Neuroimaging Part II Joseph C.

Masdeu
And R. Gilberto González (Eds.)
Visit to download the full and correct content document:
https://textbookfull.com/product/neuroimaging-part-ii-joseph-c-masdeu-and-r-gilberto-
gonzalez-eds/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Neuroimaging Part II 1st Edition Joseph C. Masdeu And

R. Gilberto González (Eds.)

https://textbookfull.com/product/neuroimaging-part-ii-1st-
edition-joseph-c-masdeu-and-r-gilberto-gonzalez-eds/

Neuroimaging Part I Joseph C. Masdeu And R. Gilberto

González (Eds.)

https://textbookfull.com/product/neuroimaging-part-i-joseph-c-
masdeu-and-r-gilberto-gonzalez-eds/

Neuroimaging Part I 1st Edition Joseph C. Masdeu And R.

Gilberto González (Eds.)

https://textbookfull.com/product/neuroimaging-part-i-1st-edition-
joseph-c-masdeu-and-r-gilberto-gonzalez-eds/

Cognitive Systems and Signal Processing 4th

International Conference ICCSIP 2018 Beijing China
November 29 December 1 2018 Revised Selected Papers
Part II Fuchun Sun
https://textbookfull.com/product/cognitive-systems-and-signal-
processing-4th-international-conference-iccsip-2018-beijing-
china-november-29-december-1-2018-revised-selected-papers-part-
Voice Therapy Clinical Case Studies Joseph C. Stemple

https://textbookfull.com/product/voice-therapy-clinical-case-
studies-joseph-c-stemple/

Digital Heritage Progress in Cultural Heritage

Documentation Preservation and Protection 7th
International Conference EuroMed 2018 Nicosia Cyprus
October 29 November 3 2018 Proceedings Part II Marinos
Ioannides
https://textbookfull.com/product/digital-heritage-progress-in-
cultural-heritage-documentation-preservation-and-protection-7th-
international-conference-euromed-2018-nicosia-cyprus-
october-29-november-3-2018-proceedings-part-ii-marinos/

Internet of Things IoT Infrastructures Second

International Summit IoT 360 2015 Rome Italy October 27
29 2015 Revised Selected Papers Part II 1st Edition
Benny Mandler
https://textbookfull.com/product/internet-of-things-iot-
infrastructures-second-international-summit-iot-360-2015-rome-
italy-october-27-29-2015-revised-selected-papers-part-ii-1st-
edition-benny-mandler/

Advances in Multimedia Information Processing – PCM

2017: 18th Pacific-Rim Conference on Multimedia,
Harbin, China, September 28-29, 2017, Revised Selected
Papers, Part II Bing Zeng
https://textbookfull.com/product/advances-in-multimedia-
information-processing-pcm-2017-18th-pacific-rim-conference-on-
multimedia-harbin-china-september-28-29-2017-revised-selected-
papers-part-ii-bing-zeng/

Neuroimaging and neurophysiology in psychiatry 1st

Edition Linden

https://textbookfull.com/product/neuroimaging-and-
neurophysiology-in-psychiatry-1st-edition-linden/
 HANDBOOK OF CLINICAL
NEUROLOGY

Series Editors

MICHAEL J. AMINOFF, FRANÇOIS BOLLER, AND DICK F. SWAAB

VOLUME 136

AMSTERDAM BOSTON HEIDELBERG LONDON NEW YORK OXFORD

PARIS SAN DIEGO SAN FRANCISCO SINGAPORE SYDNEY TOKYO
ELSEVIER
Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, USA

© 2016 Elsevier B.V. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical,
including photocopying, recording, or any information storage and retrieval system, without permission in writing
from the publisher. Details on how to seek permission, further information about the Publisher’s permissions
policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright
Licensing Agency, can be found at our website: www.elsevier.com/permissions.

This book and the individual contributions contained in it are protected under copyright by the Publisher (other than
as may be noted herein).

Notices
Knowledge and best practice in this field are constantly changing. As new research and experience
broaden our understanding, changes in research methods, professional practices, or medical treatment may
become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any
information, methods, compounds, or experiments described herein. In using such information or methods they
should be mindful of their own safety and the safety of others, including parties for whom they have a professional
responsibility.
With respect to any drug or pharmaceutical products identified, readers are advised to check the most current
information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to
verify the recommended dose or formula, the method and duration of administration, and contraindications. It is
the responsibility of practitioners, relying on their own experience and knowledge of their patients, to make
diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety
precautions.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability
for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or
from any use or operation of any methods, products, instructions, or ideas contained in the material herein.

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress

ISBN: 978-0-444-53486-6

For information on all Elsevier publications

visit our website at https://www.elsevier.com/

Publisher: Shirley Decker-lucke

Acquisition Editor: Mara Conner
Editorial Project Manager: Kristi Anderson
Production Project Manager: Sujatha Thirugnana Sambandam
Designer: Alan Studholme

Typeset by SPi Global, India

Handbook of Clinical Neurology 3rd Series

Available titles
Vol. 79, The human hypothalamus: basic and clinical aspects, Part I, D.F. Swaab, ed. ISBN 9780444513571
Vol. 80, The human hypothalamus: basic and clinical aspects, Part II, D.F. Swaab, ed. ISBN 9780444514905
Vol. 81, Pain, F. Cervero and T.S. Jensen, eds. ISBN 9780444519016
Vol. 82, Motor neurone disorders and related diseases, A.A. Eisen and P.J. Shaw, eds. ISBN 9780444518941
Vol. 83, Parkinson’s disease and related disorders, Part I, W.C. Koller and E. Melamed, eds. ISBN 9780444519009
Vol. 84, Parkinson’s disease and related disorders, Part II, W.C. Koller and E. Melamed, eds. ISBN 9780444528933
Vol. 85, HIV/AIDS and the nervous system, P. Portegies and J. Berger, eds. ISBN 9780444520104
Vol. 86, Myopathies, F.L. Mastaglia and D. Hilton Jones, eds. ISBN 9780444518996
Vol. 87, Malformations of the nervous system, H.B. Sarnat and P. Curatolo, eds. ISBN 9780444518965
Vol. 88, Neuropsychology and behavioural neurology, G. Goldenberg and B.C. Miller, eds. ISBN 9780444518972
Vol. 89, Dementias, C. Duyckaerts and I. Litvan, eds. ISBN 9780444518989
Vol. 90, Disorders of consciousness, G.B. Young and E.F.M. Wijdicks, eds. ISBN 9780444518958
Vol. 91, Neuromuscular junction disorders, A.G. Engel, ed. ISBN 9780444520081
Vol. 92, Stroke – Part I: Basic and epidemiological aspects, M. Fisher, ed. ISBN 9780444520036
Vol. 93, Stroke – Part II: Clinical manifestations and pathogenesis, M. Fisher, ed. ISBN 9780444520043
Vol. 94, Stroke – Part III: Investigations and management, M. Fisher, ed. ISBN 9780444520050
Vol. 95, History of neurology, S. Finger, F. Boller and K.L. Tyler, eds. ISBN 9780444520081
Vol. 96, Bacterial infections of the central nervous system, K.L. Roos and A.R. Tunkel, eds. ISBN 9780444520159
Vol. 97, Headache, G. Nappi and M.A. Moskowitz, eds. ISBN 9780444521392
Vol. 98, Sleep disorders Part I, P. Montagna and S. Chokroverty, eds. ISBN 9780444520067
Vol. 99, Sleep disorders Part II, P. Montagna and S. Chokroverty, eds. ISBN 9780444520074
Vol. 100, Hyperkinetic movement disorders, W.J. Weiner and E. Tolosa, eds. ISBN 9780444520142
Vol. 101, Muscular dystrophies, A. Amato and R.C. Griggs, eds. ISBN 9780080450315
Vol. 102, Neuro-ophthalmology, C. Kennard and R.J. Leigh, eds. ISBN 9780444529039
Vol. 103, Ataxic disorders, S.H. Subramony and A. Durr, eds. ISBN 9780444518927
Vol. 104, Neuro-oncology Part I, W. Grisold and R. Sofietti, eds. ISBN 9780444521385
Vol. 105, Neuro-oncology Part II, W. Grisold and R. Sofietti, eds. ISBN 9780444535023
Vol. 106, Neurobiology of psychiatric disorders, T. Schlaepfer and C.B. Nemeroff, eds. ISBN 9780444520029
Vol. 107, Epilepsy Part I, H. Stefan and W.H. Theodore, eds. ISBN 9780444528988
Vol. 108, Epilepsy Part II, H. Stefan and W.H. Theodore, eds. ISBN 9780444528995
Vol. 109, Spinal cord injury, J. Verhaagen and J.W. McDonald III, eds. ISBN 9780444521378
Vol. 110, Neurological rehabilitation, M. Barnes and D.C. Good, eds. ISBN 9780444529015
Vol. 111, Pediatric neurology Part I, O. Dulac, M. Lassonde and H.B. Sarnat, eds. ISBN 9780444528919
Vol. 112, Pediatric neurology Part II, O. Dulac, M. Lassonde and H.B. Sarnat, eds. ISBN 9780444529107
Vol. 113, Pediatric neurology Part III, O. Dulac, M. Lassonde and H.B. Sarnat, eds. ISBN 9780444595652
Vol. 114, Neuroparasitology and tropical neurology, H.H. Garcia, H.B. Tanowitz and O.H. Del Brutto, eds.
ISBN 9780444534903
Vol. 115, Peripheral nerve disorders, G. Said and C. Krarup, eds. ISBN 9780444529022
Vol. 116, Brain stimulation, A.M. Lozano and M. Hallett, eds. ISBN 9780444534972
Vol. 117, Autonomic nervous system, R.M. Buijs and D.F. Swaab, eds. ISBN 9780444534910
Vol. 118, Ethical and legal issues in neurology, J.L. Bernat and H.R. Beresford, eds. ISBN 9780444535016
Vol. 119, Neurologic aspects of systemic disease Part I, J. Biller and J.M. Ferro, eds. ISBN 9780702040863
Vol. 120, Neurologic aspects of systemic disease Part II, J. Biller and J.M. Ferro, eds. ISBN 9780702040870
Vol. 121, Neurologic aspects of systemic disease Part III, J. Biller and J.M. Ferro, eds. ISBN 9780702040887
Vol. 122, Multiple sclerosis and related disorders, D.S. Goodin, ed. ISBN 9780444520012
Vol. 123, Neurovirology, A.C. Tselis and J. Booss, eds. ISBN 9780444534880
vi AVAILABLE TITLES (Continued)
Vol. 124, Clinical neuroendocrinology, E. Fliers, M. Korbonits and J.A. Romijn, eds. ISBN 9780444596024
Vol. 125, Alcohol and the nervous system, E.V. Sullivan and A. Pfefferbaum, eds. ISBN 9780444626196
Vol. 126, Diabetes and the nervous system, D.W. Zochodne and R.A. Malik, eds. ISBN 9780444534804
Vol. 127, Traumatic brain injury Part I, J.H. Grafman and A.M. Salazar, eds. ISBN 9780444528926
Vol. 128, Traumatic brain injury Part II, J.H. Grafman and A.M. Salazar, eds. ISBN 9780444635211
Vol. 129, The human auditory system: Fundamental organization and clinical disorders, G.G. Celesia
and G. Hickok, eds. ISBN 9780444626301
Vol. 130, Neurology of sexual and bladder disorders, D.B. Vodušek and F. Boller, eds. ISBN 9780444632470
Vol. 131, Occupational neurology, M. Lotti and M.L. Bleecker, eds. ISBN 9780444626271
Vol. 132, Neurocutaneous syndromes, M.P. Islam and E.S. Roach, eds. ISBN 9780444627025
Vol. 133, Autoimmune neurology, S.J. Pittock and A. Vincent, eds. ISBN 9780444634320
Vol. 134, Gliomas, M.S. Berger and M. Weller, eds. ISBN 9780128029978
Vol. 135, Neuroimaging Part I, J.C. Masdeu and R.G. González, eds. ISBN 9780444534859
 Foreword

We are proud to present the first volumes dedicated to neuroimaging in the Handbook of Clinical Neurology series.
Neurologists, not just those in training, may wonder at the kind of medical world that existed before modern imaging.
Indeed, the neuroscience community has since its beginning attempted to understand the human mind and brain
through imaging. As far back as 1880, the Italian physiologist Angelo Mosso introduced the “human circulation
balance,” which could noninvasively measure the redistribution of blood during emotional and intellectual activity.
More recently, semi-invasive techniques such as pneumoencephalography (introduced by Dandy in 1918) and arteri-
ography (pioneered by Moniz in 1927) allowed partial visualization of the brain and its surrounding structures. New
methods enabling easier, safer, noninvasive, painless, and repeatable imaging have only been developed in the past
50 years or so, starting with computed tomography, some of whose developers won the 1979 Nobel Prize for medicine
or physiology. The many subsequent developments in neuroimaging are masterfully presented in these two volumes.
The volumes deal with a variety of neuroimaging-related topics. They include thorough descriptions of the involved
methods and their application to specific diseases of the brain, spinal cord, and peripheral nervous system. Emphasis is
given to the most common disorders, such as tumors, strokes, multiple sclerosis, movement disorders, infections,
dementia, and trauma, but less common conditions such as neurocutaneous syndromes are also discussed. The impor-
tant questions of when and where to image, as well as the differential diagnosis of imaging findings, are discussed in
the light of specific syndromes. A separate section covers pediatric neuroimaging. The volumes conclude with sections
dedicated to interventional neuroimaging as well as to postmortem imaging and neuropathologic correlations.
We have been fortunate to have as volume editors two distinguished scholars, Dr. Joseph C. Masdeu, of the Depart-
ment of Neurology, Methodist Hospital, Houston, Texas, and Dr. R. Gilberto González, from the Department of Radi-
ology, Massachusetts General Hospital in Boston. Both have been at the forefront of neuroimaging research for many
years. They have assembled a truly international group of authors with acknowledged expertise to contribute to the
texts and have produced two authoritative, comprehensive, and up-to-date volumes. Their availability electronically on
Elsevier’s Science Direct site as well as in print format should ensure their ready accessibility and facilitate searches for
specific information.
We are grateful to the volume editors and to all the contributors for their efforts in creating such an invaluable
resource. As series editors we read and commented on each of the chapters with great interest. We are therefore con-
fident that both clinicians and researchers in many different medical disciplines will find much in these volumes to
appeal to them.
And last, but not least, it is always a pleasure to acknowledge and thank Elsevier, our publisher – and, in particular,
Michael Parkinson in Scotland, and Mara Conner and Kristi Anderson in San Diego – for their unfailing and expert
assistance in the development and production of these volumes.
Michael J. Aminoff
François Boller
Dick F. Swaab
 Preface

Neuroimaging has become one of the most useful set of tools for understanding and diagnosing diseases of the ner-
vous system. Fittingly, the present two volumes of the Handbook of Clinical Neurology review the extensive advances
in the field. In the first volume, discussions of the various techniques used in neuroimaging are followed by reviews of
the imaging findings caused by brain diseases. We have chosen not to include a chapter on brain anatomy because it
would be quite long and extant atlases are excellent. The second volume begins with a description of the functional
anatomy of the spine and of the imaging findings in diseases of the spine and spinal cord. Imaging of peripheral nerve
and muscle follows. Then, there is a section on when and how to image the various clinical syndromes produced by
diseases of the nervous system. Adequacy in the use of expensive neuroimaging tools has always been a priority, but it
is becoming more acute as the application of neuroimaging expands more rapidly than the available resources. The
next section is unusual in a book of this type: a description of the various imaging findings that should lead to con-
sideration of the diseases causing them. Such information is particularly important when using techniques like com-
puted tomography and magnetic resonance imaging, which offer a panoply of findings and are extensively used in
clinical practice. Next is a section on pediatric neuroimaging, led by a chapter on imaging findings during normal
development. After three chapters on the therapeutic use of endovascular imaging, the second volume concludes with
a chapter on postmortem imaging as a tool to better define normal brain structure on imaging and its alteration by
some disorders.
We hope that this book will be useful to all those who work with clinical imaging of the nervous system, such as
neurologists, neuroradiologists, neurosurgeons, and nuclear medicine physicians. Some sections, for instance, the sec-
tions on the spine, peripheral nerve, and muscle, may be helpful to orthopedic surgeons and rehabilitation specialists.
Neuropsychologists may find helpful the chapters on neurodegenerative disorders leading to cognitive impairment.
Neuroimaging is used not only clinically, but also by those interested in clarifying the still largely undiscovered
landscape and functional intricacy of the brain. While the clinical literature of neuroimaging is extensive, even more
extensive and more widely cited is the literature of neuroimaging applied to the study of the healthy human nervous
system. Although human disease has traditionally led to a better understanding of normal structure and function,
researchers looking primarily for information on the normal nervous system should look elsewhere.
We are most thankful to the authors, who have distilled their expertise in superbly written and illustrated chapters.
Mr. Michael Parkinson, from Elsevier, has skillfully coordinated the gathering of information for these two volumes.
We are also thankful to the three series editors and, particularly, to Dr. François Boller, for their excellent suggestions.
Joseph C. Masdeu
R. Gilberto González
 Contributors
F. Agosta
Neuroimaging Research Unit, Institute of Experimental
Neurology, Division of Neuroscience, San Raffaele
Scientific Institute, Vita-Salute San Raffaele University,
Milan, Italy
C. Amlie-Lefond
Department of Neurology, Seattle Children’s Hospital,
Seattle, WA, USA
A. Atri
Ray Dolby Brain Health Center, California Pacific
Medical Center Research Institute, Sutter Health,
San Francisco, CA, USA
J.C. Augustinack
Athinoula A. Martinos Center for Biomedical Imaging,
Department of Radiology, Massachusetts General
Hospital, Charlestown, MA, USA
A. Bali
Department of Radiology, Antwerp University Hospital
and University of Antwerp, Antwerp, Belgium
A.M. Blitz
Neuro-radiology Division, Johns Hopkins University
School of Medicine, Baltimore, MD, USA
N. Bogduk
Newcastle Bone and Joint Institute, University of
Newcastle, Newcastle, Australia
A. Boon
Department of Physical Medicine and Rehabilitation and
Department of Neurology, Mayo Clinic, Rochester, MN,
USA
B.H. Brinkmann
Division of Epilepsy, Department of Neurology, Mayo
Clinic, Rochester, MN, USA
H. Brunel
Neuroradiology Service, H^opital la Timone, Marseille,
France
N.D. Bryant
Vanderbilt University Institute of Imaging Science and
the Department of Radiology and Radiological Sciences,
Vanderbilt University, Nashville, TN, USA
P.M. Capone
Medical Imaging and Neurology, Winchester Medical
Center, Winchester and Department of Neurology,
Virginia Commonwealth University, Richmond, VA, USA
G.D. Cascino
Division of Epilepsy, Department of Neurology, Mayo
Clinic, Rochester, MN, USA
M. Castillo
Division of Neuroradiology, Department of Radiology,
University of North Carolina, Chapel Hill, NC, USA
F. Cendes
University of Campinas, Department of Neurology,
Campinas, SP, Brazil
C.T. Chin
Department of Radiology, University of California,
San Francisco, CA, USA
H.M. Dahmoush
Department of Radiology, Children’s Hospital of
Philadelphia and University of Pennsylvania,
Philadelphia, PA, USA
B.M. Damon
Vanderbilt University Institute of Imaging Science and
the Department of Radiology and Radiological Sciences,
Departments of Biomedical Engineering and Molecular
Physiology and Biophysics, Vanderbilt University,
Nashville, TN, USA
xii CONTRIBUTORS
C.C. Della Santina P. Hedera
Department of Otolaryngology, Head and Neck Department of Neurology, Vanderbilt University,
Surgery, Johns Hopkins University School of Medicine, Nashville, TN, USA
Baltimore, MD, USA
J.A. Hirsch
O. Diaz Neuroendovascular Service, Massachusetts General
Neurovascular Center, Methodist Hospital, Houston, Hospital, Boston, MA, USA
TX, USA
T.A.G.M. Huisman
C. Ellenberger Division of Pediatric Radiology, Johns Hopkins
Lebanon Magnetic Imaging, Lebanon, Pennsylvania, Hospital, Baltimore, MD, USA
USA
J.V. Hunter
Department of Pediatric Radiology, Texas Children’s
A. Faje
Hospital and Baylor College of Medicine, Houston, TX,
Neuroendocrine Unit, Massachusetts General Hospital
USA
and Harvard Medical School, Boston, MA, USA
M. Kellogg
A. Faridar Department of Neurology, Oregon Health and Science
Department of Neurology, Houston Methodist Hospital, University, Portland, OR, USA
Houston, TX, USA
F.J. Kirkham
M. Filippi Neurosciences Unit, UCL Institute of Child Health,
Neuroimaging Research Unit, Institute of Experimental London; Southampton Children’s Hospital and Clinical
Neurology, Division of Neuroscience, San Raffaele and Experimental Sciences, University of Southampton,
Scientific Institute, Vita-Salute San Raffaele University, Southampton, UK
Milan, Italy
J.P. Klein
C.R. Fitz Departments of Neurology and Radiology, Brigham and
Department of Radiology, Children’s Hospital of Women’s Hospital and Harvard Medical School,
Pittsburgh, Pittsburgh, PA, USA Boston, MA, USA

A.E. Flanders A. Klibanski

Department of Radiology, Thomas Jefferson University Neuroendocrine Unit, Massachusetts General Hospital
Hospital, Philadelphia, PA, USA and Harvard Medical School, Boston, MA, USA

A.D. Furtado M. Kliot

Department of Radiology, Children’s Hospital of Department of Neurological Surgery, Northwestern
Pittsburgh, Pittsburgh, PA, USA Feinberg School of Medicine, Chicago, IL, USA

R. Gadhia M. Koob
Department of Neurology, Houston Methodist Hospital, Pediatric Radiology Imaging Service, Centre Hospitalier
Houston, TX, USA Universitaire de Strasbourg, H^opital de Hautepierre and
Laboratoire ICube, Universite de Strasbourg-CNRS,
N. Girard Strasbourg, France
Neuroradiology Service, H^ opital la Timone and Aix
Marseille Universite, Marseille, France T. Krings
Division of Neuroradiology and Neurosurgery,
L. Goerner University of Toronto, Toronto Western Hospital and
The Radiology Group, Honolulu, HI, USA University Health Network, Toronto, ON, Canada

R.G. González T. Leslie-Mazwi

Department of Radiology, Massachusetts General Neuroendovascular Service, Massachusetts General
Hospital, Boston, MA, USA Hospital, Boston, MA, USA
 CONTRIBUTORS xiii
K. Li P.M. Parizel
Vanderbilt University Institute of Imaging Science and Department of Radiology, Antwerp University Hospital
the Department of Radiology and Radiological Sciences, and University of Antwerp, Antwerp, Belgium
Vanderbilt University, Nashville, TN, USA
T.J. Pfiffner
C.W. Liang Dent Neurologic Institute, Amherst, NY, USA
Neurovascular Imaging Research Core and Department
of Neurology, University of California Los Angeles, S. Pillen
Los Angeles, CA, USA Department of Sleep Medicine, Kempenhaeghe
Expertise Center for Epileptology, Sleep Medicine and
D.S. Liebeskind Neurocognition, Heeze, The Netherlands
Neurovascular Imaging Research Core and Department
N.K. Pinter
of Neurology, University of California Los Angeles,
Dent Neurologic Institute, Amherst, NY, USA
Los Angeles, CA, USA
A. Poretti
B.O. Marquis
Division of Pediatric Radiology, Johns Hopkins
Division of Epilepsy, Department of Neurology, State
Hospital, Baltimore, MD, USA
University of New York Downstate Medical Center,
Brooklyn, NY, USA
J. Rabinov
Neuroendovascular Service, Massachusetts General
J.C. Masdeu
Hospital, Boston, MA, USA
Department of Neurology, Houston Methodist Hospital,
Houston, TX, USA
L. Rangel-Castilla
Neuroendovascular Surgery, University of Buffalo
M.J. McKeown
Neurosurgery, State University of New York, Buffalo,
Pacific Parkinson’s Research Centre and Division of
NY, USA
Neurology, Djavad Mowafaghian Centre for Brain
Health, University of British Columbia and Vancouver
H.L. Rekate
Coastal Health, Vancouver, BC, Canada
Department of Neurosurgery, Hofstra Northshore LIJ
School of Medicine, Manhasset, and The Chiari
L.L. Mechtler
Institute, Great Neck, NY, USA
Dent Neurologic Institute, Amherst, NY, USA
D. Rodriguez-Luna
E.R. Melhem
Stroke Unit, Department of Neurology, Vall d’Hebron
Department of Diagnostic Radiology and Nuclear
University Hospital, Vall d’Hebron Research Institute,
Medicine, University of Maryland, Baltimore, MD, USA
Autonomous University of Barcelona, Barcelona, Spain
C.A. Molina
R. Ruff
Stroke Unit, Department of Neurology, Vall d’Hebron
Department of Neurology, Case Western Reserve
University Hospital, Vall d’Hebron Research Institute,
University School of Medicine, Cleveland, OH, USA
Autonomous University of Barcelona, Barcelona, Spain
G. Saliou
D.E. Newman-Toker Neuroradiology Service, Centre Hospitalier
Department of Neurology and Department of Universitaire Bic^etre, Le Kremlin Bic^etre, France
Otolaryngology, Head and Neck Surgery, Johns Hopkins
University School of Medicine, Baltimore, MD, USA R. Scranton
Department of Neurosurgery, Methodist Hospital,
R.B. Olatunji Houston, TX, USA
Department of Radiology, University Hospital Vienna,
Vienna, Austria S.M. Selkirk
Spinal Cord Injury Division, Louis Stokes Cleveland
A. Panigrahy Veterans Affairs Medical Center and Department of
Department of Radiology, Children’s Hospital of Neurology, Case Western Reserve University School of
Pittsburgh, Pittsburgh, PA, USA Medicine, Cleveland, OH, USA
xiv CONTRIBUTORS
P. Shankar
Division of Neuroradiology, Department of Radiology,
University of North Carolina, Chapel Hill, NC, USA
D. Shaw
Department of Radiology, Seattle Children’s Hospital,
Seattle, WA, USA
N.G. Simon
St Vincent’s Clinical School, University of New South
Wales, Sydney, Australia
A.J. Stoessl
Pacific Parkinson’s Research Centre and Division of
Neurology, Djavad Mowafaghian Centre for Brain
Health, University of British Columbia and Vancouver
Coastal Health, Vancouver, BC, Canada
N.A. Tritos
Neuroendocrine Unit, Massachusetts General
Hospital and Harvard Medical School, Boston, MA,
USA
N. van Alfen
Department of Neurology, Radboud University Medical
Center, Nijmegen, The Netherlands
L. van den Hauwe
Department of Radiology, Antwerp University Hospital
and University of Antwerp, Antwerp, Belgium
A.J.W. van der Kouwe
Athinoula A. Martinos Center for Biomedical Imaging,
Department of Radiology, Massachusetts General
Hospital, Charlestown, MA, USA

V. Sulc J. Van Goethem

Department of Neurology, 2nd Faculty of Medicine,
Charles University in Prague and Motol University
Hospital, Czech Republic
B. Swearingen
Department of Neurosurgery, Massachusetts General
Hospital and Harvard Medical School, Boston, MA, USA
J. Talbott
Department of Radiology, University of California,
San Francisco, CA, USA
Department of Radiology, Antwerp University Hospital
and University of Antwerp, Antwerp, Belgium
A.J.L. Van Hoyweghen
Department of Radiology, Antwerp University Hospital
and University of Antwerp, Antwerp, Belgium
A. Vossough
Department of Radiology, Children’s Hospital of
Philadelphia and University of Pennsylvania,
Philadelphia, PA, USA

W.H. Theodore C. Zamora

National Institute of Neurological Disorders and Stroke, Division of Neuroradiology, Department of Radiology,
Bethesda, MD, USA University of North Carolina, Chapel Hill, NC, USA

M.M. Thurnher V.M. Zohrabian

Department of Radiology, University Hospital Vienna, Department of Diagnostic Radiology, Yale University
Vienna, Austria School of Medicine, New Haven, CT, USA
Handbook of Clinical Neurology, Vol. 136 (3rd series)
Neuroimaging, Part II
J.C. Masdeu and R.G. González, Editors
© 2016 Elsevier B.V. All rights reserved

Chapter 32

Functional anatomy of the spine

NIKOLAI BOGDUK*
Newcastle Bone and Joint Institute, University of Newcastle, Newcastle, Australia

Abstract
Among other important features of the functional anatomy of the spine, described in this chapter, is the
remarkable difference between the design and function of the cervical spine and that of the lumbar spine.
In the cervical spine, the atlas serves to transmit the load of the head to the typical cervical vertebrae. The
axis adapts the suboccipital region to the typical cervical spine. In cervical intervertebrtal discs the anulus
fibrosus is not circumferential but is crescentic, and serves as an interosseous ligament in the saddle joint
between vertebral bodies. Cervical vertebrae rotate and translate in the sagittal plane, and rotate in the
manner of an inverted cone, across an oblique coronal plane. The cervical zygapophysial joints are the
most common source of chronic neck pain. By contrast, lumbar discs are well designed to sustain com-
pression loads, but rely on posterior elements to limit axial rotation. Internal disc disruption is the most
common basis for chronic low-back pain. Spinal muscles are arranged systematically in prevertebral and
postvertebral groups. The intrinsic elements of the spine are innervated by the dorsal rami of the spinal
nerves, and by the sinuvertebral nerves. Little modern research has been conducted into the structure of
the thoracic spine, or the causes of thoracic spinal pain.

INTRODUCTION CERVICAL SPINE

In writing a chapter on anatomy for neurologists the risk
arises of being arcane or banal. Neurologists will already
be familiar with the precepts of classic anatomy, and
would not be inclined to read a chapter that repeats bor-
ing, undergraduate material. For these reasons, the pre-
sent chapter has been cast in a different manner.
Although conventional elements of anatomy are
reprised, they are permeated by several themes. New
facts are provided, stemming from modern research into
the structure of the spine, along with new perceptions
about design and function. Throughout, the focus is
on clinical relevance, particularly with respect to the
mechanisms of spinal injury and spinal pain. In that
regard, certain structures – ignored in conventional
undergraduate curricula – are promoted to epidemiolog-
ically significant, clinical importance.
The cervical spine serves as a mobile support for the sen-
sory platform of the head. It allows the sensory apparatus
for vision, hearing, and smell to be elevated or depressed
in the sagittal plane, and to scan the environment in the
horizontal plane. In order to subserve these functions,
the cervical spine has to be mobile, yet sufficiently strong
to support the weight of the head. Its vulnerability, to
either minor or major injuries, lies in being long, slender,
and carrying the large mass of the head at its summit.
Both for descriptive purposes and functionally, the
cervical spine can be divided into three zones: the suboc-
cipital zone, centered on the C1 vertebra; a transitional
zone formed by the C2 vertebra; and the typical zone,
encompassing the C–7 vertebrae (Bogduk and Mercer,
2000) (Fig. 32.1). These zones differ both in structure
and in function.

*Correspondence to: Nikolai Bogduk, PO Box 128, The Junction, New South Wales 2291, Australia. E-mail: nbogduk@bigpond.
net.au
676 N. BOGDUK

Fig. 32.1. Sagittal magnetic resonance images of the cervical spine, showing its structure and zones. (A) Median scan, showing
the vertebral bodies and interverterbral discs. The white dots mark the mean location of the axes of rotation for flexion-extension of
the vertebra above. The odontoid process (op) projects rostrally from the body of C2, to lie behind the anterior arch of the atlas (C1).
(B) Lateral scan, showing the occipital condyle (oc), the lateral mass (lm) of the atlas (C1), the articular pillars (ap), and the zyga-
pophysial joints (zj) that they form, at the segments labeled. (Courtesy of Dr. Tim Maus, Mayo Clinic, Rochester, MN.)

Suboccipital zone Upper transition zone

The C1 vertebra (the atlas) shares none of the features of
typical cervical vertebrae, and should never have been
considered cervical. In structure and in function it is
more like an occipital vertebra. In structure it resembles
the occipital bone, as can be seen in axial scans. In func-
tion, it more closely operates with the head, rather than
with the remainder of the cervical spine.
The classic description of the atlas as a ring vertebra
belies its design and function. The critical components of
the atlas are its two lateral masses (Fig. 32.2). Superiorly,
these present superior articular processes that receive the
occipital condyles, and thereby cradle the skull. Inferi-
orly, the lateral masses present inferior articular pro-
cesses that rest on the C2 vertebra, and thereby
transmit the load of the head to the remainder of the cer-
vical spine. The anterior and posterior arches of the atlas
serve little function other than holding the two lateral
masses both apart and together, while the latter do the
mechanical work of the atlas.
Upon receiving the occipital condyles into their deep
sockets, the superior articular processes of each lateral
mass form the atlanto-occipital joints (Figs 32.2 and
32.3). These synovial joints constitute the only direct con-
nection between the skull and C1. They allow a small
range of flexion-extension, but the depth of their sockets
precludes axial rotation. Therefore, as the head rotates
(in the transverse plane) the atlas is obliged to move with
it. In that respect, the atlas behaves like a passive washer,
between the skull and C2.
The upper half of the C2 vertebra (the axis) is designed to
support the atlas. Superiorly and laterally, it presents
superior articular processes that slope caudally and lat-
erally, and act like sloping shoulders on which the lateral
masses of the atlas rest (Fig. 32.2). The inferior articular
processes of the atlas have a reciprocal, caudal and lat-
eral slope. The apposed articular processes on each side
form the lateral atlantoaxial joint (Figs 32.2 and 32.3).
The caudolateral slope of the lateral atlantoaxial joint
helps stabilize the atlas in the coronal plane, but also
underlies the mechanism of Jefferson fractures. Severe
axial loads, applied to the skull, will drive the atlas cau-
dally, but its lateral masses will also spread laterally
down the lateral slope of the lateral atlantoaxial joints,
resulting in burst fractures of the anterior and posterior
arches.
Centrally, the axis presents a long odontoid process
(the dens) that projects behind the anterior arch of the
atlas, with which it forms the median atlantoaxial joint
(Figs 32.1, 32.3, and 32.4). The anterior arch is held against
the odontoid process by the transverse ligament, which
spans like a belt between the two lateral masses of the
atlas, behind the odontoid process (Figs 32.4 and 32.5).
Posterior displacement of the atlas is prevented by
impaction of the anterior arch against the odontoid pro-
cess, at the median atlantoaxial joint. Anterior displace-
ment is prevented by tension in the transverse ligament
(Fielding et al., 1974). The ligament allows up to 3 mm
normal range of separation between the odontoid
 FUNCTIONAL ANATOMY OF THE SPINE 677

Fig. 32.2. Coronal magnetic resonance images of the cervical spine, showing the structure of its components. (A) Anterior scan,
showing the occipital condyles (oc) resting in the sockets of the lateral masses (lm) of the atlas, and forming the atlanto-occipital
joints (aoj); and the lateral masses bracketing the odontoid process (op), and resting on the “shoulders” of the axis (C2), where they
form the lateral atlantoaxial joints (laaj). The vertebral bodies of C2–7 form the anterior column of the cervical spine. (B) Posterior
scan, through the synovial joints of the cervical spine. Note the wedge shape of the lateral mass (lm) between the atlanto-occipital
joint (aoj) and the lateral atlantoaxial joint (laaj). The zygapophysial joint at C2–3 slopes caudally and medially, but those at suc-
cessive levels are essentially horizontal. The dotted line illustrates the ellipsoid shape depicted by the C2–3 zygapophysial joints
and the C2–3 disc, into which the atlas (C2) nestles on to the typical cervical spine.

process and the anterior arch in adults, and 5 mm in chil-
dren. In the past, the magnitude of the interval between
the anterior arch and the odontoid process has been used
as a measure of atlantoaxial instability, but as a predictor
of neurologic compromise the posterior atlantodental
interval (Fig. 32.3) has greater sensitivity and specificity
(Wasserman et al., 2011).
Severe forces delivered anteriorly to the head can frac-
ture the odontoid process. Such fractures threaten the sag-
ittal stability of the atlas. In turn, anterior or posterior
displacement of the atlas can threaten the spinal cord.
Rheumatoid arthritis of the atlantoaxial joints can weaken
the transverse ligament of the atlas, resulting in anterior
subluxation of the atlas (Wasserman et al., 2011).

Fig. 32.3. Close-up views of sagittal magnetic resonance images of the suboccipital joints. (A) Median scan through the odon-
toid process (op) and vertebral body of C2. With the front of the odontoid process, the anterior arch (aa) of the atlas forms the
median atlantoaxial joint (maaj). The transverse arrow marks the posterior atlantodental (pa) interval. (B) Lateral scan through
the lateral mass (lm) of the atlas. With the superior sockets of the lateral mass of the atlas, the occipital condyle (oc) forms the
atlanto-occipital joint (aoj). With C2, the lateral mass forms the lateral atlantoaxial joint (laaj). Note the bi-convex shape of
the lateral atlantoaxial joint. The triangular, white signals anteriorly and posteriorly within the joint are the fibroadipose meniscoids
that it contains. (Courtesy of Dr. Tim Maus, Mayo Clinic, Rochester MN.)
678 N. BOGDUK
Fig. 32.4. A sketch of an axial (top) view of the atlas sitting on
the axis. The lateral masses of the atlas bracket the odontoid
process (op) of the axis, and are themselves joined by the ante-
rior (aa) and posterior (pa) arches of the atlas. With the anterior
arch, the odontoid process forms the median atlantoaxial joint
(maaj). The transverse ligament (tl) spans like a belt behind the
odontoid processes, between the two lateral masses.
Although the osseous articular processes of the lat-
eral atlantoaxial joint are flat, their articular cartilages
are convex (along the sagittal plane) (Koebke and
Brade, 1982) (Fig. 32.3). As a result, the atlas perches
somewhat precariously on the axis, with its convex infe-
rior articular cartilages balancing on the convexities of
the superior articular cartilages of the axis. The spaces
anteriorly and posteriorly between the convex cartilages
are filled by wedge-shaped fibrocartilaginous menis-
coids (Mercer and Bogduk, 1993).
Although a small degree of flexion and extension is
possible between the atlas and the axis (Bogduk and
Mercer, 2000), the cardinal movement between these
two vertebrae is axial rotation. During this movement
the atlas pivots at the median atlantoaxial joint, while
its lateral masses slide backwards or forwards, circum-
ferentially, at the lateral atlantoaxial joints. However,
because of the convexity of the articular cartilages in
these joints, the lateral masses also descend, down the
posterior or anterior slope of the cartilages, as they move
backwards or forwards, respectively. As a result, the
atlas settles (lowers or screws down) during axial rota-
tion, and rises when the movement is reversed
(Bogduk and Mercer, 2000). During these displace-
ments, the meniscoids of the joints cover the exposed
surfaces of the subluxating articular cartilages.
The total range of axial rotation of the atlas is consid-
erably large. It has been measured as 43  5.5° which
effectively amounts to 50% of the range of axial rotation
of the head and neck (Bogduk and Mercer, 2000). At the
extremes of this range, very little of the articular
Fig. 32.5. A sketch of the suboccipital joints and ligaments, as
view from behind with the posterior arch of the atlas resected.
The lateral mass (lm) of the atlas supports the occipital bone,
and rests of the axis, forming the atlanto-occipital joint (aoj)
above, and the lateral atlantoaxial joint (laaj) below. The trans-
verse ligament (tl) holds the atlas (al) against the odontoid pro-
cess (op). The alar ligaments bind the odontoid process to the
margins of the foramen magnum, thereby connecting the skull
to C2 but bypassing C1.
cartilages of the lateral atlantoaxial joints remain
opposed; the joint is almost dislocated.
In order to accommodate this large range of motion,
the capsules of the lateral atlantoaxial joint are loose,
and serve little to hold the atlas on the axis. That service
is provided by the alar ligaments. On each side, these lig-
aments pass essentially transversely from the upper end
of the odontoid process to the margin of the foramen
magnum (Fig. 32.5). In doing so, they bypass the atlas,
and lock the head into place on the axis, effectively clamp-
ing the atlas between the skull and C2. The alar ligaments
are the cardinal restraint to axial rotation of the head
(Dvorak et al., 1987). They are sufficiently strong to pre-
vent anterior dislocation of the head even if the transverse
ligament is completely severed (Fielding et al., 1974). Dis-
ruption of an alar ligament can result in rotatory instabil-
ity of the head and atlas (Dvorak et al., 1987).
Excessive axial rotation of the atlas can result in a lat-
eral mass dislocating at the lateral atlantoaxial joint,
causing fixed atlantoaxial deformity (Wortzman and
Dewar, 1968). A less dramatic form of torticollis can
arise if, after rotation of the head and atlas, a meniscoid
of the lateral atlantoaxial joint fails to re-enter the joint
space, catches under the capsule of the joint, and acts like
a loose body to prevent derotation of the joint (Mercer
and Bogduk, 1993).
 FUNCTIONAL ANATOMY OF THE SPINE
Fig. 32.6. A sketch of a coronal view of how forces from the
head are transmitted into the cervical spine. On each side, the
weight of the head passes through the occipital condyle (oc),
into the lateral mass (lm) of the atlas, and into the axis
(C2) through the lateral atlantoaxial joint. From there, the forces
diverge, partly into the posterior column of zygapophysial
joints, and partly into the anterior column of vertebral bodies
and discs. Half the load passes anteriorly and half posteriorly.
Increasing interest has been focused on the lateral
atlantoaxial joints as a possible source of cervicogenic
headache. This contention can be tested by controlled,
intra-articular, diagnostic blocks of the putatively pain-
ful joint (Bogduk and Bartsch, 2008; Bogduk and
Govind, 2009; Bogduk, 2014).
Lower transition zone
The lower half of the C2 vertebra has the structure of a
typical cervical vertebra (Figs 32.1 and 32.2). Centrally it
presents a vertebral body, and laterally it presents paired
inferior articular processes. Having received the lateral
masses of the atlas, the axis transmits the load of the
head along an anterior channel, through its vertebral
body to the vertebral bodies below, and along paired pos-
terior channels, through the zygapophysial joints
(Fig. 32.6). Approximately half of the axial load is trans-
mitted through the anterior channel, and half through the
two posterior channels.
Typical cervical vertebrae
The cardinal elements of a typical cervical vertebra are
its vertebral body and two articular pillars (Figs 32.1
679
and 32.2). Secondarily, transverse processes project lat-
erally from the articular pillars, and posteriorly the two
pillars are united by a pair of laminae, which support a
midline spinous process at their junction. The transverse
processes and spinous processes serve as levers upon
which act the muscles that control the position of the cer-
vical vertebrae. Along its superior, posterolateral margin
on each side, each vertebral body bears uncinate pro-
cesses. Previously enigmatic, the uncinate processes
underlie the nature of the joints between the cervical ver-
tebral bodies and how they operate.
Consecutive articular pillars are united by the zy-
gapophysial joints (Figs 32.1 and 32.2), which are synovial
joints formed by the inferior articular process of the
vertebra above and the superior articular process of
the vertebra below. Fibroadipose meniscoids intervene
between the articular cartilages of these joints (Mercer
and Bogduk, 1993). The zygapophysial joints are planar,
and at typical cervical levels are oriented at about 40° to
the coronal and transverse planes, so that they face back-
wards and upwards (Nowitzke et al., 1994). At the C2–3
level, however, the joints also face medially, such that the
pair of joints depict an ellipsoid socket into which nestles
the weight of the axis, and the load that it carries from
the head (Figs 32.2 and 32.6).
Consecutive vertebral bodies are united by interverteb-
ral discs, and by the anterior and posterior longitudinal lig-
aments (Mercer and Bogduk, 1999). The anterior ligament
connects only the typical cervical vertebrae, from C2 cau-
dally. The posterior longitudinal ligament forms a
carpet along the floor of the vertebral canal at typical cer-
vical levels, but expands into the membrana tectoria to
cover the back of the atlantoaxial region. In doing so,
the ligament separates the dural sac and spinal cord from
the mechanics of the median atlantoaxial joint.
Posterior ligaments are lacking in the cervical spine.
Interspinous ligaments are represented by only a sagittal
layer of fascia (Mercer and Bogduk, 2003). The ligamen-
tum nuchae lacks the structure of a ligament. It consists
largely of a narrow, coronal raphe, anchored to the C7
spinous process, and formed by interlacing tendons of
the splenius muscles and trapezius (Mercer and
Bogduk, 2003).
The intrinsic structure of the cervical intervertebral
discs is unlike that of lumbar discs, and differs with
age (Oda et al., 1988; Mercer and Bogduk, 1999). The
nucleus pulposus of cervical discs is gelatinous only in
children and young adults. By the age of 30 it dries
out to form a fibrocartilaginous plate (Oda et al.,
1988). Moreover, the nucleus is not surrounded by con-
centric lamellae of the anulus fibrosus (Mercer and
Bogduk, 1999). The anulus fibrosus is largely deficient
posteriorly, and consists of a thin, paramedian band of
collagen fibers that run longitudinally between the
680 N. BOGDUK

Fig. 32.7. Sketches of various views of the internal structure

of the cervical disc. In a front view, all fibers of the anterior
anulus fibrosus pass towards a point on the inferior anterior
surface of the vertebral body above. In a top view, the anulus
fibrosus (af) is crescentic in shape, thick anteriorly but tapering
at the uncinate processes (u). The nucleus pulposus (np) is a
fibrocartilaginous plate. Posteriorly the anulus is restricted
to a small bundle of paramedian, longitudinal fibers. A side
view shows the fibers of the anulus fibrosus passing upwards
and forwards. A transverse cleft runs from one uncinate pro-
cess to the other.

vertebral bodies (Fig. 32.7). Posterolaterally, the nucleus

is covered by the posterior longitudinal ligament, rather
than by anulus fibrosus. Anteriorly, the anulus fibrosus
is crescentic in shape, thin posteriorly near the uncinate
processes, but thicker anteriorly towards the midline. All
of its collagen fibers pass in a similar direction, effec-
tively aiming to a median point on the lower anterior sur-
face of the vertebral body above. This configuration
endows the anulus fibrosus with the structure of a thick
interosseous ligament that binds the anterior edges of
consecutive vertebral bodies.
The superior surface of each cervical vertebral body
presents two curvatures: a slight convex curvature along
the sagittal plane, and a deep concave curvature trans-
versely between the uncinate processes (Fig. 32.8). These
curvatures endow the vertebral body with the configura-
tion of a saddle joint (Bogduk and Mercer, 2000). Con-
sequently, the cervical interbody joints operate like a
saddle joint, with motion restricted to two planes: the
sagittal plane and an oblique coronal plane.
In the sagittal plane, the vertebral bodies can rock and
slide (rotate and translate), to provide for flexion and
extension of the neck. From above downwards, the typ-
ical cervical vertebrae exhibit progressively less transla-
tion for each degree of rotation, during flexion or
extension. This is reflected by the different locations
of their axes of movement. At higher levels the axes
Fig. 32.8. Sketches of a posterolateral view of a typical cervi-
cal vertebral body, showing the two curvatures of its superior
surface: a downward concavity along the sagittal plane, and a
second concavity, facing upwards and forwards, between the
uncinate processes (u). These two concavities endow the inter-
vertebral disc with the features of a saddle joint.
lie in the vertebral body below the moving vertebra,
but are progressively closer to their intervertebral disc
at lower levels (Amevo et al., 1991) (Fig. 32.1). These dif-
ferences correlate strongly with the height of the articu-
lar pillar at each segment (Nowitzke et al., 1994). Taller
pillars provide less space into which the vertebra can
translate once it has commenced sagittal rotation. Con-
versely, at segments with shorter pillars, sagittal rotation
lifts the inferior articular processes of the moving verte-
bra off the supporting articular pillar, and provides a
greater gap into which it can translate (Nowitzke
et al., 1994).
The second plane of movement of typical cervical
vertebrae is set at 40° forwards of the coronal plane,
and lies parallel to the plane of the zygapophysial joints
 FUNCTIONAL ANATOMY OF THE SPINE
Fig. 32.9. A sketch of a typical cervical intervertebral joint,
illustrating the mechanics of rotation of the vertebral body,
across the oblique concavity of the uncinate processes, and
around an oblique axis through the vertebral body. The motion
is like that of a cone whose apex is fixed but whose bases nev-
ertheless free to twist and spin, in the direction and plane indi-
cated by the arrow.
(Bogduk and Mercer, 2000). Across this plane, each ver-
tebra rotates like an inverted cone whose apex is fixed,
but whose base can twist (Fig. 32.9). The apex of the cone
corresponds to the anterior, median point on the verte-
bral body to which the fibers of the anulus fibrosus
are directed; and the anulus fibrosus serves to hold this
apex in place. Meanwhile the posterior, inferior edge of
the vertebral body presents a convex surface that is
cupped by the concave surface between the uncinate pro-
cesses of the vertebra below. This latter geometry is that
of an ellipsoid joint, and the posterior inferior margin of
the vertebral body is free to spin, or swing, across this
ellipsoid surface. Thus, while the anterior end of the ver-
tebral body is fixed, its posterior end and its posterior
elements are free to spin clockwise or counterclockwise
across the oblique coronal plane. During this motion, the
inferior articular processes of the zygapophysial joints
simply glide laterally across the surfaces of their sup-
porting superior articular processes.
A consequence of this mode of operation is that the
interbody joints of typical cervical vertebrae cannot tol-
erate a posterolateral anulus fibrosus, for it would
impede the spin of the posterior vertebral body across
the oblique coronal plane. Consequently, although a pos-
terior anulus is present at birth and in young children, it
gradually disappears as neck movements increase
(Tondury, 1972). By about the age of 9 years, the postero-
lateral anulus tears, and clefts appear in the region of the
uncinate processes. Progressively these clefts enlarge
centrally, until they meet in the midline, at about the
age of 30, to form a transverse cleft from one uncinate
681
Fig. 32.10. The mechanics of the early phase of whiplash
injury. As a result of a thrust from below, the cervical spine
undergoes a sigmoid deformation. Lowe segments, e.g.,
C5–6, undergo a posterior sagittal rotation around an abnor-
mally high instantaneous axis of rotation (iar), which results
in posterior elements being impacted and anterior elements
being stretched.
process to the other. This cleft is not a degenerative
change but a normal age change. The cleft effectively
forms the “joint space” across the posterior interverteb-
ral disc that allows axial rotation of the head to be accom-
modated and amplified in range by the typical cervical
vertebrae.
This structure and mechanics of the cervical spine are
of relevance to the mechanisms of injury in whiplash.
The early phase of whiplash injury involves a thrust from
below (Bogduk and Yoganandan, 2001; Bogduk, 2006).
This upward thrust deforms the cervical spine into a sig-
moid shape, within which the lower cervical vertebrae –
typically C5 and C6 – undergo an abnormal extension
(Fig. 32.10). The vertebra rocks backwards but without
translating. As a result, it rotates about an abnormally
high axis of rotation (Kaneoka et al., 1999). During this
motion, anterior elements are stretched while posterior
elements are impacted. The anterior anulus fibrosus
can be torn or avulsed, resulting in so-called rim lesions.
Impaction in the zygapophysial joints can cause impac-
tion fractures of the articular cartilages, or contusions
of the intra-articular meniscoids. During later phases,
the cervical spine rebounds into flexion, which can exces-
sively strain the capsules of the zygapophysial joints
(Curatolo et al., 2011).
Physiologic studies in laboratory animals have shown
that the capsule strains induced by whiplash injury result
in persistent nociception from the injured joint, and per-
sistent changes within the central nervous system charac-
teristic of chronic pain (Winkelstein, 2011). Clinical
studies have shown that the cervical zygapophysial joints
682 N. BOGDUK
are the single most common source of chronic neck
pain after whiplash, accounting for between 50% and
60% of cases (Bogduk, 2011). Most commonly, neck
pain – with referred pain to the shoulder girdle – stems
from the C5–6 joint, while headache stems from the
C2–3 zygapophysial joint.
Less well understood is pain from the cervical interver-
tebral discs. Conspicuously, degenerative disc disease is
not associated with neck pain. Furthermore, discogenic
pain appears to be uncommon, once zygapophysial joint
pain is taken into account (Yin and Bogduk, 2008). Per-
haps discogenic pain is caused by strains of the inteross-
eous ligament formed by the anterior anulus fibrosus, but
diagnostic techniques by which to test this proposition
have not been developed.
LUMBAR SPINE
The cardinal role of the lumbar spine is to support the
thorax and upper limbs – and any loads that they
carry – and to transmit those loads to the pelvis and lower
limbs (Bogduk, 2012a). Secondarily, the lumbar spine
accommodates a modest range of movement between
the thorax and pelvis.
In order to subserve these functions, the essential ele-
ments of the lumbar spine are the vertebral bodies of the
five lumbar vertebrae (Fig. 32.11). These are stacked into
a strong column, and are united by intervertebral discs
Fig. 32.11. Sagittal magnetic resonance images of the lumbar spine. (A) Median scan showing the vertebral bodies and spinous
processes (sp). The white dots mark the location of the axes of rotation of the vertebra above. (B) Lateral scan through the inter-
vertebral foramina and the L3–4 to L5–S1 zygapophysial joints. ped, pedicle of L3; sap, superior articular process of L4; iap,
inferior articular process of L5.
and by the anterior and posterior longitudinal ligaments
(Bogduk, 2012a). Bowing the column into a lordosis
endows the lumbar spine with the ability to absorb
dynamic axial loads (bouncing). Axial impulses deform
the lordotic curve; the energy is absorbed by the elastic
discs and longitudinal ligaments; and is returned to
restore the more upright curve, once the axial impulse
has passed (Bogduk, 2012a).
The lumbar intervertebral discs are well designed to
accommodate compression loads (Hickey and Hukins,
1980). Each consists of hydrated nucleus pulposus, sur-
rounded by an anulus fibrosus, and capped superiorly
and inferiorly by a vertebral endplate that joins the disc
to the adjacent vertebral body (Fig. 32.12). The anulus
fibrosus is formed by concentric layers of collagen
fibers, in which the fibers in any one layer run in parallel,
at about 60° to the long axis of the spine, but in succes-
sive layers that orientation alternates.
Axial compression is resisted primarily by the concen-
tric layers of the anulus fibrosus (Markolf and Morris,
1974) (Fig. 32.12). However, the tendency of the anulus
under load is to buckle, both outwards and inwards. This
buckling is resisted by the hydrostatic nucleus pulposus.
When the nucleus is compressed it exerts a radial pressure
that braces, and stiffens, the anulus, thereby preventing it
from buckling. A small range of flexion-extension is
accommodated by the discs (about 13° per segment), dur-
ing which the anulus fibrosus on the side to which
 FUNCTIONAL ANATOMY OF THE SPINE
Fig. 32.12. Close-up views of a sagittal magnetic resonance image of an L3–4 intervertebral disc. (A) The components of the disc.
np, nucleus pulposus; af, anulus fibrosus; vep, vertebral endplate. (B) The mechanics of the disc. Axial compression loads are
primarily borne by the lamellae of collagen in the anulus fibrosus. When compressed, the nucleus pulposus exerts radial pressure
to brace the anulus, and prevent it from buckling under load.
movement occurs is compressed slightly, while the anulus
on the opposite side is stretched (Bogduk, 2012a).
While strongly designed to resist compression, the
lumbar discs are poorly designed to resist axial rotation.
Because the collagen fibers of the anulus fibrosus alter-
nate in direction in successive layers, only half are avail-
able to resist axial rotation in one direction or the other.
For stability in axial rotation, the lumbar vertebral bodies
and intervertebral discs rely on the posterior elements of
the lumbar vertebrae (Bogduk, 2012a).
The posterior elements are based on an arch (Bogduk,
2012a) (Fig. 32.13). The arch is supported by stout pedi-
cles that emanate from the upper posterior surface of
each vertebral body. The pedicles serve to transmit
forces from the succeeding posterior elements to the ver-
tebral bodies, which control the position or movements
of the vertebral bodies. The arch is completed by left
and right laminae that join in the midline. From the junc-
tion of the two laminae springs a large spinous process,
and from the junction between the pedicle and lamina on
each side arises a long transverse process. These pro-
cesses serve as levers to which attach the muscles that
control the movements of the lumbar vertebrae.
At its superior and inferior lateral corners respec-
tively, each lamina bears a superior and inferior articular
process. Like large mittens, the paired superior articular
processes reach cranially to grasp the inferior articular
processes of the vertebra above, and form the zygapo-
physial joints. The plane of these joints is parallel to
the longitudinal axis of the lumbar spine. Consequently,
during flexion of the vertebral bodies, the inferior artic-
ular processes glide freely out of the sockets formed by
the superior articular processes, until movement is
arrested by tension in the joint capsules (Bogduk,
2012a). The axis of this movement typically lies in the
disc below the moving vertebra (Pearcy and Bogduk,
1988) (Fig. 32.11A), which indicates only a small amount
of translation for every degree of rotation of the moving
683
vertebra. As the inferior articular processes move, they
lift away from the superior articular process, tanta-
mount to partially subluxating the joint. Fibroadipose
meniscoids protect the exposed surfaces of the articular
cartilages during this displacement (Engel and Bogduk,
1982; Bogduk and Engel, 1984).
In axial views, the lumbar zygapophysial joints vari-
ously present flat, C-shaped, or J-shaped appearances,
which correspond to the primary functions of these
joints (Horwitz and Smith, 1940). Flat joints essentially
face medially and posteriorly. C-shaped joints have an
anterior end that faces posteriorly, and a posterior end
that faces medially. J-shaped joints have a small anterior
lip facing posteriorly, and a larger surface facing medi-
ally. The medially facing surfaces serve to resist axial
rotation of the vertebrae. Attempted axial rotation
swings the inferior articular process laterally, but this
movement is arrested by the opposing superior articular
process. The range of motion is limited to about 2° or less
per segment (Pearcy and Tibrewal, 1984), and is accom-
modated only by compression of the articular cartilage.
The surfaces that face posteriorly serve to resist forward
displacement (listhesis) of the vertebra.
Impaction of an inferior articular process against its
superior articular process tends to force the inferior pro-
cess backwards, and lift the lamina from which it arises
(like opening a hatchback). In turn this tendency stresses
the junction between the lamina and its pedicle. Repeated
impactions – particularly during repeated axial
rotation – can cause stress fractures at this point, result-
ing in pars interarticularis defects.
The lumbar zygapophysial joints can be a source of
low-back pain, but its prevalence is uncertain. It appears
to be uncommon or rare in injured workers, but is com-
mon in elderly patients (Bogduk, 2008, 2012b).
The most common cause of chronic low-back pain is
internal disc disruption (Bogduk et al., 2013). This condi-
tion is characterized by degradation of the nucleus
684 N. BOGDUK

Fig. 32.13. Sketches of the posterior elements of a lumbar vertebra. (A) Posterior view. The two laminae (la) form a quadrangular
plate, from whose corners project the superior (sap) and inferior (iap) processes. From the junction of the two laminae projects the
spinous process (sp). On each side, the inferior and superior articular processes of consecutive vertebrae form the zygapophysial
joint (zj) (B) Axial (top) view. The posterior elements are connected to the vertebral body (vb) by the pedicles (p). The transverse
process (tp) projects from the junction of the pedicle and lamina, on each side.

pulposus of the affected disc and the development of
radial fissures into the posterior or posterolateral anulus.
The condition has been produced in laboratory animals,
and pursued in numerous clinical studies. Its cause is
compression injuries that produce small fractures of
the vertebral endplate. These result in degradation of
the matrix of the nucleus pulposus. As the nucleus
becomes less able to retain water, it is no longer able
to pressurize and brace the anulus. Pressures in the
nucleus drop, but rise in the posterior anulus. The
unbraced anulus progressively delaminates, particularly
in regions of high stress where the laminae are curved: at
the posterolateral corners or the posterior paramedian
sector. Pain arises as a result of chemical irritation of
nociceptors in the anulus by degradation products from
the nucleus, and as a result of the increased mechanical
stresses on the surviving, intact laminae of anulus
(Bogduk et al., 2013). To various degrees of certainty
the condition can be diagnosed by characteristic features
on magnetic resonance imaging, such as Modic lesions in
the vertebral body or high-intensity zones in the anulus
fibrosus, and by provocation discography (Bogduk
et al., 2013). No treatment has been vindicated, but sev-
eral minimally invasive interventions are being pursued,
which encompass ablating nociceptors in the disc, inject-
ing restorative agents such as stem cells, or injecting
antagonists of inflammation.
MUSCLES
The anatomy of muscles of the cervical and lumbar spine
is made complex by the diversity of their numerous
attachments. If those specifics are ignored, the anatomy
becomes simpler.
Small muscles connect consecutive spinous processes
and transverse processes. Too small to move their verte-
brae effectively, these muscles serve as proprioceptors
for the spine (Bastide et al., 1989).
Prevertebral muscles are represented only in the cer-
vical spine (Standring, 2008). The longus cervicis con-
nects the vertebral bodies and transverse processes of
the cervical vertebrae. It is covered by the longus capitis
which anchors the skull to the cervical vertebrae. These
muscles are weak flexors of the head and neck.
Various suboccipital muscles control movements
of the head in relation to the atlas and the axis. They
are the rectus anterior and rectus lateralis anteriorly,
and the rectus capitis posterior major and minor acc-
ompanied by obliquus inferior and obliquus superior,
posteriorly (Standring, 2008). Collectively these muscles
control the orientation of the head on the atlas and axis.
The postvertebral muscles are aligned systematically,
side by side and by layers (Standring, 2008). Multifidus
is the deepest and most medial muscle. Its fascicles arise
from a spinous process and descend to various insertions
on articular processes and transverses processes one to
several segments caudally. It is flanked by the longissi-
mus system of muscles, which attach to transverse pro-
cesses near their bases, and whose components are large
at lumbar levels, but virtually miniscule at cervical levels.
Further laterally runs the iliocostalis system, which
attaches to transverse processes near their tips, and
whose components are, likewise, large at lumbar levels
but miniscule at cervical levels. A semispinalis system
is vestigial at lumbar levels but well developed at cervical
levels. Semispinalis cervicis arises from the cervical spi-
nous processes, and covers the multifidus with fascicles
longer than those of the latter muscle. Semispinalis capi-
tis arises from the occiput, and is anchored to the cervical
 FUNCTIONAL ANATOMY OF THE SPINE 685
transverse processes. It is the largest of the posterior lateral atlantoaxial joint before joining the cervical
neck muscles. It is covered by the splenius muscle, which plexus (Lazorthes and Gaubert, 1956).
passes cranially and laterally from the raphe of the liga- The C3–7 cervical spinal nerves lie above their like-
mentum nuchae to wrap around all the other posterior numbered vertebrae, enclosed in their respective inter-
muscles of the neck. Splenius cervicis reaches the upper vertebral foramina. They are joined by the C8 spinal
cervical transverse processes, while splenius capitis nerve, which lies in the C7–T1 intervertebral foramen.
reaches the superior nuchal line. Variously and collec- The ventral rami of C3 and C4 join the cervical plexus,
tively, the cervical postvertebral muscles act to extend and the lower cervical ventral rami join the brachial
the head and the cervical spine. plexus. The dorsal rami of the typical cervical spinal
Other muscles use the vertebral column adventi- nerves form lateral branches that supply the splenius,
tiously, as a base from which to act on nonspinal struc- longissimus, and iliocostalis; and medial branches that
tures. In the neck, these include the scalene muscles, supply the deeper and medial posterior neck muscles,
which act on the ribs; and levator scapulae and trapezius, and the cervical zygapophysial joints (Bogduk, 1982).
which act on the shoulder girdle. Sternocleidomastoid is The cervical medial branches have constant locations
the principal flexor and rotator of the head and neck, but on the cervical articular pillars, which allow them to be
passes directly from the manubrium and clavicle to the targeted for fluoroscopy-guided diagnostic blocks, by
head, with no connection to the cervical spine. Being which pain from the zygapophysial joints can be diag-
locked to the skull through the atlanto-occipital joints, nosed (Bogduk, 1982, 2011).
the atlas is rotated when the sternocleidomastoid rotates Gray rami communicantes, from the stellate ganglion
the head. and from the cervical ventral rami, form a plexus –
In the lumbar spine, psoas major arises from the ver- called the vertebral nerve – that accompanies the verte-
tebral bodies, discs, and transverse processes to act on bral artery through the foramina transversaria of
the femur, but does not move the lumbar spine the neck, and into the posterior cranial fossa (Bogduk
(Bogduk et al., 1992). Quadratus lumborum attaches et al., 1981a). Although migraine cervicale, or the
to the lumbar transverse processes but acts principally Barré–Lieou syndrome, has been attributed to irritation
on the 12th rib; its actions on the lumbar vertebrae are of these nerves, and spasm of the vertebral artery, labo-
effectively trivial (Phillips et al., 2008). Transversus ratory studies have shown the vertebrobasilar system to
abdominis stems from the lumbar transverse processes, be remarkable unresponsive to stimulation of the verte-
and has virtually no effect on the lumbar vertebrae bral nerve (Bogduk et al., 1981a; Lambert et al., 1984).
(Macintosh et al., 1987). Likewise, latissimus dorsi gains The cervical sinuvertebral nerves are formed by
some anchorage to the lumbar spinous processes but has somatic roots from the ventral rami and autonomic roots
a negligible action on the lumbar spine (Bogduk from the rami communicantes in the vertebral nerve. As
et al., 1998). recurrent meningeal branches they innervate the cervical
dural sac, but also innervate the cervical discs and the
posterior longitudinal ligament (Bogduk et al., 1988).
The C1–3 sinuvertebral nerves innervate the ligaments
INNERVATION
of the median atlantoaxial joint before passing through
The C1 spinal nerve is unlike other spinal nerves, which foramen magnum to supply the dura mater over the cli-
reinforces the atlas being suboccipital rather than cervi- vus (Kimmel, 1960).
cal in nature. This nerve lacks a typical dorsal root gan- The lumbar spinal nerves lie obliquely in their inter-
glion, but ganglion cells can be found amongst the vertebral foramina, each below the like-numbered verte-
rootlets of the spinal accessory nerve. The C1 dorsal bra. Their ventral rami enter the lumbar or lumbosacral
ramus appears amongst the posterior suboccipital mus- plexus. Their dorsal rami form lateral and intermediate
cles (Lazorthes and Gaubert, 1956). Sometimes it can branches that innervate the iliocostalis and longissimus
have a cutaneous branch. The C1 ventral ramus crosses muscles respectively (Bogduk et al., 1982; Bogduk,
the posterior arch of the atlas, behind the superior artic- 1983). Medial branches innervate the lumbar zygapophy-
ular process. It innervates the atlanto-occipital joint sial joints and the multifidus (Bogduk et al., 1982;
before entering the cervical plexus (Lazorthes and Bogduk, 1983). Where the medial branches cross the root
Gaubert, 1956). of the superior articular process they can be targeted for
The C2 spinal nerve lies behind the lateral atlantoaxial fluoroscopy-guided diagnostic blocks, by which pain
joint, and forms a large dorsal ramus that supplies the from the lumbar zygapophysial joints can be diagnosed
more superficial posterior neck muscles, and becomes (Bogduk, 1983, 2008, 2012b).
cutaneous as the greater occipital nerve, over the occiput At each segmental level, the lumbar sinuvertebral
(Bogduk, 1982). The C2 ventral ramus supplies the nerves arise from the ventral ramus and gray ramus
686 N. BOGDUK
Fig. 32.14. Sagittal magnetic resonance images of the thoracic
spine. (A) Median section, through the vertebral bodies, spinal
cord (sc), and spinous processes (sp) cord. (B) Paramedian sec-
tion through the zygapophysial joints (zj). Intervertebral discs
(ivd) are evidence in both sections. (Courtesy of Dr. Tim Maus,
Mayo Clinic, Rochester MN.)
communicans. Each passes back into the intervertebral
foramen to supply the dural sac, the posterior longitudi-
nal ligament, and the posterior anulus fibrosus (Bogduk
et al., 1981b; Bogduk, 1983). These nerves provide the
sensory pathway for lumbar discogenic pain.
THORACIC SPINE
There have been no substantial advances in the descrip-
tion of the anatomy of the thoracic spine since editions
of anatomy textbooks of the 19th and 18th century. In
parallel, there has been little advance in the understand-
ing of thoracic spinal pain and its sources, let alone
causes. No diagnostic or treatment procedures have been
validated. Thoracic spinal pain essentially remains a
mystery.
Like cervical and lumbar vertebrae, the thoracic ver-
tebrae have vertebral bodies that are connected by inter-
vertebral discs and longitudinal ligaments, and posterior
elements that are connected by zygapophysial joints
(Fig. 32.14). The distinction of the thoracic spine is that
it suspends the ribs. At typical thoracic levels, the head
of the rib articulates with the intervertebral disc and
demifacets on the edges of the vertebrae that bind that
disc, and the articular tubercle of the rib articulates with
the transverse process of the upper of the two vertebrae
Fig. 32.15. Axial magnetic resonance image of a typical tho-
racic spinal segment. vb, vertebral body; zj, zygapophysial
joint; sp, spinous process; cvj, costovertebral joint; ctj,
costotransverse joint. (Courtesy of Dr. Tim Maus, Mayo
Clinic, Rochester, MN.)
(Fig. 32.15). Exceptions to this arrangement occur at T1
and at T11 and T12, where the head of the rib fully artic-
ulates with the like-numbered vertebrae.
Few studies have explored the innervation of the tho-
racic spine (Bogduk, 2002). The thoracic sinuvertebral
nerves are assumed to be homologous to those at cervical
or lumbar levels. The courses of the thoracic dorsal rami
appear to differ from those at cervical and lumbar levels,
but are nevertheless homologous (Chua and Bogduk,
1995). Whereas the medial branches at cervical and lum-
bar levels wind around the base of the superior articular
process at each segmental level, at thoracic levels the
dorsal ramus stretches to the tip of the transverse pro-
cess before dividing into medial and lateral branches.
This difference is reconciled once it is realized that what
are called the transverse processes at cervical and lumbar
levels are embryologically costal elements (rudimentary
ribs), whereas the embryologic transverse elements (or
true transverse processes) are absorbed into the base
of the superior articular process. Consequently, at cervi-
cal and lumbar levels, the medial branches cross the
superior articular process because the true transverse
processes also lie there. This distinction becomes perti-
nent for minimally invasive, diagnostic, and treatment
procedures that target thoracic medial branches. The tar-
get lies on the transverse process, not on the superior
articular process (Chua and Bogduk, 1995).
A persisting curiosity pertains to the structure of tho-
racic intervertebral discs. Cervical discs differ greatly
from lumbar discs, but undiscovered is the transition
 FUNCTIONAL ANATOMY OF THE SPINE
zone. Are thoracic discs like cervical discs, or do they have
the structure of lumbar discs? Given that cervical uncinate
processes are homologous to the heads of the ribs, unpub-
lished observations suggest that discs change their struc-
ture where uncinate processes or their rib equivalent
cease. Thoracic discs become lumbar in nature at T11,
where the rib no longer articulates with the disc.
REFERENCES
Amevo B, Worth D, Bogduk N (1991). Instantaneous axes of
rotation of the typical cervical motion segments: a study in
normal volunteers. Clin Biomech 6: 111–117.
Bastide G, Zadeh J, Lefebvre D (1989). Are the ‘little muscles’
what we think they are? Surg Radiol Anat 11: 255–256.
Bogduk N (1982). The clinical anatomy of the cervical dorsal
rami. Spine 7: 319–330.
Bogduk N (1983). The innervation of the lumbar spine. Spine
8: 286–293.
Bogduk N (2002). Innervation and pain patterns of the thoracic
spine. In: R Grant (Ed.), Physical therapy of the Cervical
and Thoracic Spine, 3rd edn. Churchill Livingstone, New
York, pp. 73–81.
Bogduk N (2006). Whiplash injury. In: F Cervero, TS Jensen
(Eds.), Handbook of Clinical Neurology Vol. 81: Pain,
Elsevier, Amsterdam, pp. 791–801.
Bogduk N (2008). Evidence-informed management of chronic
back pain with facet injections and radiofrequency neurot-
omy. Spine J 8: 56–64.
Bogduk N (2011). On cervical zygapophysial joint pain after
whiplash. Spine 36: S194–S199.
Bogduk N (2012a). Clinical Anatomy of the Lumbar Spine and
Sacrum, 5th edn. Elsevier, Edinburgh.
Bogduk N (2012b). Lumbar medial branch neurotomy. In:
S Dagenais, S Haldeman (Eds.), Evidence-Based
Management of Low Back Pain, Elsevier, St Louis,
pp. 351–363.
Bogduk N (2014). The neck and headaches. Neurol Clin 32:
471–487.
Bogduk N, Bartsch T (2008). Cervicogenic headache. In:
SD Silberstein, RB Lipton, DW Dodick (Eds.), Wolff’s
Headache, 8th edn. Oxford University Press, New York,
pp. 551–570.
Bogduk N, Engel R (1984). The menisci of the lumbar zyga-
pophyseal joints. A review of their anatomy and clinical
significance. Spine 9: 454–460.
Bogduk N, Govind J (2009). Cervicogenic headache: an
assessment of the evidence on clinical diagnosis, invasive
tests, and treatment. Lancet Neurol 8: 959–968.
Bogduk N, Mercer SR (2000). Biomechanics of the cer-
vical spine. I: Normal Kinematics. Clin Biomech 15:
633–648.
Bogduk N, Yoganandan N (2001). Biomechanics of the cervi-
cal spine Part 3: minor injuries. Clin Biomech 16: 267–275.
Bogduk N, Lambert G, Duckworth JW (1981a). The anatomy
and physiology of the vertebral nerve in relation to cervical
migraine. Cephalalgia 1: 11–24.
Bogduk N, Tynan W, Wilson AS (1981b). The nerve supply to
the human lumbar intervertebral discs. J Anat 132: 39–56.
687
Bogduk N, Wilson AS, Tynan W (1982). The human lumbar
dorsal rami. J Anat 134: 383–397.
Bogduk N, Windsor M, Inglis A (1988). The innervation of the
cervical intervertebral discs. Spine 13: 2–8.
Bogduk N, Pearcy M, Hadfield G (1992). Anatomy and biome-
chanics of psoas major. Clin Biomech 7: 109–119.
Bogduk N, Johnson G, Spalding D (1998). The morphology
and biomechanics of latissimus dorsi. Clin Biomech 13:
377–385.
Bogduk N, Aprill C, Derby R (2013). Lumbar discogenic pain:
state-of-the-art review. Pain Med 14: 813–836.
Chua WH, Bogduk N (1995). The surgical anatomy of thoracic
facet denervation. Acta Neurochir 136: 140–144.
Curatolo M, Bogduk N, Ivancic PC et al. (2011). The role of
tissue damage in whiplash-associated disorders. Spine 36:
S309–S315.
Dvorak J, Hayek J, Zehnder R (1987). CT-functional diagnos-
tics of the rotatory instability of the upper cervical spine
part 2. An evaluation on healthy adults and patients with
suspected instability. Spine 12: 726–731.
Engel R, Bogduk N (1982). The menisci of the lumbar zyga-
pophysial joints. J Anat 135: 795–809.
Fielding JW, Cochran G van B, Lawsing JF et al. (1974). Tears
of the transverse ligament of the atlas. J Bone Joint Surg
56A: 1683–1691.
Hickey DS, Hukins DWL (1980). Relation between the struc-
ture of the anulus fibrosus and the function and failure of
the intervertebral disc. Spine 5: 100–116.
Horwitz T, Smith RM (1940). An anatomical, pathological
and roentgenological study of the intervertebral joints
of the lumbar spine and of the sacroiliac joints. Am
J Roentgenol 43: 173–186.
Kaneoka K, Ono K, Inami S et al. (1999). Motion analysis of
cervical vertebrae during whiplash loading. Spine 24:
763–770.
Kimmel DL (1960). Innervation of the spinal dura mater and
dura mater of the posterior cranial fossa. Neurology 10:
800–809.
Koebke J, Brade H (1982). Morphological and functional stud-
ies on the lateral joints of the first and second cervical ver-
tebrae in man. Anat Embryol 164: 265–275.
Lambert GA, Duckworth JW, Bogduk N et al. (1984). Low
pharmacological responsiveness of the vertebro-basilar cir-
culation in Macaca nemestrina monkeys. Eur J Pharmacol
102: 451–458.
Lazorthes G, Gaubert J (1956). L’innervation des articulations
interapophysaire vertebrales. Comptes Rendues de
l’Association des Anatomistes 43: 488–494.
Macintosh JE, Bogduk N, Gracovetsky S (1987). The biome-
chanics of the thoracolumbar fascia. Clin Biomech 2:
78–83.
Markolf KL, Morris JM (1974). The structural components
of the intervertebral disc. J Bone Joint Surg 56A:
675–687.
Mercer S, Bogduk N (1993). Intra-articular inclusions of the
cervical synovial joints. Br J Rheumatol 32: 705–710.
Mercer S, Bogduk N (1999). The ligaments and anulus fibro-
sus of human adult cervical intervertebral discs. Spine 24:
619–626.
688 N. BOGDUK
Mercer SR, Bogduk N (2003). Clinical anatomy of ligamen-
tum nuchae. Clin Anat 16: 484–493.
Nowitzke A, Westaway M, Bogduk N (1994). Cervical zyga-
pophyseal joints: geometrical parameters and relationship
to cervical kinematics. Clin Biomech 9: 342–348.
Oda J, Tanaka H, Tsuzuki N (1988). Intervertebral disc
changes with aging of human cervical vertebra from the
neonate to the eighties. Spine 13: 1205–1211.
Pearcy MJ, Bogduk N (1988). Instantaneous axes of rotation of
the lumbar intervertebral joints. Spine 13: 1033–1041.
Pearcy MJ, Tibrewal SB (1984). Axial rotation and lateral
bending in the normal lumbar spine measured by three-
dimensional radiography. Spine 9: 582–587.
Phillips S, Mercer S, Bogduk N (2008). Anatomy and
biomechanics of quadratus lumborum. J Eng Med 222:
151–159.
Standring S (Ed.), (2008). Gray’s Anatomy, 40th edn.
Churchill Livingstone, Edinburgh, pp. 736–743.
Tondury G (1972). The behaviour of the cervical discs during
life. In: C Hirsch, Y Zotterman (Eds.), Cervical pain,
Pergamon Press, Oxford, pp. 59–66.
Wasserman BR, Moskovitch R, Razi AE (2011). Rheumatoid
arthritis of the cervical spine. Clinical considerations. Bull
Hosp Joint Dis 68: 136–148.
Winkelstein BA (2011). How can animal models inform on the
transition to chronic symptoms in whiplash? Spine 36:
S218–S225.
Wortzman G, Dewar FP (1968). Rotary fixation of the atlan-
toaxial joint: rotational atlantoaxial subluxation.
Radiology 90: 479–487.
Yin W, Bogduk N (2008). The nature of neck pain in a private
pain clinic in the United States. Pain Med 9: 196–203.
Handbook of Clinical Neurology, Vol. 136 (3rd series)
Neuroimaging, Part II
J.C. Masdeu and R.G. González, Editors
© 2016 Elsevier B.V. All rights reserved

Chapter 33

Neuroimaging of spine tumors

NANDOR K. PINTER, THOMAS J. PFIFFNER, AND LASZLO L. MECHTLER*
Dent Neurologic Institute, Amherst, NY, USA

Abstract
Intramedullary, intradural/extramedullary, and extradural spine tumors comprise a wide range of neo-
plasms with an even wider range of clinical symptoms and prognostic features. Magnetic resonance imag-
ing (MRI), commonly used to evaluate the spine in patients presenting with pain, can further characterize
lesions that may be encountered on other imaging studies, such as bone scintigraphy or computed tomog-
raphy (CT). The advantage of the MRI is its multiplane capabilities, superior contrast agent resolution,
and flexible protocols that play an important role in assessing tumor location, extent in directing biopsy, in
planning proper therapy, and in evaluating therapeutic results. A multimodality approach can be used to
fully characterize the lesion and the combination of information obtained from the different modalities
usually narrows the diagnostic possibilities significantly. The diagnosis of spinal tumors is based on
patient age, topographic features of the tumor, and lesion pattern, as seen at CT and MRI. The shift
to high-end imaging incorporating diffusion-weighted imaging, diffusion tensor imaging, magnetic res-
onance spectroscopy, whole-body short tau inversion recovery, positron emission tomography, intrao-
perative and high-field MRI as part of the mainstream clinical imaging protocol has provided
neurologists, neuro-oncologists, and neurosurgeons a window of opportunity to assess the biologic behav-
ior of spine neoplasms. This chapter reviews neuroimaging of spine tumors, primary and secondary,
discussing routine and newer modalities that can reduce the significant morbidity associated with these
neoplasms.

INTRODUCTION advances in the care of patients with spine tumors.

The characterization of spine tumors by MRI involves
The historic classification of spine tumors is based on
determining, in the context of patient’s age and sex,
the use of myelography with three main groups, as sche-
the location of the lesion and whether or not it enhances
matically depicted in Figure 33.1: (1) extradural extrame-
after gadolinium injection. Computed tomography (CT)
dullary; (2) intradural extramedullary; and (3) intradural
best delineates osseous integrity while MRI is better
intramedullary. The incidence of metastatic disease
at assessing soft-tissue involvement. The purpose of
involving the vertebrae, epidural space, and leptome- this chapter is to describe the neuroimaging findings
ninges accounts for 97% of tumors involving the
of spine tumors based on the location of the tumor in
spine. Primary tumors of the spine, spinal cord, spinal
its relationship to the dura and spinal cord (Mechtler
meninges, and cauda equina are relatively rare (Duong
and Nandigam, 2013).
et al., 2012). Data from national registries and improved
imaging capabilities have allowed spine tumor specialists
METASTATIC TUMORS OF THE
the opportunity to study and treat these unusual and rare
VERTEBRAL COLUMN
tumors with more confidence and better results. The
introduction of magnetic resonance imaging (MRI) to The spine is the third most common site for metastatic
clinical practice has been one of the most important disease and the most common site for bone metastasis

*Correspondence to: Laszlo L. Mechtler, MD, Professor of Neurology and Oncology, Dent Neurologic Institute, 3980 Sheridan
Drive, Amherst NY 14226, USA. Tel: +1-716-250-2000, Fax: +1-716-250-2045, E-mail: lmechtler@dentinstitute.com
690 N.K. PINTER ET AL.
Fig. 33.1. Historic classification of spine tumors based on computed tomography myelography. (A) Normal, (B) extradural extra-
medullary, (C) intradural extramedullary, and (D) intradural intramedullary. From Mechtler and Nandigam (2013).
(Shah and Salzman, 2011). Metastatic disease of the ver-
tebral column is more frequent than primary neoplastic
diseases. Approximately two-thirds of cancer patients
will develop bone metastasis and symptomatic spinal
metastasis will occur in almost 10% of cancer patients.
The most common primary sites are the prostate, breast,
kidney, lung, and thyroid. The incidence of skeletal
metastases according to the primary tumor is as follows;
breast 73% (47–85%), prostate 68% (33–85%), thyroid
42% (28–60%), lung 36% (30–55%), kidney 35%
(33–40%), esophageal 6% (5–7%), and gastrointestinal
5% (3–11%) (Maccauro et al., 2011) The most common
cause of metastatic spine disease is breast cancer in
women; however, in men, prostate cancer is most com-
mon. The thoracic spine is the most commonly involved.
The majority of the lesions are extradural in location,
consisting of lesions which are localized to the epidural
space and those which are nested in the vertebral body.
Prostate, breast, and lung cancer are again the leading
cause of spinal cord compression, each accounting for
about 15–20% of the cases. The remaining cancers stem
from renal cell, non-Hodgkin’s lymphoma, multiple
myeloma, colorectal cancers, sarcomas, and unknown
tumors. Pain, the most common initial feature, occurs
in 95% of adults and 80% of children. Pain is usually
localized to the site of metastasis and is caused by
stretching the pain-sensitive bony periosteum. Radicular
pain is less frequent but is also localizing. Nocturnal pain
upon lying down is typical.
Three types of bone metastasis are distinguished:
osteolytic, osteoblastic, and mixed; 71% are osteolytic,
8% are osteoblastic, and 21% are mixed.
Osteolytic metastases typically develop in cancers
of the breast, lung, kidney, thyroid, oropharyngeal can-
cers (Shah and Salzman, 2011) and in melanoma (Sun
et al., 2013). This is a result of osteoclast activation, rather
than a direct invasion of bone tissue by tumor cells.
In osteoblastic metastases the balance of bone meta-
bolism is shifted to the benefit of bone production as
a result of pathologic activation of osteoblasts. Osteo-
blastic lesions usually occur in prostate, bladder and
nasopharyngeal cancer, medulloblastoma, neuroblasto-
mas, and bronchial carcinoid (Long et al., 2010).
Imaging of vertebral metastases
In today’s clinical practice MRI is the most important
modality in imaging of metastatic spine disease. Plain
film is no longer the routine diagnostic toolbar due to
its low sensitivity and specificity (Salvo et al., 2009;
Shah and Salzman, 2011). Nuclear medicine studies have
a well-defined role in metastasis imaging. Bone scans
have been used for screening, since the tracer accumu-
lates in metastatic sites with high sensitivity, thus reflect-
ing the increased bone turnover. The sensitivity and
specificity of bone scans were improved with single-
photon emission computed tomography (SPECT) scans
(Ryan and Fogelman, 1995). Flurodeoxyglucose (F18-
FDG) positron emission tomography (PET) alone and
PET CT can discover spinal metastases with a sensitivity
of 74% and 98%, respectively (Metser et al., 2004). F18-
FDG PET has been reported to be more sensitive in
detecting osteolytic metastases (Cook and Fogelman,
2000). CT has a lower sensitivity in detecting osseous
metastases and an inferior diagnostic accuracy com-
pared to MRI (Buhmann Kirchhoff et al., 2009). In fact,
CT is less accurate in detecting paraspinal soft tissue,
bone edema, and bone metastases that may be missed
if destruction is not present (Shah and Salzman, 2011).
Therefore, CT has a rather complementary role in
first-line imaging of spinal metastases and owns priority
only in those cases when the integrity and fine structure
of the trabecular and cortical bone are a question, preop-
erative planning is required, or when MRI is contraindi-
cated. MRI is superior to CT in all other cases.
Metastatic lesions are most commonly focal or multi-
focal and the diffuse involvement of the vertebral bodies
is less common. Focal abnormalities are hypointense
on T1 and hyperintense on T2 and short tau inversion
recovery (STIR) sequences. In general, metastases will
enhance with contrast, although it is important to always
acquire a noncontrast study for comparison. The diffuse
 NEUROIMAGING OF SPINE TUMORS 691

Fig. 33.2. Diffuse metastatic involvement of the cervical and upper thoracic spine. The T1-weighted image (A) shows heteroge-
neous signal intensity in all vertebral bodies, with mostly isointense to hypointense signal accompanied by focal hyperintense areas
(open arrow) compared to intervertebral discs. On T2-weighted image (B), similar heterogeneity can be seen with hyperintense and
isointense areas. On short tau inversion recovery (STIR) sequence (C) diffuse hyperintense signal correlates with the T1 hypoin-
tensity, typical of metastatic disease of the bone marrow. The involvement of spinous processes is highly suspicious on T2 and
becomes unquestionable on STIR (arrows). The focal hyperintensities on T1 are associated with fatty marrow and represented as
hypointense signal on STIR.

marrow involvement can be difficult to assess, because a
generally low signal intensity appearance can be mislead-
ing, giving the false impression of normal marrow. It is
helpful to compare the marrow’s signal intensity to that
of the discs and muscles. In adults it can be regarded as
abnormal if the marrow has lower signal intensity than
discs or muscles (Fig. 33.2) (Long et al., 2010). Metasta-
ses tend to occur in the posterior part of the vertebral
body, involving the pedicles. Most often metastases
are destructive and can be expansive.
In osteolytic metastases, the cancellous bone is
replaced by tumor tissue and acceleration of bone
resorption occurs due to an imbalance in metabolism.
The focal disequilibrium in calcified elements will create
the characteristic image of circumscribed translucency
on radiographs and a generally low-density area framed
by intact, partly intact, or sclerotic osseous components
on CT. On MRI hypointense T1, hyperintense T2, and
STIR signals will represent the structural and biochem-
ical (e.g., edema) changes in the bone marrow (Long
et al., 2010). Posterior cortical and pedicle destruction
is not infrequent and best visualized by CT. During
the course of antitumor therapy progressive sclerosis
may be visible on follow-up imaging, indicating positive
response to therapy. Fluid levels may occur in lytic
metastases (Jarraya et al., 2013) and might even mimic
primary tumors (Colangeli et al., 2010). The term osteo-
blastic refers to the biologic behavior of the lesions, but
from the imaging point of view, sclerotic or osteosclero-
tic may sound more practical, as they reflect the imaging
findings. Osteoblastic metastases are usually focal or
mottled and in most cases are multifocal. Diffuse
involvement of the vertebral body occurs occasionally.
On MRI the lesions show T1 and T2 hypointensity. On
T2 and STIR images a bright rim surrounding the scle-
rotic lesions may be depicted, reflecting bone marrow
edema. This is known as the “halo” sign. Edematous sig-
nal can often be observed on STIR, expanding from the
body to the pedicles, without a well-defined lesion on T1
or T2 sequences. The enhancement pattern in MRI may
vary depending on the degree of sclerosis. On CT and
plain film these lesions show high density resulting from
excessive calcification. The epidural expansion of the
tumor may create the “draped curtain sign,” which can
be depicted on axial MRI slices (Fig. 33.3).
Vertebral compression deformity
Vertebral compression deformity is frequent in elderly
patients. The cause of this abnormality can be benign
or malignant. Distinguishing between benign and patho-
logic vertebral body compression fractures is usually
possible on MRI. Chronic benign fractures typically
have marrow signal intensity that is isointense with nor-
mal vertebrae on all sequences. In addition, there is
no involvement of the posterior elements, absence of
paravertebral or epidural mass, and preservation of pos-
terior cortex. Pathologic fractures show comparatively
low signal intensity on T1- and high signal intensity
on T2-weighted sequences (Fig. 33.4) (Griffith and
Guglielmi, 2010). Pathologic compression fractures
typically enhance with contrast; however, conventional
MR techniques cannot always be used to differentiate
benign from malignant lesions due to their similar
appearances. For example, osteoporotic compression
fracture can be confused with metastatic compression
692 N.K. PINTER ET AL.

Fig. 33.3. A 48-year-old man with multiple myeloma. Sagittal (A) and axial (B) contrast-enhanced T1-weighted scans with fat-
suppression shows a large enhancing mass occupying the T12 vertebral body, extending into the spinal canal and the left pedicle
(arrows). The yellow arrows on the axial image mark as the tumor invades the anterior epidural space bilaterally and compression
on the thecal sac, creating the “draped curtain sign.” Reformatted computed tomography in the coronal and sagittal planes
(C and D) shows the osteolytic nature of this tumor. This was thought to be a solitary plasmacytoma. Further imaging revealed
“punched-out” osteolytic lesions (black arrows) characteristic of multiple myeloma.

Fig. 33.4. Compression fracture resulting from metastatic disease. T1 hypointense (A), T2 intermediate-hyperintense (B) and
short tau inversion recovery STIR hyperintense signal (C) is demonstrated in the compressed T9 vertebral body. The body shows
a vertebral plana deformity. Posteriorly focal sclerotic changes can be observed on the sagittal reconstruction of CT scan (D).

in the acute phase. Edema in an acute benign compres-
sion fracture replaces the normal marrow, resulting in
hypointensity on T1-weighted images and hyperintensity
on T2-weighted images. The vertebral body with benign
fracture may enhance with contrast. These MR signal
intensity characteristics are similar to those of metastases
and cause ambiguity, especially when only a single lesion
is present. Pathologic fractures are often multiple; other
key features of pathologic compression fractures include
loss of the posterior body height, pedicle involvement,
epidural and paraspinal mass.
Diffusion-weighted imaging (DWI) is emerging as a
powerful clinical tool for directing the care of patients
with cancer (Padhani et al., 2011). The basic biologic pre-
mise for the use of DWI is the characteristic increased
cellular content of malignant tissue, and increased water
content. These features result in higher signal intensity
of malignant disease on high b-value images with corre-
sponding low apparent diffusion coefficient (ADC)
values (Khoo et al., 2001). This is especially helpful in
patients with neuroblastomas, leukemia, lymphoma,
rhabdomyosarcoma, Ewing sarcoma, and metastatic
primitive neuroectodernal tumors. DWI can also be a
useful tool for distinguishing acute benign osteoporotic
from malignant vertebral compression fractures, result-
ing in low or isointense signal on DWI and high ADC
values on ADC maps (Duvauferrier et al., 2013;
Rumpel et al., 2013).
Whole-body STIR has also been used to monitor dis-
ease progression in patients with lymphoma, multiple
myeloma, metastatic primitive neuroectodernal tumors,
and neurofibromatosis (Fayad et al., 2013).
PRIMARY TUMORS OF THE SPINE
Benign primary spine tumors
Hemangioma is the most common primary tumor of the
spine; the incidence from autopsy studies is estimated to
 NEUROIMAGING OF SPINE TUMORS
be 10% (Ropper et al., 2011). Histologically they consist
of multiple small, thin-walled vessels interspersed with
trabeculae and infiltrating the medullary cavity. Heman-
giomas have a characteristic, polka-dot appearance on
the axial CT images, which represent the cross-section
of the prominent vertical trabeculae. On MRI lesions
show high signal intensity on T1 and T2 images, often
with a heterogeneous structure, which may also contain
flow voids as signs of vascular elements. Osteoid
osteomas occur in young adults and their most common
symptom is a painful scoliosis that worsens at night
(Theodorou et al., 2008). They usually arise from the
posterior elements. Lumbar spine involvement is typical.
On radiographs osteoid osteomas have the appearance
of a lytic lesion surrounded by sclerotic rim. The same
finding is present on CT scans. The sclerotic changes
may affect the central area or nidus. On T1 images the
nidus displays intermediate signal intensity and can have
signal voids as a result of calcification. On T2 images the
nidus appears as a low-intensity area surrounded by
high-intensity rim representing edema. Osteoblastomas
have a nonspecific MRI appearance, with T1 and T2 pro-
longation depending on the degree of matrix mineraliza-
tion. Osteoblastomas have a reactive rim similar to
osteoid osteomas, which can lead to overestimation of
lesion (Chai and Cho, 2013).
Aneurysmal bone cysts are expansile lesions contain-
ing blood-filled cysts separated by walls of bone (Long
et al., 2010). The spine is affected in 12–30% of all aneu-
rysmal bone cyst cases. On CT scans ballooning, lobu-
lated cystic lesions can be found occasionally with
693
fluid–fluid levels (Ropper et al., 2011). The latter is indic-
ative of hemorrhage and may be more sensitively visual-
ized by MRI (Fig. 33.5). Signal changes, representing
blood degradation products of different age, can be
appreciated (Vidal and Murphey, 2007). Giant cell tumor
is a benign, locally aggressive tumor that accounts for
approximately 4–5% of primary bone tumors. Most
commonly, giant cell tumors are found in long bones.
CT and MRI may reveal an associated soft-tissue com-
ponent, and contrast administration may help to differ-
entiate the bony and soft-tissue elements. On MRI, they
appear as low to intermediate signal intensity on T1- and
T2-weighted imaging. Giant cell tumors often cause the
destruction of sacral foramina. Osteochondroma is an
uncommon tumor, accounting for about 4% of solitary
spinal tumors. Osteoschondromas seem to demonstrate
distinct predilection for certain parts of the spine, specif-
ically C2. It is often difficult to discover the lesion on
radiograph or MRI images due to its small size and
the continuity with the bone it originates from. Thin-slice
CT scans may be needed to confirm the diagnosis (Vidal
and Murphey, 2007).
Malignant primary spine tumors
Chondrosarcoma is the second most common nonlym-
phoproliferative malignant tumor of the spine in adults
(Rodallec et al., 2008). The mineralized matrix is best
visualized by CT, and areas of calcification are repre-
sented as signal voids on MRI. The noncalcified tumor
shows low attenuation on CT, low to intermediate signal

Fig. 33.5. Aneurysmal bone cyst (ABC). Sagittal and axial T2-weighted images (A and B respectively) show a well-defined, mul-
tilobulated lesion in the L3 vertebral body. The lesion consists of several small cystic compartments which demonstrate sharply
separated areas of T2 hyperintense and intermediate signal, representing fluid–fluid levels due to gravitational forces. This is best
demonstrated on the axial T2 scan due to the supine position of the patient (white open arrow, B). The lesion extends to the spinal
canal and causes the compression of the dural sac (white arrows). Computed tomography (CT) scan (C) shows an almost complete
replacement of osseous tissue by the lesion with only a thin layer of cortical bone remaining on the right side (black open arrow). It
also depicts the ballooning character of the lesion and some sclerotic changes on the margins. The invasive nature of the ABC is
also highlighted by the partial destruction of the upper endplate as displayed by the sagittal T2 scan (asterisk). On sagittal recon-
struction of the CT scan (D) it is also clear that the lesion extends posteriorly into the pedicle (black arrow).
Another random document with
no related content on Scribd:
 Silence. By my troth, I was not there.
Shallow. Death is certain. Is old Double of your town living
yet?
Silence. Dead, sir.
Shallow. Jesu, Jesu, dead! a’ drew a good bow: and dead! a’
shot a fine shoot: John a Gaunt loved him well, and betted much
money on his head. Dead!—a’ would have clapped i’ the clout at
twelve score; and carried you a forehand shaft a fourteen and
fourteen and a half, that it would have done a man’s heart good
to see. How a score of ewes now?
Silence. Thereafter as they be: a score of good ewes may
be worth ten pounds.
Shallow. And is old Double dead?

You get little or none of that solemn, sweet rusticity in Dickens: nor of
the rush of England in spring with slow country-folk watching it:

The fields breathe sweet, the daisies kiss our feet,

Young lovers meet, old wives a-sunning sit;
In every street these tunes our ears do greet—
Cuckoo, jug-jug, pu-we, to-witta-woo!
Spring, the sweet Spring!

You will remember that Pickwick, in its first conception, was to deal
with the adventures and misadventures of a Sporting Club after the
fashion of the Handley Cross series by Surtees. Now Surtees—not a
great writer but to this day (at any rate to me) a most amusing one—
was, although like Dickens condemned to London and the law, a
north-country sportsman, and could ride and, it is reported, “without
riding for effect usually saw a deal of what the hounds were doing.”
The Pickwickian sportsmen had to decline that competition very
soon.
 V
But they, and a host of Dickens’ characters, are very devils for
post-chaises.
“If I had no duties, and no deference to futurity, I would spend my
life in driving briskly in a post-chaise with a pretty woman,” said Dr.
Johnson. “There are milestones on the Dover Road,” and we spin
past them. You will remember that Dickens in his apprenticeship
spent a brief but amazingly strenuous while as reporter for the
Morning Chronicle, scouring the country after political meetings by
road-vehicles in all weathers. As he told his audience, twenty years
later, at the annual dinner of the Newspaper Press Fund:

I have often transcribed for the printer, from my shorthand

notes, important public speeches in which the strictest accuracy
was required, and a mistake in which would have been, to a
young man, severely compromising, writing on the palm of my
hand, by the light of a dark lantern, in a post-chaise and four
galloping through a wild country, and through the dead of night,
at the then surprising rate of fifteen miles an hour.... Returning
home from exciting political meetings in the country (and it might
be from Exeter west, or Manchester north) to the waiting press in
London, I do verily believe I have been upset in almost every
description of vehicle known in this country. I have been, in my
time, belated in miry by-roads, forty or fifty miles from London, in
a wheel-less carriage, with exhausted horses and drunken post-
boys, and have got back in time for publication....

So, you see, this world Dickens imagined was more than crowded; it
was a hurrying, a breathless one. This sense of speed in travel, of
the wind in one’s face; of weight and impetus in darkness, with
coach lamps flaring through the steam from your good horses’
hindquarters, runs as an inspiration through much of the literature of
the early nineteenth century. De Quincey has hymned it
magnificently in The English Mail Coach, and you may enjoy a
capital drive of the sort in Tom Brown’s School Days: and always the
rush of air whets your appetite for the hot rum-and-water at the stage
hostelry or the breakfast of kidney-pie. Dickens saw the invasion of
the railway train, and lived to be disastrously mixed up in a railway
collision. But railway-train travelling at sixty miles an hour or over,
has a static convenience. For the pleasures of inconvenient travel,
without a time-table, I have recourse to a sailing-boat: but I can well
understand my fellow-creature who prefers a car or a motor-bicycle
to the motion of four horses at a stretch gallop. With the wind of God
in his face he gets there (wherever it is) before the dew is dry, does
his business, swallows his bun and Bovril and is home again with an
evening paper for the cosy gas-cooked meal, ere yet Eve has drawn
over his little place in the country her gradual dusky veil.
Rapid travel, as Dickens well knew it and how to describe it—
with crime straining from what it fears—is one of his most potent
resources. Read the flight of Carker in Dombey and Son.

VI
His is a crowded world then, tumultuous and full of fierce hurry:
but a world (let us grant it) strangely empty of questioning ideas,
subtle nuisances that haunt many thoughtful men’s souls, through
this pass of existence “still clutching the inviolable shade.” He wrote
far better novels than John Inglesant, novels far, far, better than
Robert Ellesmere; but you cannot conceive him as interested in the
matter of these books—which yet is serious matter. Still less, or at
least as little, can you imagine him pursuing the track of so
perplexed a spirit as Prince André in Tolstoy’s War and Peace.
Churches annoyed him. He will, of a christening or a marriage
service (let be a funeral), make the mouldiest ceremony in the world.
We offer the baby up; we give the blushing bride away; but in the
very act we catch ourselves longing for that subsequent chat with the
pew-opener which he seldom denies us for reward. Dickens, in
short, had little use for religious forms or religious mysteries: for he
carried his own religion about with him and it was the religion of
James—so annoying alike to the mystic and the formalist—“to visit
the fatherless and widows in their affliction, and to keep himself
unspotted from the world.” This again belongs to his “universality.” Is
it not the religion of most good fellows not vocal? It is observable
how many of his heroes and heroines—his child heroes and
heroines especially—pass through his thronged streets and keep
themselves unspotted.
But, if careless of mysteries, Dickens had a hawk’s eye for truth
of morals. You never find him mocking a good or condoning an evil
thing: here his judgment and its resultant passion of love or of hate, I
dare to say, never went wrong. Sinners—real sinners—in Dickens
have the very inferno of a time: the very forces of Nature—“fire and
hail, snow and vapours, wind and storm, fulfilling God’s word”—hunt
the murderer to the pit that yawns; till he perishes, and the sky is
clear again over holy and humble men of heart. Again, witness, here,
the elemental flight of Jonas Chuzzlewit. Carlyle never said an
unjuster thing (and that is saying a deal) than when he accused
Dickens’ theory of life as entirely wrong. “He thought men ought to
be buttered up ... and all sorts of fellows have turkey for their
Christmas dinner.” It is false. Dickens had a keener eye for sin than
Carlyle ever had; and a relentless eye: “a military eye,” said Henry
James of it, recalling his first introduction to the great man—“a
merciless military eye.” “A field-punishment eye,” I say!

VII
But this world of Dickens, you may object, was an unreal world,
a phantasmagoric world. Well, I hope to discuss that—or rather the
inference from it—in my next lecture, which shall deal, in Aristotelian
order, with his plots first and his characters next. But, for the
moment, if you will, Yes: his world was like nothing on earth: yes, it is
liker to Turner’s sunset to which the critic objected “he never saw a
sunset like that,” and was answered, “Ah, but don’t you wish you
could?” Yes, for Dickens made his world—as the proud parent said
of his son’s fiddle—“he made it, sir, entirely out of his own head!”
“Night is generally my time for walking” (thus begins Master
Humphrey, in The Old Curiosity Shop) “although I am an old man.”
So in that crowded phantasmagoric city of London, which is in
his mind, Dickens walks by night—not like Asmodeus, lifting the
roofs and peering into scandals: but like the good Caliph of his
favourite Arabian Nights, intent to learn the life of the poor and
oppressed, and as a monarch to see justice done them: a man
patterning his work on the great lines of Fulke Greville, sometime of
Jesus College, in this town; with which let me conclude to-day:

The chief use, then, in man, of that he knows,

Is his painstaking for the good of all:
Not fleshly weeping for our own-made woes,
Not laughing from a melancholy gall,
Not hating from a soul that overflows
With bitterness, breath’d out from inward thrall:
But sweetly rather to ease, loose, or bind,
As need requires, this frail, fall’n humankind.
 DICKENS (III)

I
I LEFT you, Gentlemen, with a promise to say something on
Dickens’ plots and Dickens’ characters, taking them in that
Aristotelian order. Now why Aristotle, speaking of drama, prefers Plot
to Character; if his reasons are sound; if they are all the reasons;
and, anyhow, if they can be transferred from drama and applied to
the Novel; are questions which some of you have debated with me
“in another place,” and, if without heat, yet with all the vigour
demanded by so idle a topic. But, for certain, few of you will dissent
when I say of Dickens that he is memorable and to be loved (if loved
at all) for his characters rather than for his plots. You have (say) a
general idea of Dombey and Son, a vivid recollection of Captain
Cuttle, Mr. Toots, Susan Nipper, perhaps a vivid recollection of
Carker’s long, hunted flight and its appalling end, when the pursuer,
recovering from a swoon—

saw them bringing from a distance something covered ... upon a

board, between four men, and saw that others drove some dogs
away that sniffed upon the road, and soaked his blood up, with a
train of ashes.

Or you have a general idea of Our Mutual Friend, and your memory
preserves quite a sharp impression of Silas Wegg, Mr. Boffin, the
Doll’s Dressmaker. But if suddenly asked how Carker’s flight came
about, why Boffin practised his long dissimulation, and what
precisely Wegg or the Doll’s Dressmaker had to do with it—could
you, off-hand, supply a clear answer? Some votaries can, no doubt:
but I ask it of the ordinary reader. Myself indeed may claim to be
something of a votary, with an inexplicably soft spot in my heart for
Little Dorrit: yet, and often as I have read that tale, I should be
gravelled if asked, at this moment, to tell you just what was the
secret of the old house, or just what Miss Wade and Tattycoram have
to do with the story. Somehow, in retrospect, such questions do not
seem to matter.
In truth, as I see it—and foresee it as a paradox, to be defended
—Dickens was at once, like Shakespeare in the main, careless of his
plots, and, unlike Shakespeare, over-anxious about them. I shall
stress this second point, which stabs (I think) to the truth beneath the
paradox, by and by.
But first I ask you to remember that Dickens habitually published
a novel in monthly numbers or instalments; starting it, indeed, upon a
plan, but often working at white heat to fulfil the next instalment, and
improvising as he went. Thackeray used the same method, with the
printer’s devil ever infesting the hall when the day for delivery came
around. This method of writing masterpieces may well daunt their
successors, even in this journalistic age of internal combustion with
the voice of Mr. H. G. Wells insistent that the faster anyone travels
the nearer he is ex hypothesi to that New Jerusalem in which there
shall be no night (and therefore, I presume, not a comfortable bed to
be hired), but the eternal noise of elevators and daylight-saving
made perfect. It did not daunt our forefathers: who were giants of
their time, undertook a Pendennis or a Dombey and Son, and having
accomplished a chapter or so, cheerfully went to bed and slept under
that dreadful imminent duty. You all know, who have studied
Pickwick, that Pickwick began (so to speak) in the air; that it took the
narrative, so desultory in conception, some numbers before it found
a plot at all. But how admirable is the plot, once found or—to say
better—once happened on! For a double peripeteia who could ask
better art than the charitable turn of Pickwick on Jingle in the
debtor’s prison, and the incarceration and release of Mrs. Bardell?
Consider the first. Insensibly, without premonition of ours and I dare
to say, of no long prepared purpose in the author, the story finds a
climax:
 “Come here, Sir,” said Mr. Pickwick, trying to look stern, with
four large tears running down his waistcoat. “Take that, Sir.”
Take what? In the ordinary acceptation of such language, it
should have been a blow. As the world runs, it ought to have
been a sound hearty cuff: for Mr. Pickwick had been duped,
deceived, and wronged by the destitute outcast who was now
wholly in his power. Must we tell the truth? It was something
from Mr. Pickwick’s waistcoat pocket which chinked as it was
given into Job’s hand....

But this admirable plot, with all the Bardell versus Pickwick
business, and the second most excellent “reversal of fortune” when
Mrs. Bardell, the prosecutrix, herself gets cast into prison by Dodson
and Fogg whose tool she has been, and there, confronted by her
victim and theirs, finds herself (O wonder!) pardoned—with the
simple, sudden, surprising, yet most natural and (when you come to
think of it) most Christian story of Sam Weller’s loyalty and Mr.
Weller’s aiding and abetting, so absurdly and withal so delicately
done—all this grew, as everyone knows, with the story’s growth and
grew out of fierce, rapid, improvisation. You can almost see the
crucible with the fire under it, taking heat, reddening, exhaling fumes
of milk-punch; and then, with Sam Weller and Jingle cast into it for
ingredients, boiling up and precipitating the story, to be served

as a dish
Fit for the gods

—“served,” not “carved.” You cannot carve the dish of your true
improvisatore. You cannot articulate a story of Dickens—or, if you
can, “the less Dickens he”: you may be sure it is one of his worst. A
Tale of Two Cities has a deft plot: well-knit but stagey: and, I would
add, stagey because well-knit, since (as we shall presently see)
Dickens, cast back upon plot, ever conceived it in terms of the stage;
of the stage, moreover, at its worst—of the early-Victorian stage,
before even a Robertson had preluded better things. So, when I talk
to any man of Dickens, and he ups with his first polite concession
that A Tale of Two Cities is a fine story, anyhow, I know that man’s
case to be difficult, for that he admires what is least admirable in
Dickens. Why, Gentlemen, you or I could with some pains construct
as good a plot as that of A Tale of Two Cities; as you or I could with
some pains construct a neater plot than Shakespeare invented for
The Merry Wives of Windsor or even hand out some useful
improvements on the plot of King Lear. The trouble with us is that we
cannot write a Merry Wives, a Lear; cannot touch that it which,
achieved, sets the Merry Wives and Lear, in their degrees, above
imperfection, indifferent to imperfections detectable even by a fool.
Greatness is indefinable, whether in an author or a man of affairs:
but had I to attempt the impossibility, no small part of my definition
would set up its rest on indifference—on a grand carelessness of
your past mistakes, involving a complete unconcern for those who
follow them, to batten on the bone you have thrown over your
shoulder.

II
Dickens was a great novelist—as I should contend, the greatest
of English novelists—and certainly among the greatest of all the
greatest European novelists. His failing was that he did not quite
trust his genius for the novel, but was persuaded that it could be
bettered by learning from the drama—from the bad drama of his
time. But I want you to see, Gentlemen, how honourable was the
artist’s endeavour; how creditable, if mistaken, to the man. He was a
born improvisatore. Pickwick, under your eyes, takes a shape—
conceives it, finds it—as the story goes on. Then shape he must
struggle for; the idea of “shape” has, against his genius, taken hold
on him. So Pickwick is not finished before he begins a new story,
never thinking to repeat, by similar methods, Pickwick’s
overwhelming success. No, the responsibility of that success weighs
on him; but it is a responsibility to improve. The weakness of
Pickwick, undertaken as a series of mock-sporting episodes, lies in
its desultoriness. This time we will have a well-knit plot. And so we
get Oliver Twist and Nicholas Nickleby, each with any amount of plot,
but of plot in the last degree stagey; so stagey, indeed, that in
Nickleby the critic gasps at the complacency of an author who,
having created that “nurseling of immortality” Mr. Vincent Crummles,
together with a world and the atmosphere of that world in which
Crummles breathes and moves and has his being, can work the
strings of the puppet with so fine a finger, detect its absurdities with
so sure an instinct and reveal them with so riotous a joy; yet misses
to see that he himself is committing absurdities just as preposterous,
enormities of the very same category, on page after page. The story
of Lord Frederick Verisopht and Sir Mulberry Hawke, for example, is
right Crummles from beginning to end. Crummles could have
composed it in his sleep,—and to say this, mind you, is to convey in
the very censure an implicit compliment—or, shall I use a more
modest word and say implicit homage? Crummles could have written
a great part of Nickleby: but Crummles could only have written it
after Dickens had made him. I seem to hear the two arguing it out in
some Dialogue of the Dead.

Auctor. “My dear Crummles, however did you contrive to be

what you are?”
Crummles. “Why, don’t you see, Mr. Dickens? You created
me in your image.” (sotto voce) “And, he doesn’t know it, poor
great fellow, but it seems to me I’ve been pretty smart in
returning the compliment.”

III
I have said, in a previous lecture, that Dickens, from first to last,
strove to make himself a better artist; quoting to you a sentence of
Henley’s, which I repeat here because you have almost certainly
forgotten it:
 He had in him at least as much of the French artist as of the
middle-class Englishman; and if in all his life, he never ceased
from self-education, but went unswervingly in the pursuit of
culture, it was out of love for his art and because his conscience
as an artist would not let him do otherwise.

“Unswervingly”?—no, not unswervingly. No great genius that ever

was has marched unswervingly on. As a condition of becoming a
great artist he must be more sensitive than his fellows; as a result of
that sensitiveness he will doubt, hesitate, draw back to leap the
better. The very success of his latest book, his latest picture, alarms
him. “Oh yes,” the true artist says to his heart, “popularity is sweet;
money is sweet; and I can hold both in my hand by the simple
process of repeating myself.” And the temptations are many and
great. You have on the profits of your first and second books, and a
reasonable hope of continuance, enlarged your way of life, incurred
responsibilities, built a charming house not yet paid for, married a
wife who adores you (shall we say?) and is proud of your celebrity,
but for these very reasons—and chiefly for love—will on any
diminution of your fame, fret secretly even if she does not nag
actively. Against this we have, opposed, the urge in the true artist
who—having done a thing—tosses it over his shoulder and thinks no
more of it; can only think of how to do something further and do it
better. I indicate the strength of the temptation. There are, of course,
sundry ways of getting round it. For instance, as I read the life of
Shakespeare from the few hints left to us, Shakespeare dodged it by
the Gordian-knot solution of leaving his wife and bolting to London: a
solution in this particular instance happy for us, yet not even on that
account to be recommended in general to young literary aspirants. I
mention Shakespeare here less for this, than as an exemplar of the
true artist, never content with his best, to repeat it. Why, having
written a Hamlet, an Othello, did he, instead of reproducing Hamlets
and Othellos, go on to have a shy at a Cymbeline? For the self-same
reason, Sirs, why Ulysses—if I may quote a poet none too popular
just now—could not bide at home after even such tribulations of
wandering as had become a proverb:
 I am become a name;
For always roaming with a hungry heart
Much have I seen and known; cities of men
And manners, climates, councils, governments,
Myself not least, but honour’d of them all;
And drunk delight of battle with my peers,
Far on the ringing plains of windy Troy.
I am a part of all that I have met;
Yet all experience is an arch wherethro’
Gleams that untravell’d world, whose margin fades
For ever and for ever when I move.
How dull it is to pause, to make an end,
To rust unburnish’d, not to shine in use!
As tho’ to breathe were life. Life piled on life
Were all too little, and of one to me
Little remains: but every hour is saved
From that eternal silence, something more,
A bringer of new things—

You may read the mere yearning of this, if you will, in Defoe, opening
The Further Adventures of Robinson Crusoe; or, if you will, in
Kipling’s

For to admire an’ for to see,

For to be’old this world so wide—
It never done no good to me,
But I can’t drop it if I tried

—and these express the instinct. The sanction, for us, lies in the
words

but every hour is saved

From that eternal silence, something more,
A bringer of new things.
And the desire for that—as I am sure you know—operates with no
less force of prompting in the spiritual world than in the world of
commerce and sea-travel. It carried Shakespeare at the last to that
Ariel’s isle which no commentator has ever (thank heaven!) been
able yet to locate; and it brought him home at the very last

A bringer of new things.

IV
Now if you accept no more than a much lower estimate of
Dickens than I am preaching, you will be apt to dismiss what I have
just been saying as “tall talk”: and you will be quite mistaken,
because it applies from Shakespeare down to men of infinitesimally
less desert than Dickens; to every small artist, in fact, whose
conscience will not cease harrying him until he improves on his best:
a process which obviously—and, as a matter of history, with the
great authors—never stops until they come to the grave.
At which point my now notorious discursiveness, Gentlemen,
also stops and gets back to Dickens. You see, the trouble of the
matter is that in these experiments an author can never be sure. He
takes an infinite risk, it may be against his own true genius. Where is
the critic to correct him?

V
Well, with Dickens, his own adoring public corrected him sharply
and, on the whole, with true instinct. To them he was the wand-
waving magician, the improvisatore in excelsis who had caught up
out of their midst an elderly small gentleman in spectacles and
gaiters and shot him suddenly out of Goswell Street into the
firmament, to be a star equal with Hercules—
sic fratres Helenae, lucida sidera—
“instead of which” he had turned to making plots so patently
theatrical (and of the theatre of Crummles) that the man himself was
helping everybody to see through them. So came the revenge; over-
proved by the opening chapter of Martin Chuzzlewit, which I suppose
to be about the sorriest piece of writing ever perpetrated by a great
English writer. Its perusal induces on me at any rate, something like
physical misery, not unmixed with the sort of shame any one of us
might feel if a parent behaved unbecomingly in public. I want to obey
the exhortation on Mrs. Sapsea’s monument and “with a blush
retire.”
But, note you, the general reader—that entity often abused,
seldom quite the fool that he looks—was quick to mark and punish.
Listen to Forster:

Chuzzlewit had fallen short of all the expectations formed of

it in regard to sale. By much the most masterly of his writings
hitherto, the public had rallied to it in far less numbers than to
any of its predecessors.... The primary cause of this, there is
little doubt, had been the change to weekly issues in the form of
publication of his last two stories.... The forty and fifty thousand
purchasers of Pickwick and Nickleby, the sixty and seventy
thousand of the early numbers of the enterprises in which The
Old Curiosity Shop and Barnaby Rudge appeared had fallen to
little over twenty thousand. They rose, somewhat on Martin’s
ominous announcement, at the end of the fourth number, that
he’d go to America....

They rose at once by a couple of thousand: but a serial of course

can never be easily lifted out of a rut into which it has once dropped.
The reasons for this are obvious, and the serial sales of Chuzzlewit
never over-topped twenty-three thousand. There was a very different
story when Chuzzlewit came to book form. “Its sale, since,” writes
Forster, “has ranked next after Pickwick and Copperfield.” In short,
Dickens had been, quite conscientiously, in the opening chapters of
Chuzzlewit, working against the grain of his genius. His public
recalled him to it in the brutal way the public uses. When he sat
down to write Chuzzlewit he had never an idea of carrying Martin off
to America. Suddenly, in fear of falling sales and many challenges to
make good his American Notes, he became the improvisatore again
and switched his hero across the Atlantic. Who will deny that the
American chapters of Martin Chuzzlewit are its best and, save for
any given chapter upon which Sarah Gamp knocks in, its most
memorable?

VI
None the less, and to the end, Dickens the artist is hag-ridden by
this business of “plot,” which for him meant “stage-plot.” It hampers
him in book after book, as its silly exigencies perpetually get in the
way of the reader’s pleasure, even of the reader’s understanding.
His genius did not lie that way, any more than did Shakespeare’s. I
put in this comparison, for it can never be untimely for a Professor of
English Literature to get in a word to damn the school-books which
present Shakespeare to you as chasing along his shelves for some
Italian novel to provide him with a new plot. Oh, believe me,
Gentlemen—after The Comedy of Errors and that sort of thing,
Shakespeare never bothered any more about his plots or whence he
took them. It is very right indeed for a young author to sweat his soul
over “plot” structure. But, through practice, there comes a time—
suddenly, it may be, but as sure in his development as puberty in his
physical growth—when lo! he has a hundred plots to his hand, if
heaven would but grant him time to treat them. I often wonder why
men blame the elder Dumas so severely, accepting the allegation
that he employed hirelings—viciously termed by the critic his
“ghosts” or his “devils.” Why, if you have an imagination teeming, like
Dumas’, with stories to make men happier—why, knowing how short
is life and that you cannot, on this side of the grave, tell one-fifth of
these with your own pen—why go to that grave leaving the world,
through that scruple, so much imaginatively the poorer? Only the
thing should be done frankly, openly, of course.

VII
I just raise that question. It applies to Dumas and (I think) to
most great novelists. But it applies less to Dickens than to most—
than to Trollope for instance. And in this very inapplicability lies a
secret of Dickens’ weakness which I am to suggest.
His plots are not merely stagey, melodramatic. Carefully
examined, they are seen to repeat themselves, under a wealth of
disguise, with an almost singular poverty of invention. Let us take
one most favourite trick of his—the trick of “the masked battery” as I
shall call it: the discomfiture of the villain by the betrayal of his
supposed confederate. The characters are artfully assembled for the
bad man’s triumph. Of a sudden the confederate rounds on him,
gives him away before the audience—usually in a long story, at the
end of which the baffled schemer creeps away, usually again to
destroy himself. We get this coup as early as in Oliver Twist where
Monks blurts out his story. It is repeated in Nickleby when Ralph
Nickleby is confronted with the man “Snawley” and by Squeers. In
the next novel, Martin Chuzzlewit, we get a double dose; Jonas
“given away” by an accomplice; Pecksniff explosively denounced by
Old Chuzzlewit after a long course of watchful dissimulation. This
idea of a long and careful dissimulation so catches hold of Dickens
that he goes on to rope into its service in subsequent stories two
men who, on his own showing of them, are about the very last two in
the world capable of carrying through a strategy so patient—Mr.
Micawber in David Copperfield and Mr. Boffin in Our Mutual Friend.
As a portrait, Mr. Boffin ranks pretty high even in Dickens’ gallery,
while Micawber ranks with the very best of his best. But who will
assert that either of them could have found it in his nature to behave
as the plot compels them to behave? To continue—by just the same
trick Quilp gets his exposure in The Old Curiosity Shop, Harewood
forces the revelation in Barnaby Rudge, Lady Dedlock is hunted
down in Bleak House. The more the peripeteia—the reversal of
fortune—disguises itself, the more it is the same thing.

VIII
George Santayana—he is so excellent a writer that I dispense
with “Doctor” or “Professor” or other prefix to his name—tells us that:

Dickens entered the theatre of this world by the stage door;

the shabby little adventures of the actors in their private capacity
replace for him the mock tragedies which they enact before a
dreaming public. Mediocrity of circumstance and mediocrity of
soul for ever return to the centre of his stage; a more wretched
or a grander existence is sometimes broached, but the
pendulum swings back, and we return, with the relief with which
we put on our slippers after the most romantic excursion, to a
golden mediocrity—to mutton and beer, and to love and babies
in a suburban villa with one frowsy maid.

Yes, that is true enough, but not all the truth. Dickens entered the
theatre by the stage door; but he passed through to the front, to turn
up the lights, wave his wand and create a new world—a fairy world,
let us agree: a theatrical world, as I have been attempting to show.
Yet consider—
Most of us in this room have childish recollections of green
fields, running brooks, woods in leaf, birds’ nests, cattle at pasture,
all that pageant of early summer which is going on at this moment a
few furlongs from this desk—this dead piece of timber—and at the
thought of which (if you will not think me impolite) I long to be
somewhere else at this moment. With some of us elders, not
specially imaginative, the early habit persists even after long
servitude to city life: so that still by habit our first instinct on rising
from bed is to go to the window and con the weather—how the day
is making, from what quarter the wind sets—“Is it too strong for the
fruit blossom?” “Will it be a good day for the trout?” Again, of my
experience I appeal to some of you—to those who, aware in
childhood or boyhood (quite suddenly, it may be, made aware) of the
beauty underlying this world (yes, and clothing it too), have been as
suddenly afflicted with the hopeless yearning to express it, was not
that yearning awakened, quickened in you, you knew not how, by
some casual sight—an open glade between woods, a ship with all
canvas spread, or, through the hazels,
the nesting throstle’s shining eye,
or the fish darting in the deep of a pool? Was it not some similar
moment that, though you have never yet arrived at putting it and its
underthought into words, yet so touched you that for the rest of your
days you will understand what was in Coleridge’s heart when he
wrote:

O happy living things! no tongue

Their beauty might declare:
A spring of love gush’d from my heart,
And I blessed them unaware.

Yes, and I dare say your first visit to the theatre brought you a like
delicious shock. (I can recall to this day, very distinctly, the gods and
goddesses who, between the acts of my first pantomime, danced on
the blue ceiling with baskets and festoons of roses.)
But now, bethink you that Dickens struggled through a childhood
to which green fields, trees, birds, cattle, brooks and pools, were all
denied: that the child was condemned to a squalid lodging; to spend
his days washing bottles in a dreadful blacking factory, his hours “off”
in visiting his parents in the yet more dreadful Marshalsea prison to
which his father had been committed for debt: and you will
understand not only that he had to enter the theatre of this world by
the stage door, but that the lighted theatre, when he could pay a few
pence and get to the gallery, was his one temple of beauty: that only
there—if we except a hint or two picked up in the street—from a
shabby acrobat or a stray Punch and Judy show—could he drink the
romance for which his young spirit thirsted. You have all read, I doubt
not, Charles Lamb’s paper on “My First Play,” first contributed to the
London Magazine in December, 1821, afterwards reprinted in Elia:

But when we got in, and I beheld the green curtain that
veiled a heaven to my imagination, which was soon to be
disclosed—the breathless anticipations I endured! I had seen
something like it in the plate prefixed to Troilus and Cressida in
Rowe’s Shakespeare—the tent scene with Diomede—and a
sight of that plate can always bring back in a measure the feeling
of that evening—The boxes at that time, full of well-dressed
women of quality, projected over the pit; and the pilasters
reaching down were adorned with a glistering substance (I know
not what) under glass (as it seemed), resembling—a homely
fancy—but I judged it to be sugar-candy—yet, to my raised
imagination, divested of its homelier qualities, it appeared a
glorified candy!—The orchestra lights at length arose, those ‘fair
Auroras’! Once the bell sounded. It was to ring out yet once
again—and, incapable of anticipation, I reposed my shut eyes in
a sort of resignation upon the maternal lap. It rang the second
time. The curtain drew up—I was not past six years old—and the
play was Artaxerxes!

There we have the confession of a Cockney-bred boy, more happily

placed than was Dickens at the same age or for many years later.
Lamb had his hardships, his tragedy or tragedies, in life: but in the
childhood of Dickens, most sensitively resentful, penury and
shameful occupation bit down to the bone. What other vision of
beauty had he—a born actor, as all contemporaries report—but that
which Drury Lane or Covent Garden supplied? Love, says a late
Roman singer, was born in a field:
Ipse Amor, puer Dionae, rure natus dicitur—
Pleasure planteth a field; it conceives under Pleasure, the pang of its
joy:
In a field was Dione in labour delivered of Cupid the boy:
And the field to her lap, to her fostering breast, took the rascal; he
drew
Mother’s milk from the delicate kisses of flowers and he prospered
and grew—
Now learn ye to love who loved never: now ye who have loved, love
anew!

The bad early and mid-Victorian stage hurt more than one
Victorian novelist of genius. It seriously hurt Charles Reade, for
example, who habitually sought the advice of Egeria from a fourth-
rate actress: and that should bring tears to the eyes of any critic who
knows Reade’s strong country nurture and has sized his genius. But,
with Dickens—think of that forlorn child, plotting to snatch his soul’s
sustenance in the shilling gallery of Drury Lane—at intervals how
rare! Is it any wonder that—to convert a famous phrase—coming to
power, he invoked out of the theatre a new world, to redress the
balance of his old?

IX
Moreover—and mind you this—you will never understand
Charles Dickens until you realise how exquisitely, how indignantly
the genius in this child of the blacking-warehouse felt the shame of
its lot. Dickens was never a snob: but a prouder spirit never
inhabited flesh. This shepherd boy was not one to sing in the Valley
of Humiliation. For years after success came to him he kept his
mouth closed like a steel trap upon past agonies. At length he
confided something to Forster (Life, Volume 1, Chapter 2), and few
sadder reflections have ever been implied by a grown man upon his
parents: