Data analytics - eBook PDF

Visit to download the full and correct content document:

https://ebooksecure.com/download/data-analytics-ebook-pdf/
DATA ANALYTICS
About the Author
Dr. Anil Maheshwari is a Professor of Management Infor-
mation Systems and Director of Center for Data Analytics
at Maharishi University of Management, Fairfield, Iowa,
USA.

He received a bachelor’s degree in Electrical Engineer-

ing from Indian Institute of Technology (IIT) Delhi, and
further received his MBA degree from Indian Institute
of Management (IIM) Ahmedabad. He earned his PhD
from Case Western Reserve University, Cleveland, Ohio.
He has been a Professor at the University of Cincinnati,
City University of New York, among others. His research has been published in
prestigious journals and conferences. He teaches data analytics, big data, leader-
ship, and marketing. He has authored many books in data science and leader-
ship. He has also worked in the global IT industry for over 20 years, including
leadership roles at IBM in Austin, Texas. He has completed various leadership
and marketing training programs at IBM and also won several awards.

He is a practitioner of Transcendental Meditation (TM) and TM-Sidhi techniques.

He blogs on IT and Enlightenment at anilmah.com, and is a popular speaker on
those topics. He is also a marathoner. He can be reached at akm2030@gmail.com.
DATA ANALYTICS

Dr. Anil Maheshwari

Professor of Management Information Systems and
Director of Center for Data Analytics
Maharishi University of Management,
Fairfield, Iowa, USA.

McGraw Hill Education (India) Private Limited

CHENNAI

McGraw Hill Education Offices

Chennai New York St Louis San Francisco Auckland Bogotá Caracas
Kuala Lumpur Lisbon London Madrid Mexico City Milan Montreal
San Juan Santiago Singapore Sydney Tokyo Toronto
 McGraw Hill Education (India) Private Limited

Published by McGraw Hill Education (India) Private Limited

444/1, Sri Ekambara Naicker Industrial Estate, Alapakkam, Porur, Chennai - 600 116

Data Analytics

Copyright © 2017 by Anil Maheshwari

No part of this publication may be reproduced or distributed in any form or by any means, electronic, mechanical, photocopying,
recording, or otherwise or stored in a database or retrieval system without the prior written permission of the publishers.
The program listings (if any) may be entered, stored and executed in a computer system, but they may not be reproduced for
publication.

This edition can be exported from India only by the publishers,

McGraw Hill Education (India) Private Limited

Print Edition
ISBN-13: 978-93-5260-418-0
ISBN-10: 93-5260-418-0

EBook Edition
ISBN-13: 978-93-5260-419-7
ISBN-10: 93-5260-419-9

Managing Director: Kaushik Bellani

Director—Products (Higher Education and Professional): Vibha Mahajan

Manager—Product Development: Nikhil Wadhera

Head—Production (Higher Education and Professional): Satinder S Baveja

Copy Editor: Taranpreet Kaur
Asst. Manager—Production: Suhaib Ali

Asst. General Manager—Product Management (Higher Education and Professional): Shalini Jha
Product Manager: Kartik Arora

General Manager—Production: Rajender P Ghansela

Manager—Production: Reji Kumar

Information contained in this work has been obtained by the McGraw Hill Education (India), from sources believed to
be reliable. However, neither author nor McGraw Hill Education (India) guarantee the accuracy or completeness of any
information published herein, and neither McGraw Hill Education (India) nor its authors shall be responsible for any errors,
omissions, or damages arising out of use of this information. This work is published with the understanding that McGraw Hill
Education (India) and its authors are supplying information but are not attempting to render engineering or other professional
services. If such services are required, the assistance of an appropriate professional should be sought.

Typeset at The Composers, 260, C.A. Apartment, Paschim Vihar, New Delhi 110 063 and printed at

visit us at: www.mheducation.co.in

 Dedicated to My Parents
Mrs. Meena and Mr. Ratan Lal Maheshwari
who taught me the value of Excellence and Hard Work
Preface
Data Science is a new discipline and this book on data analytics fills a need for
an accessible book on this topic. Other textbooks seem too technical and complex
for use by the students. The goal behind the making was to write a conversa-
tional book that feels easy and informative. This is a comprehensible book that
covers everything important, with concrete examples, and invites the reader to
join this field.

The book has been developed from author’s own class notes. It reflects his two
decades of global IT industry experience, as well as more than a decade of aca-
demic experience. The chapters are organized for a typical one-semester graduate
course. The book contains caselets from real world stories at the beginning of
each chapter. There is a running case study across the chapter exercises. There
are also review questions at the end of each chapter.

The book can be easily read in a short period by anyone who wants to
understand data based decision-making for their business or other or-
ganizations, but has no expertise with software tools. The text is almost
entirely devoid of complex jargon or programming code.

ORGANIZATION OF THE BOOK

Chapter 1 is designed to provide the wholeness of business intelligence and data

mining, to provide the reader with an intuition over this area of knowledge. The
rest of the book is divided into four sections as follows
■ Section 1 covers the data processing chain. Chapter 2 covers the field of
business intelligence and its applications across industries and functions.
Chapter 3 explains Data Warehousing and how does it help with data min-
ing. Chapter 4 then describes data mining with an overview of its major tools
and techniques. Chapter 5 shows insights from data that can be visualized
for better communication and consumption.
■ Section 2 is focused on core data mining techniques. Every technique is
shown through solved examples in detail. Chapter 6 shows the power and
ease of decision trees, which are the most popular data mining technique.
Chapter 7 describes statistical regression modeling techniques. Chapter 8
provides an overview of Artificial Neural Networks. Chapter 9 describes how
Cluster Analysis can help with market segmentation. Finally, Chapter 10
describes the Association Rule Mining technique, also called Market Basket
Analysis that helps find shopping patterns.
viii Preface

■ Section 3 covers more advanced topics. Chapter 11 introduces the concepts

and techniques of Text Mining that helps discover insights from text data
including social media data. Chapter 12 covers Naïve-Bayes, a classification
technique used for text mining such as spam filtering. Chapter 13 covers
Support Vector Machines, a mathematically rigorous classification technique,
also used for spam and other applications with high dimensional data.
Chapter 14 provides an overview of the growing field of web mining, which
includes mining the structure, content and usage of web sites. Chapter 15
introduces the concept of Social Network Analysis, which helps analyze both
text communication and web structure.
■ Section 4 further covers more advanced topics. Chapter 16 provides
an overview of the field of Big Data. Chapter 17 has been added as a
primer on Data Modeling, for those who do not have any background
in databases, and should be used if necessary. Chapter 18 discusses
data science careers and some additional case studies.
■ The book ends with an extensive R tutorial to help with many of the
techniques explained in the book. No prior experience with R is needed
for using this tutorial. Also few pages have been dedicated for Weka
tutorial for data mining for further subject knowledge.
Dr. Anil K. Maheshwari

WEB SUPPLEMENTS

The web supplements can be accessed at http://highered.mheducation.com/

sites/9352604180 which contains the following

For Instructors

■ PowerPoint charts for each chapter

For Students

■ Datasets that go with R tutorials as well as exercises

■ Video lectures

ADVANCE PRAISE OF THE BOOK

■ This book is a splendid and valuable addition to this subject. The whole docu-
ment is well written and I have no hesitation to recommend it as a textbook
for graduate courses in Data Analytics. The author has taken immense pain
to keep the writing style simple and easy to follow. The book explains the
 Preface ix

concepts and wholeness without going into any complicated mathematics.

It develops an intuition and generates an interest among the students in
this field. Even a casual reader of the book will be left with a keen desire
to learn more because this book brings out the practical benefits very nicely
and clearly. The caselets are good and nicely planned.

Dr. Edi R. Shivaji

B.Tech. (IIT-M), PGDM (IIM-B), Ph.D. (USA)

■ Dr. Maheshwari's book is a nice introduction to Data Analytics. He has ex-

plained the concepts in a very lucid manner. I particularly liked his chapter
on decision trees and the process to obtain them. This is a book that gives
you a nice and simple overview of the field. Overall it is a good introductory
book if one wants to pursue this field and to know what it takes to be a data
scientist. I highly recommend it!

Dr. Ramon A. Mata-Toledo

Professor of Computer Science
James Madison University, Virginia.
Acknowledgements
I bow in reverence to the Almighty God as without him this work would not have
been possible. I give my sincere thanks to him for giving me the ability, strength,
and patience to write this book.

I would like to express my gratitude to my family members, my father Mr. R.L.

Maheshwari, brother Dr. Sunil Maheshwari, wife Neerja and daughters Ankita
and Nupur, for their continuous encouragement and support while making of this
book. My colleagues Dr. Edi Shivaji, Dr. Scott Herriott, and Dr. Greg Guthrie
provided many ideas and suggestions to improve and disseminate the book; my
sincere thanks to them. Thanks to Piyush Priyadarshi of McGraw Hill Educa-
tion for reaching out to me and suggesting that I publish this book with them.
The entire MHE team worked hard on bringing this out in a short time; sincere
thanks to all of them.

I am also grateful to my numerous students at Maharishi University of Man-

agement for providing their valuable feedback and suggestions on the book. My
heartfelt thanks to Maharishi Mahesh Yogi for providing a wonderful environ-
ment of consciousness-based education that provided the inspiration for creating
this book. It is a ‘guru-dakshina’, or homage, to Maharishi.

I would like to thank my graduate students Tonmay Bhattacharjee, Marie Claire

Niyirora, and Mauro Nogueira for their valuable contribution in the making of
R tutorial for Data Mining.

I am grateful to the following reviewers who took out time and gave their valu-
able suggestions on various chapters of the book

Dr. Dharavath Ramesh

Assistant Professor, IIT Dhanbad
Sanjay Kumar Vij
Dean, Academics, ITM Universe, Vadodra
Dr. Sohail D. Pandya
Assistant Professor, SVIT, Anand, Gujarat
Regi Mathew
Professor, NMIMS, Bangalore
xii Acknowledgements

Pratosh Bansal
Professor, DAVV, Indore, Madhya Pradesh

Any suggestions for improving the book will be gratefully acknowledged.

Dr. Anil K. Maheshwari

Publisher’s Note
Constructive suggestions and criticism always go a long way in enhancing and
endeavor. We request all our readers to email us their valuable comments/
views/feedback for the betterment of the book at info.india@mheducation.
com, mentioning the title and author name in the subject line. Also, please
feel free to report any piracy of the book if spotted by you.
Contents
Preface vii
Acknowledgements xi

1. Wholeness of Data Analytics 1

Introduction 1
Business Intelligence 2
Pattern Recognition 3
Data Processing Chain 5
Terminology and Careers 16
Review Questions 16
True/False 17

SECTION 1
2. Business Intelligence Concepts and Application 21
Introduction 21
BI for Better Decisions 23
Decision Types 23
BI Tools 24
BI Skills 26
BI Applications 26
Conclusion 33
Review Questions 33
True/False 34

3. Data Warehousing 35
Introduction 35
Design Considerations for DW 36
DW Development Approaches 37
DW Architecture 38
Conclusion 41
Review Questions 41
True/False 42
xiv Contents

4. Data Mining 43
Introduction 43
Gathering and Selecting Data 45
Data Cleansing and Preparation 46
Outputs of Data Mining 47
Evaluating Data Mining Results 48
Data Mining Techniques 49
Data Mining Best Practices 53
Myths about Data Mining 55
Data Mining Mistakes 56
Conclusion 57
Review Questions 57
True/False 58

5. Data Visualization 59
Introduction 59
Excellence in Visualization 60
Types of Charts 62
Visualization Example 65
Visualization Example Phase-2 69
Tips for Data Visualization 69
Conclusion 70
Review Questions 70
True/False 71

SECTION 2
6. Decision Trees 75
Introduction 75
Decision Tree Problem 76
Decision Tree Construction 78
Lessons from Constructing Trees 82
Decision Tree Algorithms 84
Conclusion 86
Review Questions 86
True/False 87
 Contents xv

7. Regression 89
Introduction 89
Correlations and Relationships 90
Visual Look at Relationships 91
Non-linear Regression Exercise 96
Logistic Regression 98
Advantages and Disadvantages of Regression Models 98
Conclusion 100
Review Questions 100
True/False 100

8. Artificial Neural Networks 102

Introduction 102
Business Applications of ANN 103
Design Principles of an Artificial Neural Network 104
Representation of a Neural Network 105
Architecting a Neural Network 106
Developing an ANN 107
Advantages and Disadvantages of Using ANNs 107
Conclusion 108
Review Questions 108
True/False 109

9. Cluster Analysis 110

Introduction 110
Applications of Cluster Analysis 111
Definition of a Cluster 111
Representing Clusters 112
Clustering Techniques 113
Clustering Exercise 114
K-Means Algorithm for Clustering 116
Selecting the Number of Clusters 119
Advantages and Disadvantages of K-Means Algorithm 120
Conclusion 121
Review Questions 121
True/False 121
xvi Contents

10. Association Rule Mining 123

Introduction 123
Business Applications of Association Rules 124
Representing Association Rules 124
Algorithms for Association Rule 125
Apriori Algorithm 125
Association Rules Exercise 126
Creating Association Rules 128
Conclusion 130
Review Questions 130
True/False 130

SECTION 3
11. Text Mining 135
Introduction 135
Text Mining Applications 136
Text Mining Process 138
Term Document Matrix 138
Mining the TDM 140
Comparing Text Mining and Data Mining 142
Text Mining Best Practices 143
Conclusion 143
Review Questions 144
True/False 144

12. Naïve-Bayes Analysis 146

Introduction 146
Probability 147
Naïve-Bayes Model 148
Simple Classification Example 148
Text Classification Example 149
Advantages and Disadvantages of Naïve-Bayes 151
Conclusion 151
Review Questions 151
True/False 152
 Contents xvii

13. Support Vector Machines 153

Introduction 153
SVM Model 154
The Kernel Method 156
Advantages and Disadvantages of SVMs 157
Conclusion 157
Review Questions 157
True/False 158

14. Web Mining 159

Introduction 159
Web Content Mining 160
Web Structure Mining 160
Web Usage Mining 161
Web Mining Algorithms 162
Conclusion 163
Review Questions 163
True/False 163

15. Social Network Analysis 164

Introduction 164
Techniques and Algorithms 167
PageRank 172
Practical Considerations 172
Conclusion 173
Review Questions 173
True/False Questions 174

SECTION 4
16. Big Data 177
Introduction 177
Defining Big Data 178
Big Data Landscape 180
Business Implications of Big Data 181
Technology Implications of Big Data 182
Big Data Technologies 182
xviii Contents

Management of Big Data 185

Conclusion 186
Review Questions 186

17. Data Modeling Primer 187

Evolution of Data Management Systems 188
Conclusion 192
Review Questions 192

18. Statistics Primer 193

Introduction 193
Descriptive Statistics 193
Normal Distribution and Bell Curve 198
Inferential Statistics 199
Predictive Statistics 200
Conclusion 201
Review Questions 201
True/False 201

19. Data Science Careers and Additional Case Studies 202

Data Science Careers 202
Data Science Aptitude 203
Popular Skills 203
Case Studies 204

Appendix 1 R Tutorial for Data Mining 207

Appendix 2 Weka Tutorial for Data Mining 231

Additional Resources 246

Index 248
 Wholeness of Data
1 Analytics
Learning Objectives
■ Understand Business Intelligence and Data Mining Cycle
■ Learn about the tools and purpose of Business Intelligence
■ Discover what are patterns, their types, and the process of discovering patterns
■ Understand the data processing chain
■ Learn in brief about the components of the data processing chain
■ Process a simple example dataset through complete data processing chain
■ Learn about datafication and various types of data
■ Learn in brief about key terms and Data Science careers

INTRODUCTION
Business is the act of doing something productive to serve someone’s needs and
thus earn a living and make the world a better place. Business activities are
recorded on paper or using electronic media, and then these records become data.
There is more data from customers’ responses and on the industry as a whole.
All this data can be analyzed and mined using special tools and techniques to
generate patterns and intelligence, which reflect how the business is function-
ing. These ideas can then be fed back into the business so that it can evolve to
become more effective and efficient in serving customer needs; and the cycle goes
on (Figure 1.1).

Business Intelligence

Data Mining
FIGURE 1.1 Business Intelligence and Data Mining (BIDM) Cycle
Another random document with
no related content on Scribd:
is less likely to be localized, and, on the whole, it is not so severe as
the terrible torture of the neoplasm. Irregular but very decided febrile
phenomena are more likely to be present in meningitis than in tumor.
Like brain tumor, tubercular meningitis of the convexity may give
psychical disturbances, palsies, local spasms, general convulsions,
sensory disturbances, peculiar disorders of the special senses, etc.;
but these symptoms in the former usually come on more irregularly
and are accompanied less frequently with paroxysmal exacerbations
of headache, vomiting, vertigo, etc. Tubercular meningitis of the base
can be more readily distinguished from cases of tumor by the fact
that one cranial nerve after another is likely to become involved in
the diffusing inflammatory process. Tubercular meningitis is of
shorter duration than the majority of cases of brain tumor, and in it
delirium and mental confusion come on more frequently and earlier.
A history and physical evidences of more or less generalized
tuberculosis favor the diagnosis of tubercular meningitis. In both
affections the ophthalmoscope may reveal choked disc or
descending neuritis. It will be seen that the differentiation between
the affections is not always very clear, although in some cases the
decision may be quickly reached from a study of the points here
suggested.

Some of the forms of chronic hydrocephalus are difficult to

distinguish from tumors, especially gliomata. In hydrocephalus, when
not the result of, or not accompanied by, tubercular meningitis, the
disease advances more slowly and with less irritative symptoms than
in cases of tumor. Headache, vertigo, vomiting, and the other
symptoms of meningeal irritation are not so frequently present,
although the ophthalmoscopic appearances are often the same.

Rosenthal speaks of the necessity of diagnosticating brain tumor

from the chronic cerebral softening of Durand-Fardel, from acquired
cerebral atrophy, and the cerebral hypertrophy of children. An
elementary knowledge of the general symptomatology of intracranial
tumors will, however, be sufficient to prevent mistakes of
differentiation in these cases. Neither of these affections presents
the violent paroxysmal symptoms, the affections of the special
senses, or the severe motor and sensory phenomena of intracranial
growths.

Acute mania and paretic dementia are sometimes confounded with

intracranial growths. A case of brain tumor is more likely to be
regarded as one of acute mania than the reverse. In some
comparatively rare instances in the course of their sufferings the
cases of tumor become maniacal, but even a superficial study of
general symptomatology in such a case will be sufficient to clear up
the doubt.

Paretic dements are occasionally supposed to be cases of brain

tumor, because of the epileptiform attacks and isolated pareses
which occur as the disorder progresses. It is only necessary to refer
to this matter, as the mistake would not be likely to be made by one
having any familiarity with dementia paralytica.

L. J. Lautenbach, in a recent communication to the Philadelphia

Neurological Society, which embodied a large number of
ophthalmoscopic examinations of the insane at the State Insane
Hospital, Norristown, Pennsylvania, and the Insane Department of
the Philadelphia Hospital, and also the results of the investigations of
the fundus of the eye in cases of insanity by other observers,
showed that about 16 per cent. of cases of acute mania presented
well-defined papillitis—a condition which he described as one of
swelling and suffusion of the disc, corresponding to cases reported
as choked disc, descending neuritis, and severe congestion of the
optic nerve. No reports of post-mortem examinations were made of
these cases, but they did not present the clinical history of meningitis
or brain tumor. It therefore follows that the existence of papillitis in a
case of acute mania does not necessarily point to a gross lesion,
such as tumor or meningitis.

In the early stage of posterior spinal sclerosis some of the symptoms

of the initial or middle stage of intracranial growths in certain
positions are likely to be present; more particularly, such eye
symptoms as diplopia from deficiency or paresis of the ocular
muscles and disorders of the bladder may mislead. In posterior
spinal sclerosis, however, some at least of the pathognomonic
symptoms of locomotor ataxia, such as lancinating pains, absent
knee-jerk, or Argyle-Robertson pupil, will almost invariably be
present. Those tumors of the cerebellum, pons, tubercular
quadrigemina, etc. which give rise to ataxic manifestations are
usually readily discriminated from posterior spinal sclerosis by the
headache, vomiting, and other general symptoms of brain tumor,
which rarely occur in ataxia. It is far more difficult to separate non-
irritative lesions of certain cerebellar and adjoining regions from the
spinal disorder.

Strange to say, one of the most frequent mistakes of diagnosis is

that which arises from confounding brain tumor with grave hysteria.
In several of our tabulated cases the patients at different periods of
the disease and by various physicians had been set down as
suffering from hysteria. One of Hughes-Bennett's cases (Case 30), a
wayward, hysterical girl of neurotic family, had had her case
diagnosticated as hysteria by one of the highest medical authorities
of Europe, and yet after death a tumor the size of a hen's egg was
found in the cerebrum. In a case reported by Eskridge (Case 76)
hysterical excitement and special hysterical manifestations were of
frequent occurrence, and misled her physicians for a time. Eskridge
remarks, in the detailed report of this case, that to such a degree
was the emotional faculty manifest that had no ocular lesion been
present there would have been great danger of mistaking the case
for one of pure hysteria; and, indeed, a careful physician of many
years' experience, not knowing the condition of the eyes,
pronounced the woman's condition to be pregnancy complicated by
hysteria. A close study of such objective phenomena as choked
discs and paralysis will usually be of the most value.

Even malaria has been confounded in diagnosis with brain tumor.

Holt37 reports a case which presented the history of a fever, at first
periodical, with marked splenic enlargement, great muscular
soreness, and incomplete paralysis, which was diagnosticated to be
chronic malarial poison. The patient for a time improved under
quinine, but eventually grew worse, and on an autopsy a glioma-
sarcoma was found on the inferior surface of the cerebellum. Several
years since a physician about fifty years of age was brought to one
of us for consultation, and in his case a similar mistake had been
made. The case was a clear one of tumor, probably cerebellar, with
headache, neuritis, vertigo, and other general symptoms, which
pointed to an organic lesion. This patient, who came from a malarial
district in the West, had doctored himself, and had been treated by
others with enormous doses of quinine and arsenic.
37 Med. Record, March 1, 1883.

LOCAL DIAGNOSIS.—Niemeyer would hardly say to-day that the

brilliant diagnoses where the precise location of a tumor is fully
confirmed by autopsy are not usually due to the acumen of the
observer, but are cases of lucky diagnosis. It can be asserted with
confidence that the exact situation of a tumor can be indicated during
life in at least two or three locations. Great caution should be
exercised, as insisted upon by Nothnagel,38 in the localization of
tumors of the brain, because, among other reasons, of the frequent
polypus-like extension of such tumors.
38 Wien. Med. Bl., 1, 1882.

The subject of local diagnosis can be approached in several ways,

according to the method of subdividing the brain into regions. Thus,
Rosenthal discusses, in the first place, tumors of the convexity of the
brain, but as this is a very general term, covering portions of several
lobes, we can see no advantage in making such a subdivision.

A few general remarks might be made in the first place, however,

with regard to the general symptoms presented by surface or cortical
growths as compared with those which are produced by deep-seated
neoplasms. The direct or indirect involvement of the membranes in
nearly all cortical tumors makes the symptoms of irritation referable
to these envelopes very numerous and important.

The various centres so called, motor, sensory, and of the special

senses, which have their highest differentiation in the cerebral
cortex, are each and all represented by well-defined tracts of white
matter in the centrum ovale and capsules which connect these
centres with the lower brain, the spinal cord, and the periphery of the
organism. It therefore follows that symptoms produced by localized
lesions of the cortex will be reproduced in other cases by those of
the tracts which go to or come from these centres. We may thus
have a monoplegia or a hemiplegia, a partial anæsthesia or a
hemianæsthesia, a hemianopsia, a word-blindness or word-
deafness, a loss of power to perceive odors or to appreciate
gustatory sensations, from a peculiarly limited tumor or other lesions
of either the gray centres of the cortex or of the white matter of the
central area of the brain; but these specialized symptoms are more
likely to arise from cortical lesions in the case of intracranial
neoplasms, because of the much greater frequency with which these
adventitious products arise from membranes and therefore involve
the cortex.

Peculiar symptoms arise in the case of lesions of the centrum ovale

from the fact that it contains not only projection-fibres which more or
less directly connect cerebral centres with the outer world; but also a
system of commissural fibres which unite corresponding regions of
the two cerebral hemispheres by way of the corpus callosum and
commissures, and a system of association-fibres which connect
different convolutions together, in special cases even those which
are situated remotely from each other, but are associated in function.

It is evident, therefore, as asserted by Starr,39 that a peculiar set of

additional symptoms will be referable to the destruction or irritation of
these commissural and association fibres. For example, failure to
perform easily corresponding bilateral motions in face, hands, or feet
would indicate some obstruction to conduction in the commissural
fibres joining the motor convolutions. “Integrity of both occipital
lobes, and simultaneous, connected, and harmonious action in both,
are necessary to the perfect perception of the whole of any object
when the eyes are fixed upon one point of that object.” Starr gives
the following examples of the methods of detecting a lesion of such
fibres: “In the case of the fibres associating the auditory with the
motor speech-area the symptoms to be elicited seem to be very
simple. Can the patient talk correctly? Can he repeat at once a word
spoken to him? These are the questions which any one will ask who
examines a case of aphasia. But this is not all. The patient must be
further questioned. Can he read understandingly to himself, and tell
what he has read? This will test the occipito-temporal tract. Can he
read aloud? This will test the occipito-temporo-frontal tract. Can he
write what he sees? This will test his occipito-central tract. Can he
write what he hears? This will test the temporo-central tract. Can he
write what he says, speaking to himself in a whisper? This will test
his fronto-central tract. Can he name an odor or a color? Brill has
recorded40 a case of lesion of the cuneus associated with color-
blindness to green, and he states that the patient had difficulty in
naming various colors on account of the presence of a slight degree
of amnesic aphasia.... Can the patient write the name of an odor?
Can he tell how a surface feels—smooth, or warm, or heavy? Such
questions as these will suggest themselves at once to any one who
studies the association of ideas subjectively.

“Take as an example a lesion in the centrum ovale of the occipito-

temporal region. Such a lesion will produce hemianopsia, because it
involves the visual tract of the projection system. It may also produce
a peculiar mental condition known as word-blindness, in which the
patient is no longer able to associate a word or letter seen with its
corresponding sound or with the motion necessary to write it.
Charcot has reported a case of this kind.... The man, who was a very
intelligent merchant, was suddenly seized with right hemianopsia
while playing billiards, and was surprised to find that he saw but one-
half of the ball and of the table. Soon after he had occasion to write a
letter, and after writing it was surprised to find that he could not read
what he had just written. He found, however, that on tracing
individual letters with the pen or fingers he became conscious of the
letters—a few letters (r, s, t, x, y, z), however, being an exception to
this rule. When a book was given him to read he would trace out the
forms of the letters with some rapidity, and thus manage to make out
the words. If his hands were put behind him and he was asked to
read, he would still be observed to put his fingers in motion and trace
the letters in the air. Speech was in no way interfered with, but
reading aloud was only accomplished, like reading to himself, by the
aid of muscular sense. Here, then, was an example of a lesion which
had separated entirely the tract associating sight with speech—viz.
the occipito-temporal tract—but had left intact the tract associating
sight with muscular sense—viz. the occipito-central tract.”
39 Med. Record, vol. xxix. No. 7, Feb. 13, 1886.

40 Amer. Journ. of Neurology, Feb., 1883.

Our tabulated cases, although collected for the purpose of studying

inductively the phenomena of intracranial tumors from all points of
view, have been arranged to indicate, so far as is possible, the
special symptoms which are produced by growths in special
localities. Thus we have made thirteen subdivisions:

I. Superior antero-frontal region (5 cases).—The lateral and

median aspects of the hemisphere from the anterior tip
backward to the posterior thirds of the first three frontal
convolutions, the region roughly bounded by the coronal suture.

II. Inferior antero-frontal or orbital region (5 cases).—From the

anterior tip of hemisphere at the base backward to the optic
chiasm and Sylvian fissures.

III. Rolandic region or motor cortex (15 cases).—From antero-

frontal region backward nearly to mid-parietal lobe, including
posterior thirds of superior middle and inferior frontal
convolutions, ascending frontal and ascending parietal
convolutions, and anterior extremities of superior and inferior
parietal convolutions—lateral and median aspects.

IV. Centrum ovale, fronto-parietal region (5 cases).

V. Postero-parietal region (5 cases).—From Rolandic region to

parieto-occipital fissure, including posterior two-thirds of the
superior and inferior parietal convolutions and the præcuneus.
 VI. Occipital region (9 cases).—Occipital lobe—cortex and
centrum ovale.

VII. Temporo-sphenoidal region (4 cases).—Temporo-

sphenoidal lobe.

VIII. Basal ganglia and adjoining regions (19 cases).—Caudate

nucleus, lenticular nucleus, optic thalamus, internal capsule,
corpora quadrigemina, and ventricles except the fourth.

IX. Cerebellum (9 cases).

X. Floor of fourth ventricle (6 cases).—(Directly or indirectly

involved.)

XI. Pons varolii and medulla oblongata (8 cases).

XII. Crura cerebri (3 cases).

XIII. Middle region of base of brain and floor of skull (7 cases).—

In the main, from optic chiasm backward to pons, in the middle
basilar region, in some instances extending beyond this area in
special directions.

Tumors of the antero-frontal regions can be diagnosticated with

considerable certainty, partly by a study of the actual symptoms
observed and partly by a process of exclusion. Headache of the
usual type, vertigo, choked discs, inflammatory and trophic affections
of the eyes, widely varying body-temperature, and high head-
temperature are among the most positive manifestations. Mental
slowness and uncertainty seem to be greater in these cases than in
others. Mental disturbance of a peculiar character unquestionably
occurs in cases of tumor, as of other lesions, in this region. This
disturbance is exhibited chiefly in some peculiarity of character,
showing want of control or want of attention. The speech-defects
present in a number of cases were rather due to the change in
mental condition than to any involvement of speech-centres. Under
Symptomatology has been given in some detail a study of the
psychical condition in one case of antero-frontal tumor. The absence
of true paralysis and of anæsthesia is characteristic. Nystagmus and
spasm in the muscles of the neck and forearm were present in one
instance, but usually marked spasm is not to be expected. Vomiting
is less frequent than in tumors situated farther back. Facial and other
forms of paresis occasionally are present, but are not marked, and
are probably due to involvement by pressure or destruction of
surrounding tissue of neighboring motor areas. Hemianopsia, such
as was observed in Case 10, showed involvement of the orbital
region. Tumors of the inferior antero-frontal lobe give the same
positive and negative characteristics as those of the superior frontal
region, with the involvement in addition of smell and certain special
ocular symptoms, such as hemianopsia.

Tumors of the motor zone of the cerebral cortex, the region

surrounding and extending for some distance on each side of the
fissure of Rolando, can be diagnosticated with great positiveness: 15
of the 100 cases are examples of tumors of this region, and in many
of these the diagnosis of the location of the growth was accurately
made during life. Localized spasm in peripheral muscles; localized
peripheral paralysis; neuro-retinitis or choked discs; headache; pain
elicited or increased by percussion of the head near the seat of the
tumor; and elevated temperature of the head, particularly in the
region corresponding to the position of the growth,—are the
prominent indications. The spasmodic symptoms usually precede
the paralysis in these cases. The spasm is often local, and generally
begins in the same part in different attacks—in the fingers or toes or
face of one side.

A study of cases of tumor localized to the cortical motor area will

show that in almost any case a local twitching convulsion preceded
the development of paresis or paralysis. Hughlings-Jackson41 reports
a case of sarcoma, a hard osseous mass on the right side of the
head, of eighteen years' standing, subjacent to which was a tumor
the size of a small orange growing from the dura mater. The patient
was a woman aged forty-nine, whose symptoms were very severe
headache and double optic neuritis, with paresis in left leg, followed
by slighter paresis in left arm and left face. A very slow, gradual
hemiplegia came on by pressure on the cortex without any fit.
Jackson says this is the only case which he has seen in which the
hemiplegia has not followed a convulsion where the lesion has been
on the surface. In all very slowly oncoming hemiplegias which he has
seen, except this one, the tumor was in the motor tract.
41 Medical Times and Gazette, London, 1874, vol. i. 152.

As the white matter of the centrum ovale and capsules represents

simply tracts connecting cerebral centres with lower levels of the
nervous system, with each other, or with the opposite hemisphere,
lesions of this portion of the cerebrum will closely resemble those
cortical lesions to which the tracts are related. We have already
referred to the peculiar symptoms referable to involvement of
commissural and association fibres. Tumors of the centrum ovale of
the fronto-parietal region, of which five examples are reported in the
table, vary in symptomatology according to their exact location.
Those situated in the white matter in close proximity to the
ascending convolutions give symptoms closely resembling those
which result from lesions of the adjoining cortical motor centres. In
the cases of Osler, Seguin, and Pick (Cases 26, 27, 28, 29) spastic
symptoms in the limbs of one side of the body, with or without loss of
consciousness, were marked symptoms. In two of these cases some
paresis preceded the occurrence of the spasms. They did not,
however, fully bear out the idea of Jackson that the hemiparesis or
hemiplegia in tumors of the motor tract comes on slowly before the
appearance of spasm.

Tumors of the postero-parietal region present some characteristic

peculiarities. In several cases tumors were located in this region, and
in several others the white matter of the parietal lobe was softened
as the result of the obliteration of blood-vessels by the tumors. In
general terms, we might say that hemianæsthesia, partial or
complete, and impairment of sight and hearing on the side opposite
to the lesions, seemed to be the most constant peculiarities.
Tumors and other lesions of the occipital lobes have in the last few
years received extended attention, and, where possible, exact study,
because of the opportunities which they furnish for corroborating the
work of the experimental physiologists. It is unfortunate that the
records of older cases do not furnish the exact detail which would
render these tumors among the most important and interesting to be
met with in the brain: some cases have, however, been observed
with great care, and a few such are included in the table. To
understand the special significance of the symptoms of such tumors,
it will be well briefly to state some of the well-established facts about
the function of the occipital cortex. The investigations of Gratiolet
and Wernicke especially have proved that this surface of the brain is
in direct connection with the fibres (1) which are continued upward
from the posterior or sensory columns of the cord through the
posterior portion of the internal capsule, and (2) with the expansion
of the optic nerve, or the tract which passes, according to Wernicke,
from the thalamus to the occipital lobe. There is but a partial
decussation of the optic nerves at the chiasm, so that each half of
the brain receives fibres from both eyes. This arrangement is best
stated by Munk (quoted by Starr) as follows: “Each occipital lobe is in
functional relation with both eyes in such a manner that
corresponding halves of both retinal areas are projected upon the
cortex of the lobe of the like-named side; e.g. destruction of the left
lobe produces loss of function of the left halves of both retinæ.” This,
of course, causes the right halves of both fields of vision to appear
black. This condition is known as lateral homonymous hemianopsia,
and was exhibited in several of the tabulated cases (Cases 40, 41,
42, and 43). It is probable that the dimness of the right eye recorded
in Case 38 was really right lateral hemianopsia, as patients mistake
this condition for blindness of that eye alone which is on the side
upon which the visual fields are blank. It follows that this condition of
the eyes will be caused by a destructive unilateral lesion at any point
upon the optic tract behind the chiasm; and its exact nature and
location are to be inferred from other corroborating symptoms.
Among these corroborating symptoms, as will be inferred from the
other functions of the occipital cortex, is especially to be considered
partial hemiplegia and partial hemianæsthesia. This was observed in
Cases 38, 40. These most characteristic localizing symptoms of
occipital tumor have usually others, which, if not of such special
importance, yet help to form a special complexus. Among these
diffused headache is referred to by some writers as characteristic,
but it seems to us that a localized headache, with pain on percussion
over the affected region, is the only kind in this as in other regions
which could have special diagnostic importance. Affections of
hearing are recorded by some. It is not at all uncommon to have an
incomplete hemiplegia and local paralysis. In Case 41 complete
hemiplegia with facial paralysis is recorded. Local palsies, ocular and
facial, are recorded in Cases 36, 37, 38, and 39. It is doubtless by
transmitted pressure, or by extension of the tumor, or the softening
caused by it, toward the motor fibres, that these more or less
incomplete paralyses are caused. The general symptoms, such as
vertigo, vomiting, and convulsions, are frequently present with
tumors of the occipital lobes. We are at a loss to know upon what
data of theory or experience Rosenthal bases his statement that
psychic disorders are more common in occipital tumors than in those
of the anterior and middle lobes, unless he refers simply to the
hebetude and late coma which seem to come generally in these
cases.

Tumors of the temporo-sphenoidal region, so far as we have been

able to study them, present few characteristic features. Physiology
seems to point to the upper temporal convolutions as the cerebral
centres for hearing; thus, according to Starr,42 “disturbances of
hearing, either actual deafness in one ear or hallucinations of sound
on one side (voices, music, etc.), may indicate disease in the first
temporal convolution of the opposite side. Failure to recognize or
remember spoken language is characteristic of disease in the first
temporal convolution of the left side in right-handed persons, and of
the right side in left-handed persons. Failure to recognize written or
printed language has accompanied the disease of the angular gyrus
at the junction of the temporal and occipital regions of the left side in
three foreign and one American case.” In two of our four cases of
tumor in the temporo-sphenoidal region disturbances of hearing
were noted, but in none was the sense studied with sufficient care to
throw any light upon the actual character of the disorder. The case of
Allan McLane Hamilton (Case 47), already referred to under
Symptomatology, was interesting because of the presence of a
peculiar aura connected with the sense of smell. Stupidity, want of
energy, drowsiness, and general mental failure were marked in
tumors of this region.
42 American Med. Sci., N. S. vol. lxxxviii., July, 1884.

Tumors of the motor ganglia of the brain are seldom strictly localized
to one or the other of these bodies. Growths occurring in this region
usually involve one or more of the ganglia and adjacent tracts, and
can only be localized by a process of careful exclusion, assisted
perhaps by a few special symptoms. Paralysis or paresis on the side
opposite to the lesion usually occurs in cases of tumor of either the
caudate nucleus or lenticular nucleus; but whether this symptom is
due to the destruction of the ganglia themselves, or to destruction of
or pressure upon the adjoining capsule, has not yet been clearly
determined. In a case of long-standing osteoma of the left corpus
striatum (Case 49) the patient exhibited the appearance of an
atrophic hemiplegia: his arm and leg, which had been contractured
since childhood, were atrophied and shortened, marked bone-
changes having occurred. Another case showed only paresis of the
face of the opposite side. Clonic spasms were present in two cases,
in one being chiefly confined to the upper extremities of the face. In
this case paralysis was absent. Disturbances of intellect and speech
have been observed in tumors of this region. According to
Rosenthal, aphasic disturbances of speech must be due to lesions of
those fibres which enter the lenticular nucleus from the cortex of the
island of Reil.

Tumors of the optic thalamus usually cause anæsthesia or other

disturbances of sensation in the extremities of the opposite side.
They sometimes show third-nerve palsies of the same side in
association with hemiplegia on the opposite side, these symptoms
being probably due to pressure owing to the proximity of the
neighboring cerebral crus. Speech and gait in such tumors are also
often affected.

Tumors of the corpora quadrigemina give rise to disturbances of

sight and special ocular symptoms, such as difficulty in the lateral
movement of the eyes. Spasms were usually present. Automatic
repetition of words was observed in one case, nystagmus in another,
and diminished sexual inclinations in a third. In other cases peculiar
ataxic movements or a tendency to move backward were noted;
other symptoms, such as spasm, vomiting, headache, were general
phenomena of intracranial tumors; still others, such as hemiplegia,
hemiparesis, or anæsthesia, were probably simply due to the
position of the growth in the neighborhood of motor ganglia and
tracts.

Tumors of the cerebellum have some special symptoms, which also

derive importance from their characteristic grouping. The symptoms
which depend upon the lesion in the organ must be distinguished
from those which are caused by pressure upon adjacent parts,
although these latter symptoms are very important as corroborative
evidence of the location. Among the special symptoms is occipital
headache (often not present), especially when the pain is increased
by percussion about the occiput or by pressure upon the upper part
of the neck. In these cases weakness of the gait (Case 75) and other
motor phenomena, which are usually described as inco-ordination,
are of comparatively frequent occurrence. They are not so much true
inco-ordination as tremor of the limbs, rotation (which is usually only
partial), and the so-called movements of manége. These movements
were present in one-third of the cases collected by Leven and Oliver
(quoted by Rosenthal). Staggering gait is also present, and may be
dependent upon the vertigo, which is apt to be unusually intense in
this kind of intracranial tumor (Cases 69 and 71). The symptoms
caused by pressure of cerebellar tumors upon adjacent organs are
of importance, because in conjunction with the special symptoms
they acquire unusual significance. Sight and hearing are the two
special senses apt to be affected, because of pressure upon the
geniculate bodies and upon the auditory nerve or its nucleus.
Descending optic neuritis, progressing to total blindness, and varied
forms of oculo-motor paralysis may be present. Strabismus
convergens has been said to be a symptom, caused by the paralysis
of the sixth nerve. A hemiplegia and hemianæsthesia result
sometimes from pressure upon the tracts in the pons or medulla.
Continued pressure upon the medulla may eventually, toward the
termination of the case, according to Rosenthal, cause disorders of
the pulse and of respiration and deglutition. This author gives
absence of psychical symptoms as negative evidence which counts
for tumors of the cerebellum, but our table shows several instances
(Cases 70, 71, 74, and 76) in which were present hebetude,
incoherence, or hysteroidal symptoms. It is probable, however, that
such symptoms are not as common and distinct as in tumors of the
cerebrum.

Certain symptoms—or, better, groups of symptoms—characterize

tumors of the pons varolii, and serve to render the local diagnosis
comparatively certain. These depend upon the fact that the pons
combines in itself, or has on its immediate borders, nerve-tracts,
both motor and sensory, in great complexity, from or to almost every
special or general region of the body. Among these symptoms may
especially be mentioned alternating and crossed hemiplegia,
paralysis of eye-muscles (strabismus), paresis of tongue, dysphagia,
anæsthesia (sometimes of the crossed type), and painful affections
of the trigeminus. Vaso-motor disturbances have also been noted. In
one case persistent and uncontrollable epistaxis hastened the fatal
termination of the case.

Conjugate deviation of the eyes, with rotation of the head, as stated

under Symptomatology, is a condition often present in tumors of the
pons varolii as well as in the early stages of apoplectic attacks. A
paper43 has been published by one of us on a case of tumor of the
pons, and from it we will give some discussion of this subject.
43 Journal of Nervous and Mental Disease, July, 1881; Case 84 of Table.

Vulpian was probably the first to study thoroughly conjugate

deviation. The sign, when associated with disease of the pons, was
supposed by him and by others to be connected in some way with
the rotatory manifestations exhibited by animals after certain injuries
to the pons. Transverse section across the longitudinal fibres of the
anterior portions of the pons produces, according to Schiff, deviation
of the anterior limbs (as in section of a cerebral peduncle), with
extreme flexion of the body in a horizontal plane toward the opposite
side, and very imperfect movements of the posterior limbs on the
other side. Rotation in a very small circle develops in consequence
of this paralysis.44 The movements of partial rotation are caused,
according to Schiff, by a partial lesion of the most posterior of the
transverse fibres of the pons, which is followed in animals by rotation
of the cervical vertebræ (with the lateral part of the head directed
downward, the snout directed obliquely upward and to the side).
44 Rosenthal's Diseases of the Nervous System, vol. i. p. 125.

This deviation, both of head and eyes, occurs, however, not only
from lesions of the pons and cerebellar peduncles, but also from
disease or injury of various parts of the cerebrum—of the cortex,
centrum ovale, ganglia, capsules, and cerebral peduncles. It is
always a matter of interest, and sometimes of importance, with
reference especially to prognosis, to determine what is the probable
seat of lesion as indicated by the deviation and rotation.

Lockhart Clarke, Prevost, Brown-Séquard, and Bastian, among

others, have devoted considerable attention to this subject. To
Prevost we owe an interesting memoir. Bastian, in his work on
Paralysis from Brain Disease, summarizes the subject up to the date
of publication (1875). Ferrier, Priestly Smith, and Hughlings-Jackson
have investigated the relations which cortical lesions bear to the
deviation of the eyes and head.

It has been pointed out by several of the observers alluded to that

when the lesion is of the cerebrum the deviation is usually toward the
side of the brain affected, and therefore away from the side of the
body which is paralyzed. In a case of ordinary left hemiplegia it is
toward the right; in one of right hemiplegia, toward the left. In several
cases of limited disease of the pons, however, it has been observed
that the deviation has been away from the side of the lesion. In our
case (Case 84) the conjugate deviation was to the right, while the
tumor was entirely to the left of the median line, thus carrying out
what appears to be the usual rule with reference to lesions of the
pons.

During the life of the patient it was a question whether the case was
not one of oculo-motor monoplegia or monospasm from lesion of
cortical centres. It is probable, as Hughlings-Jackson believes, that
ocular and indeed all other movements are in some way represented
in the cerebral convolutions. In the British Medical Journal for June
2, 1877, Jackson discusses the subject of disorders of ocular
movements from disease of nerve-centres. The right corpus striatum
is damaged, left hemiplegia results, and the eyes and head often
turn to the right for some hours or days. The healthy nervous
arrangement for this lateral movement has been likened by Foville to
the arrangement of reins for driving two horses. What occurs in
lateral deviation is analogous to dropping one rein; the other pulls
the heads of both horses to one side. The lateral deviation shows,
according to Jackson, that after the nerve-fibres of the ocular nerve-
trunks have entered the central nervous system they are probably
redistributed into several centres. The nerve-fibres of the ocular
muscles are rearranged in each cerebral hemisphere in complete
ways for particular movements of both eyeballs. There is no such
thing as paralysis of the muscles supplied by the third nerve or sixth
nerve from disease above the crus cerebri, but the movement for
turning the two eyes is represented still higher than the corpus
striatum.

It would seem a plausible theory that we have in this conjugate

deviation of the eyes and head a distinct motor analogue to the
hemianopsia which results from certain lesions high in the optic
tracts. The fact that we never have a distinct oculo-motor
monoplegia from high lesions, but always a lateral deviation of both
eyes in the same direction, suggests that only a partial decussation
of the fibres of the motor nerves of the eyes occurs, and that each
hemisphere does not control the whole motor apparatus of the
opposite eye, but half of this apparatus in each eye.

Alternating hemiplegia, or paralysis of one side of the body followed

by a paralysis of the other side, is observed in tumors of the pons,
and is readily accounted for by the close proximity of the motor
tracts, a lesion which affects one tract first being very likely, sooner
or later, to involve, partially at least, the other, as in Case 84. Cross-
paralysis of the face and body may be seen, and like crossed
anæsthesia (seen also in Case 84) depends upon the fact that both
motor and sensory fibres to the limbs do not decussate at the same
level as these fibres to the face. Trigeminal neuralgia, from
involvement of the nerve by pressure or otherwise, is recorded in this
characteristic group of symptoms. The association of the general
with the local paralytic symptoms in the manner stated, the
involvement of sensory functions, and the deviation of the eyes and
head serve to distinguish tumors of the pons from cortical or high
cerebral local lesions. Cases 81, 84, 89, and 90 illustrate these facts
in various ways. Case 82, involving the floor of the fourth ventricle,
appears to be an exception, as the deviation is toward the side of the
lesion.

The special localizing symptoms which indicate a tumor of the crus

cerebri are paralysis of the oculo-motor nerve upon the same side as
the tumor, and especially the tendency of this paralysis to pass to the
other side later in the case; disturbance of the innervation of the
bladder; and involvement of the vaso-motor functions. In considering
these symptoms in detail it becomes very evident why we have the
alternating paralysis of the two oculo-motor nerves. As this trunk
arises from the crus, it is in direct risk of injury by the neoplasm, and
the extension of the new growth even slightly must later in the case
involve its fellow. Therefore a ptosis, followed by a similar symptom
on the other side, or other third-nerve symptoms passing from one
side to the other, with other characteristic and corroborating
symptoms, furnish strong evidence of this lesion, as in Case 93.
Rosenthal refers especially to involvement of the bladder, as
difficulty of micturition, but the three cases in the table do not present
such a symptom. He says that experiments prove that irritation of the
peduncle is followed by contraction of the bladder, and that it has
been shown that lesions of the crus abolish the influence of the will
upon micturition. As this occurs at all levels of the cord, its
occurrence with lesions of the crus is not to be considered a very
distinctive symptom. The involvement of the vaso-motor functions is
one of much interest. Its occurrence is not recorded in the cases of
tumors of the crus included in the table, but in Case 94 of twin
tumors in front of the optic chiasm it is recorded that profuse
perspiration occurred. We believe that the centres for the vaso-
motors are not well determined: they seem to be affected by various
lesions, especially about the base of the brain. Among other
corroborating symptoms may be mentioned rotatory movements and
deviation of the head: these rotatory movements are probably
caused by the action of the sound side not antagonized by the
muscles of the paralyzed side. Lateral deviation of the head is
referred to by some. Partial or complete hemiplegia, with facial
paralysis on the side opposite to the lesion, may occur; whereas the
oculo-motor palsy is seen on the same side as the lesion. Diminution
of sensibility happens on the opposite side, or occasionally pain in
the legs, as recorded in Case 92. It is of interest to note, with
Rosenthal, that the reactions of degeneration are not likely to appear
in the facial muscles in this lesion, as it occurs above the nucleus of
that nerve, and thus causes a true centric paralysis. The absence of
psychic symptoms is usually to be noted.

Tumors anywhere in the middle portion of the base of the brain and
floor of the skull, the region of the origin of the various cranial
nerves, can of course be diagnosticated with comparative ease by a
study of the various forms of paralysis and spasms in the distribution
of these nerves, in connection with other special and general
symptoms. Varieties of alternate hemiplegia are to be looked for, and
also isolated or associated palsies of the oculo-motor, pathetic,
facial, trigeminal, and other cranial nerves. In studying these palsies
it must be borne in mind that although the lesions producing them
are intracranial, the paralyses themselves are peripheral.
In most cases apparent exceptions to the ordinary rules as to
localization are capable of easy explanation; thus, for instance, in a
case of tumor of the occipital lobe (Case 44) numbness and pain
were present in the right arm, although the tumor was situated in the
right hemisphere. The tumor was of considerable size, and may
have affected by pressure the adjoining sensory tracts.

Hughlings-Jackson45 reports a case of tubercular tumor, half the size

of a filbert, in the pons under the floor of the fourth ventricle, in the
upper third of the left side. A much smaller nodule was found in the
right half of the pons. This patient, a man thirty-three years old, had
inconstant headache, a gradual incomplete hemiplegia of the right
side, with also paresis of the left masseter and right lower face.
Sensation was diminished in the right arm, leg, and trunk. The optic
discs were normal; the left pupil was smaller than the right. There
was lateral deviation of the eyes to the right. Diplopia was present in
some positions, and one image was always above the other. Aphasic
symptoms were also present. Especial interest attaches to the fact
that the facial paralysis in this case was on the same side as the
hemiplegia, opposite that of the lesion; whereas usually in lesions of
the pons facial paralysis is on the side opposite the hemiplegia. This
is explained by the fact that the tracts of the facial nerve decussate
in the pons below its upper third, and therefore in this case the lesion
caught the nerve-tracts above their decussation.
45 Med. Times and Gazette, London, 1874, p. 6.

PROGNOSIS.—The prognosis in intracranial tumors is of course

usually in the highest degree unfavorable. The early recognition of
the existence of a tumor syphilitic in origin will enable a
comparatively favorable prognosis to be made. It is far from correct,
however, to suppose that all or a majority of the cases of known
syphilitic origin are likely to have a favorable termination. Amidon46
puts this matter very correctly as follows: “Has a destructive lesion
occurred? and if so, where is it located, and what is its extent?
Indications of a destructive lesion should lead one to a cautious
prognosis as regards perfect recovery, while the prognosis for life