and Walker, J. C. (1947). Phyto-
pathology 37, 561.
Lackey, C. F. (1946). Phytopathology 36, 386.
Lackey, C. F. (1949). Phytopathology 39, 562.
Lackey, C. F. (1953). Am. J. Bot. 40,221.
Lauffer, M. A., and Price, W. C. (1945). Arch.
Biochem. 8, 449.
Limasset, P. (1951). Atti Soc. Ital. Patol. 2, 911.
Limasset, P., and Comuet, P. (1949). C. R. Acad.
Sci. (Paris) 228, 1971.
Lindner, R. C., and Kirkpatrick, H. C. (1959).
Phytopathology 49, 507.
Lindner, R. C., Kirkpatrick, H. C., and Weeks,
T. E. (1959). Phytopathology 49, 78.
Lister, R. M. (1958). Nature 182, 1814.
Lister, R. M. (1959). Nature 183, 1588.
Lister, R. M. (1960). Virology 10, 547.
Liu, H. P. (1949). Rept. Taiwan Sugar Expt. Sta.
MacClement, W. D. (1937). Parasitology 29, 266.
Maramorosch, K. (1956). Plant Disease Reptr. 40,
Martin, C. (1952). Ann. Inst. Nat. Rech. Agron.
(C) 3, 395.
Martin, C. (1954). Ann. Inst. Rech. Agron. (C)
(Ann. Epiphyt.) 5, 63.
Martin, C. (1958). Compt. Rend. Acad. Sci.
(Paris) 246, 2026.
Martin, C., and Morel, G. (1958). Compt. Rend.
Acad. Sci. (Paris) 246, 2283.
Matthews, R. E. F. (1953a). Ann. Appl. Biol. 40,
Matthews, R. E. F. (1953b). Ann. Appl. Biol. 40,
Matz, J. (1933). J. Agr. Res. 46,821.
Mayer, A. E. (1886). Die Landwirtschaftlichen
Versuchstationen 32, 451. (Phytopathological
Classic No. 7.)
McClean, A. P. D. (1948). Union S. Africa Dept.
Agr. Sci. Bull. 256.
McClintock, J. A. (1916). Science 44, 786.
McClintock, J. A. (1917). (Abstr.) Phytopathology
McKinney, H. H. (1928). Science 68, 380.
McKinney, H. H. (1948). Phytopathology 38,
McKinney, H. H. (1951). (Abstr.) Phytopathology
McKinney, H. H. (1952). Plant Disease Reptr. 36,
McKinney, H. H. (1953). Plant Disease Reptr. 37,
McKinney, H. H., and Fellows, H. (1951). Plant
Disease Reptr. 35,264.
McLean, D. M. (1941). Phytopathology 31, 420.
McLean, D. M. (1962). (Abstr.) Phytopathology
McNeal, F. H., and Afanasiev, M. M. (1955).
Plant Disease Reptr. 39, 460.
McWhorter, F. P. (1951). Phytopathology 41, 185.
McWhorter, F. P. (1953). (Abstr.) Phytopathol-
ogy 43, 479.
Medina, A. C., and Grogan, R. G. (1961). Phyto-
pathology 51, 452.
Middleton, J. T. (1944). Phytopathology 34, 405.
Milbrath, J. A. (1953). (Abstr.) Phytopathology
Misiga, S., and Valenta, V. (1957). Biologia, Bra-
tislava 12, 652.
Miyakawa, T., and Yoshii, H. (1951). Sci. Bull.
Fac. Agr., Kyushu Univ. 12(2), 143.
Mundry, K. W., and Rohmer, Irmgard (1958).
Phytopathol. Z. 31, 305.
Murphy, P. A., and McKay, R. (1926). Sci. Proc.
Roy. Dublin Soc. 18, 169.
Muzik, T. J., and LaRue, C. D. (1954). Am. J.
Botany 41, 448.
Nelson, R. (1932). Mich. State Univ. Agr. Expt.
Sta. Tech. Bull. 118.
Nelson, R., and Down, E. E. (1933). (Abstr.)
Phytopathology 23, 25.
Newhall, A. G. (1923). Phytopathology 13, 104.
Nikolic, V. J. (1956). Zasht. Bilja (Plant Prot.,
Beograd) 35, 79.
Nitzany, F. E., and Gerechter, E. K. (1962). Phy-
topathology Medit. 2, 11.
Panzer, J. D. (1957). Phytopathology 47, 337.
Panzer, J. D. (1959). Plant Disease Reptr. 43, 845.
Pfaeltzer, Hillegonda J. (1960). Tijdschr. Planten-
ziekten 66, 24.
Quanjer, H. M. (1920). Phytopathology 10, 35.
Quanjer, H. M., van der Lek, H. A. A., and Boties,
J. 0. (1916). Meded. Rijks Hoogere Land-,
Tuinen Boschbouwschool 10, 92.
Quantz, L. (1953). Phytopathol. Z. 20, 421.
Rader, W. E.. Fitzpatrick, H. F., and Hildebrand,
E. M. (1947). Phytopathology 37, 809.
Rawlins, T. E., and Tompkins, C. M. (1934).
(Abstr.) Phytopathology 24, 1147.
Rawlins, T. E., and Tompkins, C. M. (1936). Phy-
topathology 26, 578.
Reddick, D. (1931). Deuxieme Congress Internat.
de Path. Comp. 1, 363.
Reddick, D., and Stewart, N. B. (1919). Phyto-
pathology 9, 445.
Reyes, T. T. (1961). Plant Disease Reptr. 45,
Richards, B. L., Jr., and Munger, H. M. (1944).
(Abstr.) Phytopathology 34, 1010.
Roberts, R. H. (1949). Botan. Rev. 15, 423.
Ross, A. F. (1941). Phytopathology 31, 410.
Ross, A. F. (1953). Phytopathology 43, 1.
Rozendaal, A., and van der Want, J. P. H. (1948).
Tijdschr. Plantenziekten 54, 113.
Sakimura, K. (1947). Phytopathology 37, 66.
Salmon, E. S., and Ware, W. M. (1935). Ann.
Appl. Biol. 22, 728.
Samuel, G. (1931). Ann. Appl. Biol. 18, 494.
Schmelzer, K. (1955). Naturwissenschaften 42, 19.
Schmelzer, K. (1956a). Zbl. Bakt. II. Abt. 109,
Schmelzer, K. (1956b). Phytopathol. Z. 28, 1.
Schmelzer, K. (1957a). Phytopathol. Z. 28, 457.
Schmelzer, K. (1957b). Phytopathol. Z. 30, 449.
Schmelzer, K. (1960). Z. Pflanzen Krankheiten 67,
Schramm, G., Schumaker, G., and Zillig, W.
(1955). Nature 175, 549.
Schultz, E. S., and Folsom, D. (1920). J. Agr.
Res. 19, 315.
Schultz, E. S., and Folsom, D. (1921). J. Agr. Res.
Schultz, E. S., and Folsom, D. (1923). J. Agr. Res.
Schwartz, D., and Cuzin, J. (1951). Ann. Inst.
Exptl. Tab. Bergerac 1, 167.
Severin, H. H. P. (1924). Phytopathology 14, 80.
Sheffield, F. M. L. (1936). Ann. Appl. Biol. 23,
Sheffield, F. M. L. (1941). J. Roy. Microscopical
Soc. (III) 61, 30.
Sheffield, F. M. L. (1952). Ann. Appl. Biol. 39,
Shepherd, R. J., and Fulton, R. W. (1962). Phyto-
pathology 52, 489.
Sill, W. H., Jr., and Walker, J. C. (1952). Phyto-
pathology 42, 349.
Singh, G. P., Amy, D. C., and Pound, G. S.
(1960). Phytopathology 50, 290.
Skotland, C. B., and Burke, D. W. (1961). Phyto-
pathology 51, 565.
Smith, E. F. (1888). U.S. Dept. Agr. Bot. Div.
Spencer, E. L. (1935a). Phytopathology 25, 178.
Spencer, E. L. (1935b). Phytopathology 25, 493.
Sreenivasaya, M. (1930). J. Indian Inst. Sci. 13a,
Stanley, W. M. (1934). Phytopathology 24, 1055.
Stanley, W. M. (1935). Phytopathology 25, 899.
Steere, R. L. (1955). Phytopathology 45, 196.
Sutic, D. (1959). Phytopathol. Z. 36, 84.
Takahashi, W. N. (1947). Am. J. Botany 34, 496.
Taylor, R. H. (1962). Australian J. Expt. Agr.
Animal Husbandry 2, 86.
Taylor, R. H., Grogan, R. G., and Kimble, K. A.
(1961). Phytopathology 51, 837.
Thomas, W. D., Jr., and Baker, R. R. (1952).
(Abstr.) Phytopathology 42, 21.
Thomas, W. D., Jr., and Graham, R. W. (1951).
Phytopathology 41, 959.
Thornberry, H. H. (1935a). Phytopathology 25,
Thornberry, H. H. (1935b). Phytopathology 25,
Thresh, J. M. (1956). Ann. Appl. Biol. 44, 608.
Thung, T. H., and Dijkstra, J. (1958). Tijdschr.
Plantenziekten. 64, 411.
Thung, T. H., and Hadiwidiaja, T. (1950). Tijd-
schr. Plantenziekten 56, 849.
Thung, T. H., and Noordam, D. (1959). Mededel.
Landbouwhogeschool Opzoekingssta, Staat Gent
Thung, T. H., and van der Want, J. P. H. (1951).
Tijdschr. Plantenziekten 57, 173.
Timian, R. G., Peterson, C. E., and Hooker, W. J.
(1955). Am. Potato J. 32, 411.
Tinsley, T. W. (1953). Ann. Appl. Biol. 40, 750.
Toler, R. W. (1962). Dissertation Abstr. 22, 3804.
Tomlinson, N. (1955). Plant Disease Reptr. 39,
Troll, H. J. (1957). Nachrbl. Deut. Pflanzenschutz-
dienst (Berlin) 11, 218.
Troutman, J. L., and Fulton, R. W. (1958). Virol-
ogy 6, 303.
Valenta, V. (1962). Biologia, Bratislava 17, 415.
Valleau, W. D. (1932). Kentucky Agr. Expt. Sta.
Bull. 327, 43.
Valleau, W. D. (1935). Kentucky Agr. Expt. Sta.
Bull. 360, 202.
Valleau, W. D. (1939). Phytopathology 29, 549.
Van Velsen, R. J. (1961). Papua New Guinea
Agr. J. 14, 38.
Van Velsen, R. J., and Crowley, N. C. (1961).
Nature 189, 858.
Vaughan, E. K. (1956a). Tijdschr. Plantenziekten
Vaughan, E. K. (1956b). Tijdschr. Plantenziekten
Vinson, C. G., and Petre, A. W. (1931). Contrib.
Boyce Thompson Inst. 3, 131.
Wagnon, H. K., Traylor, J. A., Williams, H. E.,
and Weinberger, J. H. (1960). Plant Disease
Reptr. 44, 117.
Wallace, J. M. (1947). Phytopathology 37, 149.
Wallace, J. M., and Drake, R. J. (1962). Phyto-
pathology 52, 237.
Way, R. D., and Gilmer, R. M. (1958). Plant Dis-
ease Reptr. 42, 1222.
Weathers, L. G., and Calavan, E. C. (1959). In
J. M. Wallace (ed.), Citrus Virus Diseases. 197.
Univ. of California Press.
Williams, W. L. (1957). Dissertation Abstr. 27,
Willison, R. S., Weintraub, M., and Ferguson, J. D.
(1956). Can. J. Botany 34, 86.
Yarwood, C. E. (1952a). Nature 169, 502.
Yarwood, C. E. (1952b). Am. J. Botany 39, 119.
Yarwood, C. E. (1952c). Phytopathology 42, 137.
Yarwood, C. E. (1953a). Plant Disease Reptr. 37,
Yarwood, C. E. (1953b). Phytopathology 43,70.
Yarwood, C. E. (1954). Phytopathology 44, 230.
Yarwood, C. E. (1955a). Virology 1, 268.
Yarwood, C. E. (1955b). Hilgardia 23, 613.
Yarwood, C. E. (1956a). (Abstr.) Phytopathology 46,
Yarwood, C. E. (1956b). Phytopathology 46, 523.
Yarwood, C. E. (1957a). Phytopathology 47, 613.
Yarwood, C. E. (1957b). Advan. Virus Res. 4, 243.
Yarwood, C. E. (1958). Phytopathology 48, 39.
Yarwood, C. E. (1959). Plant Disease Reptr. 43,
Yarwood, C. E. (1962a). Phytopathology 52, 206.
Yarwood, C. E. (1962h). Plant Disease Reptr. 46,
Yarwood, C. E., Resconich, E. C., and Kado, C. I.
(1962). Virology 16, 414.
A. F. ROSS
Identification of plant viruses
PLANT VIROLOGISTS, who as a group show a
degree of heterogeneity comparable to that
shown by the viruses with which they work,
have not yet agreed upon a well-defined
species concept. Thus, there is not yet a
clear-cut unified system of classification for
the plant viruses; this absence of a firm
basis for identification means that a sys-
tematic straightforward scheme for identi-
fication has not yet been formulated. The
situation is likely to improve as more and
more viruses are characterized in detail, as
natural groupings become more evident,
and as new bases for classification and
identification are developed. The constantly
changing criteria for classification and
identification and the constantly changing
relations and values of the criteria used are
simply reflections of the fact that viruses
resemble living things in many ways.
In general, the plant virologist has based
his identifications on all information he
could obtain; thus the reliability of identifi-
cations has paralleled the growth of knowl-
edge in all phases of plant virology. The
present-day virologist has the choice of
many approaches to identification, and the
approaches used will depend largely upon
the virus or viruses concerned and the
available facilities. Some viruses can be
identified with a high degree of certainty
on the basis of precise information concern-
ing only a few properties; with others, a
judgment can be made only after the ac-
quisition of information on all of the prop-
erties that can be studied. Identification of
any plant virus on the basis of a single cri-
terion or property is rarely without its dan-
Keys for the identification of plant vi-
ruses have been difficult to devise, and no
one has been successful in devising a work-
able key embracing all known plant viruses.
Two of the most important reasons for this
are the facts that many viruses have been
incompletely characterized and that almost
all plant viruses exist as a large family of
strains or variants. These variants may dif-
fer greatly in some traits while sharing
others. Unfortunately, much of the missing
information concerns the shared character-
Historically, symptomatology has been
the most widely used criterion for the iden-
tification of plant viruses. This is the natural
consequence of the fact that the known
plant viruses, with a few recent exceptions,
have been discovered and recognized on
the basis of the diseases they cause. The
inadequacy of symptomatology as a basis
for identification has long been recognized,
and many of the early diagnoses based
solely on symptomatology are questionable
today. As information on the viruses accu-
mulated, it became apparent that a given
virus can cause widely different symptoms
in different host species, that different vi-
ruses can cause similar symptoms in the
same species, that different strains or iso-
lates of the same virus can cause entirely
different diseases, and that symptoms in-
duced by a given strain in a given host spe-
cies can vary greatly under different en-
vironmental conditions. Some of these
things were recognized many years ago,
and plant virologists began to look for other
Cross inoculations, first made in 1886
(Mayer, 1886), demonstrated that different
viruses have different host ranges, and virus
workers began to base identification in part
on host-range studies. These were conven-
iently combined with studies on symptoma-
tology, for symptoms induced in a number
of species quite logically proved better for
diagnosis than did just those symptoms in
the original host species. As further knowl-
edge accumulated, plant pathologists were
able to characterize certain viruses still fur-
ther on the basis of methods of transmission
(mechanical, specific vectors, grafting only)
and physical properties in crude juice (di-
lution end point, longevity in vitro, thermal
inactivation point). Prior to about 1930,
these were almost the only characters avail-
able for use in identification.
The demonstration of acquired immunity
(McKinney, 1929; Thung, 1931; Wingard,
1928) and of the antigenic nature of some
plant viruses (Purdy, 1929) provided two
of the most useful diagnostic tools, i.e.,
cross-protection and serological tests, re-
spectively. The development of these tools
contributed greatly to the accuracy of iden-
tification; satisfactory cross-protection tests
cannot be made with some viruses, how-
ever; and many viruses have not yet been
shown to be antigenic. More recent in vivo
reactions shown to be of diagnostic value
are the interaction of viruses in mixed in-
fections (Holmes, 1956) and the response
to infection of plant varieties containing
specific genes (Holmes, 1958).
Following the crystallization of tobacco
mosaic virus (TMV) (Stanley, 1935) and
proof that it is a nucleoprotein (Bawden
et al., 1936), a vast amount of information
has accumulated on the chemical and
physical properties of several plant viruses.
This wealth of precise information, by
which some viruses can be accurately char-
acterized in terms of morphology and
chemical structure, provides still further
criteria for use in identification. Unfortu-
nately, such biochemical and biophysical
studies can be made only with some viruses
and only in laboratories with the required
Actually, almost any fact known about a
virus may be of value in identification, for
the more information used for an identifi-
cation, the more certain is the identifica-
tion. Practically, identification of an "un-
known" consists of the demonstration of a
sufficient number of similarities between it
and a described virus to establish its iden-
tity beyond reasonable doubt. Even today,
plant virus identification is not an exact
science, and cases do occur in which iden-
tification is no more than a judgment. Au-
thorities differ on the relative weight to
give different criteria, so any identification
should be supported by statements on the
bases used for identification.
The basic information on which identi-
fications are made, and most of the meth-
ods used, will be found in other chapters
in this book. This chapter will deal largely
with some practical aspects of identifica-
tion, the application of various facts and
concepts to the problem, and an evaluation
of various tests and techniques in terms of
applicability and limitations. Methods and
special procedures will be given only in
those areas not dealt with elsewhere or
where the choice of a particular method
may be of importance.
4-2. ESTABLISHING THAT THE
"UNKNOWN" IS A VIRUS
All too often a pathogenic agent is as-
sumed to be a virus if it is shown to induce
"virus-like symptoms" following transmis-
sion by graft or by other methods commonly
used for viruses. Such assumptions have
been proved true in many cases by subse-
quent work, but they have been in error in
others. For instance, "beet latent virus" was
described several years ago (Smith, 1950,
1951a), and this virus name has appeared
Chapter 4-A. F. Ross
many times in the literature. The agent was
considered to be a virus largely on the basis
of the occurrence of the pathogen in symp-
tomless beets and mangolds, the production
of "virus-like" lesions in cowpea and other
plants rubbed with nonfiltered juice, the
ease of transmission by methods used rou-
tinely for viruses, and the production of
systemic infection of cowpea (Smith,
1951a). In 1961, Yarwood et al. (1961)
showed that "beet latent virus" is actually
not a virus but a bacterium, Pseudomonas
Viruses can be differentiated from toxins
and other nonliving agents capable of caus-
ing physiological disturbances (by their
presence or absence) on the basis of trans-
missibility in series. Viruses are usually dis-
tinguished from other pathogens on the
basis of size, for a pathogenic agent with
one dimension less than 200 mup is generally
considered to be a virus (Bawden, 1950).
The cause of a disease, therefore, can be
identified reliably as a virus on the basis of
transmissibility in the absence of pathogens
visible in the light microscope. The classical
method for testing an agent by these two
criteria is to attempt transmission with juice
passed through a filter that retains bacteria.
If juice passed through such a filter is in-
fectious, the presence of a virus is generally
considered to have been established. Fil-
terable forms of other agents do exist;
hence such evidence may not constitute
rigorous proof. Another historical approach
to the problem of size is the application of
ordinary microscopic and cultural methods
in attempts to detect presence of bacteria,
fungi, or other microscopic organisms. If
careful examination of infected tissue does
not reveal the presence of organisms, and
if no organisms can be cultured from such
tissue, the size criterion is generally con-
sidered to have been met. Modem methods
of purification and electron microscopy
have provided still another approach to this
question, for they make possible the deter-
mination of the size and shape of some vi-
ruses. Virus particles are not always easy to
find, and showing that a virus-like particle
is indeed the causal virus is a still more dif-
ficult problem; therefore these methods are
not universally applicable.
4-3. ESTABLISHING THAT ONLY
ONE VIRUS IS PRESENT
Many diseases are the result of infection
by two viruses, and indeed infection by
more than two viruses is not at all unusual.
One should always keep in mind, therefore,
that an "unknown" may actually be a mix-
ture of viruses. One of the first problems to
consider is whether one or more viruses are
present, and steps should be taken to sepa-
rate the viruses if more than one are found.
Sometimes careful examination of the
original infected plants will provide evi-
dence of a mixture. Presence of more than
one virus may be indicated by the presence
of two or more distinct types of symptoms
or by the presence of two or more types of
crystalline inclusion bodies. If the "un-
known" is collected from a field or green-
house where likely contaminants are few in
number, then specific tests can be made for
those contaminants by serological tests or
by attempted transmissions to so-called in-
dicator plants. Preliminary tests for con-
tamination should be made by inoculations
to several different species, and these
should be followed in two to three weeks
by return inoculations to the original plant
species from both the inoculated leaves and
noninoculated ones. If the return inocula-
tions do not reproduce the original disease,
then a mixture is indicated. The original
inoculations to the test species should be
made with different dilutions of juice
where possible and by different inoculation
If a mixture is detected, separation may
be accomplished by any of several methods
or a combination of methods. A single
method is generally effective only for the
separation or elimination of one virus from
a mixture; thus, several different methods
must be used to separate each component
from the others.
Separation may possibly be effected sim-
ply by inoculations to a number of different
test species and thence back to the original
species. One test species may be suscepti-
ble to one component and immune from
the others, or one component may be able
to cause systemic infection in a particular
host whereas other components may be
localized in the inoculated leaves. A host
that is systemically invaded by all compo-
nents may be useful in separation if the
viruses move at different rates; attempts to
recover virus from the tip or other remote
portions at different times after inoculation
may result in recovery of one virus free
from the others. If local-lesion hosts are
inoculated with dilute inocula, recovery
from single lesions may result in recovery
of one or more of the viruses free from the
Separation also may be accomplished by
use of different methods of transmission;
one component of the complex may be
transmitted-and others not-by mechanical
methods, by dodder, or by specific vectors.
With insect-transmitted viruses, advantage
may be taken of differences in virus-vector
relationships and of the fact that a single
vector carrying more than one virus and
given a series of test feedings on healthy
plants may transmit one virus to some
plants in the series and another to other
plants. Mechanical inoculations with highly
diluted juice may free one virus from others
of the mixture. This procedure is particu-
larly useful when one of the component vi-
ruses has a much higher dilution end point
than the others, but separation may be ef-
fected even where the component viruses
have about the same dilution end point: if
many plants are inoculated with dilutions
such that transmission is somewhat less
than 100%, some plants may become in-
fected by chance with one virus and some
with another. The viruses in a mixture also
may differ in stability; thus, heating juice
at various temperatures or allowing juice to
age at room temperature for different pe-
riods may eliminate one or more viruses
from a mixture.
Specific antisera can be used for removal
of known viruses from a mixture. Separa-
tion also may be effected by such tech-
niques as filtration through membranes
differing in size of pores (Smith, 1951b),
sedimentation in density-gradient columns
(Brakke, 1958), or electrophoresis (Brakke,
All attempts at separation should be fol-
lowed by return inoculations to the original
host species. If symptoms induced by a re-
turn inoculation differ from the original
ones, loss of one or more viruses from a mix-
ture is to be expected. Should return inocu-
lations following other procedures result in
still different symptoms, further separation
of the component viruses may have been
successful. All viruses thus separated
should be recombined and the mixture
used for inoculation of the original host spe-
cies. Reproduction of the original symptoms
would show that all components have been
accounted for. One should keep in mind,
however, that some of the procedures used
may merely have separated out one or more
minor strains from the original inoculum,
which almost certainly consisted of a mix-
ture of predominant and minor strains
(Bawden, 1950). Remixing such strains
would very likely be in a proportion differ-
ent from that in the original inoculum, in
which case the original symptoms might
not be reproduced. Whether or not this
happened will become apparent upon final
identification of the separated components.
There is no guarantee that any method
or combination of methods will result in
separation of a mixture of unidentified vi-
ruses. Some virus pairs are extremely diffi-
cult to separate; hence failure to effect a
separation is not proof of biological purity.
Furthermore, isolation of two or more types
of virus entities from an inoculum is no
guarantee that each separated component
is itself a single virus. During subsequent
attempts at identification, therefore, one
should always be alert for evidence indicat-
ing that a mixture is still being dealt with.
4-4. GENERAL APPROACH TO
Identification is perhaps best accom-
plished by a step-wise series of character-
izations at progressively higher levels of
specificity. Thus, studies on symptoma-
tology in a number of host species may be
sufficient to place a virus in one of several
large groups, e.g., mosaic, yellows, ringspot,
witches"' broom, etc. Further tests on host
range, on transmission, and on properties in
crude juice should narrow the possibilities
to just a few viruses. At this stage, tests for
specific viruses can be applied, e.g., sero-
Chapter 4-A. F. Ross
logical or cross-protection tests. If these
result in a tentative identification, they
should be followed by further confirming
tests such as examination in the electron
microscope, interaction with other viruses,
and inoculation of additional plant species.
Still further characterization may be neces-
sary if identification is not certain at this
stage or if it is desirable to carry identifica-
tion through to the strain level. The specific
tests to be used at this stage can be deter-
mined only by a careful search of the origi-
The approach suggested above is most
generally applicable to viruses that are me-
chanically transmissible. Identification of a
virus that is not mechanically transmissible
must of necessity be based largely on symp-
tomatology, host range, and (if the vector
is known) the identity of the vector and
the virus-vector relations. Fairly satisfactory
cross-protection tests can sometimes be
made where the only known method of
transmission is by grafting. If transmission
by the insect vector is dependent upon
multiplication within the insect, tests for
cross protection within the insect may pro-
vide useful information. Modern techniques
have made possible the use of other meth-
ods with some of the persistent viruses
transmitted by leafhoppers or aphids. Some
of these in plant or insect extracts can now
be transmitted by mechanical methods to
insects, which then transmit them to plants,
making it possible to measure virus stabil-
ity, to purify the viruses in some cases, and
perhaps to characterize them by means of
electron microscopy. Also, some of the leaf-
hopper-transmitted viruses have proved to
be antigenic, making it possible to use se-
rology in their identification.
Prior knowledge about the "unknown," or
of the conditions under which it was found,
may make it feasible to by-pass some of the
steps in the suggested approach. Thus if an
"unknown" were suspected to be TMV, its
reaction with TMV-antiserum might be the
most logical first step. If a strong reaction is
obtained, then a dependable identification
might be made on the basis of this fact and
a few additional confirming tests, such as
the diseases induced in a few selected
plant species, comparative effects in to-
bacco varieties with and without the N
gene from Nicotiana glutinosa L. (Holmes,
1958), thermal inactivation point, and ex-
amination in the electron microscope.
Present methods for the identification of
plant viruses have been developed for use
with viruses as they generally exist in natu-
ral infections-as intact nucleoproteins. Re-
cently, naturally occurring unstable variants
of some viruses have been found to exist
free of the usual specific protein coat
(Sanger and Brandenburg, 1961; Cadman
1962; Babos and Kassanis, 1962). These
variants differ from the parent virus in sta-
bility, in ease of transmission by conven-
tional methods (although vector relations
may be the same), in host reactions, and in
particle morphology. Being free of the spe-
cific viral protein, they are not antigenic.
Quite obviously, application of the usual
identification methods to one of these un-
stable variants would very likely fail to con-
nect it with the parent virus. This type of
virus is indicated if the "unknown" is very
difficult to transmit by ordinary mechanical
means but easily transmitted following
phenol extraction (Schlegel, 1960). Host-
range studies, cross-protection tests, and
vector relations might be used as aids in
identifying such viruses, but it is apparent
that new methods are needed.
4-5. TECHNIQUES AND TESTS
USED IN IDENTIFICATION
Usually, the objective in work with a new
virus or virus "unknown" is not simply to
establish its identity. Since the identity is
not known in advance, and since soundness
of an identification increases progressively
with the amount of information obtained,
the identification procedure normally
should be designed to provide a full char-
acterization of the virus. Usually, further
work on the virus after it is identified is to
be anticipated, which means that one will
need precise information on such things as
transmissibility, host range, symptomatol-
ogy, stability, possible local-lesion hosts,
and host species suitable for virus mainte-
nance and increase. Choice of procedures
to use, therefore, should not be limited to
the minimum needed to establish a reason-
able identification but instead should in-
clude all those that are applicable to the
virus at hand and that are possible with the
A. SYMPTOMATOLOGY AND HOST
One objective of studies on host range
and symptomatology is, of course, to char-
acterize the "unknown" in terms of the re-
actions or lack of reactions it induces in
host plants selected on the basis of their
reactions to known viruses. Such informa-
tion serves at least two purposes. First, it
may point directly to the identity of the vi-
rus or at least narrow the investigation to a
workable number of "possibilities" and thus
make possible the application of more spe-
cific tests. Second, information on symp-
tomatology and host range will almost
always be needed, after other tests have
identified the virus of which the "unknown"
is a strain, for the final determination to the
A second important objective is the dis-
covery of a suitable test or "indicator" plant.
A good test plant is needed in identification
studies for testing for possible virus increase
in inoculated plants that failed to develop
symptoms, for determining the so-called
physical properties of the virus in crude
juice, and for checking on the potency of
the inocula used in various kinds of tests.
This test plant should be one that develops
distinct symptoms soon after inoculation
and preferably one that is self-pollinated,
homozygous, and rapidly grown from seed;
for mechanical inoculations, a plant that
reacts to inoculation by the formation of lo-
cal lesions is much to be preferred.
Many times, the plant species in which
the "unknown" is discovered may be un-
satisfactory for maintenance of the culture
or as a source for high-potency inocula.
Consequently, a frequent third objective is
the discovery of a suitable species for use
as a source plant.
1. Selection of test species.-Approxi-
mately one-third of the known plant viruses
have extremely wide host ranges, including
species from widely related families; the
host ranges of some viruses include both
monocots and dicots and both herbaceous
and nonherbaceous species. Another third,
more or less, have host ranges embracing
several species in a number of genera and
perhaps a few families, whereas the re-
maining ones appear to have very restricted
host ranges, usually including only a few
closely related species in no more than a
few genera. A newly discovered virus in
one species may be a well-known virus
commonly associated with an entirely dif-
ferent type of plant species or group of spe-
cies. The list of test plants used in studies
on host range and symptomatology should
include, therefore, not only other varieties
of the source species (the species in which
the "unknown" was found) and other
closely related species but also species
known to be susceptible to a large number
of viruses. The last-named group should in-
clude species from several genera and fam-
ilies. If the source species is nonherbaceous,
for example, the test species used may logi-
cally include some herbaceous species, for
many viruses found in woody plants also
have herbaceous hosts.
Some herbaceous species have been
found to be susceptible to a large number
of viruses (Table 4-1). Another species in
the same category is Nicotiana clevelandii
Gray (Hollings, 1959). Some of these are
susceptible to a broad spectrum of viruses,
including ones normally found in woody
plants. There may be other species that
show the same range of susceptibility as
those listed, for this list may be in part a re-
flection of the frequency with which certain
species are used in host-range studies.
Whatever the reason for the apparent un-
usual susceptibility of the listed species, it
would seem advisable to follow the sugges-
tion of Thornberry (1961) that the species
listed in Table 4-1 be used in host-range
investigations. They would perhaps be
most useful where mechanical transmission
is attempted, but many of them should be
included also when other transmission
methods are used.
Several varieties of most of the species
listed in Table 4-1 are available, and some-
times the choice of varieties is highly im-
portant. Virus workers have not been par-
ticularly in agreement concerning the best
varieties to use, and choice has probably
Chapter 4-A. F. Ross
TABLE 4-1. NUMBER OF VIRUSES KNOWN TO
INFECT OR NOT INFECT SOME SPECIES OF PLANTS
(Information from Coded-Card File of Susceptible
and Insusceptible Plant Species)
Number of Viruses
Species of Plants Infecting Not infecting
Beta vulgaris L. 33 62
Brassica oleracea L. 23 46
Coste & Reyn. 47 6
Cucumis sativus L. 67 77
Datura stramonium L. 70 67
Gomphrena globosa L. 26 16
Mill. 60 76
Nicotiana glutinosa L. 78 67
Nicotiana tabacum L. 115 95
Phaseolus vulgaris L. 60 71
Pisum sativum L. 42 39
Solanum melongena L. 31 32
Solanum tuberosum L. 60 45
Vicia faba L. 35 40
Vigna sinensis Endl. 48 34
Zinnia elegans Jacq. 45 39
Reproduced from Thomberry
(1961) by per-
depended as much on easy availability as
on any other single thing. Table 4-2 lists
the varieties of some of these species that
appear to have been used most frequently
in the past or are known to offer some par-
ticular advantage. Workers with a particu-
lar type of virus may find it desirable to
substitute others or to use additional ones.
Those who work with viruses attacking a
particular kind of plant generally use more
or less the same test species, and a few at-
tempts have been made to establish stand-
ard lists. Thus, Bos et al. (1960), as a part
of a suggested procedure for the interna-
tional identification of legume viruses,
listed 30 species and recommended that
these species be used by all workers with
legume viruses throughout the world.
2. Procedure.-Systemic symptoms usu-
ally develop best in plant parts that are
growing rapidly, but local lesions sometimes
develop best in expanded mature leaves.
Sometimes the method of inoculation, e.g.,
grafting, will dictate the age of test plant to
use. In general, it is best to use young,
rapidly growing plants at about the time
two or three leaves become fully expanded
or nearly so. Some species seem to give the
best results when inoculated within a fairly
narrow age range. For example, cucumbers
are usually used before the first true leaf
has expanded appreciably, and only the
cotyledons (and perhaps one true leaf) are
usually inoculated. The primary leaves of
beans and cowpeas are generally the only
leaves inoculated mechanically, and this is
usually done just before the trifoliolate
leaves begin to expand.
The number of test plants to use is
largely determined by the availability of
plants, of greenhouse space, and of time. At
least three plants of each species or variety
should be used, and it would be preferable
to use six or more. Tests with any species
giving questionable or variable results
should be repeated, with an increase in the
number of test plants.
Ideally, all tests should be made at nor-
mal greenhouse temperatures (200-22C),
with other conditions (light, nutrition, etc.)
near those optimal for growth of the species
TABLE 4-2. SUGGESTED LIST OF VARIETIES OR
SELECTIONS FOR USE IN STUDIES ON HOST
RANGE AND SYMPTOMATOLOGY
Cucumus sativus A & C, National Pickling
Gomphrena globosa Cornell selection*
Lycopersicum esculen- Bonny Best
Nicotiana tabacum Turkish, Samsun, White
Burley, Samsun NNt
Phaseolus vulgaris Pinto, Bountiful
Solanum tuberosum USDA seedling 41956t
Vigna sinensis Black
Wilkinson and Blodgett (1948).
t Or other variety containing the N gene from
t Immune from, or resistant to, potato virus X
but likely to be infected by potato virus S.
d Immune from, or resistant to, potato viruses X
being used. Often this is not possible, be-
cause of the difficulty of doing large-scale
experiments under standardized conditions
and because the plant species themselves
may vary greatly in their requirements for
optimal growth. As temperature is perhaps
the most important single environmental
factor affecting symptoms, the practical
solution to the problem is to do the tests
when near-normal temperatures can be
maintained and when other conditions are
not particularly limiting for plant growth.
In any case, the conditions used should be
reported and the results interpreted accord-
The method of inoculation used will de-
pend upon the virus. Mechanical inocula-
tions should be attempted and used where
possible, but inoculations may be by any
feasible method (mechanical, grafting,
dodder, tissue insertion, insects, etc.). Pre-
sumably, the investigator will have to
determine something about the transmissi-
bility of the "unknown" before he under-
takes a host-range study. It should be kept
in mind, however, that the reported host
range of a virus may depend upon the
method of inoculation used, e.g., some spe-
cies may be found susceptible when inocu-
lated by grafting but not when inoculated
mechanically or by means of insects. It is
not particularly uncommon for one to en-
counter difficulty in mechanical transmis-
sion of a virus from plant to plant within the
species in which the virus is commonly
found in nature and yet encounter little
difficulty in transmitting it by mechanical
means to other species. For example,
necrotic ringspot virus of Prunus is virtually
nontransmissible by mechanical means from
cherry to cherry, but it is rather easily so
transmitted to cucumber. If transmission of
the "unknown" is seemingly restricted to
graft inoculations, it may be possible to in-
crease the number of usable test species by
use of dodder or by tissue implantations.
Results of mechanical inoculations are
likely to be inconclusive unless highly in-
fectious inocula are used. With herbaceous
species, the first systemically infected leaves
to show well-developed symptoms are gen-
erally satisfactory as source of inocula.
These should be used within a few days
after symptom development-usually 10-14
days after inoculation. Leaves inoculated
directly with highly infectious juice are gen-
erally satisfactory virus sources about a
week after the inoculation. In most of the
inoculated test species, positive reactions
should occur in either 100% or 0% of the
plants. Unless this occurs or unless tests on
local-lesion hosts indicate a satisfactory in-
oculum, steps should be taken to find a
better virus source or to improve the inocu-
Inoculated plants must be examined at
daily intervals following inoculation, so that
the sequence of symptoms may be re-
corded. Some symptoms, such as vein-clear-
ing, are transient and thus may be missed
unless frequent observations are made. In-
cubation periods, i.e., the time required
after inoculation for symptoms to develop,
may be an aid in identification. Also, some
viruses induce a characteristic sequence of
disease symptoms. Many of the ringspot
diseases, for example, are characterized by
a local phase (in inoculated leaves), a sub-
sequent acute stage (in first leaves invaded
systemically) where symptoms may be
quite severe, and eventually a chronic
phase (in leaves developing after the in-
fection is systemic) in which symptoms may
be mild or absent.
Absence of symptoms must never be
taken as evidence of lack of infection. In
other words, a symptomless plant should
not be judged insusceptible unless attempts
to recover virus from it are consistently
unsuccessful. Similarly, species developing
symptoms only in inoculated leaves must
not be judged insusceptible to systemic in-
vasion unless recovery tests from noninocu-
lated leaves are negative. Recovery tests
should be made first (about a week after
inoculation) from the inoculated leaves.
With a stable virus, sufficient inoculum may
remain on the surface of the rubbed leaf to
give a "positive" recovery test. The amount
of such virus recovered will generally be
small and thus indicated by only a few
lesions in a local-lesion test plant or by a
low percentage of "takes" in systemic hosts.
When there is any question concerning pos-
sible virus increase in the inoculated leaves,
the test should be repeated and attempts
Chapter 4-A. F. Ross
made to detect an actual increase in virus
by making quantitative tests immediately
and at various intervals following the origi-
nal inoculation. Viruses vary considerably in
the rate at which they move systemically,
hence recovery from noninoculated leaves
should be made two or more weeks follow-
Even though distinct symptoms develop
in all inoculated plants of a given species,
return inoculations should still be made to
healthy plants of the source species. This
practice not only will confirm the suscepti-
bility of the test species concerned, but it
also will serve as a check on possible separa-
tion of components of a mixture or the pos-
sible filtering out of a particular strain from
the original inoculum. Either of these things
would be indicated by failure of the return
inoculation to induce typical symptoms in
the source species.
Because of the importance of the recovery
tests in host-range studies, one should be on
the alert for a good test or indicator plant,
i.e., one that develops easily recognized
symptoms soon after inoculation. Species or
varieties that develop local lesions following
inoculation are best to use for mechanically
transmissible viruses. Hopefully, the inves-
tigator will find a suitable indicator plant
among those being used in a host-range
study, or perhaps the original source species
will be suitable. If neither is the case, it
normally would be good practice to screen
still other species or varieties for possible
use as indicator plants.
The observations or records made should
be of three types; from the standpoint of
their usefulness in identification, the order
of importance of the three types is generally
in the same order in which they will be
mentioned. First, each species or variety
should be recorded as infected or not in-
fected. Development of reproducible symp-
toms may be taken as evidence of virus
increase, but recovery tests should be made
in any case where there is any doubt. Sec-
ond, each infection should be classed as
local or systemic. Again, this may be on the
basis of symptoms supplemented where
necessary by recovery tests. Third, the
specific symptoms should be recorded in
the sequence in which they appear and
according to whether they occur in inocu-
lated or systemically invaded leaves.
The failure of a virus to infect a particu-
lar plant species may be as important to its
eventual identification as infection of an-
other species. Consequently, reports on
host-range studies should list the species
Plants should be examined for other than
external symptoms. Many viruses induce
characteristic inclusion bodies in certain
hosts. The identity of the "unknown" may
be suggested by the types of inclusion
bodies induced, by the tissues or the parts
of the cell in which the inclusion bodies are
found, and sometimes by the staining re-
actions of the inclusion bodies (McWhorter,
1941). Also, information on the type and
distribution of inclusion bodies can be used
to differentiate between some pairs of vi-
ruses (McWhorter, 1960). Certain viruses
also induce characteristic changes in spe-
cific tissues, such as phloem or xylem. The
staining characteristics of affected tissues,
or the reactions with certain chemicals, will
occasionally be indicative of the causal vi-
rus (Lindner, 1961).
Selected plant species often can be used
to differentiate between a particular virus
and another. After an identification has
been narrowed to just a few "possibilities,"
therefore, the literature should be consulted
for information on plant species that react
differently to the viruses still on the list. The
differentiation may be on the basis of sus-
ceptibility, symptoms, or systemic vs. local
3. Interpretation.-Data obtained in stud-
ies on host ranges and symptomatology
seldom should be considered sufficient for
a positive identification. Variability is the
basic reason for this. Variability due to en-
vironmental conditions is of considerable
importance from a practical standpoint, but
the difficulty could be largely overcome by
use of modem facilities for growing and
testing of plants under controlled condi-
tions. The most important source of varia-
tion, therefore, is the viruses themselves.
The multiplicity of existing strains, plus
the fact that new strains may arise at any
time, should always be kept in mind by the
virologist attempting an identification. The
fact that most new strains or mutants are at
first recognized either on the basis of symp-
tomatology or host range is ample evidence
that the ability of a virus to induce certain
symptoms or to infect certain species is not
a particularly stable characteristic.
Despite the general inability of studies on
host range or symptomatology to provide
reliable identifications, they do provide the
best criteria for identification of a few vi-
ruses. Potato spindle tuber virus, for in-
stance, is best identified on the basis of the
characteristic shape of infected potato tu-
bers. Also the ability of experienced plant
pathologists to make accurate diagnoses
solely on the basis of symptomatology
should not be discounted. This ability re-
sults from long experience with particular
viruses in specific crop plants grown under
a fairly narrow range of environmental con-
ditions. That such diagnoses can be suffi-
ciently accurate for all practical purposes
is attested to by the fact that a great many
productive field experiments and successful
control measures are based on this type of
Symptomatology and host-range studies
are of major importance in the identifica-
tion of specific strains of a virus. Differen-
tiation among virus strains has been largely
on the basis of symptoms induced or on
ability or inability to infect certain plant
species. For many strains, there are no other
methods for differentiating among them.
B. PROPERTIES N CRUDE JUICE
Years ago, Johnson (1927) pointed out
the difficulties of identifying viruses on the
basis of symptomatology and suggested the
use of other characters such as thermal
inactivation point, resistance to aging in
vitro, and dilution end point. Since then,
most first accounts of mechanically trans-
mitted viruses have included data on these
properties. The original recommendations
were concerned with properties in juice
from infected plants (Johnson and Grant,
1932), and most of the subsequent deter-
minations also have been made on this
basis. Until recently (Bos et al., 1960), no
attempt has been made toward a wide-
spread use of a standard source plant;
workers have generally used as source the
species in which the virus was first found or
the species commonly used for maintenance
and increase of the virus. Similarly, there
has been no standardization of the species
used for detecting virus activity, although
local-lesion hosts have generally been used
when available. Obviously, this lack of
standardization has resulted in much varia-
tion in results, for the crude juice of in-
fected plants of different source species is
almost certain to differ in virus concentra-
tion, in content of other constituents that
may be either protective or deleterious to
the virus, and in pH. Results of such tests
also will depend in part upon the sensi-
tivity of the test plant. As a result of these
and other sources of variation, published
data on these properties actually do vary
to a great extent. With cucumber mosaic
virus, for example, different workers have
reported thermal inactivation between 55
and 70 and dilution end points varying
from 1/1000 to 1/100,000. This may be an
extreme case; actually, there is fair agree-
ment among most workers on the in vitro
properties of many of the juice-transmitted
viruses, and such properties have proved of
considerable diagnostic value. This should
continue to be the case provided there is
general agreement that such data are not
precise, that some variation in results is to
be expected, that the data obtained do not
necessarily reflect exactly the intrinsic prop-
erties of the viruses themselves, and that
such tests are no more than aids in the
characterization of viruses. The chief value
of these in vitro properties lies in the fact
that, in general, strains (except those
variants lacking the specific protein) of a
given virus are similar with respect to
stability and to the extent to which they are
able to multiply or accumulate in a given
host species. It is true that many viruses
have similar thermal inactivation points and
dilution end points, yet there is a sufficient
range in these properties to make them
quite useful in identification. Furthermore,
information obtained from such tests is of
considerable value for other reasons; all in
vitro work with a virus begins with crude
juice, hence the virus worker should have
some prior information on virus stability
and concentration in such materials.
Chapter 4-A. F. Ross
Some care in selection of source material
is advisable. In general, source material
should consist of leaves from relatively
young plants showing typical symptoms.
Usually, those systemically infected leaves
showing the most distinct symptoms about
10-14 days following inoculation are used;
where necessary, inoculated leaves can be
used 7-10 days after inoculation. The
leaves are ground by mortar and pestle or
food chopper; the juice is separated from
the pulp by passage through cheesecloth
and used as soon as possible.
Since most of the published data have
been obtained in tests with untreated crude
juice, it seems desirable to continue on the
same basis. Bos et al. (1960) recently rec-
ommended that the tests be based on juice
diluted 1/10 with 0.01 M phosphate buffer
at pH 7. The advisability of this change in
procedure is doubtful. First, the data
obtained would not be comparable with
the bulk of the published data. Second,
phosphate is deleterious to some viruses.
Third, there is little evidence that the addi-
tion of this amount of phosphate to crude
juice would necessarily "stabilize pH and
ionic activity" in the crude juices from dif-
ferent species of plants. Precise character-
ization of the virus should be made with
purified viruses under known ionic condi-
tions; there seems little reason for attempt-
ing further rigorous standardization of these
rather crude tests that have in general
served the purpose for which they were
intended. It is important, however, that
reports of such tests carry full information
on the source and test species used and on
whether or not the crude juice was treated
in any way prior to the tests.
Slight modification may provide useful
information in special cases. Exposure of
the crude juice to oxidation will result in
the rapid inactivation of some viruses. Ad-
dition of reducing agents or of inhibitors of
oxidizing enzymes may prolong longevity,
raise the thermal inactivation point, and
increase the dilution end point of such vi-
ruses. Thus, addition of such agents in tests
with a particularly unstable virus may
materially aid in its characterization.
Until recently, the tests discussed here
have been applied for the most part to vi-
ruses that can be transmitted mechanically
from plant to plant. It is now possible to
apply these or similar tests to viruses that
can be mechanically transmitted from plant
to insect vector or from insect to insect.
Since treatments involving heating, ag-
ing, or diluting of infectious juice are often
used to separate viruses, it would be ad-
visable to make return inoculations from
plants inoculated with treated juice (but
still infectious) to the source species as a
check on the ability of the treated inoculum
to induce the original disease.
1. Thermal inactivation point.-When in-
fectious crude juice is heated, the rate of
virus inactivation is dependent not only
upon temperature but also upon virus con-
centration, pH, other materials present, etc.
The time required for complete inactiva-
tion, or inactivation to a certain point, is
affected by the same variables. Hence,
some variation due to the source species
can be expected, and dilution with water or
buffers would also introduce variations.
Consequently, it has been necessary to
choose an arbitrary time and to effect some
standardization of the type of extract used.
The most commonly accepted definition of
the thermal inactivation point of a virus (in
crude juice) is the temperature required
for the complete inactivation of the virus in
untreated crude juice during a 10-minute
In a typical test, the crude juice is pi-
petted in narrow thin-walled test tubes pre-
viously warmed to the desired temperature;
the tubes are placed in a constant-tempera-
ture water bath for exactly 10 minutes and
then removed and immediately cooled in
running water. The treated juice, and also
unheated juice, is then rubbed on the
leaves of suitable test plants, preferably
local-lesion hosts. If nothing is known about
the virus, a series of temperatures ranging
from 450 to 950 at 5 intervals is commonly
used; if something is known about the
identity of the virus, then it may be possible
to use a narrower range of temperatures.
The test should be repeated, but the sec-
ond test can be over a narrower tempera-
ture range and possibly at 2 intervals.
Five-ml serological tubes, with walls about
0.7 mm thick, are suitable. The juice (2 ml)
should be pipetted into the tubes in such a
manner that no juice is left on the inner
wall of the upper part of the tube. The
level of the water in the bath should be at
least 3 cm above the level of the juice.
In respect to their general behavior at
elevated temperatures, viruses may be
divided roughly into two groups. With those
that have a high temperature coefficient
(Qxio), such as TMV and potato virus X,
loss of infectivity is closely correlated with
denaturation (coagulation) and precedes it
by a very narrow margin. Others, such as
tobacco necrosis virus and tomato bushy
stunt virus, are characterized by a rela-
tively low Q1io, and virus inactivation pro-
ceeds well in advance of denaturation and
more or less independently of it. With these
viruses with a high Qio., temperature is the
most important single factor affecting the
amount of inactivation occurring in 10
minutes. Hence, most workers agree rea-
sonably well on the thermal inactivation
point of such viruses in crude juice. On the
other hand, temperature is of less relative
importance in determining the amount of
inactivation of the other group of viruses;
hence other factors become progressively
more important. As a result, the reported
thermal inactivation points of such viruses
vary considerably. If an "unknown" is
tentatively identified as a particular virus,
its thermal inactivation point should be in
reasonable agreement with the values re-
ported in the literature. If it is a virus like
tobacco mosaic virus or potato virus X,
agreement should usually be within 5.
With others, perhaps agreement within 100
should be expected. Large discrepancies
should be considered as good evidence but
not proof of the nonidentity of one virus
2. Other measures of heat stability.-The
stability of a virus at elevated temperatures
also can be characterized rather easily in
other ways, particularly if local-lesion as-
says are possible. From an equivalent
amount of data, for example, the velocity
constant of inactivation at a specified tem-
perature can be determined with greater
accuracy than can the thermal inactivation
point (Price, 1940). If velocity constants at
different temperatures are determined, the
temperature coefficient of inactivation, i.e.,
the ratio between velocity constants at two
different temperatures, can be calculated
(Price, 1940). The temperature coefficient
customarily determined is Qioo, the ratio of
velocity constants at temperatures differing
by 10C. Values for Qio are not constant
over any appreciable temperature range,
for they tend to increase as the temperature
is increased. Another value, the energy of
activation, is more nearly constant over dif-
ferent temperature ranges than is Qio and
can be calculated from the same data
Velocity constants, temperature coeffi-
cients, and the energy of activation have
been determined for only a few viruses
(tabulated by Pollard, 1953; and Bawden,
1950) and thus at present are of limited
value in identification. These measures of
heat stability would appear to be more
accurate and more informative than ther-
mal inactivation points and hence are of
potential use in identification. For the
present, parallel studies with an "unknown"
and known viruses, either unpurified or
purified, could be very useful in identifica-
3. Longevity in vitro.-Different viruses
behave differently when allowed to age in
vitro at room temperature. The thermal
inactivation point is not an index of ability
to retain infectivity at room temperature;
thus resistance to aging is yet another way
to characterize a virus. Loss of activity dur-
ing aging at room temperature may result
from thermal inactivation, i.e., the same
series of reactions that lead to rapid loss of
infectivity at or near the thermal inactiva-
tion point, or it may result from action by
microorganisms or from oxidative reactions.
Thus TMV is remarkably stable at room
temperature, even in crude juice, because
it is not particularly subject to any of the
three types of reactions. Tobacco necrosis
virus, which has the same thermal inactiva-
tion point as TMV, is much less stable than
is TMV at room temperature because ther-
mal inactivation proceeds at an appreciable
rate at room temperature. Potato virus X,
which is similar to TMV in that inactivation
and denaturation proceed at about the
same rate, is rather unstable at room tem-
Chapter 4-A. F. Ross
perature because it is subject to microbial
Tests on longevity in vitro are usually
made with untreated crude juice stored in
stoppered containers at room temperature
(20-22C). Samples are removed at in-
tervals and tested for infectivity on suitable
test plants. Unless something is known
about the general stability of the virus, the
first series of intervals should be at a
geometric progression (e.g., 1,2,4,8,16,32
. days) until infectivity is lost; shorter
intervals over a narrower range should then
be used in a confirmatory run.
As with thermal inactivation tests, results
may vary with a number of factors, such as
the exact temperature used, the source spe-
cies, the sensitivity of the assay, and the
microbial population of the juice. Reports
should always specify the source plant, test
plant, and temperature. Useful information
may be gained by varying the technique,
such as by adding reducing agents, by fil-
tration to sterilize the juice, or by adding
chemicals that inhibit bacteria and fungi;
if such things are done, they should be
described in detail in the report.
Small differences in longevity should not
be considered to be significant. In general,
longevity tests should be considered as
doing no more than characterizing a virus
in terms of whether its longevity in vitro
can be measured in minutes, hours, days,
weeks, months, or years.
Yarwood and Sylvester (1959) have sug-
gested that the relative stability of viruses
in vitro be measured and expressed in terms
of "half-life," i.e., the time required for loss
of one-half of the original activity. Nitzany
and Friedman (1963) investigated the ap-
plicability of the half-life concept to the
aging in vitro of several unpurified viruses
and concluded that half-life is the more ac-
curate expression of virus inactivation.
Whether or not this concept will prove
useful in identification work depends upon
how widely it is accepted and applied. It
is likely to be of limited use until the half-
life of the known viruses is reported, until
fiducial limits of the test are determined,
until reproducibility of results by different
investigators working with the same viruses
has been determined, and until a standard-
ized procedure is adopted. One should keep
in mind the fact that the accuracy of a half-
life test will be no greater than the accuracy
of the assay. Also, tests on crude juice are
still replete with uncertainties and probably
will remain of limited diagnostic value.
4. Dilution end point.-The extent to
which a virus multiplies and/or accumu-
lates in a given host species is somewhat
characteristic of the virus and most of its
strains. Hence, a simple dilution test may
serve as yet another pointer in attempts to
identify a virus.
Dilution is usually on a logarithmic scale
(i.e., 1/1, 1/10, 1/100, 1/1000, . .
1/10,000,000), and each dilution should be
rubbed on an equal number of uniform test
plants. Water is the recommended diluent.
The dilution end point is generally re-
ported as being between two dilutions, i.e.,
between the highest dilution that was still
infectious and the next highest one.
Like the other two tests on properties of
viruses in crude juice, this test also is rather
crude and is subject to several sources of
variation. In addition to the identity of host
and test species, the environmental condi-
tions under which each is grown will in-
fluence the results. The temperature at
which the source plant is maintained is of
particular importance, for temperature can
have a very marked effect on virus titer.
Hence, the experimental conditions should
always be reported. Since considerable var-
iation is to be expected, dilutions at less
than 10-fold steps are not necessary. In in-
terpreting results, one should recognize that
the test is designed to differentiate among
viruses that differ greatly in dilution end
point. The dilution end point, as deter-
mined by the prescribed method, should
fall within the range reported in the litera-
ture, and 10- or 100-fold differences should
be given no particular significance.
C. TRANSMISSION CHARACTErSTICS
The extent to which facts about the
transmission of a virus are used in identifi-
cation depends largely upon the nature of
the virus. Transmission characteristics may
play a very minor role in the identification
of a mechanically transmissible virus, for
many other tools are available. On the other
hand, transmission characteristics are of
major importance in the identification of
many viruses that are transmitted by spe-
cific vectors, that cannot be transmitted by
mechanical means, and that cannot as yet
be worked with serologically. Barley yellow
dwarf virus, for instance, can be identified
only on the basis of symptomatology, host
range, the aphid species capable of trans-
mitting it, and the virus-vector relationship.
Even with mechanically transmitted vi-
ruses, most of which are transmitted by
living vectors, data on transmissibility may
be of material aid in identification. On the
one hand, these viruses can be grouped
according to the relative ease of transmis-
sion by the usual mechanical methods and
also according to whether or not transmis-
sion is facilitated by the addition of reduc-
ing agents or by phenol extraction. On the
other hand, identification of one or more
vectors of a virus would identify the virus
as belonging to a certain group and thus
greatly narrow the possible identities that
might be considered. A tentative identifica-
tion could be strengthened significantly by
the discovery of one or more vectors, or it
could be invalidated by this.
Identification of a vector is almost an
essential step in the identification of many
vector-borne viruses, i.e., those that are not
antigenic and are difficult to characterize
in vitro. An identification of an "unknown,"
as barley yellow dwarf virus for example,
would be incomplete unless it is shown to
be transmitted by one of the known aphid
vectors of this virus.
Discovery of the vector of a virus is
usually followed by studies on virus-vector
relations; such studies are likely to aid in
the identification of the virus. Of particular
value would be information on whether the
virus is persistent or nonpersistent in the
vector, on the duration of the latent period
(if such exists), on how long after acquisi-
tion the vector retains the ability to trans-
mit, and on whether or not the virus mul-
tiplies within the vector.
The extent to which discovery of the
vector of an unknown virus will contribute
to its identification is related to the number
of viruses transmitted by that vector. The
green peach aphid, Myzus persicae Sulzer,
transmits 30 or more plant viruses, thus
demonstration that an "unknown" is trans-
mitted by this vector would be a relatively
small step toward its identification. In con-
trast, tomato spotted wilt virus is the only
one known to be transmitted by thrips;
thus only a minimum amount of further
characterization would be needed to estab-
lish the identity of a thrips-borne "un-
known" as tomato spotted wilt virus.
Transmission characteristics are some-
times of value in differentiation among
strains of the same virus. Mutants devoid of
the specific protein coat may be difficult to
transmit mechanically even though the
parent virus is easily transmitted in this
manner (Sainger and Brandenburg, 1961;
Cadman, 1962; Babos and Kassanis, 1962).
Potato virus C, a strain of potato virus Y, is
not transmitted by the green peach aphid,
the usual vector of potato virus Y (Bawden
and Kassanis, 1947). One strain of potato
yellow dwarf virus is transmitted specifi-
cally by one leafhopper species, and yet
another strain is transmitted specifically by
another species (Black, 1941). Similarly,
there are several vector-specific strains of
barley yellow dwarf virus, and these strains
can be differentiated only on the basis of
their specific aphid vectors (Rochow,
D. INTERACTION WITH OTHER VIRUSES
Various aspects of virus interaction fol-
lowing simultaneous or sequential inocula-
tion with virus pairs may be utilized to
obtain information on the identity of one
member of the pair. For the most part,
tests of this nature are designed to detect
either relatedness or unrelatedness of the
"unknown" and a "known" virus. Some-
times, however, the way in which the "un-
known" interacts with a specific virus may
help to establish the identity of the "un-
1. Cross protection.-The phenomenon of
cross protection is discussed fully by W. C.
Price, and his chapter should be consulted
for additional information. The subject will
be dealt with here in bare outline only, with
emphasis on the role of cross protection in
identification, on its limitations, and on
some aspects of interpretation.
Chapter 4-A. F. Ross
As might be expected, a plant completely
infected by a virus suffers no further dam-
age if reinoculated with the same virus. It
is not surprising, then, that prior infection
with one virus affords protection against
closely related ones. This principle is the
basis for the so-called cross-protection test,
which is one of the more important and
widely used tools for determining the
identity of a virus. Basically, the cross-
protection test is a test for the relatedness
of two viruses. The test consists essentially
of challenge inoculation with one virus of
tissue previously infected by another. Thus,
it can be used most effectively only after
the identity of an "unknown" has been
fairly well established by other methods
and the investigator is ready to test the
hypothesis that it is a particular virus. The
success of such tests depends upon several
things, the most basic of which is that the
second or challenge virus be one whose
multiplication or effects can be recognized
in the presence of the first virus.
The test would be of little value if it were
not for the additional fact that protection
does not ordinarily occur between unre-
lated viruses. A plant infected by one virus
is almost always susceptible to an unrelated
virus, i.e., the second virus is almost always
able to become established and multiply
despite the presence of the first. There is
much variation in the extent to which the
first interferes with the multiplication of the
second and with its ability to induce
characteristic symptoms. Depending upon
the virus used, the first may have very
little or no effect on the behavior of the
second, it may be appreciably inhibitory
(Thomson, 1958), or it may have a stimula-
tory effect (Thomson, 1961). In the plan-
ning and interpretation of cross-protection
tests, one should always bear in mind the
fact that the control (healthy) and the test
plant (infected by the first virus) are en-
tirely different test materials and that they
will not necessarily provide environments
equally suitable for the establishment and
propagation of the second virus.
Prior infection by one virus rarely results
in total exclusion of an unrelated virus,
hence some interference between unrelated
viruses would be of little consequence if
complete protection were always afforded
against a related virus. Unfortunately, this
is not always so. The degree of protection
infection by one strain confers against the
effects of another strain depends upon the
closeness of the relationship of the two vi-
ruses (Matthews, 1949, 1957) and on the
completeness of infection of the test tissue
by the first virus. In addition, it sometimes
depends on the host species used, the
method by which the second virus is intro-
duced, and the order in which the two vi-
ruses are used. Thus, results with some
unrelated pairs used under certain condi-
tions may be indistinguishable from those
with some pairs of related viruses.
Anyone attempting cross-protection tests
should keep in mind the fact that cross pro-
tection is relative. The fact that results are
not always clear-cut does not mean that
cross-protection tests are useless. It does
mean, however, that failure to demonstrate
cross protection does not necessarily indi-
cate unrelatedness and that such "negative
results," as well as partial or incomplete
protection, should be interpreted with cau-
tion. On the other hand, positive results,
i.e., complete protection, can usually be
depended upon as indicating relatedness.
Results of reciprocal or "mirror" tests in
particular are highly dependable. If A pro-
tects completely against B, and B gives
similar protection against A, the only valid
conclusion is that A and B are identical or
closely related. Relatedness would be
strongly indicated where protection is com-
plete in one direction but absent or incom-
plete in the other. One should remember,
however, that unilateral protection involv-
ing three unrelated viruses has been
reported from one laboratory (Bawden and
Kassanis, 1941, 1945), although the
results could not be confirmed in another
(Schmelzer et al., 1960). The safest proce-
dure, of course, is to consider results of
cross-protection tests as evidence, not abso-
lute proof, and to base final conclusions on
several different criteria.
Since positive results are so much more
dependable than negative ones, the investi-
gator should choose the best available
method for his test. Cross protection is rela-
tive, hence it is highly desirable to use a