Civilization And Climate

CO 00

64225 >m

OSMANIA UNIVERSITY LIBRARY

Call No. i?3'^r/ /f ^4 ^ Accession No.% '
Author

nut

This book should be.returrtedjt'iiu^-^fof^ ?he date hist marked below-

CIVILIZATION AND CLIMATE

CIVILIZATION AND
CLIMATE

BY
ELLSWORTH HUNTINGTON

Late Research Associate in Geography in Yale University

NEW HAVEN
YALE UNIVERSITY PRESS

LONDON GEOFFREY CUMBERLEGE OXFORD UNIVERSITY PRESS

Copyright, 1015, 1922, 1924-, by Yale University Press.
Printed In the United Stales of America.

Published, 1915.

Second Printing, 19 16.

Third Printing, 1919.

Second and Enlarged Edition, 1922.

Third Edition with Many New Chapters, \ 924.

Third Edition, Second Printing, 193*3.

Third Edition, Third Printing, 1935.

Third Edition, Fourth Printing, September, 1939.

Third Edition, Fifth Printing, April, 1945.

Third Edition, Sixth Printing, April, 1948.

All rights reserved. This book may not be reproduced, in
whole or in part, in any form (except by reviewers for the
public press), without written permission from the publishers.

TABLE OF CONTENTS

List of Illustrations vii

Preface to First Edition xi

Preface to Third Edition xv

Author's Bibliography xvii

Chapter I. Introduction 1

Chapter II. Race or Place 30

Chapter III. The White Man in the Tropics . . 56

Chapter IV*. The Effect of the Seasons .... 76
Chapter V. The Effect of Humidity and Tempera-

Jure 109

Chapter VI. Work and Weather 136

Chapter VII." Health and the Atmosphere . . . 153

Chapter VIII. Mortality, Moisture, and Variability 174

Chapter IX. Health and Weather 194

Chapter X: The Ideal Climate 220

Chapter XI. The Distribution of Civilization . . 240
Chapter XII. Vitality and Education in the United

States 275

Chapter XIII. The Conditions of Civilization . . 291

Chapter XIV. The Shifting of Climatic Zones . . 315
Chapter XV. The Pulsatory Hypothesis and Its

Critics 335

Chapter XVI. The Shifting Centers of Civilization . 347
Chapter XVII. Aboriginal America and Modern Aus-
tralia 366

vi TABLE OF CONTENTS

Chapter XVIII. The Climatic Hypothesis of Civiliza-
tion 387

Appendix 413

Index 433

LIST OF ILLUSTRATIONS

Figure 1. The Effect of the Seasons on Factory

Operatives in Connecticut and at Pittsburgh 84
Figure 2a. Human Activity and the Seasons ... 92
Figure 2b. Human Activity and the Seasons ... 93
Figure 3. Seasonal Variations of Mental Compared

with Physical Activity 105

Figure 4. Relative Humidity and Work in Connecticut 112
Figure 5. Average Weekly Temperature During the

Summers of 1910-1913 in Connecticut . . 116
Figure 6. Effect of the Days of the Week on Piece-
Workers 120

Figure 7. Variations in Daily Wages 122

Figure 8. Human Activity and Mean Temperature . 124
Figure 9. Growth of Wheat at Various Temperatures 129
Figure 10. Mean Temperature and Vital Processes in

Plants, Animals, and Man 130

Figure 11. Human Activity and Change of Mean Tem-
perature from Day to Day 140

Figure 12. The Stimulus of Storms 147

Figure 13. Seasonal Variations in Energy and Health . 154
Figure 14. Seasonal Variations in Conceptions and

Deaths in Japan, 1901-1910 .... 159
Figure 15. Post-operative Death Rate at Boston in

Relation to Humidity and Temperature . 180
Figure 16. Relation between Deaths from Pneumonia
and Influenza and Interdiurnal Changes of

Temperature 184

Figure 17. Correlation between Weather Elements and
Daily Deaths in New York City, December
to March 1882-1886 196

Vlll

LIST OF ILLUSTRATIONS

Figure 18a. Excess or Deficiency of Death Rates in

Months of Extreme Weather in American

Cities, 1900-1915 206

Figure 18b. Excess or Deficiency of Death Rates in

Months of Extreme Weather in American

Cities, 1900-1915 207

Figure 19. Correction for Effect of Other Climatic

Factors 211

Figure 20. Excess or Deficiency of Deaths in Relation

to Stormy Periods Lasting One Month, Two

Months, ancLThree Months, 1900-1915 . 216
Figure 21. Net Effect of Weather in the United States 218
Figure 22. The Effect of Climate on Human Energy as

Inferred from Work in Factories . . . 234
Figure 23. The Distribution of Civilization^n Europe 257
Figure 24. The Distribution of Civilization in Asia . 259
Figure 25. The Distribution of Civilization in Australia 260
Figure 26. The Distribution of Civilization in Africa 261
Figure 27. The Distribution of Civilization in South

America 262

Figure 28. Civilization in North America, According

to all Contributors 264

Figure 29. Civilization in North America, According to

Twenty-five Americans 265

Figure 30. Civilization in North America, According

to Seven British Contributors .... 268
Figure 31. Civilization in North America, According

to Six Germanic Europeans 269

Figure 32. Civilization in North America, According

to Six Latin Europeans and One Russian . 272
Figure 33. Civilization in North America, According

to Five Asiatics 273

Figure 34. Climatic Energy in the United States on the

Basis of Factory Work 276

LIST OF ILLUSTRATIONS

IX

Figure 35. Vitality in the United States, According to

Life Insurance Statistics 276

Figure 36. Distribution of Climatic Health on Basis of

Seasonal Variations in 33 Cities, 1900-1915 277
Figure 37. Civilization in the United States . . . 277
Figure 38. Distribution of Education in the United

States 284

Figure 39. Percentage of Gainfully Employed Persons
in the United States Engaged in Manu-
facturing, 1919 284

Figure 40. Distribution of Climatic Energy in Europe 292
Figure 41. Distribution of Health in Europe . . . 293
Figure 42. Distribution of Civilization in Europe . . 293
Figure 43. The Distribution of Human Health and

Energy on the Basis of Climate .... 295
Figure 44. The Distribution of Civilization . . . 295
Figure 45. Birthplaces of Persons of Unusual Ability

in the United States 307

Figure 46. Changes of Climate in Western Asia and

Growth of Trees in California . . . .318
Figure 47. Changes of Climate in California for 2,000

Years 321

Figure 48. The Shifting of the Storm Belt .... 358

PREFACE TO FIRST EDITION

THIS volume is a product of the new science of geography.
The old geography strove primarily to produce exact maps
of the physical features of the earth's surface. The new goes
farther. It adds to the physical maps an almost innumerable
series showing the distribution of plants, animals, and man, and
of every phase of the life of these organisms. It does this, not
as an end in itself, but for the purpose of comparing the physi-
cal and organic maps and thus determining how far vital phe-
nomena depend upon geographic environment. Among the things
to be mapped, human character as expressed in civilization is
one of the most interesting and one whose distribution most
needs explanation. The only way to explain it is to ascertain the
effect of each of many cooperating factors. Such matters as
race, religion, institutions, and the influence of men of genius
must be considered on the one hand, and geographical location,
topography, soil, climate, and similar physical conditions on
the other. This book sets aside the other factors, except inci-
dentally, and confines itself to climate. In that lie both its
strength and weakness. When the volume was first planned, I
contemplated a discussion of all the factors and an attempt to
assign to each its proper weight. The first friend whom I con-
sulted advised a directly opposite course, whereby the emphasis
should be centered upon the new climatic facts which seem to
afford ground for a revision of some of our old estimates of the
relation between man and his environment. In writing the book
I have growingly felt the wisdom of that advice, and have been
impressed with the importance of concentration upon a single
point, even at the expense of seeming to take a one-sided view.

xii PREFACE TO FIRST EDITION

If the reader feels that due weight is not given to one factor or
another, he must remember that many unmentioned phases of
the subject have been deliberately omitted to permit fuller
emphasis upon the apparent connection between a stimulating
climate and high civilization.

The materials for this volume have been derived from a great
variety of sources. Although personal observation and investi-
gation are the basis of much that is here stated, still more has
been derived from the world's general store of knowledge. Except
in a few special cases I have not attempted to give references.
To the general reader footnotes are not only useless, but often
a distraction and nuisance. The careful student, on the other
hand, cannot form a fair estimate of the hypothesis here pre-
sented without reading previous publications in which I have
set forth the reasons for many conclusions which are not fully
discussed here for lack of space. These publications contain
numerous references. Accordingly, the needs of the student will
be met by giving a brief list of books and articles which have
served as preliminary steps to the present volume. These publi-
cations form a logical series with only such repetition as is
necessary to make each a complete unit. It is scarcely necessary
to add that the rapid growth of the subject during the past ten
years has led to important modifications in some of the earlier
conclusions.

The facts set forth in this volume have by no means been
derived wholly from observation and reading. Not far from a
hundred people have given direct personal assistance. They are
so numerous that it is impossible to mention them all by name.
Therefore it seems best not to single out any for special thanks.
Many of my colleagues among the Yale faculty and among the
geographers of America have gone out of their way to offer sug-
gestions, or friendly criticisms, or to bring to my attention
publications and facts that might have escaped my notice.
Others connected with such organizations as the Carnegie Insti-

PREFACE TO FIRST EDITION xiii

tution of Washington and the United States Weather Bureau
have placed me under obligation by the kind way in which they
have taken a personal rather than official interest in answering
queries and providing data. Equally great courtesy has been
shown by officers and other members of the teaching force at
West Point and Annapolis, and by officials connected with
various factories, including some whose figures it has not yet
been possible to tabulate. Another large group comprises con-
tributors to the map of civilization, many of whom devoted to
this work time which they could ill afford. Lastly, I owe much
to personal friends who fall in none of the groups already speci-
fied. I suppose that the total time given to this book by all these
scores of people makes their contribution larger than mine. My
chief hope is that they may feel that their kindness has not been
wasted. To each and all I can only express my deep sense of
gratitude, and most of all to those whose advice from the be-
ginning has done more than anything else to keep this book true
to a single aim.

E. H.
Yale University,

New Haven, Conn.,
July, 1915.

PREFACE TO THIRD EDITION

IN preparing the third edition of this book two circumstances
have led to a radical revision. First, the World War has
receded far enough into the background so that people's minds
are once more able to concentrate upon the far-reaching prob-
lems of science, instead of the temporary details associated with
war. During the past three years this has led to a marked in-
crease in the number of publications dealing with the problems
of this book. Second, during the nine years since the first edition
was issued many new facts have come to light which amplify
and strengthen the general hypothesis of the effect of climate
upon civilization.

The present edition differs from the first in several important
respects :

1. In the first edition inheritance, physical environment, and
culture were recognized as the three main factors in determining
the distribution of civilization. Physical environment was of
course treated fully, since it is the main subject of the book.
Enough was also said about human culture to show that I fully
appreciate its importance, especially as an explanation of the
difference between aboriginal America and the Old World. In-
heritance, however, was dismissed briefly. In the present edi-
tion it receives a good deal of emphasis, especially in the first
chapter, which is almost wholly new. It would be emphasized
much more strongly were it not that in The Character of Races
I have devoted a whole book to the problem. That book and this
are so closely allied that neither is complete without the other.

2. The relation of climate to health has been much discussed
during the past nine years. Accordingly three new chapters, Vii

xvi PREFACE TO THIRD EDITION

to IX, have been added on this topic. They balance the three
preceding chapters by discussing the manifold effects of cli-
mate and weather upon man from the standpoint of disease and
death, instead of from the standpoint of the day's work.

3. In the first edition of Civilization and Climate I assumed
that historians and others would be more familiar with the evi-
dence of climatic changes during historic times than is actually
the case. Accordingly in Chapter XIV the hypothesis of pulsa-
tory climatic changes is more fully elaborated than formerly,
while Chapter XV, which is new, is devoted to criticisms of that
hypothesis.

4. Chapters XVI and XVII are devoted to some of the main
criticisms of the hypothesis that climate is one of the three main
determinants of the distribution of civilization. An especially
important new feature is a study of the white man in tropical
Australia.

5. In addition to this a large number of minor points have
been added. Hence, although certain sections such as Chapters
III to VI remain practically unchanged, the book as a whole has
a distinctly new aspect. In its present form the book does not
insist as strongly as before upon the supreme importance of
climate, but the arguments which lead to the conclusion that
climate ranks with racial inheritance and cultural development
as one of the three main factors in determining the distribution
of civilization seem much stronger than previously.

E. H.
New Haven, September, 1924.

AUTHOR'S BIBLIOGRAPHY

THIS list contains the titles of publications of the author
which deal with the problems of this book. For details as
to many points here discussed, the reader is referred to these
publications.

(A) Books and longer articles on changes of climate and their
effect on man:

(1) EXPLORATIONS IN TURKESTAN. Publications 26 and 73
of the Carnegie Institution of Washington, 1905, pp. 157-317.

(2) THE PULSE OF ASIA. Boston, Houghton Mifflin Company,
1907, 415 pp.

(3) PALESTINE AND ITS TRANSFORMATION. Boston, Hough-
ton Mifflin Company, 1911, 433 pp.

(4) THE CLIMATIC FACTOR AS ILLUSTRATED IN ARID AMER-
ICA. Publication 192 of the Carnegie Institution of Washington,
1914, 341 pp.

(5) THE SOLAR HYPOTHESIS OF CLIMATIC CHANGES. Bulletin
of the Geological Society of America, vol. 25, 1914, pp. 477-
590.

(B) Shorter articles dealing with phases of the problem of
climatic changes not treated under (A) :

(6) THE BURIAL OF OLYMPIA. Geographical Journal, Lon-
don, 1910, pp. 657-686.

(7) THE OASIS OF KHARGA. Bulletin of the American Geo-
graphical Society, New York, vol. 42, 1910, pp. 641-661.

(8) TREE GROWTH AND CLIMATIC INTERPRETATIONS. In Post-
glacial Climatic Changes, Publication No. 352, Carnegie Insti-
tution of Washington, 1924. (In press.)

(C) Books and articles dealing with climate and civilization:

(9) PHYSICAL ENVIRONMENT AS A FACTOR IN THE PRESENT

xviii AUTHOR'S BIBLIOGRAPHY

CONDITION OF TURKEY. Journal of Race Development, Worces-
ter, Mass., vol. 1, 1911, pp. 460-481.

(10) GEOGRAPHICAL ENVIRONMENT AND JAPANESE CHAR-
ACTER. Journal of Race Development, Worcester, Mass., vol. 2,
1912, pp. 256-281. (Reprinted in a volume entitled JAPAN AND
JAPANESE AMERICAN RELATIONS, edited by G. H. Blakeslee,
New York, 1912, pp. 42-67.)

(11) CHANGES OF CLIMATE AND HISTORY. American Histori-
cal Review, vol. 18, 1913, pp. 213-232.

(12) THE ADAPTABILITY OF THE WHITE MAN TO TROPICAL
AMERICA. In Latin America. Clark University Addresses, 1913,
edited by G. H. Blakeslee, pp. 360-386.

(13) THE GEOGRAPHER AND HISTORY. The Geographical
Journal, London, January, 1914.

(14) A NEGLECTED FACTOR IN RACE DEVELOPMENT. The
Journal of Race Development, Worcester, Mass., vol. 6, 1915,
pp. 167-184.

(15) CLIMATIC VARIATIONS AND ECONOMIC CYCLES. The
Geographical Review, vol. 1, March, 1916, pp. 192-201.

(16) MAYA CIVILIZATION AND CLIMATIC CHANGES. Proceed-
ings of the 19th International Congress of Americanists, Wash-
ington, 1917, pp. 150-164.

(17) CLIMATIC CHANGE AND AGRICULTURAL EXHAUSTION AS
ELEMENTS IN THE FALL OF ROME. The Quarterly Journal of
Economics, vol. 31, February, 1917, pp. 173-208.

(18) THE RED MAN'S CONTINENT. New Haven, 1919.

(19) PRINCIPLES OF HUMAN GEOGRAPHY. New York, 1920.

(20) BUSINESS GEOGRAPHY. New York, 1922.

(D) Books and articles published since the first edition of CIVI-
LIZATION AND CLIMATE and dealing largely with the relation
between climate and health:

(21) WORLD POWER AND EVOLUTION. New Haven, 1919.

(22) THE INTERPRETATION OF THE DEATH RATE BY CLIMO-
GRAPHS. Modern Medicine, vol. 1, May, 1919.

AUTHOR'S BIBLIOGRAPHY xix

(23) THE CONTROL, OF PNEUMONIA AND INFLUENZA BY THE
WEATHER. Ecology, vol. 1, January, 1920.

(24) THE PURPOSE AND METHODS OF AIR CONTROL IN HOS-
PITALS. The Modern Hospital, vol. 14, April, 1920.

(25) METHODS OF AIR CONTROL AND THEIR RESULTS. The
Modern Hospital, vol. 14, May, 1920.

(26) THE RELATION OF HEALTH TO RACIAL CAPACITY: THE
EXAMPLE OF MEXICO. The Geographical Review, vol. 11, April,
1921.

(27) AIR CONTROL AS A MEANS OF REDUCING THE POST-
OPERATIVE DEATH RATE. American Journal of Surgery, vol. 35,
July and October, 1921.

(28) INFLUENZA AND THE WEATHER IN THE UNITED STATES
IN 1918. The Scientific Monthly, vol. 17, November, 1923.

(29) CAUSES OF GEOGRAPHICAL VARIATIONS IN THE INFLU-
ENZA EPIDEMIC OF 1918 IN THE CITIES OF THE UNITED STATES.
Bulletin of the National Research Council, vol. 6, July, 1923.

(E) Books and articles dealing with natural selection and racial
character:

(30) GEOGRAPHY AND NATURAL SELECTION. Annals of the
Association of American Geographers, 1924.

(31) THE CHARACTER OF RACES. New York, 1924.

(F) Books dealing zvith the nature and causes of climatic
changes :

(32) CLIMATIC CHANGES (with S. S Visher). New Haven,
1922.

(33) EARTH AND SUN: AN HYPOTHESIS OF SUNSPOTS AND
WEATHER. New Haven, 1923.

CHAPTER 1
CIVILIZATION AND CLIMATE

INTRODUCTION

THE races of the earth are like trees. Each according to its
kind brings forth the fruit known as civilization. As russet
apples and pippins may grow from the same trunk, and as
peaches may even be grafted on a plum tree, so the culture of
allied races may be transferred from one to another. Yet no one
expects pears on cherry branches, and it is useless to look for
Slavic civilization among the Chinese. Each may borrow from
its neighbors, but will put its own stamp upon what it obtains.
The nature of a people's culture, like the flavor of a fruit, de-
pends primarily upon racial inheritance, which can be changed
only by the slow processes of biological variation and selection.
Yet inheritance is only one of the factors in the development
of civilization. Religion, education, government, and all of man's
varied occupations, customs, and institutions his inherited
culture as the anthropologists say form a second great group
of social influences whose power seems almost immeasurable.
They do for man what cultivation does for an orchard. One tree
may bear a few wormy, knotty little apples scarcely fit for the
pigs, while another of the same variety is loaded with great red-
cheeked fruit of the most toothsome description. The reason
for the difference is obvious. One tree grows in a grassy tangle
of bushes with no room to develop, little chance to get sunlight,
and scant opportunity to obtain nourishment because of the
abundance of other plants and the poverty and thinness of the
unfertilized soil. The other stands in the midst of a carefully

2 CIVILIZATION AND CLIMATE

tilled garden where it has plenty of room to expand and enjoy
the sun, and where its roots can spread widely in a deep, mellow,
well-fertilized soil. Moreover, one tree is burdened with dead
wood and suckers, and infested with insects and other parasites ;
while the other is carefully pruned, scraped, and sprayed.

In spite of the most careful and intelligent cultivation, a tree
of the finest variety may fail to produce good fruit. Too much
rain or too little ; prolonged heat or constant cloudiness ; frost
when the blossoms are opening, or violent wind and hail may all
be disastrous. The choicest tree without water is worth less than
the poorest where the temperature and rainfall are propitious.
Its health is ruined, and it can bear no fruit. Here, as in the
preceding case, the great need of the tree is health in the fullest
and broadest sense. A good climate, good cultivation, and good
nourishment are merely means of giving the tree perfect health
and thus allowing the fullest development of its inheritance.
Thus the two great factors which really determine the quality
of the fruit are inheritance and health. The other factors,
namely, food, climate, parasites, and cultivation, are important
chiefly as means whereby health, or, perchance, inheritance, is
modified.

Does the fruit known as civilization depend upon these same
conditions? It seems to me that it does. Few would question that
a race with a superb mental and physical inheritance and en-
dowed with perfect health is capable of adding indefinitely to
the cultural inheritance received from its ancestors, and thus
may attain the highest civilization. But if that same cultural
inheritance were given to a sickly race with a weak inheritance
of both mind and body, there would almost surely be degenera-
tion. Aside from biological inheritance, the main factors in
determining health are climate, food, parasitic diseases, and a
people's stage of culture, which corresponds to the cultivation
of the tree. Moreover, these same four factors, through their
potency in selecting some types for preservation and others for

CIVILIZATION AND CLIMATE 3

destruction, and perhaps through their power to cause muta-
tions, are among the main agencies in determining inheritance.
Climate stands first, not because it is the most important, but
merely because it is the most fundamental. It is fundamental
by reason of its vital influence upon the quantity and quality not
only of man's food but of most of his other resources ; it plays
a large part in determining the distribution and virulence of
the parasites which cause the majority of diseases ; and through
its effect upon human occupations, modes of life, and habits, it
is one of the main determinants of culture. On the other hand,
neither food, disease, nor culture has any appreciable effect
upon climate, although they may modify its influence. Moreover,
climate has a direct effect upon health in addition to its indirect
effect through food, disease, and mode of life. Hence although
climate may be no more important than other factors in deter-
mining the relative degree of progress in different parts of the
world, it is more fundamental in the sense that it is a cause
rather than a result of the other factors.

In studying climate it is essential to draw a sharp distinction
between three types of influences. In the first place climate has
a direct effect upon man's health and activity. Second, it has
a strong indirect but immediate effect through food and other
resources, through parasites, and through mode of life. Third,
by its combined direct and indirect effects in the past it has
been a strong factor some would say, the strongest in caus-
ing migration, racial mixture, and natural selection ; and it may
have had something to do with producing the variations which
the biologists call mutations. Thus it has had a powerful effect
upon inheritance.

From the days of Aristotle to those of Montesquieu and
Buckle, there have been good thinkers who have believed that
the direct effect of climate is the most important factor in de-
termining the differences between the degree of progress in vari-
ous parts of th? earth. Others have held that wherever food is

4 CIVILIZATION AND CLIMATE

available for a moderately dense population and man can avoid
diseases like tropical malaria, human culture can rise to the
highest levels. The location of the world's great nations seems
to them largely a matter of accident.

The majority of people reject both of these extreme views.
Few doubt that climate has an important relation to civilization,
but the majority consider it less important than racial inherit-
ance, proper food, or good institutions in the form of church,
state, and home. We realize that a dense and progressive popula-
tion does not live in the far North or in deserts simply because
l 1 he difficulty of getting a living grinds men down and keeps
them isolated. We know that the denizens of the torrid zone are
slow and backward, and we almost universally agree that this
is connected with the damp, steady heat. We continually give
concrete expression to our faith in climate. Not only do we talk
about the weather more than about any other one topic, but we
visit the seashore or the mountains for a change of air. We go
South in winter, and to cool places in summer. We are depressed
by a series of cloudy days, and feel exuberant on a clear, brac-
ing morning after a storm. Yet, in spite of this universal recog-
lition of the importance of climate, we rarely assign to it a
'oremost place as a condition of civilization. We point out that
cpreat nations have developed in such widely diverse climates as
v he hot plains of Mesopotamia and Yucatan and the cool hill
country of Norway and Switzerland. Moreover, although Illi-
nois and southern Mongolia lie in the same latitude and have
the same mean temperature, they differ enormously in civiliza-
tion. To put the matter in another way, we recognize two great
sets of facts which are apparently contradictory. We are con-
scious of being stimulated or depressed by climatic conditions,
*nd we know that as one goes northward or southward, the dis-
tribution of civilization is generally in harmony with what we
should expect on the basis of our own climatic experiences.
Nevertheless, even in our own day, regions which lie in the same

CIVILIZATION AND CLIMATE 5

latitude and apparently have equally stimulating climates differ
greatly in their degree of civilization. When we compare the
past with the present, we find the same contradiction still more
distinctly marked. Hence our confusion. From personal experi-
ence we know that the direct effects of climate are of tremendous
importance. Yet many facts seem to indicate that this impor-
tance is less than our observation would lead us to anticipate.

The reason for this doubtful attitude can easily be discovered.
The things that we call facts are often not well established.
Although we believe in the influence of climate, we know little
of the particular climatic elements which are most stimulating
or depressing. How much do we know of the relative importance
of barometric pressure, wind, temperature, or humidity ? What
about the comparative effects of the climates of England and
southeastern Russia? In addition to this, we are far from know-
ing what type of climate prevailed in Egypt, Greece, or Mespo-
tamia when they rose to eminence. Many good authorities have
asserted that the climate of those regions was the same two or
three thousand years ago as now. This view is rapidly losing
ground, but those who believe in a change are not certain of
its nature. They are not yet wholly agreed as to whether it has
produced an important influence upon the particular climatic
elements which are most stimulating to the human system.

This book has been written because two recent lines of in-
vestigation apparently combine to explain at least part of the
contradictions which have hitherto proved so puzzling. In the
first place a prolonged study of past and present climatic varia-
tions led to the conclusion that the climate of the past was dif-
ferent from that of the present. In early historic times, for
example, some parts of the world appear to have been drier than
now* and other parts moister. In any given place, however, the
change from the past to the present has not consisted of
~* eadily progressive tendency in one direction but of variations.
In the places that were formerly moister than now there appear

6 CIVILIZATION AND CLIMATE

to have been alternate changes toward dryness and then toward
moisture ; while in the places that are drier than in the past there
have been corresponding variations of the opposite types. In
a word, the climate of historic times seems to have undergone
a pronounced series of pulsations which have varied in char-
acter from one part of the earth to another. The second line of
investigation which originally led to the writing of this book
was a study of the climatic conditions under which people of
European races are able to accomplish the most work and have
the best health. This investigation led to the conclusion that the
principle of climatic optima applies to man quite as fully as to
plants and animals. According to this principle each living
species has the best health and is most active under certain
definite conditions of temperature, humidity, wind movement,
storminess, variability, and sunlight, or, more exactly, under
certain combinations of these conditions. Any departure from
the optimum conditions leads to a decrease of activity and effi-
ciency.

During the last ten years the importance of racial inheritance
and racial selection has been strongly emphasized. In the first
edition of Civilization and Climate the importance of race is
strongly emphasized, but I failed to see how important a part
has been played by climatic changes in selecting certain types
of people for destruction or preservation. Such selection is ap-
parently one of the chief ways in which the character of races
is altered. The climatic pulsations of the glacial, post-glacial,
and historic periods appear to have exerted a profound influence
upon the degree of habitability of different parts of the earth.
Thus famine, distress, and disease have arisen, and the pressure
of population has led to migration, racial mixture, and the pres-
ervation of one type of people in one place and another some-
where else. Natural selection under the stress of climate goes far
toward explaining many of the cases where the distribution of
civilization does not agree with what would be expected on the

CIVILIZATION AND CLIMATE 7

basis of the direct effect of climate. So important is this that
I have written a book on The Character of Races for the ex-
press purpose of applying the principles of natural selection to
the history of racial development. That book might well have
been called Civilization and Race in order to emphasize the
fact that it is a continuation of the present work. The change
in my own realization of the part played by climatic changes is
one of the chief reasons why the present edition of Civilization
and Climate is in some respects almost a new book.

A large part of the reasoning of this book stands or falls with
the hypothesis of climatic pulsations in historic times. The steps
which led to the hypothesis may be briefly sketched as follows :
In 1903, under the inspiration of the broad vision of Raphael
Pumpelly and the careful scientific methods of William Morris
Davis, I began to study the climate of the past. Two years' work
with the Pumpelly Expedition sent to Turkestan by the Carnegie
Institution of Washington led to the conviction that Reclus,
Kropotkin, and others are correct in believing that two or three
thousand years ago the climate of Central Asia was moister than
now, a view which I advocated in Explorations in Turkestan.
Later, during the Barrett Expedition to Chinese Turkestan, it
became evident that the scientists who hold that the ancient
climate in those regions was as dry as that of today also have
much strong evidence to support their view. It soon appeared,
however, that this apparent contradiction is fully explained
by the fact that throughout the dry regions of Central Asia and
the eastern Mediterranean the evidences of moist and dry con-
ditions, respectively, are grouped in distinct periods ; the begin-
ning of the Christian era was moist, for example, and the seventh
century dry. This led to what I have called the "pulsatory
hypothesis," which furnished a name for The Pulse of Asia.
According to this hypothesis, although the historic and pre-
historic past in those particular regions was in general moister
than the present, the change from moist to dry has taken place

8 CIVILIZATION AND CLIMATE

irregularly in great waves. Even in early historic times certain
centuries were apparently drier than today, while others not
long ago were moist. In 1909, this view was confirmed during
the Yale Expedition to Palestine, the results of which are set
forth in Palestine and Its Transformation.

Then a series of journeys in the drier parts of the United
States and in Mexico and Central America, in cooperation with
D. T. MacDougal of the Desert Botanical Laboratory of the
Carnegie Institution, showed that the main features of previous
conclusions apparently apply to the New World as well as the
Old. The most important feature of this work in America was
the measurement of the thickness of the annual rings of growth
of some four hundred and fifty of the Big Trees of California
the Sequoia Washingtoniana, which grows high in the Sierras.
These measurements made it possible to form a fairly reliable
climatic curve for 2000 years and an approximate curve for
another thousand. The final data as to the Big Trees were pub-
lished in The Climatic Factor, which appeared in 1915, at the
same time as Civilization and Climate. The agreement of the
California curve with the climatic curve of western Asia as pre-
viously worked out, and the constantly growing evidence as to
the reliability of tree growth as a measure of climate have done
far more than anything else to cause the hypothesis of climatic
mlsations to be widely accepted. Here is the way the matter
is summed up by the British meteorologist, C. E. P. Brooks, in
his book on The Evolution of Climate (1922) : "The question of
climatic changes during the historic period has been the subject
of much discussion, and several great meteorologists and geog-
raphers have endeavoured to prove that at least since about 500
B.C. there has been no appreciable variation. It is admitted that
there have been shiftings of the centers of population and civi-
lization, first from Egypt and Mesopotamia to the Mediterra-
nean regions, and later to northern and western Europe, but
these have been attributed chiefly to political causes and es-

CIVILIZATION AND CLIMATE 9

pecially to the rise of Islam and the rule of the 'accursed Turk. 1 *
Recently, however, there has arisen a class of evidence which
cannot be explained away on political grounds, and which ap-
pears to have decided the battle in favour of the supporters of
change; I refer to the evidence of the trees. The conclusions
derived from the big trees of California have fallen admirably
into line with archaeological work in Central America, in central
Asia and other regions, and have shown that the larger varia-
tions even of comparatively recent times have been very exten-
sive, if not world-wide, in their development."

Another important factor in perfecting the pulsatory hy-
pothesis has been the study of the Maya ruins in Yucatan and
Guatemala. They join with other evidence in suggesting that
changes of climate are of different types in different parts of
the world. Central America seems to have been relatively dry
at times when the Big Trees of the Sierras suggest that Cali-
fornia was moist. This is an important modification of some of
the conclusions which I have seemed to imply in earlier books.
At practically the same time when this newer conclusion was
published, an almost identical idea was presented by J. W.
Gregory of England, whose article 7s the Earth Drying Up? in
the Geographical Journal for 1914 is the strongest criticism of
my climatic theories that has ever appeared. It will be discussed
more fully later. The fact that two investigators who seemed to
be opposed should independently publish the same conclusion
without knowing what the other was doing greatly strengthens
the force of that conclusion.

Having reached the conclusion that pulsatory climatic
changes have taken place during historic times and have differed
in type from region to region, the next step was to study the
mechanism and cause of the supposed changes. In Palestine and
the eastern Mediterranean the conditions of vegetation, espe-
cially the palm and vine, as Gregory has well shown, make it
almost certain that variations in storminess and rainfall, rather

10 CIVILIZATION AND CLIMATE

than in temperature, have been the primary factor. The recorded
observations upon the mild climatic pulsations of the past hun-
dred years support this conclusion. Various lines of evidence
also indicate that climatic pulsations probably consist of a
shifting of the earth's climatic zones or at least of the areas of
cyclonic storms alternately toward and away from the equator.
The idea as to zones, although not as to storms, was announced
by the German geologist, Penck, at essentially the same time
that I announced it, but the two conclusions were wholly inde-
pendent and were based on quite different data. In both cases it
was specifically recognized that the same kinds of climatic
shiftings have taken place both in prehistoric and historic times,
although the earlier changes were of greater magnitude. In these
shiftings the temperate zone of storms appears to have been
shoved irregularly back and forth. When it was farther south
than at present, the subtropical countries which now are subarid
must have been relatively moist. At the same time the subtropical
arid belt was apparently sliif ted toward the equator so that on
the borders of the torrid zone certain lands which now are wet
were then relatively dry. When the shifting of zones took place
in the opposite direction the reverse changes of climate appar-
ently took place.

It was only after the preceding conclusions as to climatic
pulsations had reached essentially their present form that I
began the next phase of the investigation, namely, the study of
the actual effect of present climates upon human health and
activity. This is important because some critics have supposed
that I have unduly emphasized the importance of climatic
changes, or have even formed a theory in regard to them for the
purpose of bolstering up a preconceived idea that differences of
climate from place to place are a main cause of the present dis-
tribution of human progress. On the contrary, up to this period
my reasoning had been somewhat as follows : If climatic changes
have occurred during historic times, they must have had some

CIVILIZATION AND CLIMATE 11

economic effect because such changes alter the capacity of a
region to support population. The economic changes in their
turn must have led to political disturbances and migrations.
Is there any evidence of such events at times when the climate
suffered unusually great or rapid changes?

The possibility of such a connection between climate and
history has deeply interested a great number of students. Kro-
potkin, for instance, has vividly portrayed the wav in which a
gradual desiccation of Asia presumably drove into Europe the
hordes of barbarians whose invasions were so important a fea-
ture of the Dark Ages. If the change from the climate of the
past to that of the present has been marked by pulsations rather
than by a progressive change in only one direction and if there
have been certain periods of rather rapid change and of great,
though temporary, extremes, as seems highly probable, the cor-
respondence between historic events and climatic vicissitudes
may be closer than would otherwise seem credible. Indeed, as
soon as I had framed a preliminary outline of the curve of
climatic changes during historic times, it appeared as though
many of the great nations of antiquity, had risen or fallen in
harmony with favorable or unfavorable conditions of climate.
During periods of drought, not only are the people of the drier
regions forced to migrate, especially if they are nomads, but
increasing aridity, even in more favored places such as Greece,
must cause economic distress, and thus engender famine, misery,
and general discontent and lawlessness. A recent journey to
China, which gave an opportunity for a study of the famines and
barbarian invasions that have afflicted that country for two
thousand years, has added greatly to the already abundant evi-
dence of the truth of this view. It has also emphasized the re-
markably intimate connection between economic distress and
political discontent, a connection which is obvious in advanced
countries like the United States, Australia, and Europe as well
as in backward regions like China, Persia, India, and Mexico.

12 CIVILIZATION AND CLIMATE

While these economic and political effects of climatic changes
were being studied I became more and more impressed by the
fact that when each country rose to a high level of civilization,
it appears in a general way to have enjoyed a climate which
approached more closely than now to certain well-defined con-
ditions. These conditions appeared to resemble, but by no means
duplicate, those now prevailing in most of the regions where
civilization is highest. In spite of marked variations, the general
tendency during periods of high civilization has apparently been
toward cool, but not extremely cold, winters, and toward sum-
mers which though warm or even hot for several months are
generally varied by storms or at least by winds which produce
frequent changes of temperature. It became especially evident
that a relatively high degree of storminess and a relatively long
duration of the season of cyclonic storms have apparently been
characteristic of the places where civilization has risen to high
levels both in the past and at present. Hence such places experi-
ence much variability, a condition which later work has led me
to believe highly beneficial.

Up to this point in my investigations, I saw no ground for
appealing to anything except economic and political factors in
explanation of the apparent connection between civilization and
climate. Then a little book on Malaria: A Neglected Factor in
the History of Greece and Rome, by W. H. S. Jones, convinced
me that climatic changes have altered the conditions of health
as well as the economic situation. Later studies indicate that in
other countries such as Central America, Indo-China, Java, and
Egypt, as well as Greece and Rome, changes in the amount and
virulence of such diseases as malaria and yellow fever may have
been potent factors in diminishing the vitality of a nation. In
fact it now seems probable that through their effect on bacteria,
on the water supply, on the breeding places of insects, on the
quality of the food, and perhaps in other ways climatic changes

CIVILIZATION AND CLIMATE 13

may exert quite as much effect as through the more direct eco-
nomic channels.

The study of diseases was the natural prelude to a closer
inquiry into the fact that at times of favorable climate in coun-
tries such as Egypt and Greece the people were apparently filled
with a virile energy which they do not now possess. Many
authorities attribute the loss of this to an inevitable decay which
must overtake a nation as old age overtakes an individual. Others
ascribe it to the lack of adaptability in various institutions, to
increasing luxury, to contact with inferior civilizations, to a
change in racial inheritance, or to various other factors, most
of which are doubtless of importance. Previous to 1911 a few
authorities such as 0. Fraas had connected the decline of energy
in Egypt, for example, with climatic changes, but they gave so
few reasons and the whole matter seemed so doubtful that I had
little faith in their suggestions. At that time, Charles J. Kull-
mer of Syracuse University sent me a manuscript calling atten-
tion to the remarkable similarity between the distribution of
cyclonic storms and of civilization. His article was never pub-
lished, but was presented at a meeting of the Association of
American Geographers. He advanced the idea that the baro-
metric changes which are the primary cause of storms, or
perhaps some electrical phenomena which accompany them, may
produce a stimulus which has much to do with the advancement
of civilization. Although he presented no definite proof, his sug-
gestion seemed so important that I determined to carry out a
plan which had long been in mind. This was to ascertain the
exact effect of different types of climate by means of precise
measurements. Dexter, in his book on Weather Influences,
had made a beginning. Lehmann and Pedersen had made a small
series of measurements whose highly suggestive results have been
published under the title Das Wetter und imsere Arbeit. A
few physicians and students of child psychology were also at
work, and their results have been summed up in such publica-

14 CIVILIZATION AND CLIMATE

tions as Hellpach's GeopsycJiische Erscheinungen and Berliner's
Einfluss von Klima, Wetter und Jahreszeit auf das N erven- und
Seelen-leben. Nevertheless, there existed no large series of meas-
urements of the actual efficiency of ordinary people under dif-
ferent conditions of climate.

The ideal way to determine the effect of climate would be to
take a given group of people and measure their activity daily
for a long period, first in one climate, and then in another. This,
however, would not be practicable because of the great expense,
and still more because the results would be open to question. If
people were thus moved from place to place, it would be almost
impossible to be sure that all conditions except climate remained
uniform. If the climate differed markedly in the two places, the
houses, food, and clothing would also have to be different. Social
conditions would change. New interests would stimulate some
people and depress others. Hence no such experiment now seems
practicable.

The most available method is apparently to take a group of
people who live in a variable climate, and test them at all seasons.
The best test is a man's daily work, the thing to which he devotes
most of his time and energy. Accordingly, I took the records of
four groups of people ; namely, some five hundred factory opera-
tives in three Connecticut cities, New Haven, New Britain, and
Bridgeport; nearly nine thousand operatives at Pittsburgh;
three or four thousand operatives in southern cities from Vir-
ginia to Florida; and over seventeen hundred students at the
United States Naval Academy at Annapolis, and the Military
Academy at West Point. In most cases each person's record
covered an entire year, or at least the academic year. All the
records were compared with the weather, as explained in later
chapters. The results were surprising. Changes in the barometer
seemed to have little effect. Humidity apparently possesses con-
siderable importance, but the most important element is clearly
temperature. The people here considered were phvsicallv most

CIVILIZATION AND CLIMATE 15

active when the average temperature ranged from 60 to 65 F.,
that is, when the noon temperature rose to 70 or more, while the
night temperature fell to 55 or so. This is higher than many of
us would expect. Mental activity, however, reached a maximum
when the outside temperature averaged about 38 F., that is,
when there were mild frosts at night. Another highly important
climatic condition seemed to be the change of temperature from
one day to the next. People did not work well when the tempera-
ture remained constant. Great changes were also unfavorable.
The ideal condition, or optimum, seemed to be mild winters with
some frosts, mild summers with the temperature rarely above
75 F., and a constant succession of mild storms and moderate
changes of weather from day to day.

The facts just stated seem to be of great significance as will
be fully explained in this book. They suggest that the weather
exerts a rather large and easily measurable effect upon man's
capacity for both physical and mental work. This conclusion
naturally led to a query as to how the climates in different parts
of the world vary in their effect on human efficiency. Accord-
ingly, I constructed a map showing how human energy would
be distributed throughout the world if all the earth's inhabitants
were influenced as were the fifteen thousand people of the four
American groups mentioned above. This map was found to agree
to a remarkable extent with a map of the present distribution
of civilization based on the opinions of about fifty geographers
and other widely informed men in a dozen countries of America,
Europe, and Asia. Moreover, it agreed with the conclusions
previously drawn as to the relation of climatic changes to the
civilization of the past. Take, for example, the decadent coun-
tries as to whose past climate we have some definite idea. In
practically every case the climate during their more flourishing
periods appears to have approached the optimum, as determined
in the United States, more nearly than at present. This does
not mean that the climate of Egypt, North India, China,

16 CIVILIZATION AND CLIMATE

Greece, or the Maya regions in Guatemala was ever like that of
either New York or California. It merely means that it ap-
proached more closely than at present to one or the other of these
American types. Hence at the time of its greatness each region
apparently enjoyed more than its present advantages in eco-
nomic conditions, in freedom from parasitic diseases, and in
direct climatic stimulus.

When my investigations had reached this point, the first edi-
tion of Civilization and Climate was written. During the nine or
ten years that have since elapsed not only has much new evidence
come to light, but my own point of view has changed consider-
ably. The changes are set forth in the series of books and articles
listed in the preface to the present (third) edition. They are
referred to at some length in later chapters, but may here be
briefly summarized.

The comments on Civilization and Climate by historians and
others make it more and more evident that the crux of the hy-
pothesis of this book lies in changes of climate. The question,
however, is not whether the climate of ancient Egypt, for ex-
ample, was like that of modern England. It certainly was not
and never could be. The contrast of the two countries in latitude,
topography, and relation to land and sea makes any such close
resemblance impossible. The real question is this : When Egypt
rose to its greatest heights did its climate approach appreciably
nearer than now to the type which provides the optimum condi-
tions of energy, health, and economic strength for a people in
the early Egyptian stage of development? Bear in mind that
when the ancient Egyptians first rose to a state approaching
civilization, they had not yet learned to use iron tools. Even in
later days they had nothing like our modern skill in using wood
and coal for heating houses, in manufacturing cotton and woolen
cloth on such a scale that even the poor can be warmly clothed,
in building houses that are proof against wind, rain, and cold ;
nor had they our skill in combating disease. Hence their stage

CIVILIZATION AND CLIMATE 17

of development caused the optimum climate to be warmer for
them than for us. We are able to guard ourselves against low
temperature and exposure, and thus gain an important stimulus
without suffering much harm. They could not withstand cold
winters.

Bearing in mind then, that the optimum climate varies accord-
ing to a nation's stage of civilization and also that there is
doubtless some difference in the optimum from race to race, our
problem becomes to determine how far the climate of the past in
any given region was like that which is best for the stage of
human progress, and perhaps the race, with which we happen
to be dealing. This point of view is slightly different from that
of the first edition of this book. The change is due largely to
GilFillan's article on The Coldward Course of Progress in the
Political Science Quarterly, 1920.

So far as changes of climate are concerned, the ten years since
this book was written have seen considerable new evidence as to
the reality and nature of historic as well as prehistoric pulsa-
tions. As an entire new chapter will be devoted to this matter,
and as the comment of Brooks on the convincing character of
the evidence has already been quoted, it is enough to point out
here an interesting fact as to the kind of people who have ac-
cepted the conclusions of this book. The non-scientific public,
which has doubtless been the widest audience of Civilization and
Climate, has accepted the book on the reasonableness of its main
hypothesis, and with an open mind as to future proof or dis-
proof. Geologists, archaeologists, and those geographers who
have had a geological training are the types of scientists who
have found the hypothesis most convincing. This is because
most of the methods of reasoning and lines of evidence employed
in discussions of climatic changes are such as are commonly used
in geology, archaeology, and geography. The evidence consists
chiefly of ancient lakes and streams, old roads in deserts, dead
vegetation and its rings of annual growth, abandoned fields and

18 CIVILIZATION AND CLIMATE

irrigation systems, and especially ruins and other traces of man,
which are really human fossils. Such evidence appeals to geolo-
gists, archaeologists, and geologically trained geographers.

Anthropologists, economists, and historians, on the other
hand, have been slow to believe that climatic changes have had
much influence upon human history. They accept, indeed, the
geologists' conclusion that previous to recorded history great
climatic changes drove man this way and that, destroyed ancient
types of culture, and either wiped whole races out of existence,
or profoundly modified them, physically, mentally, and socially.
They apparently have no difficulty in accepting the geological
evidence that among primitive men, as among plants and
animals, climate has been one of the most powerful factors in
determining the distribution and vigor of different types. But
when it comes to the period of written history many historians,
and some anthropologists and economists, no longer trust the
geological methods of reasoning. Their opinions are more or
less unconsciously molded by two widely accepted assumptions.

The first assumption is that climatic uniformity is a normal
condition. This idea seems wholly untenable in view of the con-
stantly growing evidence of numerous and important glacial
periods and other extreme types of climate at all stages in geo-
logical history. The more we know of the geological record, the
more clear it becomes that change, not uniformity, is the rule.
Even in the long periods when the larger types of climatic
changes have been absent, there is abundant evidence of minor
fluctuations and pulsations. The second assumption is equally
doubtful. It holds that written records and statistics are more
reliable than the geological type of evidence. Of course written
records and statistics are far more reliable than any other types
of evidence if they are sufficiently full, if they can be trusted,
and if they are prepared by people who are conscious of the
purposes for which they are to be used. But the written and
recorded evidence as to the climate of the past consists of mere

CIVILIZATION AND CLIMATE 19

scraps of information set down in most cases accidentally and
with no idea of their possible significance in the distant future.
Such evidence has, of course, great value, and must be studied
assiduously. Nevertheless, it is inevitably subordinate to the
geological type of evidence, which may be either physiographic,
botanical, or archaeological.

It seems clear, then, that the ultimate decision as to whether
climatic changes have taken place on a large scale during his-
torical times does not rest with historians. It rests primarily
with persons who are trained in climatological, and especially
geological, methods. During the last ten years geographers as
a whole, in spite of some exceptions, seem to have become per-
suaded that the historical period has witnessed a series of cli-
matic pulsations like those of the prehistoric or post-glacial
period although on a smaller scale. According to their own
written statement in answer to a questionnaire, over nine tenths
of the geographers of America, if we may judge from the fifty
who have expressed their opinions, hold this view, although they
are not quite so fully agreed as to how great the effect of these
pulsations upon man may have been. Of course it is possible
that a few geographers failed to answer the questionnaire be-
cause they did not wish to express an opinion contrary to mine.
I think, however, that the number of such is very limited, for
most of those who are known to hold views unlike my own ex-
pressed them freely. Even when all due allowance is made for
failures to answer, it seems clear that among the people who
are best competent to weigh the evidence the great majority
believe in pulsations of climate. If these geographers with a
geological training are right, there seems to be no escape from
the conclusion that during certain periods of ancient history the
climate of places like Egypt, Mesopotamia, and North India
approached the optimum more closely than at present. Since the
lower stage of culture in those early days presumably caused
the optimum temperature for human progress to be higher than

20 CIVILIZATION AND CLIMATE

is now the case among the most advanced races, the climatic
conditions were even more favorable than I realized when this
book was first written. If this conclusion is well grounded, it
becomes a basic fact which the historian must take into account
just as every careful student of early man now continually takes
account of the fact that primitive man was greatly influenced
by the vicissitudes of the successive glacial epochs, and just as
every economist recognizes that modern man, on a smaller scale,
is profoundly influenced by good or bad crops. History can
never be written correctly until its physical basis is thoroughly
understood, and until it is recognized that economic conditions,
and human health and energy vary from age to age almost as
much as do the conditions of politics, religion, and personality.
Another line where there has been much progress in the last
ten years is the determination of the nature and importance of
the climatic optimum for man's physical development and health
as opposed to his economic development. After Civilization and
Climate had been completed, I undertook a study of the rela-
tion of deaths to climate. The results appeared in World Power
and Evolution. Some eight and a half million deaths in about
sixty cities in the United States and thirty in France and Italy
were analyzed according to the average weather of each month
during periods which in most cases amounted to at least ten
years. Except for about 400,000 in the United States, all the
deaths occurred during the normal period immediately preced-
ing the Great War. In addition to this, about 50,000,000 deaths
in other countries were analyzed less intensively, but with es-
sentially the same results. On the other hand, about 700,000
deaths in New York from 1877 to 1888 have been very minutely
analyzed according to the day of death. For the six years from
1883 to 1888 the number of deaths each day has been compared
with the weather day by day during the two weeks ending with
the day of death. In another investigation 7200 deaths from
lobar pneumonia in New York in 1913 were compared with the

CIVILIZATION AND CLIMATE 21

weather for each day. Again, in the two largest hospitals in
Boston the relation of the weather to 2300 deaths succeeding
operations was looked into. At a later date the death rate during
the influenza epidemic of 1918 in the United States was analyzed
by still a different method in order to determine whether the
weather had any effect in altering the rate from city to city.

The net results of these investigations, as shown in World
Power and Evolution and in various technical articles agree
closely with those of the investigation of factory workers and
students. They confirm the work of other investigators in show-
ing almost beyond question that there is a distinct optimum
condition of climate for man just as for plants and animals.
This optimum varies relatively little from one set of people to
another or from place to place. Even for negroes the departure
from the white standard is by no means so large as would be
expected, though it is unmistakable. Any departure from the
optimum for a given race or individual seems to render people
not only less efficient, but also more susceptible to disease and
hence an easier prey to bacteria and other parasites. Moreover,
as appears in World Power and Evolution, there is some evidence
that departures from the optimum climate render people less
buoyant, less capable of prolonged and steady mental activity,
and correspondingly less likely to make progress. The signifi-
cant fact about the whole matter is that, so far as I am aware,
in every case where large bodies of people have been carefully
analyzed, the same major responses to climate become evident,
even though there may be differences in details. Thus the prog-
ress of the last ten years seems to add appreciably to the general
certainty as to the nature and importance of climatic optima,
and as to the effect of departures therefrom upon health, effi-
ciency, and progress.

The difference between the present and first editions of this
book in respect to natural selection and racial inheritance is
especially important. It may be illustrated by ancient Greece.

22 CIVILIZATION AND CLIMATE

In the first edition I supposed that the reason for the peculiar
ability of Greece was a mystery which there was no immediate
prospect of solving. It was the result of some unexplained bio-
logical process. I believed that no matter where those particular
people might have gone or at what period they had happened
to live they would have achieved much more than any ordinary
people. As events actually shaped themselves, the ancient Greeks
migrated to a land near more ancient centers of civilization
whence they could receive the inspiration of the greatest pre-
ceding cultures. By reason of the numerous gulfs and harbors of
their land the Greeks were easily able to get what they wanted
from other countries, provided they had energy enough to sail
abroad for material riches, and capacity enough to absorb the
mental riches with which they came in contact. So far as climate
is concerned, Greece appears to have enjoyed unusually favor-
able conditions throughout most of the period from perhaps
1000 to 300 B.C., and especially about 400 B.C. Previous to 1100
B.C., however, there seems to have been an unfavorable period
culminating perhaps 1300 to 1200 years before Christ, while
at a later date a period of rapid climatic degeneration from
800 to 200 B.C. was followed by highly unfavorable conditions
during the succeeding century. The favorable climate during the
period of the greatest Grecian development apparently ren-
dered the economic conditions distinctly more favorable than
those of today, and stimulated the Greeks to a high degree of
physical and mental energy. At the same time it rendered the
environment unfavorable not only to the anopheles mosquito
which causes malaria, but to other disease-bearing organisms.
Thus the Greeks with their high inheritance enjoyed an environ-
ment which gave full opportunity for the development of the
best that was in them. The fall of Greece, according to the hy-
pothesis set forth in this book ten years ago, was greatly in-
fluenced by a rapid deterioration of climate. In the third
centurv B.C., a decrease in rainfall caused a most serious diminu-

CIVILIZATION AND CLIMATE 23

tion in the capacity of Greece to support population. This in-
creased the political difficulties while at the same time the ability
of the people to cope with such difficulties was diminished by a
decline in the stimulating qualities of the climate. At the same
time the increase in marshes and in stagnant pools, because of
changed conditions of rainfall, made the environment favorable
for malarial mosquitoes, while poverty, a poor food supply, and
other adverse conditions also fostered disease. In general, the
poor economic and political situation, and the unfavorable con-
ditions of health tended to kill off the blonde Norse invaders
who seem to have been the leaders among the early Greeks.

At present my view of the rise and fall of Greece differs from
the one set forth above in only one respect, but that is highly
important. Instead of recognizing natural selection as an im-
portant cause merely of the decline of Greece, I now regard it
as one of the main causes of the rise of that country. The people
who made Greece great, as I have shown in detail in The Char-
acter of Races, were primarily the Dorians and especially the
lonians. These people appear to have originated far to the
north, perhaps in the great plains of Russia. They carne to
Greece in a series of migrations, the first of which may have been
that of the Achseans in the fourteenth century before Christ,
while a later invasion culminated in the Dorian invasion two or
three centuries later. The lonians appear to have been largely
Greeks of the old Achaean and Minoan upper classes, who were
led to migrate by the prolonged disturbances and fighting which
harassed Greece for generations and perhaps centuries after
the Dorian invasion. Attica had previously been almost depopu-
lated and did not attract the Dorians because it was infertile
and poverty-stricken by reason of its dryness. Later, however,
perhaps in part because of an amelioration of climate, it became
the refuge of large numbers of the old upper classes from the
rest of Greece. When its population became too dense, some of
the settlers, together with many of the more able and energetic

24 CIVILIZATION AND CLIMATE

people from other parts of Greece, went across the
Sea eastward to the Ionian coast of Asia Minor. Now we are
expressly told by Thucydides and others not only that the
settlers in Attica were a highly selected group from the leading
families of the rest of Greece, but that they strenuously kept
themselves apart from the lower classes. Even if an Athenian
married a woman of the lower classes his children followed the
social status of the mother and not of the father. Thus among
the citizens of Athens a selected inheritance was kept almost
unimpaired for many centuries.

But note another important fact. Not only was there a rigid
selection of good material when the lonians settled in Athens,
but there appears to have been a perhaps more rigorous selec-
tion during the earlier migrations of the people who finally be-
came Greeks. Every migration is more or less selective because
the weak, the feeble, the cowardly, and those lacking the spirit
of adventure, together with those who lack determination, are
gradually weeded out. The longer and more difficult the migra-
tion, the more strenuous is the selection, especially among women
and children. All but the most vigorous, both mentally and
physically, are weeded out with peculiar rapidity. Thus the net
result of practically every long or difficult migration in which
women as well as men take part is to pick out a relatively small
group of people of unusual capacity. The group may contain
only one out of twenty, or one out of one hundred or a thousand
of the original stock. Now the Athenians, as we have just seen,
went through this selection twice, and may have gone through
it at earlier times also. They also maintained the quality of the
original stock more or less completely by refraining from inter-
marriage with any but the elite of other regions and by various
practices such as infanticide, which eliminated weaklings. Such
selection with various modifications has been, I believe, one of
the most effective factors in producing competent races.

But what has this to do with climate? Much, because a large

CIVILIZATION AND CLIMATE 25

number of the migrations of history appear to have been more
or less directly started by climatic vicissitudes. For example,
the Irish migration to America during the nineteenth century
was due largely to the fact that a series of good potato harvests
in the early decades of the century permitted the population of
Ireland to increase enormously. Then came a series of unduly
rainy years accompanied by bad crops. Famine and distress en-
sued, and immediately an enormous migration to America set in.
In later years, as is shown by Bruckner's data which I have
cited in an article on A Neglected Factor in Race Development
(Journal of Race Development, October, 1915), the migrations
from Ireland and likewise from Germany to the United States
have varied in close harmony with the climatic conditions and
consequent agricultural prosperity of both the Old World and
the New. When the crops are bad in Europe but good in Amer-
ica, as is not uncommon, emigration from the Old World is
greatly stimulated for a few years. On the contrary a period of
bad crops in America coinciding with good crops in Europe
gradually diminishes the pressure which leads to migration.
Similar examples on a large scale have occurred time after time
in the history of China, Central Asia, and other regions.

Contrary to the common belief, most parts of the world
normally contain practically as large a population as they are
capable of holding under the social and economic conditions
which happen to prevail at any given time and place. The birth
rate among mankind, as Carr-Saunders convincingly shows in
The Population Problem, is so large and responds so quickly to
economic changes that only a few decades or generations are re-
quired for even a relatively depopulated country to fill up to
what Woodruff in his Expansion of Races has called the satura-
tion point. Thus the great majority of migrations can be under-
stood only by thinking of all parts of the world as normally
having nearly their full quota of population. Among primitive
hunting tribes, for example, this quota may mean only one

26 CIVILIZATION AND CLIMATE

person per square mile, while in a modern highly civilized com-
munity it may mean a thousand per square mile. The conditions
are like those of a bucket filled to the brim with water. As soon
as any object is dropped into the bucket some water runs over.
In the same way, in most parts of the world any deterioration of
economic conditions such as frequently arises from bad crops,
immediately necessitates either a lowering of the standard of
living or a diminution of population through an increased death
rate by reason of poverty, famine, or war, or through migration.
A migration, as a rule, is merely a slow drifting of people with
unusual energy and initiativefrom unfavorable to favorable
districts. But not infrequently, and especially after one of the
sudden climatic vicissitudes which have been common in history,
it becomes a violent stream of invasion such as many of the bar-
barian outpourings about 1400 to 1200 years before Christ
and again at various later periods.

The original Greeks apparently took part in extensive wan-
derings of both the mild and violent kind, and natural selection
presumably worked among them effectively. The net result was
a highly selected race which gave the world a marvelous group
of famous men. These men, it appears, accomplished great things,
not only because of their high innate qualities, but because grow-
ingly favorable climatic conditions from six to four hundred
years before Christ improved their economic situation and
helped to give them great energy and splendid health. According
to this view, Greece apparently lost her ability partly because
the fine original stock became mingled with weaker elements,
and partly because the change of climate which began soon after
400 B.C. and which took place most rapidly from 300 to 200
B.C., not only enabled malaria to spread with baleful rapidity,
but introduced a stage of diminishing resources, scanty food
supply, overpopulation, and lessened climatic stimulus. These
factors presumably combined with other well-known conditions
to produce poor health and perhaps restriction of families among

CIVILIZATION AND CLIMATE 27

the upper classes, thus killing off the old dominant stock of
northern origin. Hence the racial elements to which Greece owed
her greatness disappeared, and the country fell into intellectual
insignificance. Other factors undoubtedly cooperated, but such
conditions as political corruption, social degeneracy, and undue
personal ambition are probably results of racial decay and poor
health quite as much as causes. The essential point is that at
the beginning a time of climatic stress among nomads in rela-
tively cool grasslands seems to have led to migration and a
favorable type of selection. Another period of similar stress
amid a highly developed people in a relatively southern land
where there was less opportunity or incentive for migration led
to the dying off of the more able people. Yet even in this case
many of the most competent Greeks migrated, thus still farther
impoverishing the mother country and giving rise to highly
gifted colonies like the Alexandrian community in Egypt.

To sum up the whole hypothesis of the relation of climate to
civilization, here are the factors as I see them at present. Most
parts of the world are so well populated that any adverse eco-
nomic change tends to cause distress, disease, and a high death
rate ; migration ensues among the more energetic and adventur-
ous people. Perhaps the commonest cause of economic distress is
variations in weather or climate which lead to bad crops or to
dearth of grass and water for animals. Such economic distress
almost inevitably leads to political disturbances and this again
is a potent cause of migrations. The people who migrate perforce
expose themselves to hardships and their numbers diminish until
only a selected group of unusually high quality remains. Such
people, either as warlike invaders or in small bands, enter a new
country. They may find it well populated and merely impose
themselves as a new ruling class, as seems to have happened
several times in India, or they may find it depleted of people as
in Attica. When the period of climatic stress is ended and the
climate improves, the dominant newcomers not only possess an

28 CIVILIZATION AND CLIMATE

unusually strong inheritance, but are stimulated by unusually
good economic conditions and by improved conditions of health
and energy. Moreover since the population is apt to remain
below the saturation point so long as the climate improves, the
standards of living tend to rise and to become relatively high.
Thus many people are freed from the mere necessity of making
a living and have the opportunity to devote themselves to the
development of new ideas in literature, art, science, politics, and
other lines of progress. The repeated coincidence between periods
of improving climate and periods of cultural progress appears
to be due not only to the direct stimulus of climate, as I sup-
posed in the first edition of this book, but to that stimulus com-
bined with a high racial inheritance due to natural selection.
This, I am well aware, by no means offers a complete explanation
of history, for many other elements must also be considered. But
it helps to explain many historic events which have hitherto been
only partially understood.

Here, once more, is the sequence climatic changes produce
economic results through an increase or a diminution of the food
supply. Thus there arises either a temporary condition of under-
population with comparative political tranquillity and oppor-
tunities for the growth and expansion of civilization, or a con-
dition of overpopulation with consequent political turmoil,
war, migration, and the repression of civilization. This point
of view was dominant when I wrote The Pulse of Asia and
Palestine and Its Transformation. Climatic changes also appear
to have a direct effect in stimulating or repressing man's physi-
cal activity, a viewpoint which dominated the first edition of
the present book. It is obvious that through their effect upon
food, insects, bacteria, and man's own powers of resistance
climatic changes must exert a great influence upon disease.
Hence I was led to write World Power and Evolution from the
standpoint of disease and the death rate. But this does not end
the matter, for climate apparently exerts a direct selective

CIVILIZATION AND CLIMATE 29

effect in preserving certain types of people and destroying
others, and it certainly exerts these effects indirectly through
various conditions already mentioned. Therefore it was only
logical that The Character of Races should center around
natural selection, especially in its climatic aspects. The next
step is obviously a study of the relation of climate to mutations
and thus to the origin of the new types among which natural
selection makes its choice, but that is as yet impossible. Beyond
this lies the synthesis of the effects which climate produces
through economic and political conditions, through war and
migration, food and natural resources, energy and health, and
through natural selection and mutations. And all these results of
the climatic environment must be put into due relation not only
with the results of the other factors of physical environment,
but with the opposite side of the shield, that is, with the purely
human factors such as institutions, customs, ideas, and all man's
passions, ideals, and aspirations. Then it will be possible to form
a true philosophy of history. Meanwhile the present edition of
Civilization and Climate tries to take a broader and deeper view
of human progress than its predecessor, but it makes no claim
to deal exhaustively with more than one small phase of the
matter, namely, the direct effects of climate upon human ^health
and energy.

CHAPTER 11
RACE OR PLACE

THE problem which confronts us is primarily to separate the
direct effects of climate from those of inheritance, regard-
less of whether the inheritance has been influenced by the climate
of the past. It may be made concrete by comparing two sharply
contrasted races, Teutons and negroes. Suppose that there were
two uninhabited Egypts, exactly alike, and that one could be
filled with negroes and the other with Teutons. Suppose that
these settlers were average members of their races, and were
equipped with the same religion, education, government, social
institutions, and inventions. This might easily happen if the
negroes came from the United States. Suppose, further, that
neither race received new settlers from without, or lost any
except through natural selection. Which would succeed best?
"The Teutons, of course," is the answer. "What a foolish ques-
tion." But is it so foolish? You are thinking of the first genera-
tions. I am thinking of the twentieth or later. Does anyone know
what five hundred or a thousand years of life in Egypt would
do for either Teutons or negroes if no new blood were intro-
duced ?

At the end of that time the two sets of people would assuredly
be different, for the effect of a diverse inheritance would last
indefinitely. The advantage in this respect would presumably be
on the side of the Teutons. I wish to emphasize this matter, for I
shall have much to say about the effect of climate, and I want
to make it perfectly clear that I do not underrate the impor-
tance of race. Although the matter is by no means settled, many

RACE OR PLACE 31

authorities think that the brain of the white man is more com-
plex than that of his black brother. Strong, in the Pedagogical
Seminary for 1913, and Morse, in the Popular Science Monthly
for 1914, have shown that in Columbia, South Carolina, the
white children are mentally more advanced than the colored. By
applying the Binet tests to 225 children in two white schools and
to 125 children in a colored school, they obtained the following
table, showing the amount by which the two races exceeded or
fell short of what would be expected.

COLORED WHITE

More than one year backward, 29.4% 10.2%

Satisfactory, 69.8% 84.4%

More than one year advanced, 0.8% 5.3%

Among the white children those from the middle classes made
a better showing than those of factory operatives, but both
were ahead of the colored. So far as home environment is con-
cerned, the factory children have almost no advantage over the
colored children. A slight advantage may possibly arise from
the fact that when the Binet tests were originally devised, they
were designed to measure the capacities of white children. The
negro race may have capacities which the white does not possess
and which do not play a part in the tests. In appreciation of
humor, for example, and in equability of temperament there
can be little question that the black rnan surpasses the white.
These things, however, can scarcely account for the fact that
29.4 per cent of the colored children showed a mental develop-
ment more than a year behind that which would be expected from
their age, while only 10.2 per cent of the white children were
equally backward.

So far as I am aware, every exact test which has been made
on a large scale indicates mental superiority on the part of the
white race, even when the two races have equal opportunities.
For example, in Washington the colored children remain in

32 CIVILIZATION AND CLIMATE

school quite as long as the white, but they do not accomplish so
much in the way of study and do not reach so high a grade. In
the cities of the South, Mayo and Loram find that where the
races are given essentially the same instruction, the proportion
of whites who are promoted is greater than that of negroes.
Moreover, the difference seems to increase with years, which
suggests that the average colored child not only stands below
the average white child in mental development at all ages, but
ceases to develop at an earlier age. In the high schools of New
York, the superiority of the white race is shown by Mayo's
examination of the average marks. By the time the children
reach the high school, the processes of promotion have weeded
out a much larger proportion of colored children than of white.
Hence, the negroes form a specially selected group whose supe-
riority to the average of their race is more marked than the
superiority of the white high school children when compared
with the rest of the white race. Nevertheless, the average marks
of the white children are distinctly higher than those of the
colored.

In order to test the capacity of the two races in a wholly
different way, I have made a comparison of white and colored
workmen employed under precisely similar conditions. The first
case was a cigar factory at Jacksonville, Florida. The employees
were practically all Cubans. Both the whites and the blacks have
very little education, and their home environment in Cuba differs
to only the smallest extent. They earn good wages, but are often
out of work, and are generally shiftless and unreliable. There is,
of course, no color line in Cuba, and the same is true in the cigar
factories. Black men and white work side by side at the same
tables. In such a factory, if the black man is as capable as the
white he has exactly as good a chance, for he is paid by the piece,
and his earnings depend entirely on himself. What, then, do we
find? Taking all the operatives, we have 39 white and 65 negroes.
Their average earnings, as measured by the wages of two weeks,

RACE OR PLACE 33

are in the ratio of 100 for the whites to only 51 for the negroes.
To make the comparison more favorable to the negroes, let us
eliminate those who roll low-grade cigars where little skill is
required and the pay is low. We then have 39 white men and 44
negroes. They are doing exactly the same work under exactly
the same conditions, but the whites earn a dollar where the
negroes earn 75 cents. At a similar factory at Tampa, Florida,
17 colored men were at work and 303 white. In this case prac-
tically all of the few negroes happened to be men of long experi-
ence, while many of the whites were comparatively new. Never-
theless, the whites are still on a par with the colored men, the
ratio being 100 to 99.8.

One of the best places for comparing the two races is the
Bahama Islands. For reasons which I shall present later, the
process of making "poor whites" has probably gone farther in
the Bahamas than in almost any other Anglo-Saxon community.
Part of the white people are like their race in other regions, but
a large portion have unmistakably degenerated. Witness their
intense and bigoted speech, their sunken cheeks and eyes, their
sallow complexion, and their inert way of working. In spite of
racial prejudice, there is no real color line in the Bahamas.
Persons with more or less negro blood are worthy occupants of
the highest positions, and are universally accepted in the most
exclusive social circles. The British government gives the negro
every possible opportunity. The state of affairs may be judged
from the remarks of a "poor white" sailor, who said to me : "You
want to know why I likes the southern states better than the
North. It's because they hates a nigger and I hates him, too.
What kind of a place is this where they do everything for the
nigger and nothing for the white man? It's bad enough to have
to go to jail, but it's damned hard for a white man to be taken
there by a nigger constable." In one Bahaman village I saw
negro girls teaching white children in the public schools. In that
same village a number of the leading white men cannot read or

34 CIVILIZATION AND CLIMATE

write. When they were children their parents would not send
them to school with negroes. The despised negroes learned to
read and write, but have now largely forgotten those accomplish-
ments. The proud whites grew up in abject ignorance. Today
the same thing is going on. I visited two villages where the white
children are staying away from school because they will not go
to negro teachers. The homes of such whites are scarcely better
than those of their colored neighbors, and their fathers are
called "Jim" and "Jack" by the black men with whom they work.
Racial prejudice apparently works more harm to the whites
than the blacks. So far as occupations go there is no difference,
for all alike till the soil, sail boats, and gather sponges.

When the lumber industry was introduced into the islands,
whites and blacks were equally ignorant of the various kinds of
work involved in cutting trees and converting them into lumber.
The managers did not care who did the work so long as it was
done. They wanted three things : strength, docility or faithful-
ness, and brains. They soon found that in the first two the
negroes were superior. Time and again persons in authority,
chiefly Americans, but also some of the more capable native
whites, told me that if they wanted a crew of men to load a boat
or some such thing, they would prefer negroes every time. The
poor white shirks more than the colored man, he is not so strong,
and he is proud and touchy. Other things being equal, the negro
receives the preference. But other things are not equal. The very
men who praise the negroes generally added : "But you can't use
a negro for everything. They can't seem to learn some things,
and they don't know how to boss a job." The payroll reflects
this. Even though the negroes receive the preference, the 4*00
who are employed earn on an average only about 60 per cent as
much as the 57 white men. If we take only the 57 most competent
negroes, their average daily wages are still only 88 per cent as
great as those of the native whites. The difference is purely a
matter of brains. Although the white man may be ignorant and

RACE OR PLACE 35

inefficient, with no more training than the negro, and although
his father and grandfather were scarcely better, he possesses an
inheritance of mental quickness and initiative which comes into
evidence at the first opportunity.

All these considerations seem to point to an ineradicable racial
difference in mentality. As the plum differs from the apple not
only in outward form and color, but in inward flavor, so the
negro seems to differ from the white man not only in feature and
complexion, but in the workings of the mind. No amount of
training can eradicate the difference. Cultivation may give us
superb plums, but they will never take the place of apples. We
have tried to convert the black man into an inferior white man,
but it cannot be done. Initiative, inventiveness, versatility, and
the power of leadership are the qualities which give flavor to the
Teutonic race. Good humor, patience, loyalty, and the power
of self-sacrifice give flavor to the negro. With proper training he
can accomplish wonders. No one can go to a place like Hampton
Institute without feeling that there is almost no limit to what
may be achieved by cultivation. In an orderly, quiet way, those
negro boys and girls go about their daily tasks and give one the
feeling that they are making a real contribution to the world's
welfare. To be sure, they work slowly, they are not brilliant in
their classes, they rarely have new ideas in their manual work,
but yet they are faithful. The willing, happy spirit of their work
is something that we nervous, worried white people need sorely
to learn. Once in a long time there comes a leader, a man to
whom both white and black look up, but such leadership is
scarcely the genius of the race. Yet leadership is what the black
man must have. At such places as Hampton he gets it, and one
realizes that the white man's initiative joined to the Christian
spirit which is there so dominant can give a training which
overcomes much of the handicap of race.

Having turned aside to pay tribute to the potency of race,
education, and religion in determining the status of civilization,

36 CIVILIZATION AND CLIMATE

let us come back to physical environment. What part does this
play? Is it so important that a strong race in an unfavorable
climate is likely to make no better showing than a weak race
in a favorable climate? How far can a bad climate undo the effect
of a good training?

In answer to these questions, we may well compare the Teu-
tonic and negro races when each is removed from the climate in
which it originally developed. Before proceeding to this a word
should be added to forestall any possible misunderstanding of
rny attitude toward the southern parts of the United States
and toward other progressive regions which, nevertheless, suffer
somewhat from climatic handicaps. In searching for the truth
I shall be forced to say some things which may not be wholly
pleasing to residents of such regions. It must be clearly under-
stood, however, that these are not stated on my own authority.
All are based either on the consensus of opinion among a large
number of persons including many southerners, or upon the
exact figures of the United States census or other equally re-
liable sources. My part has been simply to interpret them. Be-
lieving that the South contains a great number of people who in
all essential respects have an inheritance equal to that of the
best northern stocks, I have tried to find out why the southern
part of the United States has prospered less than the northern.
This does not mean that I reject the old ideas as to the cause,
but simply that I emphasize another which has not received
sufficient consideration. It does not discredit the South nor its
people. It does not alter the fact that southerners possess a
courtesy and thoughtfulness which we of the worried and hurried
North need greatly to imitate. Nor does it mean that men of
genius are not as likely to be born in one section as another.
Instead of this, it merely indicates that in addition to the many
efforts now being made to foster progress in the South by other
means, we should add a most vigorous attempt to discover ways
of overcoming the handicap of climate. This book is written with

RACE OR PLACE 37

the profound hope that the truth which it endeavors to discover
may especially help those parts of the world whose climate, al-
though favorable, does not afford the high degree of stimulation
which in certain other restricted areas is so helpful.

Let us first undertake a study of what the census shows as to
negroes and whites in different parts of the United States. The
only people whom we can compare with accuracy are the farmers,
for they are the only ones for whom exact statistics are avail-
able. Fortunately they are the part of the community where
social prejudices and other hampering conditions have the
smallest influence. The prosperity of the farmer, more than that
of almost any other class of society, depends upon his own indi-
vidual effort. If he is industrious, he need never fear that he and
his family will not have a roof over their heads and something
to eat. Even when the crops are bad, he rarely is in danger of
suffering as factory operatives often suffer, at least not in the
eastern United States, with which alone we are now concerned.
Moreover, the prejudice against colored people has little effect
upon farmers. No one hesitates to buy vegetables peddled by a
darkey farmer. Finally, farming is the occupation in which the
South has been least hampered as compared with the North. For
over half a century the negro has been able to buy land freely in
any part of the country. The southerners, whether white or
black, have suffered economically because of slavery and the con-
sequent war, but they have a good soil and a climate far better
for agriculture than that of the North, and they have peculiarly
good opportunities to raise tobacco and cotton, two of the great-
est money-making crops in the world. Taken all in all, the
farmers of the country ought to show the relative capacities of
different races and of the same race under different conditions
better than almost any other class of people.

In 1904 the United States Census Bureau published a bulle-
tin on the negro. From that I have prepared Table 1 showing
the relative conditions in four groups of states in 1900. The first

s

E

&

S3 B

EC 4) o

* 2 |

* I

w r tj

w o

eg

Q

- B

S I

a i

x x

X wi

Q o
X X i

"a ^
I

!f s

a si

J3.3

o .-3

2 ^C

I .S *

t yi Op >

^-s
<y PH

> -< to .o c: ,

-H o CO >p CO >

^8

3 . .^ o> e~oi

^.co ^S W&fc

-"

AQ CTs O
#- T* -*

CO ^-r-1 ^O^-^ ^-O ^ , - ^.
i O *O O *O O O> O *O O O Q i < (.

.f e> - o o c& o & o _r o ^ <
^^^^^^ M -S^ 1 '

O 5C O

3 fl -.3"
s|l||

i^o ii

^ ^^ A a

*H 0) bO^ "^

O bD as C 2

^ ^ p 2

Total numb
Average acr
Average a
per farm,
land whic

RACE OR PLACE 39

row of numbers (line 1) shows the total number of white and
colored farmers. The second row shows that the farms of the
northern white men average about 100 acres in size, while those
of the southern white men are larger. The colored farms, on the
other hand, have an average size of about 50 acres. In the next
row of figures, line 3, we notice that the northerners forge ahead.
Even in the relatively hilly states of New York and Pennsyl-
vania, the white farmers have improved 63.5 acres per farm, or
69 per cent of the whole, leaving only 31 per cent in the rough
state of bushes or woods. The northern negroes do exactly as
well in proportion to their holdings, for they have cleared 33.0
acres, which is also 69 per cent of the average farm. In the
Carolinas, Georgia, and Florida, on the contrary, the white men
have improved only 34 per cent of their land, and the colored
men 58 per cent. For the states farther west (comparison B),
approximately similar conditions prevail. The negroes are
obliged to clear a larger percentage than the whites because
their small holdings would not otherwise furnish a living. The
significance of the figures lies in the fact that the northerners,
whether white or black, show more energy in improving their
land than do the southerners of the same kind.

Since this table was in print the corresponding data for 1910
and 1920 have appeared. Unfortunately they are less full than
for 1900 and do not include the value of products, line 9. For
line 4 the percentages for each of the three census years are as
follows :

TABLE 2.

Comparison A

Comparison B

N.W.

N.N.

S.W.

S.N.

N.W.

N.N.

S.W.

S.N.

1900

100

59

28

11

100

49

32

13

1910

100

81

41

24

100

48

25

14

1920

100

74

64

20

100

41

31

16

In comparison A, the gain of the other groups in relation to
the northern white farmers is noticeable. This, however, does

40 CIVILIZATION AND CLIMATE

not mean merely that agriculture is improving in the South, but
also that it is declining in the Middle Atlantic States. In B, the
percentages are almost unchanged.

In both comparisons of Table 1 items 4 and 9 are the most
significant. They show the value of the farms and the value of
the annual products. In each item of Table 1 the values are stated
in dollars as given in the census report, while underneath I have
added percentages. In computing the percentages the highest
value is reckoned as 100 per cent and the rest figured accord-
ingly. In each item of both comparisons for all three census
years, as given in Table 2, the northern whites stand at the top.
In general, taking both comparisons into account, the northern
white man's farm is worth twice as much as that of his colored
neighbor, and he gets twice as much from it. The southern white
man has a farm worth less than that of the northern negro, but
he gets from it approximately the same amount of products. The
southern negro's farm is worth less than half as much as the
southern white man's and he gets from it about two thirds as
much. Taking all the farms from our four groups of states
and reckoning them according to the value of what they actually
produced in 1900 and of their value in 1920, the census ranks
them as shown in Table 3.

TABLE 3. THE RELATIVE EFFICIENCY OF WHITE AND NEGRO
FARMERS IN THE NORTH AND SOUTH

Per value of Per value of

products in 1900 farms in 1920

Northern whites, 100 100

Southern whites, 51 44

Northern negroes, 49 49

Southern negroes, 34 18

This little table possesses profound significance. It shows un-
mistakably two types of contrast. First, there is the racial con-

RACE OR PLACE 41

trast, the result of long inheritance. That, apparently, is what
makes the negroes fall below the whites in both the North and
the South. There is also a climatic contrast. That, apparently,
is why the negroes who come to the North rise above the usual
level of their race, while the whites of the South fall below the
level of theirs. I realize that the contrast between the two sec-
tions is explained in a hundred ways by as many different people.
One ascribes it to the fact that slavery was a poor system eco-
nomically. Another says that the South is cursed for having
consented to the sin of slavery. Again, we are told that the pre-
dicament of the South is due to the War of Secession, the failure
to develop manufactures, the absence of roads and railroads,
bad methods of farming, the presence of the negro making the
white man despise labor, and many other equally important
causes which cannot here be named. Still other authorities
ascribe the condition of the South to its supposed settlement by
adventurers, whereas the North had its Pilgrims.

I would not minimize the importance of these factors. All are
of real significance, and if any had been different, the South
would not be quite what it is. All depend upon the two funda-
mental conditions of race or inheritance, and place or climate.
Yet in the contrast between the North and South, the climatic
effects seem to be the more potent. Slavery failed to flourish in
the North not because of any moral objection to it, for the most
godly Puritans held slaves, but because the climate made it un-
profitable. In a climate where the white man was tremendously
energetic and where a living could be procured only by hard and
unremitting work, it did not pay to keep slaves, for the labor of
such incompetent people scarcely sufficed to provide even them-
selves with a living, and left little profit for their masters. In the
South slavery was profitable because even the work of an ineffi-
cient negro more than sufficed to produce enough to support him.
Moreover, the white man was not energetic, and his manual work
was not of much more value than that of a negro. Hence, it was

42 CIVILIZATION AND CLIMATE

easy to fall into the habit of using his superior brain, and letting
the black man perform the physical labor. If the Puritans had
settled in Georgia, it is probable that they would have become
proud slave-holders, despising manual work.

So far as inheritance is concerned, the white southerners,
according to the generally accepted principles of biology, must
be essentially as well off as the white men of the North. New
England has probably had a certain advantage from the strong
fiber of her early settlers, but that section is excluded from our
comparison because it has so few colored farmers. In New York,
Pennsylvania, New Jersey, and the states farther west, the white
farmers in 1900 were of highly mixed origin, and there is little
reason to think that they inherit any greater capacity than do
the white men of the South. Hence, we infer that the difference
shown by the census is largely a matter of climate. It has arisen
partly by indirect means such as slavery and disease, partly by
direct means such as the disinclination to physical exertion. This
demands emphasis, for we are told that the South needs nothing
but a fair opportunity, plenty of capital, and abundant roads,
railroads, and factories, or else it needs only education, a new
respect of one race for the other, cooperation between the two
for the sake of the common good, and a deeper application of
the principles of Christ. All these things are sadly needed, but
it is doubtful whether they can work their full effect unless sup-
plemented by a new knowledge of how to neutralize the climatic
influences which seem to underlie so many southern problems. In
the climate of the South a part of the white population becomes
a prey to malaria, the hookworm, and other debilitating ail-
ments. People cease to be careful about food and sanitation.
Even those who are in good health do not feel the eager zest for
work which is so notable in the parts of the world where the
climatic stimulus is at a maximum. Thus one thing joins with
another to cause a part of the people to fall far below the level
of their race, and to become "Poor Whites," or "Crackers."

RACE OR PLACE 43

These increase in number as one passes from a more to a less
favorable climate. It is their run-down, unkempt farms which
bring the average of the southern whites so dangerously near the
level of the negroes. The best farms of the South vie with those
of the North. They show what could be done if all the inhabitants
could be instilled with the energy and wisdom of the best.

Aside from North America the only large area where Teutons
and negroes come into direct contact as permanent inhabitants
is South Africa. There they meet on practically equal terms. The
English and Boers began to settle in South Africa in large num-
bers only in the first half of the nineteenth century. In 1921 the
South African Republic contained about 1,500,000 Europeans,
4,700,000 Bantu natives, and 700,000 other natives and
Asiatics. A large proportion of the white men were not born
there, and hence the new conditions have not had time to pro-
duce their full effect. The majority of the natives are Zulus, but
the most capable appear to be the Basutos, an allied race who
have preserved a large measure of independence in the Draken-
berg mountains. Both the Zulus and the Basutos came from
the North a few generations ago. Some preceded the white man
and some have come since his arrival.

The colored people are most numerous in the north and east
of the Republic, that is, in Rhodesia and Natal. The white men
are most abundant in the south and in the central plateau, that
is, in Cape Colony, Orange River Colony, and Transvaal. With
ever increasing force, however, the blacks are pushing into the
white man's country. They are brought as laborers for the
mines ; they are wanted for the farms ; they are in demand as
servants ; and they are themselves taking up farms and success-
fully cultivating them. They are doing more than this, however,
for they are actually ousting the Europeans. In 1902 the Eng-
lish and the Boers finished a bitter war. Ten years later their
enmity had almost vanished in the common fear of the negro.
Aside from the disturbances due to the European War of 1914,

*4 CIVILIZATION AND CLIMATE

the great political question has long been the black man. One
party advocates segregation, with a white man's South Africa
in the highlands from Transvaal southward, and a black man's
South Africa in Natal and Rhodesia. No black man, they say,
should be allowed to live permanently outside his own country,
although he might go elsewhere to work temporarily. The other
party holds that such measures are too radical, but it also
recognizes the gravity of the situation.

The problem presents itself under an economic guise. The
colored men have a lower standard of living than the whites.
Hence they work more cheaply. They furnish so abundant a
supply of labor that white laborers have no chance. Thus a
large number of the Europeans even a tenth according to
ardent believers in the future of South Africa are "poor
whites." They are a shiftless set, living from hand to mouth,
proud of their race, yet less efficient than the blacks. The prob-
lem of preventing them from becoming an immediate charge
upon the community is serious. They lack the push and energy
which characterize the rest of the white population. According
to Stevens, in his book White and Black, 5 per cent of the
white population in certain regions have fallen so low that they
would rather resort to crime than work in competition with the
black man. These figures have been questioned, but they are
abundantly confirmed by Dr. Andrew Balfour, Director-in-
Chief of the Wellcome Bureau of Scientific Research in London.
In some lectures on Sojourners in the Tropics and Problems of
Acclimatisation, published in The Lancet in 1923, he states that
he "referred the matter to Colonel P. G. Stock, of the Ministry
of Health, who knows South Africa intimately, and he confirmed
Huntington's statement, pointing out, however, that in parts of
the Transvaal chronic malaria may be to blame." The most
sinister fact is that these "poor whites" appear to have been
largely born in the country. The newcomers are on the whole
more energetic. They find employment, and if they have difficulty

RACE OR PLACE 45

in one place, move on to another. The poor whites lack the ini-
tiative to do this. If they fall into difficulties, they tend to lie
down and give up. They need higher wages than the blacks in
order to maintain their traditional standard of living. They are
not efficient enough to get higher wages. If they had the restless
energy which characterizes the children and grandchildren of
emigrants from Europe in Canada, for example, they would
scarcely fall into such straits.

Since the problem is economic, the South Africans are striv-
ing to apply economic remedies. This is wise, but success is
doubtful unless other factors are also considered. Back of the
economic facts, and in many ways conditioning them, lies the
climate. South Africa is supposed to have a climate admirably
adapted to Europeans. I shared the common opinion until I
began to gather statistics of the eff ect of climate upon efficiency.
These, as will be shown later, indicate that although the South
African climate is pleasant, it lacks the stimulating qualities
which are so important in Europe and North America. This
lack of stimulus increases rapidly as one goes from south to
north. Here, then, is the situation that confronts us : In South
Africa the white men settled first in the regions most favorable
from a climatic point of view and then pushed northward into
worse conditions. Even the best parts of South Africa cannot
approach England and Holland in the excellence of their cli-
mate. Hence, the white settlers are everywhere at a disadvantage.
On the other hand, the Bantu negroes have come into South
Africa from the north, where the climate is far less favorable
than in their new homes. Thus the two races face each other
under conditions which lessen the white man's energy, while they
stimulate the black man. The whites are still far ahead, and will
doubtless continue to be so indefinitely. Nevertheless, the weaker
ones are being weeded out and prepared for destruction. What
the final result will be, no man can say. It depends upon whether
we can discover a means of preventing the deterioration which

46 CIVILIZATION AND CLIMATE

now seems to attack a portion of the population when people
move from a good climate to a worse.

A more striking case than that of South Africa is found in
the Bahama Islands. At the time of the American Revolution a
considerable number of Loyalists were so faithful to England
that they sacrificed their all in order to escape from the new
flag with its stars and stripes. Leaving their homes in Georgia
and other southern states they sought the British territory of
the Bahamas. Other colonists came from Great Britain. Now,
after from three to five generations, the new environment has
had more opportunity than in South Africa to produce its full
effect. Almost nowhere else in all the world have people of the
English race lived as genuine colonists for several generations
in so tropical a climate. What has been the result? There can
be but one answer. It has been disastrous. Compare the Bahamas
with Canada. The same sort of people went to both places.
Today the descendants of the Loyalists in Canada are one of
the strongest elements in causing that country to be conspicu-
ously well governed and law-abiding, and the descendants of
other colonists, both British and French, vie with them in this
matter. In the Bahamas the descendants of the same type of
people show today a larger proportion of poor whites than can
probably be found in any other Anglo-Saxon community. Al-
though no figures are available, my own observations lead to
the conclusion that the average white farmer is scarcely ahead
of the average negro.

Whatever the exact figures may be, there can be no question
that in the Bahamas the two races tend to approach the same
level. This seems to indicate a marked retrogression of the white
race in regions which are climatically unsuitable. Let me hasten
to say that many of the more intelligent Bahamans do not differ
from the corresponding portions of the Anglo-Saxon race else-
where. At home they feel themselves handicapped, but when the
young people go away to the northern United States or Eng-

RACE OR PLACE 47

land, they frequently show marked ability. Their inheritance is
still good. As to the poor whites, who were described in connec-
tion with the lumber industry, it is not so certain that their in-
heritance remains unimpaired, for in some villages genuine ab-
normalities both of body and mind are seen. This, however, may
be due to the intermarriage of cousins which has been common
in certain communities.

The inefficiency of many of the white Bahamans, however, is
not due to intermarriage, as is sometimes implied, for villages
where this prevails are scarcely worse than those where it is no
more common than in America. Nor is the inefficiency due to
disease. The hookworm is practically unknown. According to
a report of Dr. McHattie, Chief Medical Officer of the Islands,
only two cases had been reported up to October, 1913. In this
report, for which I am indebted to the courtesy of Dr. J. A.
Ferrell of the Rockefeller International Health Commission, the
author points out that "the remarkably rapid manner in which
the soil . . . dries after even the heaviest rain" prevents the
development of the infective larvae. For similar reasons malaria
is no more prevalent than in Delaware, for instance, and in gen-
eral the islands are decidedly healthful. A monotonous diet may
be another detrimental factor, but it is scarcely the root of the
matter. Many of the people are well fed, and all could be so if
they displayed any energy. Indeed, many people say that life
is altogether too easy in the Bahamas. The soil is wonderfully
fertile, crops of some kind will ripen at all seasons, and a man
can work less than half his time and still readily procure an
abundance to eat and wear for himself and his family. On the
other hand, we are often told that the difficulties of life have
broken the spirit of the inhabitants. The soil, in spite of its rich-
ness, is thin, and rocks are so abundant that the plough is almost
unknown. Hand agriculture in little patches in the midst of
naked limestone is the rule. It cannot be denied that there are
difficulties in comparison with many other tropical countries.

48 CIVILIZATION AND CLIMATE

For instance, I was talking with a negro whose parents were in
a slave ship bound from Africa to Cuba when a British warship
captured it. The slaves were taken to the Bahamas and liberated.
In answer to a question as to how his parents liked the islands
compared with Africa, the son said: "They didn't like it. They
used to say, 'In Africa one could lie around all day and do
nothing and always find something to eat. Here one has to work
or else starve.* " The truth seems to be that compared with
North Prussia or Maine the Bahamas are a very easy place in
which to make a living, but that much more work is needed than
in some other tropical regions. They are at the happy mean.
Other difficulties such as the tropical hurricanes which sweep
over the country once in every few years ; insect pests, which
are neither more nor less harmful than in other countries ; the
American tariff; competition with Cuba, and above all the iso-
lated position of the islands are frequently cited as causes of
the constant Bahaman failures. The islands are always suffering
from bad luck, and something must be to blame.

All these various factors doubtless play a part in retarding
the development of the Bahamas. Back of them, however, lies
a factor of even greater import, namely, an inertia due to the
climate. It does not cause the difficulties mentioned above, but
it aggravates them and makes it almost impossible to overcome
them. I talked about this with perhaps fifty of the more intelli-
gent people, including both natives and foreigners who have
been there a number of years. Almost without exception they
said the same thing. "This climate is one of the best in the world.
You can see that for yourself. It is very healthful, and we have
very few sicknesses. The only trouble is that it does not make
one feel like work. In winter it is all right, although even then we
cannot fly around the way you Americans do. We always feel
lazy, and in summer we want to sit around all the time." As an
American picturesquely put it : "Until I came to the Bahamas
I never appreciated posts. Now I want to lean against every one

RACE OR PLACE 49

that I see." Many of the men and almost all the women com-
plained of feeling tired. Even the children are listless. One young
man stated the case very strongly, "We go to bed tired in sum-
mer and we get up more tired, and the summer lasts from April
to October." Again and again people said : "Oh, it's all very well
for you to think we're lazy, but try living here six months or a
year and you'll be as lazy as we are. It's something in the air.
Just look at these young ministers who come out from England.
At first they are full of energy, but after a year or two it oozes
out, though their spirit is still as zealous as ever." Two of the
ministers spoke of the fact that when they came out they thought
nothing of walking twenty miles, but now they dread the thought
of two. Several of the most thoughtful and intelligent islanders,
men who have succeeded in business and whose judgment would
be respected anywhere, said: "We know that we are physically
ttnable to do what English and Americans can do. We are weaker
than our fathers, and they were weaker than theirs. It is a grief
to send our children away, but in our hearts we know that this
is not a white man's country." All this, it must be remembered,
is not due to any specific disease, so far as we are aware. Indeed,
I met several people who said that a stay of a few years in the
Bahamas had improved their health, but at the same time had
made them feel inefficient.

Aside from extremely ignorant persons whose opinion is of
little value, the only men who spoke of the climate more hope-
fully were five or six highly trained officials and others occupy-
ing positions of authority. These men, without exception, can
control their own time. In most cases their office hours are from
9 or 10 a.m. to 2 p.m., or less. They are men of naturally strong
physique ; they have the opportunity and the will to take regular
exercise ; and, most important of all, they make long and fre-
quent visits to the United States or England.

The benefit to be derived from a visit to a more bracing cli-
mate is astonishing. The contrast between the dull, sallow com-

50 CIVILIZATION AND CLIMATE

plcxions and thin cheeks of the women and girls who have always
lived on the islands and the round, rosy cheeks of those who
have recently come back from a long stay at the North is most
striking. According to a local saying you cannot tell whether a
Bahaman woman is pretty until she goes away, and has a chance
to fill out her cheeks and get some color. It is by no means strange
that the stronger, more energetic young white people are fast
leaving the islands. I asked a Bahaman girl, who had been study-
ing nursing in New York, whether she enjoyed life more in the
United States than in the Bahamas. "How can one help enjoying
it more there?" she answered. "There one feels like doing things.
Here one never feels like anything." Like almost everyone else
she was sure that it was the climate even more than the new social
environment which made the difference.

One thing that surprised me was to hear the Bahamans speak
of the stimulus of living in Florida. A native merchant remarked:
"If I hire a new man I don't have to ask whether he has been to
Florida. I know it by the way he works, but it docs not last
long." Here again the social environment is an important factor,
but various people told me that the air somehow makes them feel
more capable of work in Florida than at home. The women of
Florida I heard them say it themselves are pale and wan
compared with their northern sisters. One of them, whose color
still shows her northern origin, remarked: "When I come home
after a summer in the North, I am full of energy and see all sorts
of things that I want to change about the house. But after a
month or two I don't care whether things are fixed or not." One
hears the same sort of thing everywhere. A factory superin-
tendent from Atlanta, Georgia, told me that the Florida work-
men, even the most skillful mechanics, drive him frantic because
they are so shiftless and so ready to take a day off whenever
they feel like it far more so than at Atlanta, even though
Atlanta seems slow to northerners. Yet, in spite of all these
things, Florida is a more stimulating place than the Bahamas.

RACE OR PLACE 51

Its summers are not much better, but its winters are sometimes
frosty, while in the Bahamas the thermometer practically never
goes below 50 F. Perhaps of greater importance, as we shall
see later, is the fact that in Florida the temperature from day to
day varies much more rapidly than in the Bahamas, even though
both places are in the same latitude. Hence, the mainland is
blessed with a genuine climatic stimulus compared with the uni-
form islands.

The last thing to be said about the Bahamas concerns the
effect of the climate on mental activity. Practically all the
islanders with whom I talked thought that the effect of the cli-
mate on mental activity is at least as great as on physical.
Several of the more thoughtful, without any suggestion on my
part, put the matter in this way: "The worst thing about this
climate is the effect on the mind. Not that people do not have as
good minds as elsewhere, but one soon gets weary of hard mental
effort. It is extremely difficult to concentrate one's thoughts.
At night one cannot seem to make himself read anything serious
nothing but the lightest kind of stories." In our own southern
states one hears the same complaint. Even in Virginia the book-
sellers say that during the long summer almost no one touches a
serious book. One feels it everywhere, for on the trains, at the
railroad stations, and at the newsdealers it is generally difficult
to find the higher grade of magazines. Time and again during
a recent journey of three months in the southern states I tried
to get such papers as the Outlook, Independent, Harper's, At-
lantic* Review of Reviews, The Century, and so forth but
all that I could find was trashy story magazines. The dealers
rarely keep the better magazines because people will not read
them. Lack of training surely has something to do with the
matter, but mental inertia due to lack of climatic stimulus seems
to be at least equally important.

Let us return now to our question as to a Teutonic and a
negro Egypt. The farmers of the northern and southern states,

52 CIVILIZATION AND CLIMATE

the race problem of South Africa, and the backwardness of the
Bahamas, all seem to point to the same conclusion. When the
white man migrates to climates less stimulating than those of
his original home, he appears to lose in both physical and mental
energy. This leads to carelessness in matters of sanitation and
food, and thus gives greater scope to the diseases which under
any circumstances would find an easy prey in the weakened
bodies. The combination of mental inertia and physical weakness
makes it difficult to overcome the difficulties arising from isola-
tion, from natural disasters, or from the presence of an inferior
race, and this in turn leads to ignorance, prejudice, and idleness.
Thus there arises a vicious circle which keeps on incessantly.
From its revolving edge a part of the community is thrown off as
poor whites, whose number increases in proportion to the ener-
vating effect of the climate and the consequent speed with which
the circle revolves. That climate is the original force which sets
the wheel in motion seems to be evident, because it is only in
adverse climates that we find the "cracker" type of "poor white
trash" developing in appreciable numbers. If white men lived a
thousand years in Egypt it seems probable that a large propor-
tion of them would degenerate to this type. Whether they would
still retain an inheritance of health and mentality sufficient to
keep them ahead of a similar body of negroes can scarcely be
determined.

The chief reason for doubt in this respect is that we do not
yet know just how natural selection would work in such a case.
It would almost certainly act in two ways. First, many of the
abler young people of the white race would presumably migrate,
as they do in the Bahamas. This tendency is generally strongest
in the upper classes who can afford to send their children away
to school. It is also strong among the young people of all classes
who possess more than the average initiative, ambition, and
physical strength. It becomes weaker and weaker as one goes
down the scale, and almost ceases among the "poor whites," who

RACE OR PLACE 53

have so little mental capacity and so much physical inertia
that in spite of much grumbling they remain where they are
and compete with the colored people. The other kind of natural
selection consists of a selective death rate. Children who inherit
certain physical and mental traits are more likely to die than
are children who do not possess those traits. What the traits
are which cause extermination we do not know. A fair skin, a
nervous temperament, an excess of activity, and an unwilling-
ness or incapacity to get sufficient rest may be qualities which
doom certain white stocks to gradual extinction outside their
own climate. In places like South Africa and the Bahamas the
temperament which is willing to intermarry with the colored
people helps certain types of white people to perpetuate their
inheritance, but at the same time it gradually eliminates the
qualities of energy, initiative, and inventiveness which seem to
be so much more characteristic of Nordics than of negroes.

It must not be forgotten that theoretically it may be possible
that some day a carefully controlled series of crosses between
whites and blacks may eliminate the weak traits of each and
combine the good traits. Thus a race may arise which resembles
negroes in its good temper and its capacity to withstand a
tropical climate, but which will have the progressive, executive,
and inventive capacities of the white race. Such crosses have
been made among animals. For example, Mr. M. F. C. Honore
of the Transvaal has sent me the following quotation which he
believes to be prophetic of what will some day happen in South
Africa. It is from Winston Churchill, the British Cabinet Minis-
ter. "At Naivasha [practically on the equator in British East
Africa] there is the Government stock farm. One may see in their
various flocks, the native sheep, the half-bred English, the three
quarter bred, etc. The improvement is amazing. The native
sheep is a hairy animal, looking to the unpractised eye more
like a goat than a sheep. Crossed with Sussex or Australian
blood, his descendant is transformed into a woollen beast of

54 CIVILIZATION AND CLIMATE

familiar aspect. At the next cross the progeny is almost indis-
tinguishable from the purebred English in appearance, but
better adapted to the African sun and climate. It is the same
with the cattle. In the first generation the hurnp of the African
ox vanishes. In the second he emerges a respectable English
shorthorn."

Such carefully controlled crossbreeding may perhaps be
possible among mankind after hundreds or thousands of years.
But first we must know what human qualities are "unit char-
acters" so that they are inherited according to the Mendelian
law and are not due to the combination of a series of such char-
acters. Then we must learn what qualities are dominant over
others so that the presence of one hides the other. Another highly
complex problem is to determine what qualities are linked with
others so that one cannot be inherited without the other. The
fact that linked qualities are very common may mean that cer-
tain good qualities like the tolerance of the negro for a hot cli-
mate can never be inherited without certain undesirable qualities
like the lack of care for the future which is one of the chief causes
of negro shiftlessness. Even if such linkage is not an insuperable
barrier to the production of a really new race by intelligent
crossbreeding there still remains the almost insurmountable
obstacle of deep-seated human customs, racial antipathies, and
modern ideas of individual liberty. Nevertheless, it is worth while
to reflect on the following dream of Laf cadio Hearn : "It is
neither unscientific nor unreasonable to suppose the world even-
tually peopled by a race different from any now existing yet
created by the blending of the best types of all races ; uniting
western energy with far eastern patience, northern vigor with
southern sensibility, the highest ethical feelings developed by all
great religions with the largest mental faculties evolved by all
civilizations ; speaking a single tongue composed from the richest
and strongest elements of all preexisting human speech, and

RACE OR PLACE 55

forming a society unimaginably superior yet unimaginably un-
like to anything which now is or will ever be."

This is an inspiring dream even though most biologists now
regard it as impossible. So far as climate is concerned the hard
reality seems to be that at present, both by its direct action and
through natural selection a warm, monotonous, and unstimu-
lating climate tends to reduce human activity both physical and
mental, regardless of race.

CHAPTER 111
THE WHITE MAN IN THE TROPICS

THUS far we have dealt with the temperate zone. Even the
Bahamas lie north of the Tropic of Cancer. Let us now
turn to the torrid zone, which contains the world's richest and
most inviting fields of future development. Let us inquire into
the effect of that region upon Europeans who attempt to live
there permanently. The isolation of the tropical regions, their
lack of facilities for transportation, and the great difficulties of
agriculture will doubtless be overcome, but that will by no means
solve the problem. Two great obstacles will still remain the
native inhabitants and the white man's own mind and body.

Whatever may be the cause, it is generally agreed that the
native races within the tropics are dull in thought and slow in
action. This is true not only of the African negroes, the South
American Indians, and the people of the East Indies, but of the
inhabitants of southern India and the Malay peninsula. Perhaps
they will change, but the fact that the Indians both of Asia and
South America have been influenced so little by from one to four
hundred years of contact with the white man affords little
ground for hope. Judging from the past, there is scant reason
to think that their character is likely to change for many gen-
erations. Until that time comes they will be one of the white
man's greatest obstacles. Experience shows that the presence
of an inferior race in large numbers tends constantly to lower
the standards of the dominant race. Here in America, although
the negro forms only a tenth part of the population, he is one
of our gravest problems. Yet he is not so great a handicap as

THE WHITE MAN IN THE TROPICS 57

are the native races of the tropics. Whatever the negro may
have been when he was first brought to America, he is now less
stolid and indifferent, more subject to stimulating influences
than he was when he came, or than the Indians of tropical
America. It is literally true in South America, for instance, that
the more an Indian is paid the less he will work. If one day's pay
will buy two days' food, he will work half the time ; if the pay is
increased so that one day's pay will buy food for three days, he
will work one third of the time. The experiment has been tried
again and again. The most considerate employers of tropical
labor agree with the most inconsiderate that in general it is use-
less to attempt to spur the Indians by any motive beyond the
actual demands of food and shelter. Kindness and consideration
on the part of the employer undoubtedly promote faithfulness,
but they seem rarely to arouse ambition or energy. With the
negro in Africa, as everyone knows, much the same condition
prevails, but where he has been brought to the United States this
is by no means so true. For example, in Central America it is
generally thought that a negro from Jamaica is more efficient
than an Indian, while a negro from the United States is much
more efficient. The negro in the United States is generally con-
sidered more efficient than he was in Africa, whereas his stay-at-
home brother and the Indian of tropical America, remaining in
their old environment, do not seem to have changed.

Doubtless the change in the negro is due to a new social en-
vironment quite as much as to a new physical environment, and
many authorities believe that the social change is the more im-
portant. This, however, does not materially alter the case. As
conditions are now, it is extremely difficult to change the physical
environment of tropical races so long as they remain in their
present habitat, and it seems to be equally difficult to change
their social environment. Those who dwell permanently in the
white man's cities are influenced somewhat, but the results are
often disastrous. Here as almost everywhere within twenty de-

58 CIVILIZATION AND CLIMATE

grees of the equator, the general tendency seems to be to revert
to the original condition as soon as immediate contact with the
white man is removed.

This does not mean that contact with a higher civilization
will never benefit the people of the tropics, but merely that the
process is bound to be slow. The aborigines of tropical America,
for example, show little sign either of disappearing, or being
swallowed up by a multitude of immigrants, as has been the case
in temperate latitudes, nor do they appear to be changing their
character. On the contrary, in Latin America, the only tropical
region except Australia where the white man has settled in large
numbers, the proportion of Indian blood is apparently increas-
ing at the expense of the white, and the Indians act and think
almost like their ancestors three or four centuries ago. This is
largely because the white man, except in a few favored places,
suffers from tropical diseases more than does the native, and his
children tend to move away if strong, or to be weaklings who die
young and leave few children. It is notorious that India contains
almost no fourth generation of Indian-born British. The British
children are either sent back to Europe to recover their health,
or else become enfeebled and their descendants die out. Even
with the help of modern medical science, it is not yet certain that
the permanent white population can increase greatly, although
sojourners are sure to become numerous. In Australia, to be
sure, the white man seems to be succeeding within the tropics,
but he is still new there and has the inestimable advantage of
active natural selection and freedom from contact with the
natives.

In many well-populated tropical countries modern science
lowers the death rate among the natives, and thus increases
their numbers. The white man has permitted the native popula-
tion of India to double and that of Java to increase seven-fold,
partly by conquering diseases and partly by the prevention of
famine and war. If the conclusion just reached is correct, it

THE WHITE MAN IN THE TROPICS 59

seems probable that tropical countries will long continue to
maintain a dull, unprogressive population. Contact with such a
population constantly exposes the white man to a most deterio-
rating influence. For example, the inferior mental ability of the
lower race, and its incapacity for effective organization lead to
the abuse of its labor and to its exploitation in some form of
peonage, even though the fact may be disguised by legal phrase-
ology. Again, the presence of a despised race is almost certain
to lead to low sexual morality. In the same way, political equal-
ity becomes a mere form of speech, for the dominant race will
not permit the other to gain rights at its expense. Manual labor,
too, is despised, for it is associated with the idea of an inferior
race. All these things may be looked upon as disadvantages of
the lower race, but I believe that the higher reaps by far the
greater injury. The conditions just mentioned appear to be
among the most potent factors in rendering it difficult for the
white man to attain as much success in tropical regions as in
those farther to the north or south. Their evil effect is roughly
proportional to the difference between the two races. That dif-
ference is at a maximum where a low tropical race remains in
its original, unstimulating environment, and is brought into
contact with immigrants of a highly developed race who com-
pletely change their environment. The newcomers are released
from old restraints at a time when they stand in peculiar need
of them. Instead of being stimulated to greater love of political
freedom and equality, sterner morality, and more intense in-
dustry, as was the case among the settlers in New England, the
immigrants are in danger of being weakened in all of these re-
spects. The effect on the original immigrants is bad enough,
but on their children it is far worse. The settler, or European
colonist, who is possessed of wealth and power, can to a slight
degree shield his family, but even in such cases the children are
'"n constant contact with servants. They grow up with a supreme
contempt for the natives, and at the same time with the feeling

60 CIVILIZATION AND CLIMATE

that they can treat them as they choose. If poorer people, that
is, colonists in the ordinary sense of the word, attempt to live in
the tropics, especially if they are people who work with their
hands, their children are exposed still more to all the con-
taminating influences of contact with the natives. Hence, the
second and third generations, and the fourth and fifth, if there
are any, suffer more than their ancestors.

The degree to which the indirect or external handicaps of
tropical countries are effective in lowering the standards of
civilization depends largely upon the amount of energy and
will power possessed by the inhabitants. This, in turn, depends
upon physiological conditions. Obviously, diseases have much
to do with the matter. This subject has been so much discussed
that I shall here refer to it only briefly. There can be little
doubt that malaria and the many other diseases which are
characteristic of tropical countries play an important part in
causing a low state of civilization. The old idea that the people
who live in tropical regions are immune to local diseases is no
longer accepted by students of tropical medicine. Adults, to be
sure, are often immune, but apparently this is only partially true
of children. Vast numbers of children die in infancy and early
childhood from the diseases which prevent the white man from
permanently living within the tropics. Others suffer, but recover.
They bear the results with them to the grave, however, in the
form of enlarged spleens, or other injuries to the internal organs
of the body. The world has of late been astonished at the ravages
of pellagra and of other diseases due to such organisms as the
hookworm. People who are subject to them cannot be highly
competent. Their mental processes, as well as their physical
activity, are dulled. So long as a community is constantly
afflicted with such disorders, it can scarcely rise high in the
scale of civilization. Nothing is more hopeful for the tropics
than the rapid progress in the control of these diseases. If they
could be eliminated, not only might the white man be able to

THE WHITE MAN IN THE TROPICS 61

live permanently where now he can be only a sojourner, but the
native races would probably be greatly benefited. How great
this benefit would be we cannot yet tell, but the elimination of
the diseases which especially affect children would probably do
much to increase vitality, energy, and initiative. This in itself
would be an immeasurable boon not only to the natives, but to
the white man, who would thereby be freed in part from some of
his worst social dangers.

This highly desirable result cannot be obtained quickly. The
achievements of the United States in Panama are sometimes said
to prove that diseases can be eliminated anywhere in tropical
countries. This is true, but it must be remembered that Panama
is a highly specialized case. During the building of the Canal
a great number of people were collected in a small area, and
enormous sums of money were freely expended. Everyone was
subject to strict, semi-military rule, and similar conditions still
continue. Such methods cannot be applied to millions of square
miles. The expense would be prohibitive. The ordinary farmer
in tropical regions cannot expect to be protected by his govern-
ment. He must protect himself. In the long run even tropical
races may learn this lesson, but it will be a difficult and expensive
task, and will require a radical change in the people themselves.
Such a change will doubtless come, but not for generations, and
not until a long selective process has gone on whereby those
who do not adopt modern medical methods will gradually be
eliminated, while those who adopt them will persist.

There has been so much misunderstanding of Panama and so
many wild statements that it may be well to set forth the exact
facts. The Health Department of the Panama Canal, as it is now
called, has charge of three districts, whose population in 1917
was as follows: the city of Panama, 61,074; the city of Colon,
25,386 ; and the Canal Zone, 27,543. For purposes of health and
sanitation all are under the control of the United States, and no
expense is spared to make them as healthful as possible. In order

62 CIVILIZATION AND CLIMATE

to avoid the complications due to the influenza epidemic of 1918,
let us take the period from 1912 to 1917. By 1912 the health
measures of the United States Army had reached such perfection
that the death rate had been reduced 50 per cent. The improve-
ment still continues, but it is now slow and apparently does little
more than keep pace with the similar improvement in the ad-
vanced parts of the world. The two cities of Panama and Colon
contain the ordinary mixed population of tropical seaports :
negroes from the West Indies ; Mestizos half Spanish, half
Indian from the neighboring parts of Central and South
America ; a few Chinese and other Asiatics ; some Europeans and
Americans. A considerable number of the employees of the Canal
live there. The Canal Zone, on the other hand, contains a large
proportion of Canal employees, chiefly Americans, West Indian
negroes, and Europeans. Among all of these the percentage of
men between twenty and fifty years of age is large. The follow-
ing figures show the crude death rates from 1912 to 1917 among
the civilian population, excluding soldiers, in the three districts
of Panama and in certain other areas with which comparisons
may profitably be made.

Panama . 30.5 Chile . . . 27.9 Bombay (1910-1912) . . 37.0
Colon . . 24.8 Spain . . . 22.0 Calcutta (1910-1912) . . 26.1
Canal Zone. 13.6 United States . 13.9 Amsterdam (1901-1913) . 12.6

A multitude of other figures might be presented all of which
would show that while the work done in Panama has been ad-
mirable, the general conditions of health in the cities of Panama
and Colon are still twice as bad as in the advanced parts of the
world. They are about on a par with those of similar cities of
India, for Bombay and Calcutta, by reason of their size and
desperate overcrowding, presumably have higher death rates
than do Indian cities as small as Panama and Colon. The death
rate for infants under one year bears out this general conclusion,
as appears from the following figures showing the deaths per
one thousand births in 1915, 1916, and 1917:

THE WHITE MAN IN THE TROPICS 63

Panama, Colon, and the Canal Zone 232

Colored people in New York 182

Colored people in United States Registration Area .... 172

White people in United States Registration Area .... 96

White people in Minnesota 69

The area where births are registered in the United States in-
cludes only a small part of the South, so that the death rate
among colored infants as a whole is higher than appears above.
In Richmond, Virginia, during 1917, 1918, and 1919, it aver-
aged 198. In cities farther south it doubtless reaches a level as
high as that for people of all sorts at Panama.

The foregoing data make it obvious that the widespread idea
as to the healthfullness of Panama is based solely on the small
number of people in the Canal Zone. But the death rate of 13.6
given above for the Canal Zone has by no means the significance
that is usually supposed. Its use for comparative purposes is
vitiated by two facts ; first, the number of deaths by violence,
chiefly by accident, is unusually high in the Canal Zone ; and
second, the inhabitants of the Canal Zone are a highly selected
group mostly of good physique and in the prime of life and hence
bound to have a relatively low death rate no matter where they
live. The best way to make a fair comparison is to take people
of the same age, sex, and occupation who have otherwise also
been selected by the same method, and compare the death rates
in different places. But this is impossible. As the next best thing
let us take the death rate from 1912 to 1917 among the Canal's
white employees from the United States and compare it with the
rate for men of similar age elsewhere. If we assume that the pro-
portion of white men of different ages in the Canal Zone is the
same as among the white employees from the United States and
was also the same at the census of 1920 as in the period from
1912 to 1917, both of which are essentially the case, it is easy
to compute the relative death rate in other regions on the same
basis as the rate for Panama. Using the data prepared by the
International Institute of Statistics together with the records

64 CIVILIZATION AND CLIMATE

of Yale University, we find that if the proportion of men of vari-
ous ages were the same in the other places as among the white
American employees of the Panama Canal, the death rates
among such men would be as follows :

Approximate death

rate when deaths
Death rate from due to violence
all causes are eliminated

New York State, 1906-1915 12.3 10.6

Connecticut, 1906-1915 10.8 9.3

Washington State, 1908-1913 7.8 5.3

New Zealand, 1906-1915 6.7 4.8

White Canal employees from the United

States, 1912-1917 4.9 2.7

Students in Yale University, 1912-1917 . . 1.9 1.7

Does this mean that the climate of the states of New York and
Connecticut is relatively bad, while that of Panama and the home
of Yale University at New Haven is remarkably good? Not at
all. It simply means that the two states have relatively normal
death rates for their particular climates and for a comparatively
unselected population. They are handicapped by their numer-
ous unhealthful factories and cities and by the great number of
their immigrants, many of whom are poor, ignorant, and of low
caliber mentally. Moreover, many of the more energetic young
people have migrated westward. The adventurous and persistent
qualities which lead to migration are partly due to health and
physical vigor, and partly to mental initiative, adaptability, and
readiness to try a new life and new methods. It needs no demon-
stration to show that such people are sure to have a low death
rate, especially when they are highly prosperous, as in the state
of Washington. They have the physical vigor to withstand dis-
ease, they have the good sense to take care of themselves, and
they have the means wherewith to purchase good food, good
shelter, good sanitation, and good medical service. Since New
Zealand is harder to reach than Washington, its immigrants

THE WHITE MAN IN TPIE TROPICS 65

have been even more highly selected for thrift, health, and
physical and mental vigor.

Panama, like Washington and New Zealand, attracts chiefly
the more vigorous type of people. The man who is organically
diseased rarely thinks of going there. Moreover, in Panama
white employees come from America as adults. On the contrary,
many of the people of Washington and New Zealand were born
there and have remained regardless of whether they possess the
pioneer vigor and initiative of their parents. Again, even if they
have the brave spirit that overcomes physical handicaps, the or-
ganically weak are not allowed to go to Panama as employees.
If they try to go, they are weeded out by physical examinations.
Even that, however, does not end the matter, for the examina-
tions are repeated each year. Every individual who shows signs
of weakness is advised to leave Panama as soon as possible;
many are ordered home ; and not a few are deported, especially
those suffering from mental disorders. During the three years
for which I have been able to find data (1914, 1915, and 1917)
the deportations on account of disease among employees of all
sorts, both white and colored, amounted to approximately 40
per cent of the total deaths from disease. If these people had
stayed in Panama, as they stay in New York, Connecticut,
Washington, or New Zealand, many of them would soon have
died. Inasmuch as such deportations have been going on for
years, it is practically certain that without them the death rate
at Panama would be decidedly larger than now. In addition to
the persons who are deported, a far larger number go home
voluntarily on the advice of their physicians. Moreover, many
who show no immediate signs of disease remain at home after
one of their earlier furloughs because they find the climate at
Panama uncomfortable.

In addition to all this it must be remembered that the white
employees at Panama are practically all officials or clerks ; they
belong to a class of society which by reason of its intelligence is

66 CIVILIZATION AND CLIMATE

able to take care of itself, so that its death rate is normally much
lower than that of the great body of men of similar age. More-
over, the employees are well paid and admirably housed. They
likewise have long and frequent vacations at home whereby the
effect of the tropical climate is partially neutralized. All these
conditions, even without the excellent medical care which the
employees receive, free of cost, would insure a degree of health
in Panama much better than in most tropical regions. On the
other hand, if the population of Panama were an ordinary un-
selected type and if none of the weak or sick were sent away, it
seems probable that in spite of admirable sanitation and medical
care the death rate would be larger than in New York among
people of similar age. This last statement is merely an opinion,
since by its very nature it is not susceptible of actual proof. We
know as a fact, however, that the death rate at Panama is
greatly lowered by the selection of healthy, intelligent employees
as well as by good medical care.

The conditions at Yale or any other university suggest that
in such cases selection is even more important than medical care.
From June, 1912, to June, 1917, the average number of under-
graduates at Yale was 2476, among whom the total number of
deaths was nine through disease and one by accident. This gives
an annual death rate of 0.8 per thousand against 4.5 for all
the young men of Connecticut between 15 and 24 years of age.
In other words, an unselected young man in Connecticut is 5.6
times as likely to die as is a selected Yale student. If a similar
ratio prevailed among the University men up to the age of about
55 years, and if the proportion of men at each age were the same
as at Panama, their death rate, barring accidents, would be only
1.7 against 2.7 among the white American employees at Panama.
Now as a matter of fact, from the standpoint of health the
employees at Panama are far more rigidly selected than are the
Yale students. No medical examination is required for entrance
to the University, no one is actually sent away because of his

THE WHITE MAN IN THE TROPICS 67

health, and the amount of medical attention is less on the whole
than at Panama. Other things being equal, this ought to give a
higher death rate at Yale than at Panama. Yet, as a matter of
fact, it actually gives Panama the higher rate by 60 per cent.
This turning of the tables against Panama seems to be due to the
adverse climate.

The net result of the preceding investigation is this : There
can be no doubt of the great value and success of the medical and
sanitary work at Panama. It has cut the death rate in halves at
the cities of Panama and Colon. Nevertheless, the death rate in
those cities is still extremely high, about twice that of the United
States as a whole. So far as the white people at Panama are con-
cerned, the death rate is very low, but that proves nothing about
the climate. It merely proves that it is possible to obtain prac-
tically any death rate by selecting the cases. One could go to
hospitals and select the critical cases. That might give a death
rate of eight or nine hundred. One might select college athletes
and from time to time throw out any who showed signs of illness,
and the death rate would be zero. But to use such death rates as
evidence concerning the climate would be highly misleading. It
is poor policy to use any such reasoning in respect to Panama,
Northern Australia, or any other region where the climate pos-
sesses disadvantages. To do so encourages false hopes. When
these are disappointed, people tend to blame the whole science
of tropical medicine. That science is doing wonderful things,
but as yet there is no evidence that it has overcome the effects of
climate, although it has certainly mitigated them. We shall re-
turn to this subject in connection with Australia. There, as in
Panama, the tropical death rate is lower than those of better
climates, but this is due primarily to the selection of certain
types of residents. I have dwelt on this matter because there is a
vast deal of misapprehension and very little realization of the
importance of selection.

Let us return now to our main question. Suppose that the

68 CIVILIZATION AND CLIMATE

white man should succeed in cultivating the tropical forests,
transversing the waste places, and conquering the diseases. Sup-
pose also that he should eliminate the deteriorating influences of
low social and moral standards among the natives. But suppose
also that there were no selection of the white colonists. If all this
were suddenly done, and average unselected white men were set
down in a tropical garden of Eden, would they be able to hold
their own among the peoples of the world? Would Teutons or
Latins under such circumstances be able permanently to main-
tain as high a standard of civilization as is maintained by their
brothers in Europe? Or would there be a change in some of the
traits which we are wont to call racial? Clearly we are back at
the point where we started, and are confronted by the question
of race versus place. We must determine how much of our Euro-
pean and American energy, initiative, persistence, and other
qualities upon which we so much pride ourselves is due to racial
inheritance, and how much to residence under highly stimulating
conditions of climate.

One of the lines along which we may seek for an answer is by
a comparison of the character of Europeans in tropical coun-
tries with their character in the temperate zone. Whatever differ-
ences we may find are presumably due partly to physiological
and partly to sociological causes, but they manifest themselves
chiefly through the will. In tropical countries weakness of will
is unfortunately displayed not only by the natives, but by a
large proportion of the northerner sojourners. It manifests
itself in many ways. Four of these, namely, lack of industry, an
irascible temper, drunkenness, and sexual indulgence are par-
ticularly prominent, and may be taken as typical. Others, such
as proneness to gambling and disregard for the truth, might
equally well be considered if space allowed.

In the quality of industry the difference between people in
tropical and other countries is well known. We have already
touched on it in the Bahamas, but let us amplify it further.

THE WHITE MAN IN THE TROPICS 69

Practically every northerner who goes to the torrid zone says at
first that he works as well as at home, and that he finds the climate
delightful. He may even be stimulated to unusual exertion. Little
by little, however, even though he retains perfect health, he slows
down. He does not work so hard as before, nor does the spirit of
ambition prick him so keenly. On the low, damp seacoast, and
still more in the lowland forests, the process of deterioration is
relatively rapid, although its duration may vary enormously in
different individuals. In the dry interior the process is slower,
and on the high plateaus it may take many years. Both in books
and in conversation with inhabitants of tropical regions one
finds practical unanimity as to this tropical inertia, and it
applies both to body and mind. After long sojourn in the tropics
it is hard to spur one's self to the physical effort of a mountain
climb, and equally hard to think out the steps in a long chain of
reasoning. The mind, like the body, wants rest. Both can be
spurred to activity, but this exhausts vitality. The common ex-
planations of tropical inertia are diverse. One man says that
within the tropics hard work is unnecessary, because salaries
are high ; another asserts that it is because servants are cheap ;
still another claims that hard work is dangerous to the health ;
and almost all agree that "anyhow, one doesn't feel like working
down here." Probably all four of these factors cooperate, and
each, doubtless, produces pronounced results, but the last two,
health and "feeling," seem the most important.

In spite of individual exceptions, white men who spur them-
selves to exert their minds as earnestly and steadily within the
tropics as at home are in great danger of breaking down in
health. They become nervous and enfeebled, and readily suc-
cumb to tropical diseases. This is one of the most powerful
deterrents to the development of an efficient white population
in equatorial regions. If the more intellectual members of the
community ruin their health, they are almost sure to die before
their time, or else to go back to the North. In either case they

70 CIVILIZATION AND CLIMATE

are not likely to leave many children to perpetuate their char-
acteristics. Thus if white colonization takes place on a large
scale within the tropics, there is grave danger that the physically
strong but mentally lethargic elements will be the ones to be-
come the ancestors of the future population. In the past this
factor must have operated to weed out the more intellectual
members of each of the many races that have migrated toward
the equator. The inertia which prevents the less competent mem-
bers of a tropical community from overworking may perhaps
be interpreted by teleologists as a merciful provision of Provi-
dence to warn man that he must not work too hard in the torrid
zone, but that will scarcely help to advance civilization. Few
people will question the reality of the tropical inertia. It is the
same lassitude which everyone feels on a hot summer day the
inclination to sit down and dream, the tendency to hesitate
before beginning a piece of work, and to refrain from plunging
into the midst of it in the energetic way which seems natural
under more stimulating conditions.

Lack of will power is shown by northerners in tropical regions
not only in loss of energy and ambition, but in fits of anger. The
English official who returns from India is commonly described
as "choleric." Every traveler in tropical countries knows that
he sometimes bursts into anger in a way that makes him utterly
ashamed, and which he would scarcely believe possible at home.
Almost any American or European who has traveled or resided
within the tropics will confess that he has occasionally flown
into a passion, and perhaps used physical violence, under cir-
cumstances which at home would merely have made him vexed.
This is due apparently to four chief causes. One is the ordinary
tropical diseases, for when a man has a touch of fever, his temper
is apt to get the better of him. In the second place, the slowness
of tropical people is terribly exasperating. The impatient north-
erner uses every possible means to make the natives hurry, or to
compel them to keep their word. His energy is usually wasted

THE WHITE MAN IN THE TROPICS 71

the native remains unmoved, and the only visible result is an
angry and ridiculous foreigner. Yet a show of anger and violence
often seems to be the only way of getting things done, and this is
frequently used as an excuse for lack of self-control. In the
third place, the consequences of becoming angry arc less dan-
gerous than elsewhere. The inert people of tropical countries
often submit to indignities which an ordinary white man would
bitterly resent. Of course they object to ill treatment, and will
retaliate if possible, but they generally do not have sufficient
energy or cunning to make their vengeance effective against the
powerful white man. Finally, those who have lived in the tropics
generally find that, even when things go smoothly, and they are
in contact with people of their own kind and are in compara-
tively good health, they are more irritable than at home. In
other words, their power of self-control is enfeebled. Of course
there are many exceptions, but that does not affect the genera]
principle.

Drunkenness, our third evidence of lack of self-control, need
scarcely be discussed. Within the tropics, the white man's alco-
hol in the form of rum is scarcely more injurious to the natives
of Africa than it is in other forms to himself. In places such as
Guatemala and parts of Mexico, drunken men and women may
be seen upon the streets at almost any time of day. Nowhere else,
during extensive travels in America, Europe, and Asia, have I
seen so much drunkenness as in Guatemala. Among white men a
large number drink as badly as the natives. Here is an example ;
a railway conductor was telling me about drinks in Guatemala:

"They've got something here called 'white-eye,' " he remarked.
"You know that Mexican 'mescal,' and how strong it is? Well,
white-eye's got mescal chained to a telegraph pole. Yes, I drink
it. A man's got to drink something. The first time I tried it, I
got crazy drunk and smashed things up the way they all do. I
was arrested and fined fifty dollars. [This is really only two and
a half, for Guatemalan currency consists of non-redeemable

72 CIVILIZATION AND CLIMATE

paper, which at that time was worth about five cents on a dollar
a characteristic evidence of tropical incapacity.] I got fined
several times that way and didn't like it. Then one day when I
was going to get drunk, I said to myself, Til go and pay my fine
now and then they won't bother me.' I did that several times, and
the *jefe politico' liked it [presumably because it was an easy
way of pocketing the money]. Then he said to the police : 'Don't
bother this man. Just let him get drunk all he likes, and he'll pay
his fines at the proper time.' I tell you, white-eye is bad stuff.
The only proper way to drink it is to take a quart bottle in the
morning. Find a place that will stay shady all day. Drink the
whole thing right down and get so dead drunk that you will
sleep till night."

I do not cite this man as typical of all the white men in the
tropics. Far from it. Many conduct themselves with sobriety
and industry, but such men almost invariably make frequent and
protracted visits to the better climate of the North. If a white
man stays steadily for long periods in the tropics, however, and
if his character has any weak spots, they are almost sure to be
exaggerated. The drunkenness of the tropical white man arises
in part from the constant heat, which makes people want some-
thing to drink at all times, partly from the monotony of life,
and still more from the absence of the social restraints which
exercise so powerful an inhibitory influence at home. Back of all
these things, however, among both the white men and natives,
there seem to lie two conditions which arc directly connected
with the climate. One is the same cnfecblement of the will which
makes a man burst into anger. The other is a constant feeling
of inefficiency which makes a man crave something to brace
him up.

The last of the ways in which weakness of will is evident in
tropical countries is the relation of the sexes. Its importance can
scarcely be overestimated. It leads to the ruin of thousands of
northerners, even though they do not yield to drink, to anger,

THE WHITE MAN IN THE TROPICS 73

or to laziness. When once they have fallen into pronounced im-
morality the other weaknesses soon follow. The condition of the
native races is still worse. Everywhere within the tropics mis-
sionaries say that their converts can be taught honesty, in-
dustry, and many other virtues, but that even the strongest find
it almost impossible to resist the temptations of sex. Many
Europeans condone this. They say that it is natural, and that
the natives had better be left to their own conventional ways of
restricting but not preventing sexual intercourse. Perhaps they
are right, although I cannot be certain. That is not the point,
however. We are at present concerned with the effect which free
indulgence has upon civilization and upon the capacity for
progress. This may be illustrated by what Gouldsbury and
Sheane, for example, say of the Zulus in northern Rhodesia.
They hold that one of the greatest reasons why these people
remain so backward is that their thought and energy are largely
swallowed up in matters of sex. During the years when the
young men ought to be getting new ideas and thinking out the
many little projects and the few great ones which combine to
cause progress, the vast majority are thinking of women, and
planning to gain possession of some new woman or girl. Under
such circumstances no race can rise to any high position.

The causes of these conditions are various. Many writers dis-
miss the matter by saying that the social standards of tropical
people are low and tend to cause northerners to conform to them.
This is true, but it explains nothing. A real, though minor, rea-
son for the lowness of the standards is found in the free, open
life which is almost universal within the tropics. People are out
of doors so much and it is so easy to meet in secret that tempta-
tion arises very frequently. Much more important is the scanty
dress of the women, and its character, which calls attention to
their sex. Livingstone speaks with disgust of the way in which
his carriers, hour after hour, discussed the breasts of the half-
naked women whom they met. Even in the North women seem to

74 CIVILIZATION AND CLIMATE

be strangely indifferent to the effect of their mode of dress upon
men. They do not seem to think that they are responsible if their
low-necked gowns and the making of their clothing in such a
way that each little movement of their bodies can be detected,
stir men's passions. They appear oblivious to the fact that the
display of their beauty often means that some other woman must
pay the penalty. Within the tropics these conditions are exag-
gerated. I believe I am speaking within bounds when I say that
any young man of European race with red blood in his veins is
in more danger of deteriorating in character and efficiency be-
cause of the women of the tropics than from any other single
cause.

The strength of this deteriorating force is not merely exter-
nal. Either the actual temptation to sexual excess is greater
in the tropics than elsewhere, or else the inhibitory forces are
weakened by the same processes which cause people to drink to
excess, to become unduly angry, and to work slowly. Hcllpach
states that it is said that in southern Italy sexual irregularities
increase greatly at times when the hot, damp wind known as
the sirocco blows across the Mediterranean from the deserts
of northern Africa. This is so well recognized among the people
themselves that offenses committed under such circumstances
are in a measure condoned. Violence, too, is more common at
such times, for self-control of every kind is weakened. In eastern
Turkey the hot desert winds cause the whole community to be-
come cross and irritable. I have there seen a missionary, a man of
unusual strength of character, shut himself up in his study all
day, because he knew that he was in danger of saying something
disagreeable. I cite this case because, among the people whom I
have known, missionaries are, on the whole, most completely
masters of themselves and the least likely to let minor circum-
stances turn them from the Christlike lives which they are striv-
ing to live day by day before the native communities. For this
same reason, to return to our immediate subject, I quote the

THE WHITE MAN IN THE TROPICS 75

remark of a missionary in Central America when we were dis-
cussing the morality of the country. He was a most austere man,
a member of a small and extremely devout sect, and his whole
being was devoted to preaching the gospel. Speaking of his own
experiences, he said:

"When I am in this country, evil spirits seem to attack me.
I suppose you would call them something else, but that is what
I think they are. When I am at home in the United States I feel
pure and true, but when I come here, it seems as if lust were
written in the very faces of the people."

In all the evils which have just been mentioned laziness,
anger, drunkenness, and immorality social causes undoubtedly
play an important part. A strong public opinion would save
many a young northerner from drink and immorality, and would
keep him faithful to his work. A clear religious faith or a high
ideal of duty would do the same thing. Good homes, proper dress,
and many other material changes would help greatly. So, too,
would a study of how it has come to pass that certain tropical
races, in spite of their environment, have developed compara-
tively high moral codes to which they strictly adhere, while a
few have actually learned the lesson of industry. Along with the
social aspect of the question, however, and neither more nor
less important, goes the physical. We must discover to exactly
what extent physical conditions help or hinder the development
of strong character. That is the purpose of the chapters that
follow.

CHAFTEK IV

THE EFFECT OF THE SEASONS

IN comparing Teutons with negroes, or tropical people with
those of the temperate zone, we have been following a method
as old as the days of Aristotle. Such comparisons have led to
most interesting generalizations not only at the hands of Aris-
totle himself, but of many other men such as Montesquieu, Hurn-
boldt, and Ratzel. Yet the importance of climate as a factor in
civilization is still in doubt. For instance, no one denies that
South Africa is confronted by a grave race problem, but many
say that it is purely economic, and has nothing to do with cli-
mate. They support this view by strong arguments. Thus we
are left in uncertainty. The only way to remove this is to devise
some method whereby to separate the effects of climate from
those due to all other causes, whether economic, historic, social,
religious, racial, or something else. Accordingly, the rest of this
volume will be devoted to an investigation of the exact effect of
various climatic factors upon selected groups of people, and to
an attempt to discover how human energy and other qualities
would be distributed if all the earth's inhabitants were influenced
like these particular groups.

In the study of climate one of the most puzzling features is
the diversity of opinion among persons of good judgment. For
instance, at what season do people work fastest in the northern
United States? Some will say the winter, some the spring, a
considerable number the fall, and a few the summer. Most will
say that they are least efficient in summer, but others believe
that they are at their worst in the early spring or late winter.

EFFECT OF THE SEASONS 77

Again, ask a dozen friends whether they work best on clear days
or cloudy. The majority will probably answer that the first clear
day after a storm is by all means the best. A small number will
perhaps think the matter over more carefully, and then say that
after a storm the clearness of the air and the brightness of the
sun are certainly inspiring, but one really accomplishes more
when it rains.

This divergence of opinion is due largely to the fact that
climatic effects are of two kinds, psychological and physiologi-
cal. We are always conscious of the first, but often unconscious
of the second. The two are admirably distinguished in Hellpach's
book on Geopsychische Erscheinungen. An example will make
the matter clear. It is well known that at high altitudes the num-
ber of red corpuscles in the blood increases enormously, and
the capacity to absorb oxygen and to give out carbon dioxide
is correspondingly modified. Yet many people can go to altitudes
of 5000 feet or more without realizing that their physiological
functions have been altered. To cite my own case, up to the age
of twenty-one I had never been a thousand feet above the sea.
Then I went to live at an altitude of 4500 feet. The only physio-
logical effect of which I was conscious was unusual sleepiness for
the first few months, but whether this was due to the altitude or
to the dryness of the air, I do not know. For two or three years
I never thought of the physiological effect of the altitude until
one day, happening to have climbed to a height of 7000 feet, I
began to run up hill. I lost my breath and became tired so quickly
that I was alarmed and thought I must be sick. I was much re-
lieved when it occurred to me that the altitude was not favorable
for running up hill. Manifestly my physiological functions were
different from what they were at sea level, although I was un-
conscious of it. On the other hand, psychologically I was daily
conscious of living in a place where the air was extraordinarily
clear, and where the mountains were always in sight across a
splendid plain twelve hundred feet below us. Presumably both

78 CIVILIZATION AND CLIMATE

the physiological and psychological conditions had an appre-
ciable effect upon the work of every day, but which was the
greater it is impossible to tell.

In this connection Lchmann and Pcdersen state an interesting
fact. In Denmark and Norway they made a series of daily tests
of the strength of three individuals by means of a dynamometer.
They found that the change of atmospheric pressure due to an
ascent of two or three thousand feet makes no appreciable dif-
ference. A similar descent, however, is accompanied by a marked
increase of strength which disappears within three or four days.
They suggest that this may be due to the persistence of abun-
dant red corpuscles when people come down from high places.
The red corpuscles multiply very rapidly under the influence
of low pressure, but are slower in disappearing when the pres-
sure once more increases. Thus, for the first day or two after a
person has come down from the mountains, more than the normal
amount of oxygen may be absorbed, and muscular strength
correspondingly increased. Possibly this is why mountaineers are
generally so irresistible when they descend upon the plains in
sudden raids. My colleague, Prof. H. E. Gregory, suggests
that this may account for the fact that in the horse-races of
the pioneer days of the southwestern United States, the poor,
scrawny animals brought down from the mountains by the In-
dians usually belied their appearance and outran the better-
looking animals of the white men. They may have had an excess
of red corpuscles. Professor Gregory adds that in some of the
highland regions of South America there is a strict rule that
before a race the competing horses must spend a certain number
of days at the race course. This may have arisen because the
animals which race directly after coming from the mountains
are apt to win. There is considerable doubt as to the truth of
this theory, but it illustrates the possibility that we may be
deeply influenced by atmospheric conditions of which we are
almost unconscious.

EFFECT OF THE SEASONS 79

In our opinions as to the effect of the seasons or of daily
changes of weather the relation between psychological and
physiological influences is probably the same as in the case of
altitude. The external conditions which we see and feel make
a greater impression than those which prevail within our bodies.
For example, most of us think that in the northern United States
we work fast in winter. As a matter of fact, the statistics of ten
thousand people show that we work slowly. The ordinary im-
pression is apparently psychological. In order to keep warm
out of doors in winter we walk fast and this leads us to think
that we do everything rapidly.

Again, the blue sky, clear air, bright sunshine, and fresh
colors of the first day after a storm are unquestionably inspir-
ing, but does that inspiration make us work any better? May it
not lead to a nervous excitement which actually hinders our
work, by causing us to look out at the beauties of nature or to
be less concentrated in other ways? The actual figures show that,
taking the year as a whole, on dull days, especially the second
such day when a storm begins to clear, we accomplish more than
on bright days, even though we grumble about the clouds and
the dampness. A bright day certainly makes us cheerful, but its
chief helpfulness, so far as our work is concerned, is felt when
it is a change from the monotony of a series of dull days. Clouds
and rain produce exactly as much rejoicing when they succeed
prolonged clear weather of the kind that we praise so highly. In
America I have never seen so much rejoicing over a bright day
as I have seen in Turkey when the first rain fell after the long
subtropical summer with its truly superb weather. The rejoic-
ing was in part due to the fact that the coming of the rains means
good crops, but I have again and again seen exuberant joy
among people to whom the crops made no difference whatever.
I have seen Americans shout for joy because the clouds had
come, and run out into the rain to let the cool drops refresh
their faces.

80 CIVILIZATION AND CLIMATE

The questions which have just been asked and the possibili-
ties that have been suggested show how indefinite are our ideas
of the effect of climate. We understand its psychological effects
fairly well. We know little of its physiological effects, however,
except when they are extreme or unusual, or when people are
sick or are in some other pathological condition. We need to
determine how ordinary people are influenced by ordinary con-
ditions of weather. That is the purpose of our present discus-
sion. The most feasible way to do this, as has already been said,
is to take groups of people who live in a variable climate, and
measure their efficiency under different conditions of weather.
The best and fullest test of efficiency is a person's daily work.
If the subject does not know that he is being tested, so much
the better. Piece-workers in factories are doing exactly what is
required for our purpose. Accordingly, to begin with New Eng-
land, I have taken the daily records of about 300 men and 250
girls, most of them for a complete year. The records are dis-
tributed over the four years from 1910 to 1913. The 550 people
were emplo^yed in three factories in the cities of Bridgeport,
New Britain, and New Haven, in the southwestern part of Con-
necticut. In all cases the officials in charge of the factories were
most courteous and helpful in assisting me to obtain the neces-
sary data, and I wish most warmly to express my gratitude to
all concerned.

In the selection of operatives for such a purpose, various con-
ditions must be fulfilled. In the first place, they must be piece-
workers who are paid according to their work and not at a fixed
rate per day. In the second place, they must be employed in
factories where their output is not limited by restrictions im-
posed by unions, or by the fear that if they earn too much, wages
will be reduced. They must be doing work that is of essentially
the same kind every day, so that their wages will not vary much
because thev are sometimes engaged upon new and unfamiliar
tasks, or upon easy tasks at some times and hard ones at others.

EFFECT OF THE SEASONS 81

Furthermore, the same people must work steadily for month
after month, throughout the year, if possible, and without tak-
ing much time off, as is such a common practice among factory
hands. Finally, they must be working where there is abundant
incentive to steady, faithful work, where the conditions of air
and light are reasonably good, and where accurate daily records
make it possible to determine not only the daily wages of each
individual but the average efficiency per hour or per day of
standard length. The number of factories where all these condi-
tions are fulfilled is small, for they demand special types of
occupation and a high standard of management. The three
factories from which data have been obtained all meet the re-
quirements. I explained what I wanted to the superintendent
or to some other responsible official in each case. He then selected
the group or groups of operatives whom he thought proper, and
placed the figures in my hands. There was no selection on my
part, and in each case I have used all the figures, omitting only
a few obvious errors amounting to perhaps a quarter of one
per cent.

An investigation such as is here set forth may follow two
modes of procedure. One is to take a few persons and investigate
each minutely in order to eliminate all accidental variations.
The other is to take many people and get rid of the personal
variations by averages. The wages of a workman depend upon
many factors aside from the weather. One man has been scolded
by his wife because he did not earn enough last week, another
wants to buy some clothes for his little boy, and a third was
drunk last night. A sore toe may have far more influence than
any possible climatic variation. To ferret out all these accidental
circumstances is out of the question. Fortunately, they do not
occur every day, and most people work weeks at a time without
being much influenced by them. Moreover, when large numbers
of people work in different cities and during different years, the
individual circumstances neutralize one another. The dav that

82 CIVILIZATION AND CLIMATE

John Jenkins is disturbed because his boy has run away, Tony
Albano is working hard because he is going to be married. Hence,
by taking five hundred people we are able to eliminate accidental
and individual circumstances and thus to reach a reliable result.
All three of the factories whence our data are obtained make
hardware, but the work varies greatly. In one factory where
Italians are the predominant nationality, brass sockets for elec-
tric lights, and other little brass fittings are made. One group of
people was here engaged in tending machines. Some were turn-
ing out screws, others were putting pieces of sheet brass into
automatic machines which turn out perforated plates. The work
requires little skill, but much quickness and concentration. An-
other group, composed largely of Italians, was engaged in roll-
ing and drawing hot brass, a heavy and somewhat difficult kind
of work, requiring considerable strength. It is difficult because
the brass must be used hot, and hence the men must work in
abnormally high temperatures. At another factory, the one
from which the largest number of records was obtained during
three successive years, there were two main groups of men and
two of women. The girls, from sixteen to twenty years of age,
were Americans by birth, but of varied descent, being chiefly
Irish, Germans, Scandinavians, English, and other north Euro-
peans. Their work was the packing of hinges and screws, which
are first wrapped in tissue paper and then placed in pasteboard
boxes. This is a light, easy task in which dexterity and accuracy
in picking up the right number of pieces are particularly im-
portant. For the first week or two when screws are packed, the
tips of the fingers become sore, which makes the work proceed
slowly. If a girl is changed from packing hinges to packing
screws, her wages fall off for a time, but such changes are not
frequent, and do not appreciably influence our figures. The men
at this factory were of all ages, and were of the same races as
the girls. They were engaged in grinding and buffing the hinges.
The first operation is hard, heavy work. The hinges are held

EFFECT OF THE SEASONS 83

upon rapidly revolving emery wheels in order to grind them to
a smooth surface. The other operation, buffing, is similar except
that it is easier, for the hinges after being ground are polished
upon rapidly revolving cloth buffs covered with emery dust. In
the third factory, the operatives were of north European de-
scent, almost all being native-born. Practically all, both girls and
boys, were young, only a few being much over twenty years of
age. The older girls leave to be married, and the boys, who are
comparatively few in number, go elsewhere to find harder and
hence better paid work. The work consists of the preparation
of armatures and other wire coils for electrical purposes. Some
operatives wind the wire upon rapidly revolving spools. Others
put together the various parts of an armature. The work is
light and not tiresome. It requires much dexterity and accuracy.
Strings have to be tied at particular spots, pieces of paper must
be inserted, the machines must be stopped when the right point
has been reached, and little ends have to be grasped and inserted
in their proper places. Taking our three factories together, the
work ranges from the hardest to the lightest. It is of many
kinds, requiring different degrees of strength and skill. The
wages depend not only upon the amount of work completed, but
upon the number of pieces rejected. In other words, the wages
represent not only speed, but accuracy.

Let us now turn to the actual performance of the operatives.
This is summed up in Figure 1. The four upper solid lines repre-
sent the work done week after week, each year from 1910 to
1913. In Figure 1 the work of only about 410 people has been
used. The rest have been omitted because the figures are not
complete for a whole year. In only one case has there been a
deliberate omission of figures which cover an entire year. That
was the Italians who draw hot brass and hence are subject to
abnormal conditions of temperature. The method of procedure
has been to find for each working day the average hourly wages
for each group of operatives. Hourly wages have been used in-

Jan Feh Mck Apr May Jirte. Jify Avy&pt Oet Nw.D**.

zo

Figure 1. 'I'hc Effect of the Seasons on Factory Operatives in
Connecticut (solid lines) and at Pittsburgh (dotted lines)

EFFECT OF THE SEASONS 85

stead of daity, so as to make it possible to compare half-days
with whole. If part of the operatives were absent on any par-
ticular day, they were simply omitted, and the average for the
rest was taken. When the daily averages had been found, they
were averaged together by weeks. In doing this, a half-day, such
as the Saturdays in summer, was given only half as much weight
as a whole day, and days when part of the operatives were absent
or when the machinery was shut down for a while, were given a
correspondingly smaller weight. Thus allowance is everywhere
made for irregularities in the number of employees and the length
of time that they work. The final process consisted of combining
the different groups. In order that each individual may have the
same importance, all the figures have been reduced to per-
centages. In this way if a girl earned a maximum wage of twelve
cents an hour, it is called 100 per cent, while if a man's maxi-
mum wages were thirty cents, this sum also is called 100 per cent.
Thus the variations in the wages of the girl and the man have
the same weight in our final computations. Because of the enor-
mous amount of work which would have been entailed, it was
not possible to reduce the wages of each individual to percent-
ages, but only those of each group. Had it been possible to work
out each individual's wages separately, the results shown in our
curves would probably have been more striking than is now
the case.

In Figure 1 the height of the curves indicates the efficiency
of the operatives at various seasons for four successive years.
The fifth curve, heavier than the others, is the average of the
preceding four. Turning to the upper line, we see that in early
January, 1910, the efficiency of about 60 factory operatives in
Bridgeport was 88 per cent as much as during the week of
maximum efficiency that year. By the middle of the month it
had fallen to 86 per cent. Later it rose fairly steadily to 96
per cent at the end of April. Then it dropped a little, rose still
higher in June, and fell off distinctly during the summer, but

86 CIVILIZATION AND CLIMATE

not so low as in winter. During the autumn it rose steadily until
early November, when it reached the highest point of the year,
after which it fell rapidly. In the same way each curve may be
traced week by week. I shall return to them shortly. Meanwhile,
it would be advantageous for the reader to look them over and
draw his own conclusions, picking out the features which are
common to all, and noting those which show different degrees
of intensity from year to year.

In Figure 1 it will be noticed that the solid lines never reach
100 per cent. This is partly because they have been smoothed,
and partly because they have been corrected to compensate for
the increased efficiency due to practice. The process of smooth-
ing, as everyone knows, is used by mathematicians to eliminate
minor variations and thus permit the main trend of a curve to
be more apparent. It merely takes off the high points and the
low. The figures for three weeks are averaged, and the average
is used instead of the original figure for the middle week. In the
present case, and in practically all the curves in this book, the
process of smoothing has been performed twice on each curve.
If the letters a to e represent the average wages for five suc-
cessive weeks, the figure actually used for the middle week, c,
is obtained from the following equation :

a -f 2b + 3c + 2d + e
9

This process of smoothing can add nothing to a curve ; it simply
takes away the less important details. If carried far enough it
would produce straight lines.

In addition to smoothing the curves I have corrected them
for the effects of practice. The curves for 1911, and 1912, and
1913 are all based on the same factory at New Britain. When
the wages for each year are averaged, we find that those for

1912 were 1.5 per cent higher than for 1911, and those for

1913 were 1.5 per cent higher than for 1912. This means that

EFFECT OF THE SEASONS 87

constant practice caused the average employee, including both
old hands and new, to be 1.5 per cent more skillful at the end of
the year than at the beginning. Hence, from January onward,
the curve rises a little until in December it is 1.5 per cent higher
than it would be if the operatives had not grown more skillful.
To eliminate this we simply tip the entire curve, raising the
January end by three quarters of one per cent and depressing
the December end by the same amount. The fluctuations, of
course, remain unchanged.

In Figure 1, if there had been no correction, the highest and
lowest points of the upper curve would lie at the points indi-
cated by the crosses, and the other curves would be changed in
corresponding ratios, there being no change at the end of June.

Turning to less technical matters, let us consider the degree
of resemblance in the four upper solid lines of Figure 1. All are
unmistakably low in January. Then from February to June
we note a general rise, varied by minor fluctuations which differ
from year to year. At the middle or end of June all reach a
distinct maximum, although in 1912 and 1913 it is of slight
proportions. Next we have a drop during the summer, pro-
nounced in 1910 and 1911, but not at all prominent in 1912, and
scarcely noticeable in 1913. Following this there comes a series
of irregular fluctuations, differing from curve to curve, but in
each case culminating in a strong maximum at the end of Octo-
ber or the beginning of November. Six weeks later, in the middle
of December, another slight maximum is suggested, and then
all the curves drop suddenly. In the average curve the minor
fluctuations tend to disappear. They are more or less accidental,
and represent peculiar conditions which pertain to one year but
not to others. The features that have been named, however, show
no sign of disappearing. They are five in number, namely, an
extremely low place in midwinter, and a less pronounced low
place in midsummer ; a high point in June, a still higher point at
the end of October, and a hump in mid-December. Much the most

88 CIVILIZATION AND CLIMATE

variable feature is the low place in summer. This is highly sig-
nificant, as we shall shortly see.

Before we discuss the causes of the variability of the summers
let us consider the meaning of the curves as a whole. In the first
place, it is evident that, although details may vary from year
to year, the general course of events is uniformly from low in
the winter to high in the fall with a drop of more or less mag-
nitude in summer. To what can this be due? Did the factories
shut down in January, or run on part time, or decrease work
because of lack of orders, or to overhaul the machinery and so
forth? Do the high wages in October and November indicate a
special rush of orders at that time? Of course, any variations
in the way in which the factor}^ is running must be reflected in
the wages of the operatives, but in the present case this does not
apply to the main variations, although it may apply to minor
details. In neither of the two factories here considered were the
responsible heads able to offer any explanation of the peculi-
arities of the curves on the basis of facto ry management or the
exigencies of business. Both are engaged in making staple arti-
cles, the chief demand for which comes in the spring when build-
ing operations begin. There is no Christmas rush on hinges and
electric light sockets. After Christmas the factories shut down
for a few days at the beginning of the year, but that ought to
increase rather than diminish the hourly earnings. When opera-
tives are working only part time they feel the need of earning
as much as possible each hour. If part of the hands are laid off,
that would also increase the average hourly wages, for the
weaker ones would be dropped, and the average ability of those
who remain would be high.

In this connection, it is important to understand that in
these factories a man is free to work as hard as he wishes at any
time of the year. The managers have deliberately adopted the
policy of getting as much work as possible out of each opera-
tive. Overhead charges for interest, superintendence, bookkeep-

EFFECT OF THE SEASONS 89

ing, salesmen, and other outside expenses, and also the charges
for unproductive labor such as engineers, janitors, and the like
are no greater no matter how hard the productive employees
work. If the producing operatives should double their output,
most of the other expenses would scarcely increase at all. Hence,
it would not only be possible to pay double wages for double
work, but it would be profitable to the factory even if it paid
perhaps $2.50 where now it pays $1.00. In view of these condi-
tions, both factories have adopted systems whose special object
is to encourage extra exertion. In one case, part of the men work
upon what is known as the "premium" plan. The management
and the men have agreed that the various tasks shall be rated
according to the number of hours which they may fairly be sup-
posed to require. If a man performs an eight-hour task, he is to
be paid for eight hours' work, no matter whether he does it in
six hours or ten. If, however, he finishes the work in less than the
stipulated time, he goes to work at another task for the rest of
the period. For half of this time he is to be paid, while the factory
gets the benefit of the other half. For example, if an eight-hour
task is finished in six, the operative works two more hours. He
is then paid for nine hours although he has only worked eight,
while the factory gets ten hours' work and pays for nine. Thus
both are the gainers. In one case the managers made a mistake
in deciding upon the number of hours needed for a certain task.
It had never been done quickly, and no one knew how rapidly it
might be done. The man who does it soon earned ten or twelve
dollars a day, where he formerly earned perhaps two and a half
or three. Inasmuch as the management had agreed not to change
the rates, they stuck to their bargain. The task only occupies
one day each month, and the matter is not serious. Moreover,
even though the operative earns such high wages, the work
actually costs the factory less than when he was earning two
dollars and a half.

In the other factory the girls are stimulated by bonuses. That

90 CIVILIZATION AND CLIMATE

is, they are not only paid for their work, but if they do more
than is expected they are paid a bonus. For example, if a girPs
wages averaged about a dollar a day, and she did work worth
$1.20, she did not receive $1.20, but $1.25 or even $1.40. The
factory finds this worth while because so much more can be pro-
duced without any increase in charges for interest, office work,
and other overhead expenses. When this bonus system was first
introduced, it produced only a slight effect. The girls did not
seem to care about the bonuses and made little effort to get
them. Then the management realized that the parents were get-
ting the extra money, and so it made no difference to the girls,
most of whom gave their pay envelopes unopened to their fathers
or mothers. Thereafter the bonus was not put in the pay envel-
ope, but was handed out in loose change. The girls kept it and
began to work hard. In the third factory, whose figures are not
extensive enough to be used in Figure 1, but which enter into
other computations, a similar system is employed. A limit is
set for each task. If the work is performed within that time a
bonus is paid. Otherwise the operatives receive only the regular
pay, no matter how much time they spend. The introduction of
this system has increased the output of the factory enormously.
Inasmuch as the various systems of bonuses and premiums are
equally applicable at all times of the year, it seems impossible
to find in the factories themselves any reason why earnings
should be very low in January, moderately low in July, high in
June, and very high in November.

We seem forced to search outside of the factories for the
reasons for our seasonal fluctuations of wages. Such things as
panics, hard times, or strikes would certainly cause a general
change in the conditions of work, but nothing of the kind oc-
curred during the period under consideration. Moreover, such
events do not recur at the same time each year. Aside from the
seasons, the only event which recurs regularly year after year
at the same time and which is important enough to cause varia-

EFFECT OF THE SEASONS 91

tions in wages is Christmas. Its effect can be seen unmistakably
in each of the solid year-curves. In that for 1910 it appears in
the little hump which culminates during the next to the last
week in December. In the other three it comes a week earlier be-
cause this factory does not pay the week's wages on the Satur-
day of the week in question, but a week later, after there has
been time to check up the work and make allowances for that
which is poorly done. Hence, money for Christmas must be
earned before the middle of December. If there were no such
thing as Christmas, the wages would probably drop off in the
way shown by the dash line in the average curve of Figure 1.
After Christmas the wages probably drop somewhat lower than
would otherwise be the case, for there must be a reaction from
the previous effort, but it is noticeable that the wages do not
reach their lowest ebb directly after Christmas, but keep on
falling for nearly a month. Something else keeps them low. The
Christmas hump is significant chiefly because it shows unmis-
takably that an outside stimulus which applies to all the opera-
tives produces a distinct result. We may properly infer that
the other permanent features of our curves are also due to some
outside force which influences all the operatives. That force must
be connected with the seasons, and it must be far more powerful
than Christmas, for its effects are far greater. There seems to
be no recourse except to ascribe the fluctuations of the curves
to climate.

The verity of the conclusion just reached is strongly con-
firmed by comparison with other regions and other types of
human activity. Figure 2, which, for convenience, is here divided
into two overlapping portions, presents a series of curves ar-
ranged according to climate, those from regions with cold win-
ters and cool summers being at the top, and cool winters and hot
summers at the bottom. The curves range from the Adirondacks
in northern New York to Tampa in southern Florida, and in-
clude one from Denmark. With them I have repeated some of

Figured. Human g-S'S&^g-gsltiSS
Activity and the ^^^feS^^^^^^O 25

Seasons

yT

"X

.

98

I

\

A. Gain in Weight of
1200 Tubercular 90
Patients at Sara- A

1

\

nac Lake, N. Y., g5
1898-1902.

80

\

\

^^

s

^

S

/

s

J

^

\

100

98

B. Work of 160 Fac-

/

x^,

s

tory Operatives in

s

S

Connecticut, 1913.

S"

-S

Repeated from **

/

Figure 1.

V

/*

\^y

r\

98

\x

f \

'

f~*

C. Work of 60 Fac- %
tory Operatives in

f

^s

^

^

^

V

A

Connecticut, 1911. 94
Repeated from
Figure 2. G

D. Deaths in the

X

/

ti

,

/

\

90
95

State of New York,

j

/

100

1892-1906. In- D
verted. In the
scale of this curve,

v

y

1

/

-

105

100 represents the

Ss ^

- .**

\ j

average death

w

"

110

rate.

/->

E. Strength of 90 110
School Children in ^

,

/

V

/

\

Denmark, 1904-6.

X

/

^^

/

105

X

^-S

/

\

-

100

/~\

/

/

\

\

98

f \

/

^/

\

96

F. Work of 410 Fac-

-J

^

/

\

tory Operatives in

/

94

Connecticut, 1910- p

/

1913. Repeated

y

^/

from Figure 1.

\

/"

-

92

100

-\

^.

98

-

/

J

\

-

90

G. Work of 65 Girls at %
Winston-Salem, N. Q

^

/*

\

r

j

% ^

C., 1914.

^^^

*

f^

\

/

11

,Q & c * , b *ti+ J > g Figure 2b. Human

^feS<JJ3 ^ i < co O Q Activity and the

"^

**"'

/

\

^t,.

100

r\

s

y

\

98
F. Work of 410 Fac-

J

s^

/

\

96 tory Operatives in

/

*~s

^s.

s

Connecticut, 1910-

/

94 1913. Repeated

^^x

from Figure i.

/

91

~\,

/

\

9a

/

v/

\

G. Work of 65 Girls at

X

\y

\

/

\

Winston Salem, N.

V

\

/*

J

\

C., 1914.

xx

S\

/

\

/*

X

\

100

98 H. Work of 120 Opera
lives in Cot ton Fac-

^

/

*>*-*'

tories at Columbia,

^

'

./

J

96 S. C., 19 12- J 4.

^f^

94

x"

^^

/

/^^

\

I. Work of 120 Opera
tives in Cotton Fac-

w-X

v^x

v

f

tories near A u-

\,

/

/*

V

100 gusta, Ga., 1912-14.

/

\

no

J. Work of 57 Carpen-

^-\

^

ters at Jackson-

\^

x

\

S

-. -

96 ville, Fla., 1911 14.

^

^ '

\

\

^^.

^

/

c

^

\

K. Work of 400 Cipir-

\|

makers at Jackson-

\

I

ville, Fla., 1911.

e/^

J

I

^

^

1913, 1914.

\
I

v

>

"\

R'

X

J

r-

I/

'

~

A

100
98

96 L. Work of 2300 Cigar-
makers at Tampa,

\

Fla.. 1912-14.

V

**

94

r^

/

,

92

\/

*^~

'

90

NOTE. In Figures 2a and 2b the unit is a year's work of an individual. Thus "120 Opera-
tives, 1912-14," means an average of 40 per year for three years.

94 CIVILIZATION AND CLIMATE

the curves of Figure 1 for the sake of comparison. The most
remarkable feature of this series is that, although there is great
diversity of place and of activity, all the curves harmonize with
what would be expected on the basis of Figure 1.

The first curve, A, is based on the work of Lawrason Brown,
a physician who has published records of the weight gained by
patients suffering from pulmonary tuberculosis at a sanatorium
at Saranac Lake in the Adirondacks. A gain in weight in this
disease is a favorable symptom, for one of the most marked
effects of tuberculosis is to cause a wasting away of the flesh.
In the present tabulation the patients who lost weight are not
included, and a drop in the curve does not indicate loss of
weight but merely a decreased rate of gain. If the patients who
lost weight were also included, however, the form of the curve
would still be the same, according to Brown. The Adirondacks,
as everyone knows, have long cold winters, while the summers
are delightfully bracing, being warm enough to be pleasant, but
never hot enough to be debilitating. Hence, from about the first
of April to the end of September the sick people make a marked
gain. During the other six months, although they may gain more
than would be the case in their own homes, they do not find the
climate nearly so advantageous as in summer, and the disad-
vantage increases until the snow disappears. The next curve, B,
is a repetition of the Connecticut curve for 1913. That year the
winter was by no means so severe as is ordinarily the case in
the Adirondacks. Hence, the curve does not remain low quite so
long as does A, and does not begin to fall so soon. The summer,
however, was almost as cool as among the Adirondacks, and
hence there is no drop during July.

The next pair of curves represents a year with a hot summer
in Connecticut, C, and the death rate for fifteen years in the
state of New York, D. The curve for deaths has been turned
upside down, so that high places represent few deaths, that is,
high vitality corresponding to high energy in the factory opera-

EFFECT OF THE SEASONS 95

lives. In New York State as a whole the effect of the summers
is very different from what it is in the Adirondacks. The cities
swelter for a few weeks in July, and that sends the death rate
up enormously, especially among children, who are quickly
taken sick, and who either die after a few days' illness, or re-
cover. That is why the curve drops so sharply in mid-summer.
In the winter, on the contrary, although it drops almost equally
low, the maximum number of deaths per day does not come till
March, although by that time the average energy of operatives
lias risen considerably. This is because people become sick in
January and February, especially those who are elderly, and
finally die after lingering illnesses quite unlike those of children.
The death rate of other places might be used quite as well as
that of New York. The Japanese rate, for instance, is as fol-
lows, the figures being those for the ten years beginning with
1899. The figures represent percentages of the normal. Those
for the state of New York, computed on the same basis, are added
in parentheses :

January, 104 (105)

February, 108 (108)

March, 103 (109)

April, 90 (106)

May, 85 (100)

June, 84 (90)

July, 97(110)

August, 116 (104)

September, 118 (97)

October, 102 (89)

November, 96 (86)

December, 98 (93)

Here the course of events is almost the same as in New York,
but with significant differences which harmonize with the cli-
mates of the two places. Winter in Japan is less severe than in
New York, and its effects do not last so long, for the highest
mortality is in February instead of March. The Japanese sum-
mers, on the contrary, are characterized by prolonged heat, and
also by great humidity, especially during the rainy season from
July to September. At the end of this period the mortality is at
a maximum. The debilitating effect of the summer lasts so long
that November and December have a higher death rate than
May and June. The late spring is especially favorable, not only

96 CIVILIZATION AND CLIMATE

because of its own excellent character, but because it follows a
winter which is not severe enough to be highly disadvantageous.

Curves E and F represent the strength of ninety school chil-
dren in Copenhagen as measured by Lehmann and Pedersen, and
the average energy of factory operatives for four years in Con-
necticut. The Danish measurements were carried on during the
school years of 1904-1905 when sixty children were tested
weekly, and 1905-1906 when ten were tested daily. By com-
bining the two years into one and making allowance for the fact
that children grow stronger from month to month just as
factory operatives grow more skillful, we obtain curve E in
Figure 2. Since neither summer nor winter is especially severe in
Denmark the dip at the two seasons is the same. The maxima in
June and November are almost synchronous with those in Con-
necticut. The minima are both delayed six or seven weeks, but
the winter minimum in March agrees with the maximum death
rate in New York. The summer minimum ought possibly to come
in July or August, but the figures for those months are not ob-
tainable, for during that time the schools in Copenhagen have
vacation. In addition to this we should expect the Danish curve
to lag a little behind that of Connecticut because of the maritime
climate. Inasmuch as Denmark is constantly swept by west winds
from the ocean it does not so quickly grow cool in winter nor
warm in summer as does Connecticut, where the prevailing winds
are from the continental interior, which of course becomes
rapidly warm in summer and cool in winter. Thus it appears that
the strength of Danish children and the energy of factory opera-
tives in Connecticut have an almost identical relation to sea-
sonal variations of climate.

Judging by curves C to F in Figure 2 one might hazard the
hypothesis that man is subject to a seasonal rhythm which re-
peats itself wherever he goes without regard to the climate. On
this basis one would expect maxima of efficiency in June and
November in all parts of the world. In curves A and B, however,

EFFECT OF THE SEASONS 97

we have already seen that where the summers are particularly
favorable and the winters unfavorable this rhythm breaks down,
and the June maximum and summer minimum disappear. If we
go farther south to places where the winters are favorable and
the summers very hot, we find a change in the opposite direction,
for the winter minimum tends to disappear, and the summer
minimum greatly increases and shoves the two maxima more and
more into the winter until the two coalesce. This is evident in
curves G to L. These represent variations in the wages of piece-
workers in southern factories, compiled according to the method
used in Connecticut. Curve G shows the work of sixty-five Anglo-
Saxon girls in a tobacco factory in Winston-Salem, North Caro-
lina. They were pasting labels on cans. Notice how their winter
minimum comes in early May instead of June. In September the
curve drops suddenly. This is because at that time the effect of
the war began to be felt, the price of cotton fell so low that the
South was in great distress, and the sale of the goods made by
this factory began to be curtailed. Therefore the girls were not
given as much work as they could do.

Curves H and I are from cotton mills in South Carolina and
Georgia, and each represents two mills. In South Carolina the
two mills are close together at Columbia, while in the other case
they are fifteen or twenty miles apart, one being in Georgia
near Augusta, and the other across the Savannah River in South
Carolina. The operatives in all cases are of pure Anglo-Saxon
stock, chiefly of the "poor white" class. Men and women are
included in nearly equal numbers. Part are weavers, while others,
engaged in the occupations known as slubbing, spooling, and
speeding, tend machines which spin the thread and wind it on
bobbins ready for the weavers. In all cotton factories the air
in the weaving room, and to a less extent in the others, is kept at
a high temperature and a high humidity. This is necessary be-
cause when the air becomes cool, or especially when it becomes
dry, the thread is apt to break and cause blemishes in the cloth.

98 CIVILIZATION AND CLIMATE

Hence, in factories where high-grade goods are manufactured
the inside temperature is so abnormal and the amount of goods
produced depends so largely on the breakage that it is almost
impossible to obtain satisfactory figures. In the factories here
considered, however, nothing but coarse cloth is manufactured.
The breaking of the thread does little harm, and relatively slight
attention is paid to the temperature and humidity of the weav-
ing rooms. Moreover, for slubbing, speeding, and spooling, the
temperature and humidity make far less difference than for
weaving. Hence, the variations in the amount of goods produced
per person depend largely on the energy of the operatives in
watching their machines and preventing them from standing
idle because of broken threads, empty bobbins, or other acci-
dents. The exigencies of business, that is, the demands for goods,
make no difference to the operatives so far as their production
per hour is concerned, for the machines run at a uniform speed
whether the factory runs one day a week or six. The cotton mill
curves arc essentially the same as that of the tobacco factory.
In H there is a double spring maximum, due to accidental cir-
cumstances, but the true maximum would probably come about
the end of April. In I the spring maximum comes still earlier,
that is in mid-April, as is appropriate to a place so far south.
The autumn maxima, on the other hand, come later than in
Connecticut, one being in early December and the other toward
the end of November.

The work of carpenters in Jacksonville, as shown in curve J,
is different from anything else that is here considered because it
is performed out of doors. The fifteen men per year whose
records are here used were engaged in making the same kind of
repairs time after time. A careful record of the hours that they
spend is kept, but the number varies greatly on account of the
weather. If it rains they cannot work. Summer is the rainiest
period, but that does not tend to diminish the amount of work
done per hour. In fact it increases it. The rain comes in hard

EFFECT OF THE SEASONS 99

showers, and while it is falling the men rarely try to work, and
the time is not reckoned. When the rain is over they work better
than before because the air is cooler, although still far from
being cool. In winter, on the contrary, from December to March,
the rain is a pronounced hindrance. It often comes in the form
of a drizzle, and the carpenters try to keep on working while it
is falling. Moreover, after the rain the wood is wet, there is apt
to be a chilly wind, the hands feel numb, and everything is op-
posed to great efficiency. Yet in spite of this, more work per hour
is done in February, the worst winter month, than in May, June,
July, or August. If these men were at work in well-protected
sheds which were heated on the occasional cool days, there is
little doubt that in December their curve would reach a maxi-
mum higher than that now reached in November, while even if
the following months were not still better, they would at least
show no pronounced drop.

The lower two curves, K and L, represent the work of cigar-
makers at Jacksonville in northern Florida and Tampa in the
southern part of the state. Those in Jacksonville were mostly
Cubans, nearly two thirds being negroes, and the rest of Span-
ish descent. At Tampa only a handful of negroes is included,
but a large sprinkling of real Spaniards is found among the
Spanish Cubans. The curves for the cigar factories are com-
piled on a different basis from the others. The reason is that
there are no definite hours. The factories are open twelve hours
a day, usually from 6 a.m. to 6 p.m. The operatives saunter in
as they please, provided they do not come later than 8 a.m., and
leave when they choose, although an attempt is made to let no
one depart before 4 p.m. While at work they sit close together
at tables, and talk volubly except when a hired reader is vocifer-
ating the news from a Spanish newspaper. At some time in the
morning they go out for a lunch, but are rarely gone as much as
half an hour. Otherwise they stay at their work till it is finished.

Since there are no fixed hours, we cannot measure the exact

100 CIVILIZATION AND CLIMATE

earnings per hour, as we have done in other cases, but only the
earnings in proportion to the time that a man might have
worked if he had chosen to do so. In other words, we measure
partly the actual capacity for work, and partly the inclination
to work. In general the two seem to vary together, but the work
of the New York State Commission on School Ventilation has
shown that during short periods of high temperature the ca-
pacity may remain unimpaired, while the inclination declines.
In the practical work of life a lack of inclination is almost worse
than a lack of capacity.

During the warmer half of the year the possible working time
in the Florida cigar factories may be properly reckoned as
eleven and a half hours. In winter, however, the light at morn-
ing and evening is not adequate for the somewhat exacting work
of cigar-making. Therefore the men are not allowed to begin so
early as in summer, nor to work so late. The exact time depends
on the degree of cloudiness as well as the height of the sun. The
factory managers say that in December the working time is cur-
tailed an hour and a quarter or more for the month as a whole.
In order not to make the winter production appear unduly large,
I have reckoned that during the shortest week -not month

o

the working time is an hour and nine minutes, that is, 10 per
cent less than in summer. Before and after that date it steadily
increases to the solstices, when it reaches the normal. Thus we
get the lower curve for Tampa. It drops low in summer and
rises to a single maximum in winter. At Jacksonville the varia-
tions in the length of the working day on account of light are
less than at Tampa because a lower grade of cigars is made, and
hence the men are allowed to work under less favorable condi-
tions of light. Inasmuch as the exact effort of dark mornings
and evenings cannot be determined, I have drawn two lines at
each end of the curve. The lower shows the wages if no allow-
ance is made for light, and the upper if the full Tampa allow-
ance is made. The actual truth lies between the two. For our

EFFECT OF THE SEASONS 101

present purpose this uncertainty makes no difference, since in
either case we have the summer minimum and winter maximum
which all our other studies would lead us to expect in this lati-
tude.

The exigencies of business have more effect on the work of the
cigar-makers than on that of the other operatives employed in
Figure 2, but they do not determine the main fluctuations of the
curves here used. In some cigar factories, to be sure, if business
is slack the employees are often not allowed to make more than
half or two thirds the usual number of cigars. For this reason I
have omitted two factories whose figures I worked up, but whose
curves I finally found to be almost wholly controlled by the
supply and demand of the business. In the three factories which
were finally used, however, that is, one at Jacksonville and two
at Tampa, the operatives are only rarely placed on a limit. It
is too expensive, especially where high-priced cigars are made,
for four cigars a day have to be allowed to each man for
"smokes." Each man smokes his full number, if not more, no
matter whether he makes one hundred cigars or two hundred.
The rush season for cigars begins in June or July and becomes
increasingly intense until about the middle of November, by
which time most of the Christmas orders have been received.
Business is dullest in January and February. The operatives,
however, know nothing about this, except as they see that men
are taken on or discharged. The frequency of changes in the
number of employees makes the cigar-maker's life hard, and
accounts for much of his proverbial shiftlessness.

Another thing which affects the wages of cigar-makers is the
dampness of the air. During the warm, damp days so character-
istic of the Florida summer, the tobacco is very pliable and
easily worked, while on dry, winter days its brittleness causes it
to break so that the work is hampered. If it were not for this
the difference between summer and winter would be intensified.

The most striking proof of the effect of the seasons is yet to

102 CIVILIZATION AND CLIMATE

be recorded. It consists of a series of data corresponding to those
of the Connecticut factories, but based on the work of operatives
in a large factory engaged in making electrical apparatus at
Pittsburgh, Pennsylvania. The employees whose wages were
investigated were employed in winding wire coils, assembling
the parts of motors, and other similar operations which demand
accuracy and speed. The admirable way in which the records of
this company are kept renders the figures of great value, but
lack of time and funds has made it necessary to limit the present
investigation to monthly, or, in 1912, bi-weekly averages of
hourly earnings. For this reason the resulting curves, which
have been inserted as fine dotted lines in Figure 1 (page 59), are
smoother than those of Connecticut where the daily earnings
have been utilized. The number of piece-workers on which these
Pittsburgh curves are based is shown in the following table:

1910. Approximately 950 men and girls in winding section.

1911. Approximately 750 men and girls in winding section.

1912. Twenty-seven girls, winders; 42 men, tinners, black-
smiths, painters. In this case all the operatives were especially
steady hands who worked throughout the year. In the few cases
where they were absent, interpolation has been resorted to.
Hence this year's curve is more reliable than the others which
are based on all the operatives in a given section or in the whole
factory without regard to whether they worked steadily.

1913. Approximately 7000 men and girls in the entire fac-
tory.

The general form of the curves for Pittsburgh and Connec-
ticut is obviously the same. In 1910 notice the deep dip in Janu-
ary, and the moderate drop in summer. The next year, 1911,
presents quite a different aspect. Because of the hot summer,
the depressions in January and July are almost equally deep,
the difference between the highest and lowest points is less than

EFFECT OF THE SEASONS 103

in most years, and the autumn maximum does not rise high above
that of May or June, as is usually the case. The curves for 1912
both show a deep depression in winter which lasts unusually
long. During the summer, on the contrary, there is not so great
a decrease in efficiency as during the previous two years. Finally,
in 1913, both curves rise almost steadily from midwinter to late
fall, with only a slight drop in summer.

The agreement between the curves for Connecticut and Penn-
sylvania is far too close to be accidental. At Pittsburgh, just
as at the other factories, variations in the total number of em-
ployees form an accurate measure of the demand for work, but
these by no means vary in harmony with the actual production
per operative. Often the average amount of work done by a
given group of individuals, or by all the piece-workers, declines
when the number of operatives increases, but quite as often the
reverse is true. Hence the conditions under which the factories
are run do not explain the variations in wages. Moreover, it
stands to reason that the same irregular variations would not
occur season after season in an electric factory in Pittsburgh
and in brass and hinge factories in Connecticut 400 miles away
unless all were under the same control. The only common con-
trolling factor which varies in harmony with the curves of
Figure 1 is the general character of the seasons. This is essen-
tially the same in both places.

We have now seen that from New England to Florida physi-
cal strength and health vary in accordance with the seasons.
Extremes seem to produce the same effect everywhere. The next
question is whether mental activity varies in the same way. Leh-
mann and Peclersen made a series of tests of the ability of school
children in addition. Their general conclusion is that mental
work varies in the same way as physical, but reaches its highest
efficiency at a lower temperature. This agrees with the investi-
gations of a few other scientists, and with the general conclu-
sions of the world as summed up in the old adage, "No one is

104i CIVILIZATION AND CLIMATE

worth a tinker's dam on whom the snow does not fall." Before
we can accept this, however, tests are needed on a large scale.
The most feasible method at present seems to be by means of the
marks of students in such schools as West Point and Annapolis.
There the young men live an extremely regular life with a mini-
mum of outside distractions. Their recitations are graded with
great severity and regularity, and a given subject is often
taught six days in the week. The marks are handed to the heads
of departments at frequent intervals and are posted where the
students can see them. No class is taught in divisions of more
more than ten or twelve, so that every student has a full oppor-
tunity to show how well he is prepared. In order to avoid all
chance of favoritism the instructors do not keep the same divi-
sion month after month, but change every few weeks. Altogether
it would be hard to devise a system which more thoroughly
eliminates the human and accidental factors. As an instructor at
West Point put it: "We are not really teachers. We are just
put here as officers to see whether the cadets have studied their
books, and to decide how many marks to take off." This is pre-
eminently true in mathematics, where the solution of a problem
is either right or wrong and can be marked accordingly.

When I broached my plan to the superintendents at the two
academies, it was received with much interest, and every facility
was placed at my disposal. I take this opportunity to express
my warm appreciation of their courtesy. Some of the instructors
were commissioned to see that the proper records were available.
The marks of individuals were, of course, not necessary. The
various marks for each day or week were merely added, and
averaged. The data here employed embrace the following: (1)
The weekly averages in mathematics for the first-year or enter-
ing class at Annapolis for the six academic years beginning with
1907-1908 and ending with 1912-1913. These classes recite six
times a week. (2) The daily marks for the first-year class in
English at Annapolis for the year 1912-1913. This class recites

EFFECT OF THE SEASONS

105

four times a week. (3) The daily marks in mathematics for a
year and a half for the classes entering West Point in 1909 and
1910. Recitations are held six days a week. The classes at
Annapolis average about 220 in number and those at West Point
about 120. The entire number of students whose marks have
been used is between seventeen and eighteen hundred, but as
some of the marks cover a period of a year and a half, the total
is equivalent to about nineteen hundred students for a single
year.

All these marks have been combined into the three lower
curves of Figure 8. Before discussing them a few words should

Oct Nov. Dec, Jan Feb. Mar flpr. Hay June

470 Operative*

Z40 Student* in
Mathematics at
West Point

22O Students in
English at
Annapolis

MOOS
Mathematics at
Annapolis

Figure 3. Seasonal Variations of Mental Compared with
Physical Activity

106 CIVILIZATION AND CLIMATE

be said as to the method of preparation. The systems of mark-
ing at the two academies are quite different. At Annapolis the
department of mathematics tries to keep the average as nearly
uniform as possible. If the instructors discover that the average
is rising or falling they mark more severely or leniently to
counteract it. At West Point, on the other hand, the marks
regularly begin high at the opening of the term and fall steadily
toward the end. There is no attempt to keep them at a uniform
level, but the instructors merely mark harder and harder or
give more and more work as time goes on. Both systems tend
to mask the effect of the seasons. The influence of the deliberate
attempt to keep the marks at a uniform level at Annapolis is
largely overcome by using a series of six years. The irregulari-
ties of one year counteract those of another except where special
circumstances such as vacations interpose a disturbing element
at the same time each year. In the English department at Annap-
olis there is less stringency about keeping the marks at a uni-
form level, and those of a single year show clearly the normal
seasonal trend. At the end of the year, however, I have omitted
the two weeks before examinations because there was then a
sudden spurt accompanied by abnormally high marks. Other-
wise all the Annapolis marks without exception have been em-
ployed in computing the curves of Figure 3.

At West Point it has been necessary to eliminate the effect of
the steady fall. The method is the same as in the correction for
increasing practice. In order to eliminate the effect of such
things as football games, holidays, examinations, reprimands,
or other circumstances which clearly have nothing to do with
climate, I have omitted all the days whose marks fall more than
10 per cent above or below what would be expected at that par-
ticular date. Omissions of this sort are such a common procedure
in astronomical and physical measurements that the mathema-
tician requires nothing more than a mere mention of what has
been done. To the layman it may seem that they are of great

EFFECT OF THE SEASONS 107

importance. In reality they rarely alter the general form of the
final curves, for exceptionally high figures balance exceptionally
low. In the second curve of Figure 3 the effect is slight except
upon the first weeks in January. There the minor maximum
which occurs just after the Christmas recess is only about half
as large as it would be if no data were omitted. At Annapolis it
is not necessary to omit the days of special events because the
marks are not subject to such wide fluctuations. It is interesting
to notice that the classes in mathematics there are influenced
by the vacation, which comes at the end of January, just as at
West Point. The English marks, on the contrary, are uninflu-
enced, probably because English is an easier subject than
mathematics. Moreover, as it is taught fewer days per week,
and hence has less weight in determining the final marks for the
work of the whole year, the students do not devote so much
energy to it.

By this time the reader has doubtless interpreted Figure 3
for himself. The upper line is the standard average curve for
factory operatives in Connecticut. It is the same as the average
curve of Figure 1, except that it begins in September instead of
January. It is placed here to permit a comparison of the physi-
cal work with mental. The curves of mental activity all resemble
it in having two main maxima, in fall and spring. At West Point,
where the climate is essentially the same as in Connecticut, the
mental maximum in the fall comes about ten days later than the
factory maximum, while the spring maximum comes two and a
half months earlier. Both occur when the mean temperature is
a little above 40 F. At Annapolis the maxima are, as it were,
pressed toward the winter. The fall maximum in English, to be
sure, begins early in November, but lasts till the middle of De-
cember. Since it represents the work of only a single year, it is
less important than the curve of mathematics, whose fall maxi-
mum does not come till the first half of December. The spring
maxima of both curves come in the middle of March. At Annap-

108 CIVILIZATION AND CLIMATE

oils, just as at West Point, the time of best work is when the
mean temperature is not far from forty degrees.

Summing up the matter, we find that the results of investiga-
tions in Denmark, Japan, Connecticut, Pennsylvania, New
York, Maryland, the Carolinas, Georgia, and Florida are in
harmony. They all show that except in Florida neither the
winter nor the summer is the most favorable season. Both physi-
cal and mental activity reach pronounced maxima in the spring
and fall, with minima in midwinter and midsummer. The con-
sistency of our results is of great importance. It leads to the
belief that in all parts of the world the climate is exercising an
influence which can readily be measured, and can be subjected to
statistical analysis. It justifies us in going on with confidence
to ascertain exactly what effect is produced by each of the
climatic elements, such as temperature, humidity, and pressure.

CHAPTER V
THE EFFECT OF HUMIDITY AND TEMPERATURE

HAVING seen that both physical and mental energy vary
from season to season according to well-defined laws, let
us now investigate the special features of seasonal change which
are most effective. Temperature is far the most important, but
before considering it, let us discuss those of minor importance.
One of these is light. Many students have ascribed great influ-
ence to sunlight, and to its variations from season to season, or
from one part of the world to another. For example, C. W.
Woodruff, an army surgeon, has written an interesting book
on The Effect of Tropical Light on White Men. Its main thesis
is that the backwardness of tropical countries is due to excessive
sunlight. The actinic rays at the blue end of the spectrum,
especially those beyond the limits of vision, possess great chemi-
cal power, as is evident from the fact that by their aid photo-
graphs can be taken even when no light is visible to the naked
eye. Such rays, when they fall upon the human body, are thought
to stimulate the cells to greater activity. At first this is bene-
ficial : if it goes to excess the cells apparently break down. The
process is analogous to the ripening of fruit. A moderate change
in the green tissues produces the highly favorable condition of
ripeness: more brings on decay. Thus while the return of the
light after the winter of the temperate zone may be beneficial,
excessive light may be highly injurious.

So far as our factory operatives are concerned, no effect of
light is to be discerned in the South, while in Connecticut it is
at best only slight. The heavy line next to the bottom in Figure

110 CIVILIZATION AND CLIMATE

1 (page 34) shows that from mid-September to the middle of
November the amount of work increases, although the days are
growing shorter. This is exactly opposite to what would be
expected if the shortness of the days were of primary impor-
tance. Moreover, in June when the days are longest we find a
sudden drop. If the length of the days had much to do with the
matter, there is no reason why more work should be done in
November than in June. Nor should we find that a shortening
of the days during September is accompanied by the same kind
of increase in efficiency which is seen in March when the days,
although of the same length as in September, are growing longer
instead of shorter. For all these reasons we assign only slight
importance to variations in the amount of light. Nevertheless,
some effect can apparently be detected. Compare the two lower
curves of Figure 1. In spite of the low efficiency occasioned by
the winter's cold, the curve of work begins to rise sooner than
does the curve of temperature which is placed below it. The first
appreciable lengthening of the days in January may cause this
by its cheering and stimulating influence.

The line of reasoning applied to light applies also to the pos-
sibility that the variations of the curve of work depend on the
extent to which people are shut up in the house. Obviously, this
has nothing to do with the two maxima in November and May,
nor with the minimum in July. In November people's houses
have been shut up for a month, more or less, while in May and
July they are wide open, or at least as wide open as they ever
are. The extremely low minimum in January, however, is prob-
ably due at least in part to the necessity of shutting up the house
in winter. In October the weather becomes so cold that people
begin to shut up their houses ; they live in stuffy, unventilated
quarters, and fail to take exercise in the open air. By the middle
of November this has had time to produce an effect which natu-
rally becomes more and more marked as the weeks go on. This
would harmonize with the decline of energy from November to

HUMIDITY AND TEMPERATURE 111

the middle of January. In January, however, the decline ought
not to cease if it is due chiefly to confinement within the house.
It ought to continue until about the middle of March, for not
till that time do people in Connecticut begin to let in the outside
air, and not even then to any great degree. As the curve of work
has risen distinctly by that time, some other factor must inter-
vene, presumably the increase of light to a slight extent, and the
rise of the temperature to a larger extent.

A third factor to be considered at this point is the relative
humidity of the atmosphere. A sharp distinction must be drawn
between the humidity of the outside air and that which prevails
within doors. Physicians, students of factory management,
school superintendents, and many other people have repeatedly
discussed the supposed harmful effects of the dry air in our
buildings during the winter. A much more fully attested fact is
the harmful influence of great humidity during hot weather.
We are more conscious of this than of the harm arising from
excessive dryness. This does not necessarily mean that the total
effect is worse than that of dryness, however, for hot, humid days
are much rarer than the winter days when the air in our houses
is drier than that of the majority of deserts.

So far as our curves of work are concerned, humidity does not
seem to be responsible for the fluctuations except as it is influ-
enced by temperature. In other words, the average humidity of
the outside air from season to season does not vary in such a way
as to cause maxima in May and November, and minima in Janu-
ary and July. The average humidity of the outside air in No-
vember and in January is not greatly different. Nevertheless, the
inside humidity is probably an important factor in causing the
low efficiency of midwinter.

The relation of work and humidity among the factory opera-
tives of Connecticut is illustrated in Figure 4. There the year
has been divided in three parts: (1) winter, (2) spring and
autumn, and (3) summer. In each part all the days having a

112

CIVILIZATION AND CLIMATE

given humidity have been averaged together, and the smoothed
results have been plotted. The heavy, solid lines represent what
I believe to be the true conditions when other disturbing elements
are removed, while the dotted lines show the actual figures. In
winter the dampest days are unmistakably the times of greatest
efficiency. We may shiver when the air is raw, but we work well.

50 55 60 65 70 75 80 85 90 95 100% Rel. Hum.

100%

99

98

97

100

99

100%

99

98

97

Figure 4. Relative Humidity and Work in Connecticut

a. Winter. b. Spring and Fall. c. Summer.

This is partly because in winter the dampest third of the De-
cembers, Januaries, etc., averages nearly 2 F. warmer than the
driest third. Moreover, moist air at any given temperature feels
warmer than dry and hence is less likely to cause people to
overheat their houses.

In the spring and fall, when the temperature ranges from
freezing to 70, with an average of about 50 F., the best work is
performed with a relative humidity of about 75 per cent. In
other words, neither the dry nor the wet days are the best. The

HUMIDITY AND TEMPERATURE 118

summer curve is the most complex of the three. It rises first to
a maximum at 60 or 65 per cent, then falls, and once more rises
to a higher maximum. The first maximum seems to be due to
humidity, the second to temperature. A hot, damp day is un-
questionably debilitating. The majority of the dampest days
in summer, however, are comparatively cool, for they accom-
pany storms. The coolness counterbalances the humidity, and
people's efficiency increases. Hence, we disregard the right-hand
maximum and conclude that with an average temperature of 65
to 70 a relative humidity of about 60 per cent is desirable.

The most unmistakable feature of the curves as a whole is
that they show a diminution of work in very dry weather. This
presumably has a bearing on the low level of the curve of energy
in winter. At that season the air in our houses ought to have a
humidity of 60 or 65 per cent, but most of the time the figure
is only 20 or 30. On very cold days the percentage is still lower.
For instance, if the outside air has a temperature of 14 F.
( 10 C.) and contains all the moisture it can hold, which is
usually not the case, its relative humidity when it is warmed to
70 F. will be only 12 per cent. Even on days when the outside
humidity rises to 100 per cent and the temperature is 40, the
air in an ordinary steam-heated house has a relative humidity
of only 85 per cent, which is far below the optimum. Apparently,
this extreme aridity is debilitating. It probably dries up the
mucous membranes in such a way as to increase our suscepti-
bility to colds. In this way it may be an important factor in
causing February and March to have the highest death rate of
the year. There has been a good deal of discussion as to the
actual importance of atmospheric humidity, and no small
amount of disagreement. We shall return to the matter later
when we study health.

While the effects of light, of closed houses, and of excessive
dryness explain part of the fluctuations of the curve of work,
they have little bearing on any season except the winter. An-

114 CIVILIZATION AND CLIMATE

other matter which may be suggested in this connection is vaca-
tions. These, like many other conditions of human life, are
largely seasonal. Do people work fast in the fall because they
have been rested by vacations? In professional occupations and
in business this certainly seems to be the case, but not among
factory operatives. As a rule such people do not take summer
vacations. They usually stop work at irregular intervals, or
else after Christmas when many factories shut down or work on
part time for a few days to prepare for the new year. The form
of our main curve, however, shows that neither at this time nor
in summer do vacations produce any appreciable stimulating
results. If they were the cause of fast work, the curve ought to
be the highest within a few weeks after the people return to work,
but this is not the case. During the vacation period of July and
August the amount of work is moderately low, and in early
January, after the Christmas break, very low. At the end of
August it begins to increase, and increases steadily for two and
a half months. The maximum in November is so long after the
vacation period that it can hardly have anything to do with it.
What has just been said has an important practical appli-
cation. There is a common idea that people need vacations in
summer. Of course there are strong arguments for this, since
pleasant recreation is then possible out of doors. Nevertheless,
the need is apparently greater in winter than in summer. To
meet this it is probably wise that work should be light during
the winter. Already, as everyone knows, many factories run on
part time during the first few weeks of the year, and now we see
that there are strong physical reasons for this. Another impor-
tant suggestion afforded by our curves is this : If the operatives
of a factory, or people engaged in any other kind of work, are
to be speeded up, the time to do it is when nature lends her aid.
To speed up at the end of January is analogous to taking a
tired horse and expecting him to win a race. Later in the year,
however, during the spring, especially in May, people may ap-

HUMIDITY AND TEMPERATURE 115

parently be pushed to the limit, and will not suffer, because their
energies are naturally increasing. This is still more the case in
October and early November. After the middle of November
pressure may produce important results, as we see at Christmas.
Nevertheless, the chances are that if continued it will produce
undue exhaustion, followed by a serious reaction. Possibly the
nervousness of Americans is due partly to the fact that although
we relax somewhat in summer, we keep ourselves at high pressure
through the winter when the need of relaxation is greatest.

Turning now to temperature, we see that in Figure 1 (page
84) the lower curve, showing the march of temperature through
the year, and the Connecticut curve just above it are similar in
many ways. Both are low in midwinter. From February onward
they rise together until about the middle of June. Then the effi-
ciency curve falls while the other goes on rising, a condition
which fully accords with ordinary experience. The fall of the
efficiency curve begins when the average temperature has risen
to about 68. When the temperature stops rising, the work stops
falling, and then remains nearly steady through July. At the
end of July the mean temperature has fallen to about 71. Dur-
ing the succeeding period of favorable temperature the two
curves disagree, for the amount of work goes up while the tem-
perature falls. When the average temperature falls below 48,
however, and begins apparently to be unfavorable to physical
exertion, the curve of work turns downward. Thereafter, if we
omit the Christmas hump and use the dotted line, the tempera-
ture and the amount of work decline together until they reach
the lowest point in January. It is worth while once more to call
attention to the somewhat surprising fact that in southern New
England, contrary to our ordinary opinion, low temperature
seems to be much more injurious than high.

This by no means indicates that high temperature is favor-
able. Let us consider the effect of the high temperatures of the
four successive summers shown in Figure 1 (page 84). Compare

x/fJvy&Jvty August Sept

Figure 5. Average Weekly Temperature During the
Summers of 1910-13 in Connecticut

HUMIDITY AND TEMPERATURE 117

the summer dip in the Connecticut curve, that is, the area below
the horizontal lines, with the heavily shaded areas of Figure 5,
which shows the average temperature each week during the four
summers from 1910 to 1913. The black portions indicate weeks
having an average temperature night and day of over 73. The
size and distribution of these periods of extreme heat are in close
correspondence with the amounts by which the curves of Figure
1 drop below the horizontal lines during the summers. This is
illustrated in the following little table. The line marked "defi-
ciency in work" indicates the amount by which the efficiency of
the operatives diminished because of the hot weather, that is,
the area below the horizontal lines of Figure 1. The year when
the diminution was greatest is reckoned as 100 and the others
in corresponding ratios. The other numbers show the area of
the heavy black shading in Figure 5 and represent the intensity
and duration of the hot weather. Here, too, the year of maxi-
mum heat is represented by 100, and the others by proportional
values.

YEAR. 1910 1911 1912 1913

Deficiency in work, 58 100 8 2

Severity of heat, 52 100 50 34

In each case 1911 stands highest, 1910 next, and then 1912
and 1913. In 1911 the heat not only was extreme, but lasted
long, three weeks at one time and two at another. The death rate
for July, 1911, in Massachusetts was 50 per cent greater than
in the preceding June. In 1910 the hot weather was not so severe,
it lasted four weeks instead of five, and was divided into three
parts instead of two. In 1912 the number of hot weeks was the
same as in 1910. One was extremely hot, but the rest were not
bad. Moreover, they did not come together, and the last was
separated from the others by three cool weeks during which
people had time to recover, which was not the case in 1910.

118 CIVILIZATION AND CLIMATE

Finally, 1913 was a very mild year with only two extreme weeks
which were separated by three moderate weeks.

An examination of Figure 5 makes it clear that only the ex-
treme weeks are harmful. Thus 1911 was a truly terrible summer
and 1913 a delightful one. Yet during 1911 the temperature
remained above 69 for only eight weeks while in 1913 it re-
mained above that figure for twelve weeks. Thus it appears that
if the average temperature does not rise above about 70, and
if the noon temperature rarely exceeds 80, the physical capac-
ity of European races in the United States does not suffer any
serious diminution. A slight further rise however only four or
five degrees produces disastrous consequences. A single week
of such weather does no great harm, but when several weeks
come together people rapidly become weakened. The weakening
is greater than appears in our diagrams, for during hot spells
many of the operatives, particularly the girls, stop work entirely
or stay at home in the afternoon. Those who remain are the
stronger ones, and naturally their wages are higher than the
general average. Moreover, in 1911 the heat was so intense that
the factory shut down for two or three days. Thus, if allowance
is made for these facts, the difference which a few degrees make
between two summers such as 1911 and 1913 becomes even more
pronounced. The full effect of a hot summer, especially when it
is very damp, may be gauged by the death rate in Japan (page
95). September is there 18 per cent worse than the average, in-
stead of 3 per cent better as in New York.

The relation between the temperature and the amount of work
in winter during the four years under discussion is not so pro-
nounced as in summer, but can easily be detected. The hot sum-
mer of 1911 was followed, as frequently happens, by an uncom-
monly cold winter. The reason for both is the same. Usually hot
weather in New England is commonly due to the movement of
heated air from the interior toward the coast, particularly from
the southwest to the northeast. Cold winters are due to a similar

HUMIDITY AND TEMPERATURE 119

transportation of air from the interior, this time from the north-
west. The interior of a continent, as is well known, cools off very
rapidly in winter and becomes hot rapidly in summer. When
these conditions are carried from the interior to the coasts, they
bring to New England what climatologists call a continental
climate instead of the more maritime climate which otherwise
prevails.

The effect of the cold winter of 1911-1912 can easily be seen
in the curve for 1912 in Figure 1 (page 84). That year the aver-
age temperature where the factories are located was 19.0 for
the first five weeks compared with an average of 32.7 for the
three other years whose curves are given. For the next five weeks
the temperature was 24.4 compared with 35.3. The effect of
this is seen in the low position of the 1912 curve of work far into
the spring. The fact that the energy of the operatives remained
low after the temperature began to rise suggests that the eifect
of extreme conditions may last long after more normal condi-
tions begin to prevail. The same thing is suggested by the fact
that after the summer of 1911 the curve of work does not rise
so high in November as in the preceding May. During each of
the other three years the November maximum is higher than its
predecessor. Although a single winter and a single summer are
not enough to prove that the effect of extreme conditions does
thus persist for many months, they suggest that a long stay in
an adverse climate may produce results which last for years. In
spite of a previous statement, it appears that our plan of es-
caping from possible extreme heat by taking summer vacations
in the mountains or at the seaside is wise. Equally wise is the
growing habit of getting away from the severe cold for a while
in winter. The only trouble is that those who most need such a
change are rarely the ones who get it. If people could spend the
summer on the Maine coast, the winter in Georgia, and the rest
of the year in New York, they ought to be able to do the best

Mon. Tve.

Thvr Fri.

J

4-

5

^^
-**.

e

I00

98 <&

97

Figure 6. Effect of the Days of the Week on Piece- Workw

1. 60 Men, April-July, 1912.

2. 60 Men, August-November, 1011.
. 49 Girls, 1912.

4. 31 Men, 1012.

5. 14 Girls, 1912.

6. 24 Men, 1912.

7. 60 Men, January-March, 1910.

8. Weighted Average of Nos. 1-7, or approximately MO People for On* Year*

HUMIDITY AND TEMPERATURE 121

kind of work at all seasons almost without the necessity of a
vacation.

The effect of temperature ma}' be shown in more ways than
have yet been presented. Let us determine how fast people work
on days having various temperatures, no matter in what month
they occur. The very cold days, of course, all come in winter,
but may be in December, January, or February. The very hot
days come anywhere from May to September, while days with a
temperature of about 50 occur in almost every month of the
year.

The method can be illustrated by taking all the Mondays,
all the Tuesdays, the Wednesdays, and so forth, and averaging
the work of each day of the week. This has been done for 230
people. The results are shown in Figure 6, which is inserted to
show exactly how our results are obtained, and how necessary it
is to have a large number of people. We are striving to separate
the effects of one single condition from those of a vast number.
We start with the wages of individuals which vary from day to
day for hundreds of reasons wholly unconnected with the day
of the week or the weather. The variations are so great that even
if a man is influenced by the approach of pay-day, for example,
we should probably not be able to detect it if we merely looked at
his wages for a month or two. Therefore we average all the
people of a department together, and obtain results such as
appear in Figure 7. This shows the actual wages in percent-
ages of the maximum which were earned by 170 people divided
into five departments during five weeks in January and Febru-
ary, 1913. There is little uniformity in the different lines. Where
one goes up the other goes down. Yet closer examination shows
that in at least four out of the five departments the wages during
the last two weeks were a little larger than during the earlier
weeks. The variations of the different curves are in part due
to the persistence of individual vagaries which have not yet been
averaged out, and in part to conditions affecting whole depart-

122

CIVILIZATION AND CLIMATE

ments. For example, a foreman is cross one day and good-
natured the next ; a belt breaks and delays work ; or some of the
operatives converse so much that their work suffers appreciably.
If a number of departments are averaged together these acci-

January
Monday

Monday Monday

February
Monday Monday

7. Variations in Daily Wages. Five Departments
(170 people) at New Britain, Conn.

HUMIDITY AND TEMPERATURE 123

dents, as well as those which pertain to individuals, disappear,
but not until a great many people are considered.

To find the effect of the days of the week, we take data such
as are illustrated in Figure 7, select all the Mondays, Tuesdays,
and so forth, and average each day. This gives the curves of
Figure 6. Here we begin to detect a certain degree of uniformity,
although the accidents and peculiarities of each department
are still in evidence. On the whole, however, the curves are higher
at the end of the week than at the beginning. All, to be sure, are
irregular, and the two lower not counting the heavy line
slope in the opposite direction to the rest. The fact that the re-
maining five slope in the same direction shows, however, that
these different people in different factories and during different
years were subject to a common influence. Finally, we average
all the departmental curves, giving each a weight proportional
to the number of operatives. Thus we obtain the heavy lower
line of Figure 6. This is still irregular, for although 230 people
are included, all influences other than that of the days of the
week are not yet eliminated. Nevertheless, the wages clearly in-
crease toward the end of the week. If the operatives were paid by
the day instead of by the piece, this would probably not be the
case. They would work slowly at the end of the week by reason
of being tired. With the piece-workers, on the contrary, other
considerations are dominant. If they work a trifle slowly on
Monday, they can make it up tomorrow. On Tuesday they can
be slow and make it up on Wednesday, but a few who fell behind
on Monday are beginning to work harder. So it goes from day
to day until on Friday and especially Saturday many feel that
their earnings for the week are insufficient, and hence make an
extra effort. In some cases this may not be true, as in the curve
next above the average curve. Yet it remains a general truth,
and the lower curve of Figure 6 is a concrete expression of the
fact that in the factory under discussion there is a difference
of at least 2 per cent between Monday and Saturday. Possibly

15 20 25 30 35 40 45 50 55 60 65 70 75

- 4-1

Figure 8. Human Activity and Mean Temperature

A. 300 Men in Two Connecticut Factories, 1010-13.

B. 196 Girls in One Connecticut Factory, 1911-13.

C. One Man (P) in Denmark, June-December, 190(1.

D. One Man (L) in Denmark, June- December, 1906.

E. 880 Cigar-makers in Factory B at Tampa, Fla., 1913.

F. 400 Cigar-makers in Factory A at Tampa, Fla., 1913.

G. 3 Children Typewriting in New York, 1905-6.

H. 880 Cigar-makers in Factory B at Tampa, Fla,, 1912.

I. 1560 Students in Mathematics and English at West Point and Annapolis, 1909-191&

NOTE. All the curves except G and 1
every case is reckoned as 100.

The maximum ill

HUMIDITY AND TEMPERATURE 125

the real difference is greater, and is obscured by other circum-
stances. In the cigar factories of Florida it rises to a far greater
value, for the Cubans are much disinclined to work after a holi-
day. Not only are about 10 per cent of the operatives absent on
Mondays, but those who are present come so late or are so in-
disposed to work that they accomplish only about 80 per cent
as much work as on other days. This is so important a matter
that allowance for it has been made in computations where in-
dividual days rather than weeks are concerned. The figures for
each day of the week for 780 men at Tampa are as follows :
Monday, 81.9 per cent; Tuesday, 98.7 per cent; Wednesday,
99.8 per cent; Thursday, 100 per cent; Friday, 98.3 per cent;
and Saturday, 97.9 per cent. The other days are reckoned as of
equal weight, but the figures for Monday have been increased
in the ratio of 82 to 100.

By the employment of a method similar to that used with the
days of the week we obtain the curves shown in Figure 8. These
are based on varying numbers of people, from one to over 700.
Yet all show the same general character. With the exception of
G and H, which are distinctly the least reliable, the physical
group all reach maxima at a temperature between 59 and 65.
Even the two less reliable curves reach their maxima within the
next four degrees. All the curves decline at low temperatures,
that is, on the left, and also at high. The irregularities at the
extreme limits are largely due to the fact that there the number
of days is so small that exact results cannot be hoped for.

Figure 8, with the brief statements which accompany the
respective curves, tells the whole story so plainly that it scarcely
seems worth while to amplify it. Several points, however, may
well be emphasized. For instance, below a certain temperature,
which varies from curve to curve, a further reduction does not
seem to produce much effect. People apparently become some-
what hardened, or else the conditions within the warmed houses
do not change much in spite of a change in the outside air. An-

126 CIVILIZATION AND CLIMATE

other noticeable thing is that the curve for girls has greater
amplitude than that for men in the same region. Part of this
is due to the inclusion of the group of Italians, already referred
to, who are engaged in drawing hot brass and hence are benefited
by the coldest kind of weather. Even if they were omitted, how-
ever, the girls' curve would still vary more than that of the men.
This seems to indicate that either because of their sex or because
of their age, girls are more sensitive than men.

Another point brought out by the curves is that as we go to
more southerly climes the optimum temperature of the human
race becomes higher. It is important to note, however, that the
variation in the optimum is slight compared with the variation
in the mean temperature of the places in question. For instance,
in Connecticut the optimum seems to be about 60 for people of
north European stock. This is about ten degrees higher than
the mean temperature for the year as a whole. In Florida, on the
other hand, the optimum for Cubans is about 65, which is five
degrees lower than the mean temperature for the year at Tampa.
In other words, with a difference of twenty degrees in the mean
annual temperature, and with a distinctly northern race com-
pared with a southern, we find that the optimum differs only
about 5 F. This seems to mean that for the entire human race
the optimum temperature probably does not vary more than
ten or fifteen degrees.

We have not yet pointed out all the important matters sug-
gested by the curves of Figure 8. Above the optimum the curves
in general begin to decline quite rapidly, but then cease to do
so and at high temperatures are not so low as would be expected.
This is largely because in hot weather many operatives, espe-
cially the girls and the Cubans, do not feel like work, and so
stay away from the factories. Those who come in spite of the
heat are the strongest and most efficient. Naturally, their aver-
age wages are higher than those of the ones who stay away, and
hence the general level of our curves is too high in the portions

HUMIDITY AND TEMPERATURE 127

based on the hottest weather. The mental curve, however, falls
off very rapidly at high temperatures. This is because the stu-
dents are obliged to be present on hot days just as on others.
They must recite whether they wish or not. Hence, their curve
is more reliable than the others. In this connection some ex-
periments carried on by the New York State Commission on
Ventilation are of interest. In an attempt to determine the most
favorable conditions of ventilation the Commission placed sev-
eral groups of persons in rooms where the temperature and hu-
midity were under exact control, and measured their strength,
mental activity, food consumption, and other conditions. The
experiments lasted six or eight hours a day, and each set of
subjects was tested for several weeks. Three temperatures were
used, namely, 68, 75, and 85. No appreciable effect upon
strength could be detected, nor upon mental activity, and vari-
ous other functions. This is probably because the experiments
were not sufficiently prolonged. That is, the subjects were in the
experimental rooms only a third or a quarter of each day, and
hence their condition did not have time to change appreciably.
Although the subjects did not lose in actual strength, however,
their inclination to work declined at high temperatures even
within six or eight hours.

Thus far we have been dealing with large bodies of people. It
is peculiarly important to find that no matter how small the
number, the same relation to temperature is discernible. One of
the curves in Figure 8 shows the speed and accuracy of three
children who wrote upon the typewriter a few stanzas from the
"Faerie Queen" or a page from George Eliot daily for a year,
and weekly for another year. Their records were kindly placed
at my disposal by Prof. J. McK. Cattell. I have corrected them
for the effects of practice, and have combined speed and ac-
curacy in such a way that each has the same weight. At one
period, for some unknown cause, the efficiency of the children
declined greatly for two months or more. If this were eliminated

128 CIVILIZATION AND CLIMATE

their maximum would come at a lower temperature than now
appears, probably not much above 60. In the curves of indi-
viduals, we are fortunate in having careful tests made by two
psychologists, Lehmann and Pedersen, at Copenhagen. They
tested their own strength daily with the dynamometer, and their
curves, copied directly from their monograph, are before us.
One is uncommonly regular with a maximum at 64<. The other,
less regular, has its maximum at 59. The agreement of Danish
curves based on single individuals with New England curves
based on hundreds is highly important.

The last thing to be considered in Figure 8 is the mental curve
at the bottom. It is based on so large a number of people, and
is so regular, that its general reliability seems great, although
I think that future studies may show the optimum to be a few
degrees higher than is here indicated. It agrees with the results
of Lehmann and Pedersen. Furthermore, from general observa-
tion we are most of us aware that we are mentally more active
in comparatively cool weather. Perhaps "spring fever" is a
mental state far more than a physical. Apparently people do
the best mental work on days when the thermometer ranges from
freezing to about 50 that is, when the mean temperature is
not far from 40. Inasmuch as human progress depends upon a
coordination of mental and physical activity it may be that the
greatest total efficiency occurs halfway between the mental and
physical optima, that is, with a mean temperature of about 50.

Curves such as those of Figure 8 are not peculiar to man
alone. They are apparently characteristic of all types of living
creatures. To begin with plants, many experiments have deter-
mined the rate of growth of seedlings at various temperatures.
The commonest method has been to grow different sets of seed-
lings in large numbers under conditions which are identical ex-
cept in temperature, and then to measure the average length of
the shoots. In all cases growth is slow at low temperatures, in-
creases gradually with higher temperatures, reaches a maximum

HUMIDITY AND TEMPERATURE

129

like that of man, and then falls off quickly. The course of events,
however, is not always so regular as here indicated. The curve
of wheat, for example, as worked out by MacDougal is given in
Figure 9. The peculiar double maximum there seen appears in
each case where careful tests are made. It seems to be due to
some inherent quality of the plant, and is of especial interest in

Figure 9. Growth of Wheat at Various Temperatures
After MacDougal

The figures on the left indicate growth in mm. during 48 hours

our present study because we shall soon come upon an analogous
case in man. When many species are averaged, such irregulari-
ties disappear, and we obtain the curve at the bottom of Figure
10, which has been prepared by MacDougal on the basis of his
own measurements and others given in such works as Pfeffer's
Physiology of Plants. Many of the lower plants, such as marine
algae, have their optima at lower temperatures than those here
indicated, and the same is probably true of Arctic species. On
the other hand, certain low algae which grow in hot springs must
have their optimum at a temperature above that of ordinary
plants. These differences are immaterial. We are now concerned

130

CIVILIZATION AND CLIMATE

only with the fact that so far as plants have been measured, their
response to temperature resembles that of man.

Apparently, we have to do with a quality which pertains to
all kinds of living beings, and is presumably an inherent char-
acteristic of protoplasm. The nearest approach to pure proto-
plasm is found in unicellular organisms whose bodies show only
the beginnings of differentiation into parts having separate
functions. The infusoria furnish a good example. One of these,
paramoecium, has been carefully studied by L. L. Woodruff. His

Menial
Energy

Mental and
rriysiccilEnefqy
Combined

Physical
Energy

Absorbhon of
Oxyoen by
Crayfish

Raie of fission
of infusoria

Growth
j/a/rts'

Figure 10. Mean Temperature and Vital Processes in Plants,
Animals and Man

HUMIDITY AND TEMPERATURE 131

original purpose was to determine whether it was possible for
this organism to keep on reproducing itself without conjugation
for any great length of time. Under the conditions of nature the
small motile cells often spontaneously develop a median cell wall
and ultimately divide into two new individuals, thus reproducing
the species. This process, however, does not go on indefinitely,
for when two cells come in contact they fuse with one another,
and then begin a process of fission which, like the other process,
ends in two individuals. Thus we have two types of reproduc-
tion, asexual and sexual, which apparently give rise to the same
kind of paramcecia. Woodruff's purpose was to determine
whether asexual reproduction can persist indefinitely, or whether
it leads in time to extinction. He has shown that if the media of
nutrition contain a sufficient number of elements, paramcecium
can reproduce itself indefinitely by the asexual method. Between
May 1, 1907, and May 14, 1914, he had carried his cultures
through 4417 generations without conjugation. In the spring of
1924 the paramoecia were still thriving after about fourteen
thousand asexual generations. In the course of this work he has
found that the rate of cell-division is an accurate test of the
conditions under which protoplasm exists. For example, when
extracts from nephritic kidneys or certain other diseased organs
are added to the nutrient solution, even though they are present
in such small quantities that they cannot be detected by chemical
analysis, they make their presence evident by a falling off in the
rate of fission.

One of Woodruff's most important lines of work has been to
test the relation of his infusoria to temperature. From many
experiments he finds that their activity corresponds closely to
van't Hoff's law of chemical activity. According to this well-
established law, chemical reactions of most kinds at ordinary
temperatures become nearly three times as active with every rise
of 10 C. Even in inorganic chemical reactions, however, and
far more in those of the living cell, there is a distinct limit where

132 CIVILIZATION AND CLIMATE

this rule breaks down. This limit forms the optimum of the
species. At higher temperatures the degree of activity declines,
and finally death ensues. On the basis of these conclusions,
Woodruff's data permit us to draw the second curve from the
bottom in Figure 10.

The next higher curve shows the amount of oxygen absorbed
by the common crayfish at various temperatures. The most ex-
tensive work on this subject appears to have been done by Brun-
now. The facts here given are taken from the summary by Putter
in his Vergleichende Physiologie. The amount of oxygen ab-
sorbed by an animal is an excellent measure of its physical
activity. When supplemented by measurements of the amount
of carbon dioxide given off, and of the speed with which certain
other metabolic or katabolic processes take place, it gives a
true picture of the animal's general condition. Apparently, these
various processes follow van't Hoff's law just as do the growth
of plants and the cell-division of the infusoria. The optimum in
the three cases does not vary greatly, that for plants being
about 86, for paramoecium 83, and for the crayfish 74 F.

Physiologists are not yet fully agreed as to the cause of the
phenomena shown in these curves, although there is little doubt
as to the general facts that they imply. One hypothesis may be
briefly stated. According to Putter's summary, the most prob-
able explanation is that activity goes on increasing according
to the ordinary chemical law until it becomes so great that the
organism is not capable of absorbing the necessary oxygen.
That is, at a low temperature the creature easily gets what oxy-
gen it needs, and gives it out again in the form of carbon dioxide
or of other oxidized products which remove the waste substances
from the body. As the temperature rises, the normal increase in
chemical activity takes place, the animal is still able to get rid
of all its waste products, and thus its life processes are strength-
ened. With a further rise of temperature a change sets in. The
chemical processes which break down the tissues of the body be-

HUMIDITY AND TEMPERATURE 133

come still more active, but the supply of waste products to be
eliminated by oxidation becomes so great that they cannot all
be removed. This is because in every organism there is a distinct
limit to the amount of oxygen which the creature can mechani-
cally convey to different portions within a specified time. If the
supply of oxygen is not sufficient to oxidize all the waste prod-
ucts, some of these will remain in the system. They act as
poisons. Their first effect is to diminish the organism's activity.
If they accumulate to too great an extent death ensues.

The discussion of this hypothesis must be left to the physiolo-
gists. They must decide whether the hypothesis which explains
the curves of cold-blooded animals and plants is also applicable
to warm-blooded animals. There can be little doubt, however,
that variations in the rate at which metabolism takes place in
the human body play a part in the variations in efficiency which
we are here studying. The researches of Thomson illustrate the
way in which we are beginning to discover the truth. In Man-
chester, England, from April to July, 1910, and again in March,
1913, he measured the percentage of CO2 given off in the breath
of four individuals under different conditions of temperature,
humidity, and pressure. From his figures, given in the Man-
chester Memoirs, I have compiled the following tables :

I. PERCENTAGE OF CC>2 EXHALED BY FOUR PERSONS UNDER
DIFFERENT CONDITIONS OF TEMPERATURE

Temperature, 50-51 52-53 54-55 56-57 58-59 60-61 62-63 64-65 98

4<7r 4 * r2 4t66 4<T1 4i61 4<41 4<88 4 * 62 4 -

II. PERCENTAGE OF C(>2 EXHALED BY FOUR PERSONS UNDER
DIFFERENT CONDITIONS OF HUMIDITY

Relative Humidity, 70-75% 76-80% 81-85% 86-90%
Percentage of CO 2 , 4.75 4.60 4.60 4.45

134 CIVILIZATION AND CLIMATE

The interpretation of these tables is difficult, and I can merely
offer a suggestion. An increase in the proportion of CO2 ex-
haled from the lungs obviously indicates an acceleration of the
metabolic processes which break down and consume the bodily
tissues. This liberates energy which may manifest itself in at
least three ways and possibly more. It may give rise to heat
which is used to maintain the body at the normal temperature ;
it may be used to accomplish physical or mental work; and it
may cause an excess of heat which gives rise to further metabo-
lism of a harmful nature. In the first part of the table the per-
centage of C02 is comparatively high at the lowest temperature
recorded by Thomson, and decreases with only slight irregu-
larity till the thermometer reaches 62 F. This is close to the
temperature which we have found to be the optimum. Below that
point the increased metabolism is probably needed to keep the
body warm. At higher temperatures increased production of
CO2 is again apparent. This perhaps means that too much
chemical activity is taking place, and that toxic substances are
accumulating in the way suggested by Putter. At the optimum,
according to this interpretation, the body does not have to use
an undue portion of its strength in keeping warm, nor is it
injured by too great stimulation. Thus it is in the best condition
for work.

The second part of the table shows that in the driest weather
which England enjoys, metabolism is more active than in wet
weather. Perhaps part of this is due to the fact that in dry
air the body loses water and is cooled by evaporation, and hence
requires more heat than in wet air of similar temperature. There
is more to the matter than this, however, but further measure-
ments are needed before an adequate explanation can be offered.
All that can be done here is to point out the fact that in man, as
in the lower organisms, activity varies according to tempera-
ture. This is evident in Figure 10, where the dotted upper line
is the curve of mental activity, while the accompanying solid

HUMIDITY AND TEMPERATURE 135

line shows the conditions if all accidental irregularities could
be removed. The third line in the same way represents the physi-
cal activity of both men and women in Connecticut. I have not
used the figures from the South because they are not quite so
reliable as those from Connecticut. Finally, the second line from
the top shows physical and mental activity combined, each
being given the same weight. It may be taken as representing
man's actual productive activity in the things that make for a
high civilization. The resemblance of the human curves to those
of the lower organisms is obvious. In general, the lower types of
life, or the lower forms of activity, seem to reach their optima
at higher temperatures than do the more advanced types and
the more lofty functions such as mentality. The whole trend of
biological thought is toward the conclusion that the same laws
apply to all forms of life. They differ in application, but not in
principle. The law of optimum temperature apparently controls
the phenomena of life from the lowest activities of protoplasm
to the highest activities of the human intellect.

CHAPTER VI
WORK AND WEATHER

THE effect of a given climate depends on two primary fac-
tors. One is the character of the seasons as expressed in
averages such as are furnished by our weather bureaus. The
other is the changes from day to day, that is, the weather. The
boy quoted by Mark Twain was nearly right when he defined
the difference between weather and climate as being that "Cli-
mate lasts all the time and weather only a few days." Two
climates may be almost identical in their seasonal averages, and
yet differ enormously in their effect on life, because in one the
change from day to day is scarcely noticeable, while in the
other there are all sorts of rapid variations. The old Irish-
woman who was driving her pigs to market in a pouring rain
did not realize it, but she gave expression to a truth of the
greatest importance, when a friend pitied her for being out in
such weather, and she replied, "Indade it's bad, but sure it's
thankful I am to have any kind of weather."

The changes from one day to another depend largely upon
our ordinary cyclonic storms. In such storms the barometer
goes down and then up ; the wind changes in direction and ve-
locity ; the air becomes humid, clouds gather, rain usually falls,
and then clear skies and dry air prevail; the temperature also
changes, often rising before a storm and falling afterward,
although the exact sequence depends on the location of a region
in respect to the ocean and to the center of the storm ; the daily
range of temperature also varies, for in damp or cloudy weather
the nights do not become so cool nor the days so warm as when

WORK AND WEATHER 137

the air is clear. To understand the influence of the weather all
these conditions must be investigated. Most of them, however,
appear to be of relatively slight importance when considered by
themselves. For instance, Lehmann and Pedersen could find no
appreciable effect of the pressure of the atmosphere except
where low pressure prevails a long time. The decrease in effi-
ciency at such times, however, is probably due more to pro-
longed cloudiness and its attendant circumstances than to the
barometric conditions. My own work leads to the same result.
The curves of efficiency compared with pressure are so contra-
dictory that it does not seem worth while to publish them. The
same is true of the range of temperature from day to night, and
of the direction and force of the winds. I have no doubt that all
these matters are important, and that some day their effect will
be worked out. In general, however, their influence is exerted
indirectly through changes in temperature and humidity. In
hot weather a great range from day to night is unquestionably
highly favorable, but at ordinary temperatures it seems to make
no special difference, except through its effect upon the mean
temperature.

As to the winds, Dexter, in his book on Weather Influences,
shows that they produce a marked effect upon the nerves, as is
indicated by the unruliness of school children in Denver when
high south winds prevail. Part of this is doubtless due directly
to the wind, but the unseasonably high temperature and extreme
dryness which accompany it are probably more important. Yet
we are all conscious of the effect of a steady high wind. Some
people are stimulated. I have seen a small boy, who was usually
very quiet, climb to the top of a tall tree when a violent wind
came up, and swing in the branches, singing at the top of his
voice. For a while such stimulation is probably beneficial, but if
continued day after day it makes people excitable and cross. A
striking example of the effect of a prolonged wind is seen in
eastern Persia in the basin of Seistan. During the summer, from

138 CIVILIZATION AND CLIMATE

June to September, the so-called "Wind of One Hundred and
Twenty Days" blows so violently from the north that in the
oases trees cannot grow except under the lee of high walls. The
acrid wild melon, which ripens its beautiful little green and
yellow fruit in the desert, does not spread its slender branches in
all directions after the common fashion of plants. The gales
crowd the branches into a sheaf which points so uniformly in
one direction, a little to the west of north, that it can safely be
used as a compass. When Europeans have to endure this wind
they say that it is one of the most trying experiences imaginable.
Not only does it render them irritable, but it deadens their ini-
tiative and makes them want to stay idly in the shelter of the
house. The natives, although possessed of many good qualities,
are inert and inefficient even in comparison with their fellow
Persians who live farther to the north and west. On the whole,
we may probably conclude that occasional short-lived gales and
frequent light or moderate winds are beneficial, while long
periods either of steady calms or of gales are depressing.

Aside from the conditions of weather already mentioned,
there are two whose effect appears plainly when curves are con-
structed according to the method described above. One is the
change of temperature from one day to another, and the other
is the character of the day as to clouds and sunshine. In con-
sidering changes of temperature from one day to the next, we
deal with the mean temperature for each day and not with the
extremes. A change of as much as 15 is rare. Suppose that the
thermometer stands at 60 at sunrise, rises to 80 by two
o'clock in the afternoon, then falls rapidly to 50 at sunset and
to 40 by midnight. Suppose also that the next day the tempera-
ture is 40 at sunrise, rises a little above 55 during the day,
and falls again to 45 at night. The two days would be very
different, and we should speak of them as being marked by a very
great change of temperature, a difference of 40 within ten
hours. Yet the average of the first day would be about 64 and

WORK AND WEATHER 139

of the second 49, a difference of only 15 in the mean tempera-
ture.

On the basis of this supposition the reader can estimate the
importance of the various degrees of change indicated in Figure
11. At the left the curves show the average efficiency on days
when the temperature has fallen ; in the middle are the days with
no change ; and at the right are the days characterized by a rise.
Taking only the two upper curves, those for men and girls in
Connecticut factories, the resemblance is striking. When we con-
sider the heterogeneous character of the original materials the
resemblance is still more important. The men's curve is based
on 120 men at Bridgeport in 1910 and 1911, and on 180 men
at New Britain in 1911, 1912, and 1913. The girls' curve is
based on 196 girls at New Britain in 1911, 1912, and 1913, and
on 60 girls at New Haven in 1913 and 1914. Even when the girls
and men are working in the same factory, there is no reason,
aside from the weather, why their wages should be high on the
same day. The chief difference between the two curves is that
the one for the girls varies more than that for the men, and
reaches its maximum slightly farther to the right. Apparently,
here, just as in the case of mean temperature, the girls because
of their age or sex are more subject to the influence of the
weather than are the men, and hence their curve dips deeper.

Let us now interpret the upper curves, beginning at the
middle. There they fall to their lowest level. This means that
when the temperature of today is the same as that of yesterday,
people work more slowly than after a change, no matter whether
the change is upward or downward. A variable climate is there-
fore highly desirable if people are to be efficient. Perhaps the
most surprising feature is that the lowest point of the physical
curve, and a depression of the mental curve, C, come not at 0,
but at 1. The zero point is low, lower than any point of the
physical curves except 1. Hence, our conclusion as to the
injurious effect of uniform temperature is justified, but that

Fall of Temperature
-14F -12 -10 -8 -6 -4 -2

+2

Rise of Temperature
"H +6 +8 +10

100%

98

100

99

97

100

99

Figure 11 Human Activity and Changes of Mean Temperature
from Day to Day

A. 300 Men in Two Connecticut Factories, 1910-13.

B. 259 Girls in Two Connecticut Factories, 191 1-18.

C. 460 Students in Mathematics and English at West Point and Annapolis, 1900-1911

D. 760 Cigar-makers at Tampa, Fla., in Winter (October-March), 1912 and 1013. Factory A.

E. 400 Cigar-makers at Tampa in Winter, 1913. Fac'- B.

F. 400 Cigar-makers at Tampa in Summer (April-September), 1913. Factory B.

G. 380 Cigar-makers at Tampa in Summer, 1912. Factory A.
H. 880 Cigar-makers at Tampa in Summer, 1913. Factory A.

A

B

-M

100

100

WORK AND WEATHER 141

does not explain the curious dip at 1 . The repetition of the
same phenomenon in each of the three upper curves, and a simi-
lar occurrence at 2 and 3, respectively, in the two curves
for the winter in Florida strongly suggest that we are con-
fronted by a peculiarity which pertains to man as a species, in
the same way that a double optimum of mean temperature per-
tains to wheat as shown in Figure 9. Possibly, a slight fall in
temperature causes people to shiver, as it were, and only when
the fall is slightly larger is the circulation of the blood so stimu-
lated as to increase the activity of the various organs. In the
South it may be that people's blood is more sluggish than in the
North, so that the reaction due to cooler weather does not
follow quite so soon, and hence the period of shivering is not
over until the fall in mean temperature amounts to more than
about 3. I do not assert that this is so, but it is the only expla-
nation that comes to mind.

To go on with our interpretation of the physical curves, a
slight rise of temperature seems to be favorable, but beyond
that the favorable effects of increased heat, which are strong in
cold weather, are neutralized by the unfavorable effects in warm
weather. In fact, our personal experience tells us that even
when the heat is not extreme, a sudden rise may make us un-
comfortable and lazy, as often occurs in the spring. In spite of
this, however, a rise is in general better than uniformity. When
the temperature falls, on the other hand, a distinct stimulus is
received, provided the fall amounts to as much as 4. The best
effects are seen with a fall of from 6 to 9 with girls and of
7 to 11 with men. Here again the implication is that men are
on the whole less sensitive than girls. An extreme drop is not
so favorable as one of more moderate dimensions, especially for
the girls. Taking the physical curves as a whole, the greatest
amount of energy would be expected in climates where the mean
temperature first rises 2 or 3 a day for a few days and then
drops 4 to 8 a day. If the changes are greater than this, the

142 CIVILIZATION AND CLIMATE

effect is still stimulating, but not so beneficial as under the more
moderate conditions. If there is practically no change, on the
contrary, the level of efficiency lies within the low central de-
pressions of our curves, and is less than under either of the other
conditions.

Mental work resembles physical, but with interesting differ-
ences. When the temperature falls greatly, mental work seems
to suffer more than physical, and declines as much as when there
is no change. It receives a little stimulus from a slight warming
of the air, but appears to be adversely affected when the air
becomes warm rapidly. This last statement, however, must be
qualified. The physical curves are based on the complete year,
and the conditions of summer have an opportunity to balance
those of winter. The results show the net effect for all seasons
combined. The mental curves, on the other hand, do not include
the summer vacation, which lasts from the middle of June to the
first of September at West Point, and from the middle of May
to the first of October at Annapolis. If this were included, the
effect of a pronounced lowering of the temperature would be
more noticeable than at present, for such a lowering is naturally
more stimulating in July than in January. In another respect,
also, the curve of mental efficiency needs modification. It is based
on figures from two climatic provinces, namely, southern New
York and Maryland. The great decline at times when the tem-
perature rises rapidly is due largely to conditions in Maryland,
where the hot days of the spring are much more debilitating than
in New York. The students belong to a race which has never
learned to endure sudden heat. Hence they feel it strongly. If
allowance is made for the two conditions just mentioned, the
mental curve will approach much more closely to the physical.
A drop of temperature amounting to 8 or more will appear
more stimulating than now seems to be the case, and a rapid
rise will not seem so harmful. Hence, the general conclusion for
both physical and mental activity will be essentially the same.

WORK AND WEATHER 143

It may be summed up thus : Taking the year as a whole, uni-
formity of temperature causes low energy ; a slight rise is bene-
ficial, but a further rise is of no particular value ; the beginning
of a fall of temperature is harmful, but when the fall becomes
a little larger it is much more stimulating than a rise ; when it
becomes extreme, however, its beneficial qualities begin to decline.
This conclusion must, of course, be appropriately modified ac-
cording to the season. A cold wave in January is very different
from one in July. In our curves we have given January and July
an opportunity to neutralize one another. They have not done
so. This means that after all allowances have been made for the
seasons, the total effect of cold waves is decidedly beneficial, and
of warm waves slightly so. Frequent changes, therefore, are
highly desirable.

Let us pass on now to the Florida curves. Here we find a
curious difference between summer and winter which is not easy
to understand. Let us leave that for the moment, however, and
consider only the two winter curves. Their general resemblance
is marked. The differences at the extremities are not important
because the number of days there concerned is very small. It must
be remembered that the two curves are from independent and
rival factories. The position of any particular point in either
curve depends upon a number of days scattered irregularly
through the months from October to March. Aside from a genu-
ine effect of climate, there seems to be no possible way in which
400 men in one factory in 1913 could be made to work so that
their curve would be the same as that of 380 men in another
factory in the two years 1912 and 1913. Here, as in Connecticut,
West Point, and Annapolis, we are apparently dealing with a
peculiar quality which is inherent in the human species.

One of the Florida curves, E, is low at 0, while the other is
medium. This means that days when there is no change of
temperature are not particularly favorable. At plus 2 to plus
4, however, both are fairly high, which indicates that a moder-

CIVILIZATION AND CLIMATE

ate rise of temperature is favorable. A further rise seems to be
harmful. The effect of a slight fall of the thermometer has al-
ready been discussed. A further fall is beneficial. The most
notable thing about curves D and E is the maximum from 4<
to 7. It comes at about the same place as the mental maxi-
mum, and is similar to the Connecticut maximum except that the
people in the far South do not seem to be able to stand such
extreme changes as do those in the North. In fact, it seems most
significant that the Connecticut men, who are the strongest of
our various groups, are most stimulated by a strong change of
temperature. The Connecticut girls come next, but, being less
sturdy, they do not profit quite so much by rigorous conditions.
The mental curve is largely determined by Annapolis, and as the
climate there is less severe than in Connecticut, the students seem
to feel more keenly the effects of extreme changes, although they
are stimulated by those of moderate dimensions. The same is
still more true of the people of Florida in winter. Finally, during
the summer the Floridans are stimulated by a slight drop of
temperature, provided it is not enough to make them feel chilly,
but enough to start their blood in motion. A greater drop makes
them feel cold, while even the slightest rise of temperature in
their long monotonous summer is unfavorable.

We are ready now to sum up our results. The outstanding
point is that changes of temperature, provided they are not too
great, are more stimulating than uniformity, while a fall is more
stimulating than a rise in the latitudes now under consideration.
The effect of changes depends largely upon the degree to which
people are inured to them. When they are weakened by a long hot
period like that of the Florida summer, even a slight cooling of
the air brings relief and activity, provided it does not go so far
as to make people feel chilly. When the same Floridans become
wonted to the somewhat sterner, albeit mild air of their winter,
the first effect of a lowering of the temperature may be to make
them shiver, but soon they are stimulated, and work fast. They

WORK AND WEATHER 145

are not so tough, however, as to be able to get benefit from the
occasional days when really strong cold waves sweep down
upon them. On the other hand, a rise of temperature stimulates
them, unless it is of considerable severity. Farther north the
same applies except that, being tougher, the people are more
benefited by strong changes. Judging by the difference between
summer and winter in Florida, it looks as if a little hardening
would cause even the Cubans to respond favorably to changes
at least as severe as those in Maryland, thus making the left-
hand part of their curve like C in Figure 11. Taking it all in all,
the one thing that stands out preeminently is that a fall of from
4 to 7 is everywhere stimulating, provided people are accus-
tomed to it.

Man is not the only organism that is benefited by changes of
temperature. Numerous experiments have shown that plants
are subject to a similar influence. If a plant is subjected to un-
duly low or high temperature, its growth is retarded. As the
temperature approaches the optimum, the rate of growth in-
creases. When the optimum is maintained steadily, however, not
only does the increase cease, but retrogression sets in, and the
rate of growth declines. A moderate change of temperature away
from the optimum and then back again after a few hours checks
this decline, and keeps the plant at a maximum degree of ac-
tivity. Thus conditions where the thermometer swings back and
forth on cither side of the optimum are distinctly better than
where the optimum is maintained steadily. Thus it seems to be
a law of organic life that variable temperature is better than
uniformity.

The physiological process by which frequent changes of tem-
perature affect the body is not yet known. The best suggestion
seems to be that of Dr. W. B. James. It is universally recognized
that one of the most important of the bodily functions is the
circulation of the blood. The more active and unrestricted it is,
the more thoroughly is the whole system nourished and purified.

146 CIVILIZATION AND CLIMATE

Provided it does not impose an undue strain on the heart or
arteries, anything that stimulates the circulation appears to be
helpful. Changes of temperature are a powerful agent to this
end. Witness the effect of a bath, either cold or very hot. Few
things are more stimulating than a Swedish bath. An attendant
holds two hoses, one with cold water and the other with hot, and
plays them alternately upon the patient. A man goes into such a
bath with hanging head and dragging feet. He comes out with
head erect and a new spring in his walk. Apparently, frequent
changes of the temperature of the air produce much the same
effect. No one change produces so pronounced an effect as a
Swedish bath, but the succession of stimuli due to repeated
changes throughout the year must be of great importance.

Before leaving this subject, let us test the effect of changes
in still another way. Let us see what happens during an average
series of days such as make up our common succession of weather
in New England. The ordinary course of events is first a day or
two of clear weather, then a day or two of partly cloudy weather,
next a cloudy day with or without rain, and finally another
cloudy day during which rain falls. Then the sky clears in prepa-
ration for another similar series. On this basis I have formed the
six groups indicated at the top of Figure 12. At the left, the
efficiency on all clear days which follow cloudy or partly cloudy
days has been plotted, just as in another diagram we plotted
the efficiency on Mondays. Next come the clear days which follow
another clear day. If several of these follow in unbroken suc-
cession, they are all included, but a third or fourth clear day is
rare. In the next group come the partly cloudy days which
follow either a clear or a cloudy day. The great majority follow
clear days. A second partly cloudy day is much rarer than a
second clear day, and a third is still rarer. The first cloudy day,
the fifth column, includes cloudy days which follow either clear
or partly cloudy days. Finally, the sixth column includes not
only the second cloudy day, but the third and fourth if such are

WORK AND WEATHER

147

recorded. In general, this column represents days when a storm
comes to an end, while the one to the left of it represents the
time when a storm first becomes well established. The rest of the
diagram, to the right of the sixth column, is merely a repetition
of the part already described. It is inserted to show how an ideal
series of storms would repeat itself.

Figure 12 discloses some surprising facts. For instance, the
first clear day is characterized by the slowest work in the two

100%

99

100

100

99

Figure 12. T'ie Stimulus of Storms

(1) 60 Men at Bridgeport, 1910.

(2) 60 Men at Bridgeport, August, 1911-July, 1912.

(3) 170 Men and Girls at New Britain, 1913.

(4) Weighted Average of (1), (2), and (.!), Ktiual to 2PO People for One Tear.

U)

(2)

(3)

(4)

148 CIVILIZATION AND CLIMATE

upper curves and by almost the slowest in the third. Our im-
pression of the stimulus of the bright, clear air after a storm
receives a flat contradiction. It is apparently psychological,
not physical. The second clear day makes a better showing than
the first. It stands high in two curves, and low in only one. The
first partly cloudy day is high in one curve, and medium in two.
The second partly cloudy day is medium in all three. The same
is true of the first cloudy day. The last cloudy day is as sur-
prising as the first clear day. In each of the three curves it
stands highest. People work fastest at the end of a storm. In the
lower curve of Figure 12, the whole matter is summed up in a
single line. Here we see that during an average "spell of weather"
people are least efficient on the clear days ; moderately efficient on
the partly cloudy days, and on the first cloudy day ; and most
efficient at the end of a storm. We may tell ourselves that this is
unreasonable, but when we think it over, we are likely to be
aware of its truth. Before a storm we may feel depressed, but at
the end, when the rain or snow is almost over and the air begins
to have that excellent quality which makes us forget all about
it, we bend to our work with a steadiness and concentration
which are much less common at other times. Hellpach empha-
sizes this in his book on the psychological effect of geographical
conditions. We fail to appreciate it largely because the aesthetic
impressions of a beautiful, clear day are felt much more con-
sciously than are the physiological conditions which throw us
vigorously into our work. Each storm, with its changing skies,
varying humidity, and slow rise and rapid fall of temperature,
is a stimulant. Each raises our efficiency.

This ends our survey of the effect of climate upon daily work
in the eastern United States. We have considered the influence of
the seasons, of mean temperature, of humidity, of winds, of
changes of temperature from day to day, and of the character
of each day and its relation to storms. We have also seen that
although different races, or people under decidedly diverse cli-

WORK AND WEATHER 149

matic environments, are at their best at slightly different tem-
peratures, the differences are inconsiderable, and changes of
temperature are as valuable to one as to the other. The question
now arises whether the climatic effects are really of great im-
portance. In Figure 12, the stimulus of the succession of clear
and cloudy days amounts to only 1 per cent. In Figure 11
changes of temperature from day to day produce a variation
of only a little over 2 per cent, if we omit the irregular and un-
reliable extremities of the curves. In Figure 4, the maximum
effect of humidity appears to be only 3 per cent. In Figure 8,
however, the differences are greater, for the effect of mean
temperature upon the girls in Connecticut is 7 per cent. Finally,
in Figure 1, the effect of the seasons reaches nearly 9 per cent
when four years are averaged, and nearly 15 per cent for indi-
vidual years.

These figures are far from representing the full importance
of the various factors. This will readily appear from a little
consideration. In the preceding paragraph, the percentages in-
crease in proportion to two conditions, first, the degree to which
the influence of a single factor is separated from the influence
of all other factors, and second, the length of time during which
each factor is able to exert its influence. The smallest figure, 1
per cent in Figure 12, does not represent any individual factor,
unless it be cloudiness. It does not even represent the fluctua-
tions which attend an individual storm, for the days were selected
without regard to their position in a cyclonic disturbance, but
simply according to their cloudiness. The variations shown in
the curve are due to many factors, including mean temperature,
changes of temperature, relative humidity, and others of minor
importance. As no two of these are necessarily at their maximum
at the same time, they neutralize one another. Moreover, a given
condition lasts only a day in most cases, and so has no oppor-
tunity to produce any great effect. In the curve of changes of
temperature from day to day, which shows the next larger

150 CIVILIZATION AND CLIMATE

effect, a single factor is singled out. Its full force can by no
means be seen, however, for the humidity often varies in such a
way as to neutralize it. Moreover, the effects of especially low
or high temperatures may often completely overshadow any
stimulus arising from the mere fact of a change. Furthermore,
the effect of changes of temperature rarely continues more than
two days. For example, if the thermometer averages six degrees
lower on one day than on the preceding, it may happen that there
will be a further drop before the next day, but there is far more
chance that the temperature will rise a little or remain station-
ary, or fall so little that it will not be stimulating. Hence, the
effect is rarely cumulative, and the influence of each single day
must usually stand by itself. Much the same is true of relative
humidity, except that by heating our houses we artificially in-
duce long periods of great aridity. The effects of mean tempera-
ture, on the other hand, have greater opportunity to show their
full importance, though they, too, are hampered. Relatively low
or high temperatures last many weeks, which makes it possible
for the effect of day after day to accumulate. Yet our curves
by no means show the full effect, for a cold day with a mean
temperature of 30 may come in November at a time when effi-
ciency is still at its highest. It produces its normal effect, but a
single unpropitious day, or even a week, does not suffice to de-
press people's vitality to a degree at all approaching the low
limit reached after two months of cold weather. Likewise, a day
with the most favorable temperature, not far from 60, may be
sandwiched between very hot days in July, or between two cold
days in March. Hence, people will display little energy on those
particular days, and the average efficiency at the optimum tem-
perature will appear correspondingly lower than it ought.
Finally, the seasons have more opportunity than the individual
climatic elements to produce their full effect. Even here, how-
ever, the variability of our climate does not allow any special
combination of circumstances to work long unimpeded. Warm

WORK AND WEATHER 151

waves break the cold periods of winter, and cool waves come in
summer. Storms are more active in winter than in summer, and
hence their stimulus works toward overcoming the effect of pro-
longed cold. Moreover, no single season is of great duration, and
extreme conditions do not last long enough to produce their full
effect. From all this we may conclude that the total influence of
climate upon energy is much greater than appears in any one
of our curves.

The difficulty of determining the exact proportions of any
individual influence may be made clear by an example. We know
that man's power to work depends upon food, drink, sleep, and
clothing. Suppose that while he was still supplied with these in
normal quantities we were to try to measure the effect of each.
We should test his strength at stated intervals after he had
eaten his meals, or after he had had a drink. We should find out
how many hours he slept each night and compare that with his
work. We should measure his achievements before and after he
put on his spring underwear or fall overcoat. We might get re-
sults, but it is highly doubtful whether they would be as distinct
as those here discussed. We have no difficulty in measuring the
effect of food, drink, sleep, and clothing, for we can easily vary
them to suit the needs of our experiment. With climate the case
is different. We must take it as we find it, and must experiment
on people who are constantly subject to its influence. Some day
we shall test people first in one climate and then in another, but
that will be difficult because it takes a considerable time for cli-
mate to produce its full effect. Being obliged to search for the
effects of climate without being able to change them in accord-
ance with the needs of our experiment, we are in almost as diffi-
cult a case as the experimenter who should desire to determine
the effect of the amount and kind of food consumed by a group
of individuals, but who had no control over how much they ate.
They might allow him to measure what was set before them at
each meal and what remained when it was over, but they would

152 CIVILIZATION AND CLIMATE

eat as much as they liked and when they liked. He would get
results, if he did his work carefully, but they would by no means
represent the full effect of food.

The influence of climate upon men may be likened to that of
a driver upon his horse. Some drivers let their horses go as they
please. Now and then a horse may run away, but the average
pace is slow. Such drivers are like an unstimulating climate.
Others whip their horses and urge them to the limit all the time.
They make rapid progress for a while, but in the end they ex-
haust their animals. They resemble climates which are always
stimulating. In such climates nervous exhaustion is likely to
prevail and insanity becomes common. A third type of drivers
first whip their horse to a great speed for a mile or two, and then
let them walk slowly for another mile or two. They often think
that they are accomplishing great things, and they are better
off than the two types already mentioned, but they still have
much to learn. They are like a climate which has a strong con-
trast of seasons, one being favorable and the other unfavor-
able. Still a fourth kind of driver may whip his horse sometimes
and sometimes let him walk, but what he does chiefly is to urge
the animal gently with the voice, then check him a little with the
rein. By alternate urging and checking he conserves the animal's
strength, and in the long run can cover more distance and do it
more rapidly than any of the others. Such a driver resembles a
climate which has enough contrast of seasons to be stimulating
but not to create nervous tension, and which also possesses fre-
quent storms whose function is to furnish the slight urging and
checking which are so valuable in the total effect, although each
individual impulse is almost unnoticeable.

CHAPTER VII
HEALTH AND THE ATMOSPHERE*

WE have investigated the relation of the weather to human
energy. Let us do the same for health. In a previous
chapter we saw that variations in the death rate in New York
and Japan and the gain in weight among tubercular patients
exhibit seasonal fluctuations much like those of workers in fac-
tories. As a means of measuring health, deaths are more im-
portant than disease for two reasons. First, experience has
shown that when several years are averaged together, the death
rate is an almost perfect measure of the number and severity of
the diseases which afflict a community. Second, the records of
disease are very scanty and imperfect ; they have never been
tabulated on any large scale for the entire population. Only in
rare cases can the records of certain diseases be used as well
as those of deaths. Accurate mortality records, on the contrary,
have now been kept for many years.

The health of practically every community varies in response
to the seasons. In the northern United States most physicians
are far busier in February and March than in May and June.
In July and August the demand for their services increases
again, especially among children. Then come the best months of
the year, especially October, when good health and good spirits
abound. Different types of disease, to be sure, display different
seasonal adaptations. Those of the respiratory organs, for ex-

* This chapter and the two that follow are entirely new. They are based
largely on publications mentioned in the list at the end of the prefaces, but
none of the material in Chapter IX has hitherto been published.

154

CIVILIZATION AND CLIMATE

ample, reach a maximum in winter, while those of the digestive
tract are more numerous in summer.

Admitting, then, that both energy and health show marked
seasonal variations, our aim is to determine how closely the two
sets of variations are in harmony. Part of the answer is illus-
trated in Figure 13. The upper line, A, represents variations in

1010 mi 1*12 1913

J ASOND J FMAM J J A 8 OND J FMAM J J ASOND J F MAM J J AS ON D

Figure 13. Seasonal Variations in Energy and Health

A. Work of factory operatives in Connecticut.

B. Work of factory operatives in Pennsylvania.

C. Health (death rate inverted) in Connecticut.

D. Health (death rate inverted) in Pennsylvania.

Scale for A and B on left, for C and D on right. B and D are placed below A and C for con-
venience, although belonging at essentially the same level.

the efficiency of factory workers in Connecticut from January,
1910, to December, 1914, as already given in Figure 1. The
second line, B, is the similar curve for Pittsburgh. The two lower
lines illustrate fluctuations in health in Connecticut (C) and
Pennsylvania (D). They are the curves of the death rate in-
verted so that high parts indicate good conditions or few deaths,
and low parts, poor conditions or many deaths, thus per-
mitting easy comparison with the efficiency curves. It is ob-

HEALTH AND THE ATMOSPHERE 155

vious not only that the two efficiency curves and the two health
curves run almost parallel, but that there is also a close parallel-
ism between health and efficiency. Aside from the weather all
possible causes of such parallelism seem to be excluded, for epi-
demics, business disturbances, and the like, did not occur in such
a way as to explain the similar fluctuations in two diverse phe-
nomena hundreds of miles apart.

Note how closely the four curves agree even in details. Low
efficiency during January, 1910, is followed in a month or two
by very poor health. During the spring both efficiency and
health improve. Then comes the summer with a mild tendency
toward a drop in all the curves, and the autumn with the main
maximum of the year. In 1911 the parallelism of the four curves
is again evident, as is the lag of the mortality curve after that
of efficiency. In 1912 and 1913 the sag of all the curves in sum-
mer diminishes and disappears ; for those years, it will be re-
membered, had only short periods of really hot weather. The
similarity of the four curves, especially in the summer, would
be still more marked were it not that the deaths of children under
two years of age have been omitted in the mortality curves.

Since many other observations point in the same direction,
we conclude that unfavorable weather, such as commonly pre-
vails in January, has an immediate effect in reducing people's
vitality and energy. Hence their work falls off. At the same time
they become more susceptible to disease. Accordingly, in due time
the number of deaths increases. Naturally the greatest mortality
lags several weeks after the lowest efficiency ; it takes time for
bacteria to produce infection, and for infection to lead to death.
The lag is longer in winter when respiratory diseases are the
chief enemy than in summer when digestive diseases with their
more rapid course are the chief foes. The agreement between
health and energy is thus so close that both appear to depend
npon essentially the same fluctuations of the weather.

It is especially important to determine the relation of the

156 CIVILIZATION AND CLIMATE

weather to mental as well as physical health. Hence peculiar
interest attaches to certain studies of mental abnormalities
carried on by Norbury.* He finds that the admissions to psy-
chiatric hospitals show that "Mental disorders in their incipi-
ency and recurrence parallel the efficiency curves of Huntington.
(Maximum in the spring, minimum in the autumn.)" His curves
show that certain maxima of admissions for mental disorders oc<
curred at the following periods :

Civil Hospitals of New York State, 1916-1921, June.
Norbury Sanatorium in Jacksonville, Illinois, 1900-1923, May.
Massachusetts State Hospital, 1922-1928, May and June.
State Hospitals, northern United States, 1922-1923, March.

After the maximum the diminution in admissions is in all cases
very rapid. The number of admissions fell to the lowest point at
the following periods :

New York State Hospitals, September-February.
Dr. Norbury's Sanatorium, October-February.
Massachusetts State Hospital, November-February.
State Hospitals, northern United States, August-February.

Across the Atlantic insanity in London reaches a distinct maxi-
mum in May and is low from July to February. Almost identical
conditions prevail as to suicide, except that the fall from the
high point in June is not so rapid as in the case of insanity, al-
though the minimum is reached earlier, that is, in November and
December, instead of from July to February. As to the nervous
disorders in continental Europe, Gamier states that general
paralysis in Paris follows a seasonal course almost identical
with that of insanity in London. Now nervous breakdowns, in-
sanity, suicide, and paralysis, as Norbury shows, are all due

* Frank Parsons Norbury: Seasonal Curves in Mental Disorders. Medi-
cal Journal and Record, vol. 119, 1924.

HEALTH AND THE ATMOSPHERE 157

mainly to the same cause, namely fatigue of the nerves. Such
fatigue, he says, is apparently controlled to a large degree by
the seasons. But just as the maximum number of deaths lags
some weeks or even a month or two after the time when the
weather produces the lowest efficiency in factories, so the maxi-
mum effect of fatigue of the nerves lags still more, and the great-
est number of nervous breakdowns may occur three or four
months after the period of least efficiency as measured by daily
work.

The universality with which the bodily functions respond to
the seasons may be judged from two other recent investigations.
In one case Hess* has shown that among infants the phosphates
of the blood, which are an essential element for growth, show a
pronounced seasonal tide. During the year covered by his ob-
servations the percentage of phosphates in the blood stood at
4.34 mg. per cent during June and July, 1921. It may have
risen higher during the succeeding months, but no records were
kept. In December it had fallen to 3.92 and in March to 3.58.
Pi'esumably it would have fallen still lower, but whenever the
phosphates fell below 3.75 the children were treated with ultra-
violet light, which effectively increases not only the phosphates,
but also the calcium and probably other important elements of
the blood. It has been found by many authorities that the chil-
dren's disease known as rickets follows a seasonal course like
that of the phosphates and is connected more or less closely with
the amount of ultra-violet light. For our present purposes, how-
ever, the important point is that the essential phosphates in
children show the same general seasonal variation as the death
rate, and as mental breakdowns among adults, the lag being
perhaps greater than in the death rate but less than in mental
collapses.

"Alfred F. Hess: A Seasonal Tide of Blood Phosphate in Infants. Th*
Journal of the American Medical Association, December 30, 1922, vol. 79,
pp. 2210-2212.

158 CIVILIZATION AND CLIMATE

Another case of seasonal fluctuations is discussed by Porter*
of the Harvard Medical School. In cooperation with the Health
Department of Boston, monthly records of the weight of several
thousand of the youngest school children were begun in 1909 and
continued until 1919. Among the boys born in 1905 the average
increase in weight from month to month during the years 1911
to 1918 was as follows :

January to February -{-0.18 Ibs. July to August -f0.801bs.f

February to March -{-0.4/7 August to September -}-0.93t

March to April -f-0.22 September to October -f 0.96

April to May 0.16 October to November -f 0.61

May to June -j-0.05 November to December -|-0.63

June to July -fO.SOf December to January -fO.98

In interpreting this table, allowance must be made for cloth-
ing. In May the children exchange their winter clothes for those
which are somewhat lighter, while about the end of September
the opposite change takes place. Allowance must also be made
for the long summer vacation with its opportunities for out-
door play which naturally causes the children's weight to in-
crease rapidly. If allowance is made for these two facts the
regularity of the seasonal trend of growth is intensified. From
January onward the children grow slowly ; in the spring after
the phosphates have reached the minimum and at the very time
when grown people are most subject to mental breakdowns, the
children practically cease to grow. Not until the summer vaca-
tion begins do they recover from the effect of the winter. In the
summer the fourfold advantages of freedom from school, favor-
able weather, much outdoor life, and a more varied and healthful
diet than at other seasons cause rapid gain in weight. This gain

* W. T. Porter: The Seasonal Variations in the Growth of Boston School
Children. American Journal of Physiology, May, 1920.

f The boys were not weighed during the summer vacation. The average
increase in weight from June to September was 2.23 pounds, and this has
been allotted to the summer months in what seem to be reasonable propor-
tions.

HEALTH AND THE ATMOSPHERE

159

seems to be checked somewhat when school begins, but is resumed
during the late fall and early winter. In spite of confinement in
school, little outdoor play, and a diet relatively poor in vita-
mines and other important elements, the children gain weight
more rapidly in December than in any other month except
perhaps at the end of the summer vacation. The chief favorable
factor appears to be the climate, the stimulus of which does not
disappear until the advent of really cold weather. The sudden
decline in the children's rate of gain during January appears to
correspond closely with the drop in the efficiency of factory
workers at about the same time, but the effect on growth lasts
much longer than does the more direct effect upon activity.

Among the many other instances of seasonal fluctuations in
human health, one of the most remarkable is illustrated in
Figure 14. The upper line indicates for each month the average

Apr May June July Aug Sept Oci

Deo Jcu>

200

Conceptions

DealHs

body
Increase

of
Population

Figure 14. Seasonal Variations in Conceptions and Deaths in Japan,

1901-1910

160 CIVILIZATION AND CLIMATE

daily number of conceptions which resulted in the birth of living
children in Japan during the ten years from 1901 to 1910. A
pronounced maximum in June is followed by a diminution of 46
per cent, which culminates in September at the end of the long,
hot, humid summer. Then comes a recovery which is checked but
not reversed during the winter, and which resumes its course
during the delightful spring weather of April, May, and June.
The second curve shows the average number of deaths per day.
It is almost exactly the reverse of the curve of conceptions. In
other words, during the months when many people are sick and
die, the number of conceptions is either very low or else a large
number of conceptions result in miscarriages or still-births. The
most extraordinary feature of Figure 14 is the fact that the
curves of conception and mortality cross one another in Sep-
tember. The same fact is illustrated in the lowest curve, which
indicates the excess of conceptions over deaths. In June the
conceptions which give rise to living children outnumber the
deaths by nearly 2.5 to 1. In September the conceptions are less
numerous than the deaths. No seasonal variations in farm work
or social customs seem competent to explain more than an in-
significant part of the great contrast between May and Sep-
tember. The explanation seems to be that the hot, humid summer
saps the vitality of the Japanese, especially the women, so that
they are physically unable to reproduce themselves. If the
weather which prevails in July and August should prevail
throughout the year, the Japanese as a race would apparently
diminish in numbers instead of increasing with disquieting rapid-
ity. Under such circumstances natural selection would presum-
ably work with great vigor. A race might arise capable of
withstanding the most intense tropical conditions, but it would
presumably differ from the present Japanese in many qualities
such as energy and initiative.

It would be easy to multiply examples, but space forbids. All
sorts of physiological conditions appear to vary from season

HEALTH AND THE ATMOSPHERE 161

to season in essentially the same way except that some responses,
such as the energy of people in good health, lag only a little after
the climatic conditions, others, such as diseases and the rate of
reproduction, lag several weeks ; and still others, such as the
growth of children and the occurrence of nervous breakdowns
and suicides, lag still farther. In the case of the mental disturb-
ances the lag is so great that it almost seems as if the onset of
stimulating weather after a period of unfavorable weather,
had the effect of causing a sudden collapse. In reality the real
state of affairs may perhaps be this : the winter months produce
an effect like that of driving a horse without rest and as rapidly
as possible over a bad road. When a stretch of good road is
reached the driver, to whom we may liken the weather, whips the
animal to his topmost speed and the tired beast soon breaks
down.

Let us now try to analyze the effect of the seasons upon health,
and determine the relative part played by temperature, humid-
ity, and variability. A study of about nine million deaths by
means of climographs as described in World Power and Evolu-
tion, leads to the conclusion that the optimum or most favorable
condition for human health is an average outside temperature
of about 64 F. (18 C.) for day and night, a relative humidity
of about 80 per cent, and a fairly high degree of storminess, or
at least of variability from day to day. This means a climate in
which the midday temperature rises to 70 more or less, while
that of the night falls below 60. With the rise in temperature
the noonday relative humidity declines to perhaps 60 per cent,
while during the cool night it rises high enough so that dew is
precipitated. But a constant succession of clear days is not
desirable. There must be occasional rains and variations in tem-
perature, wind, and cloudiness from day to day. Tampa experi-
ences such conditions at the end of February, New Orleans in
March, Asheville in May, Atlantic City in early June, Seattle
in August, Nantucket and Boston in September, and Portland,

162

CIVILIZATION AND CLIMATE

Oregon, in October. At the seasons when people go in largest
numbers to many famous health resorts the majority of such
places enjoy climatic conditions closely approaching those
which are ideal for physical health.

EFFECT OF TEMPERATURE ON HEALTH AND STRENGTH

D
Effect

B

C

of change

A

No. of

Best

of 10F. on

Nature of criteria

persons

temperature

death rate

1. Deaths, north Italy, 1899-1913

781,000

58

21%

2. Piece-work, men, two Connecticut

factories, 1910-1913

300

59

3. Piece-work, girls, one Connecticut

factory, 1911-1913

200

60

4. Deaths, British Columbia, 1914-

1916,

8,000

62

25%

5. Deaths, southeastern United States,

1900-1912

122,000

62

10%

6. Piece-work, Pittsburgh, 1910-1913,

9,000

63

__

7. Deaths, southern Italy, 1899-1913

752,000

63

27%

8. Cigar-making, Cubans, Tampa,

Florida, 1913

380

64

9. Deaths, southern France, 1901-1910

838,000

64

17%

10. Deaths, northeastern United States,

1900-1912

2,500,000

64

15%

11. Deaths, whites, eastern United

States, 1912-1915

921,000

64

12%

12. Deaths, dry interior of United

States, 1900-1912

71,000

64

15%

13. Deaths, northern France, 1901-1910

1,315,000

65

17%

14. Deaths, east central United States,

1901-1910

739,000

65

18%

15. Cigar-making, Cubans, Tampa,

Florida, 1913

400

65

__

16. Deaths, negroes, eastern United

States, 1912-1915

167,000

68

12%

17. Cigar-making, Cubans, Tampa,

Florida, 1912

380

68

18. Deaths, California, 1900-1912

142,000

70

18%

HEALTH AND THE ATMOSPHERE 163

The preceding table illustrates the type of statistical evidence
on which is based the conclusion that a mean temperature of
64 F. is the optimum. The experimental evidence will be illus-
trated later.

Columns A, B, and C explain themselves. Column D indicates
the approximate percentage by which a change of 10 F. raises
or lowers the death rate when the temperature ranges between
30 and 60. When all seasons are taken together the net effect
of a rise of temperature under such conditions is to lower the
death rate, while a fall increases the death rate. This, however,
applies only to a rise or fall in which the new condition of
temperature endures for some time, as in the change from season
to season.

In twelve of the eighteen cases in this table the optimum out-
side temperature was from 62 to 65 F., and in five cases
64 F. Two of the cases where the optimum falls to 60 or lower
were piece-work in Connecticut factories. This may be because
such work involves mental as well as physical activity. We have
found some evidence that the optimum temperature for mental
work is considerably below that for physical work. Naturally
an occupation where mental and physical alertness are both
needed would be most favored by a temperature between the best
temperatures for the body and the mind. As to the low optimum
of the north Italians, 58, I have no explanation. The cause
could perhaps be detected by a study of the other elements of
the weather such as humidity and wind, or of local diseases such
as malaria. It is worth noting, however, that Campani* from
an analysis of 24,500 deaths at Milan obtains results closely
similar to those here set forth. His results, to be sure, are espe-
cially important in respect to variability rather than tempera-
ture. He finds that deaths are least numerous just after storms
while the wind is blowing. They are most numerous in still air
during periods of stagnation and after periods with little change

* A. Campani: Oazetta degli Ospeddiedelle Cliniche, Milan.

164 CIVILIZATION AND CLIMATE

of temperature. Changes of temperature are beneficial in north
Italy just as in America.

Among the three cases of the preceding table where the opti-
mum temperature is above 65, one represents Cubans of
Spanish descent but with a good deal of colored admixture, and
a second represents negroes. In the first case life in a tropical
climate has presumably raised the optimum temperature some-
what. In the second, although the negroes here dealt with lived
largely in the parts of the United States from Maryland north-
ward, they probably still retained an ancestral adaptation to a
slightly warmer climate than that which is best for the white
race. It is worth noting, however, that the Cubans have spent
practically their whole lives where the coolest month averages
about 70 and the hottest over 80, while the ancestors of the
negroes have dwelt for untold generations in regions still
warmer. Nevertheless the optimum for both groups, 68, seems
to be lower not only than the average temperature of their
homes, but than the average for the coolest months in those
homes. I might add that for negroes the optimum humidity
seems to be a little higher than for white men. This again may be
an inheritance from a former environment. Such differences
between diverse races suggest that permanent physiological
changes take place whereby races become adjusted to diverse
climates. The slightness of the differences, however, suggests that
such adjustment is very slow and incomplete. An earlier and
more fundamental adjustment to climate appears still to be
largely dominant, and may represent the climate under which
man's chief physical evolution took place.

As for the extreme variation in California, where 70 appears
to be the best temperature, I am inclined to think that it is due
to accident. The California results depend largely on two cities,
San Francisco where the mean monthly temperature never
reaches 70, and Los Angeles where conditions of wind and
humidity may account for the favorable conditions during the

HEALTH AND THE ATMOSPHERE 165

months of high temperature. It is interesting to note that the
average of the six extreme cases in the table is 64.2, against
63.9 for the twelve medium cases. It must be borne in mind,
however, that the best temperature is not the same in dry
climates as in moist. We shall return to this later.

The conclusion that an outside mean temperature of about
64 F. is the optimum for Europeans agrees closely with the con-
clusion now widely accepted that an inside temperature not
above 68, and preferably lower, is the ideal. The effects of de-
viations from this ideal have been the subject of many careful
experimental investigations, among which those carried on by
the New York State Ventilation Commission under the chair-
manship of Prof. C.-E. A. Winslow are especially notable. The
results of these experiments, as set forth in a volume entitled
Ventilation, are so important that I shall conclude this chapter
by quoting several pages which pertain not only to temperature
but to other atmospheric conditions.

"The work of previous investigators from Hermans to Leon-
ard Hill has made it abundantly clear that extreme high atmos-
pheric temperatures are highly prejudicial to human health and
comfort, and that it is to such temperatures rather than to
chemical pollution that the most serious effects of bad air are
due. The most important result of our experiments has perhaps
been the demonstration that even moderately high temperatures
between 24 and 30 C. (75-86 F.) are accompanied by
demonstrable harmful results.

"A. Body Temperature and Circulatory Phenomena. We
find that the rectal body temperature exhibits a definite relation
to the temperature of the atmospheric environment, the body
temperature when observed at 8 a. m. during the summer time
showing a fairly close parallelism with the average temperature
of the outdoor air for the preceding night. In our experimental
chamber we found that at 68 F. the rectal temperature and
heart rate tended to fall, the Crampton index of vasotone in-

166 CIVILIZATION AND CLIMATE

creased, the resistance of the peripheral portion of the circu-
latory system rose, and the velocity of the blood flow was
correspondingly lessened. At 75 F. the rectal temperature,
heart rate, and Crampton index changed but slightly. At 86 F.
with 80 per cent relative humidity, the rectal temperature and
heart rate rose, the Crampton index fell, the peripheral resist-
ance decreased and the velocity of blood flow increased.

"The final average results attained under the three atmos-
pheric conditions were as indicated below.

20 C. (68 F.)
50 per cent
relative
humidity

24 C. (75 F.)
50 per cent
relative
humidity

30 C. (86 F.)
80 per cent
relative
humidity

Rectal temperature
Heart rate, reclining
Heart rate, standing
Crampton index

36.7 C.
66.0
77.0
60.0

37.0 C.
81.0
105.0
45.0

37.4 C.
74.0
99.0
34.0

"We are not prepared to say whether the maximum heart
rates attained at 75 F. as compared with 86 F. are significant ;
but it is clear that rectal body temperature bears a direct, and
the Crampton index an inverse, relation to atmospheric tem-
perature. After vigorous physical work the return of the heart
rate to normal was somewhat more prompt at 68 F. than at
75.

"Systolic and diastolic blood pressure showed no very definite
relation to temperature between 68 F. and 86 F. ; but at
101 F. blood pressure as well as rectal temperature and heart
rate showed marked increases.

"B. Other Physiological Phenomena. The rate of respiration
was slightly increased by moderately high temperatures, from
an average of 17.9 at 68 F. to 19.3 at 75 F. and 19.7 at
86 F. ; while at 101 F. the increase was very marked. The
dead space of the lungs, the volume of the supplemental air, the
alkaline reserve of the blood, the respiratory quotient, the car-

HEALTH AND THE ATMOSPHERE 167

bohydrate metabolism, the protein metabolism and the total
metabolism showed no demonstrable relation to atmospheric
temperature under the conditions studied in our experiments.

"C. Comfort and Mental Efficiency. So far as the sensations
of the subjects are concerned, as evidenced by their votes as to
the comfortableness of the experimental chamber, the difference
between 68 F. and 75 F. is comparatively slight; but a tem-
perature of 86 F. with 80 per cent relative humidity is dis-
tinctly uncomfortable, the average votes falling sharply when-
ever this condition is reached. This discomfort is not, however,
accompanied by any inability to perform mental work ; such as
naming of colors and opposites, cancellation, mental multiplica-
tion, and addition. Subjects urged to maximal performance did
equally well under both conditions, the conditions being main-
tained for four hours a day on five consecutive days, or for eight
hours a day on four consecutive days. Longer hours of exposure
continued for a prolonged period might of course yield different
results. Even when the subject was left free to work or not, as
much mental work was accomplished at 75 F. as at 68 F.
although in one short experiment a temperature of 86 F. did
seem to diminish the inclination to do mental work. At 75 F.
typewriting (which involves a certain amount of neuro-muscular
activity) seemed to be slightly diminished while performance in
mental multiplication was actually increased. In general, how-
ever, we have no clear evidence that moderate overheating im-
pairs mental efficiency.

"7X Influence of Atmospheric Humidity. Somewhat exhaus-
tive studies of the alleged influence of atmospheric humidity
upon mental achievement and comfort have yielded entirely
negative results. With relative humidities of 50 per cent and 25
per cent, respectively (all other conditions being the same),
there was no significant difference in the votes of the subjects
as to their subjective sensations of comfort, no difference in
temperature of the air next to the chest, or in pulse rate, and

168 CIVILIZATION AND CLIMATE

no difference in the performance of a long series of complex
neuro-muscular tasks specially designed to test the alleged in-
fluence of a dry atmosphere upon nervousness and efficiency.
Longer periods of exposure might of course produce effects not
detected by us ; but it seems clear that exposure to a relative
humidity as low as 25 per cent at a temperature of 75 F. for
eight hours a day on five days a week does not produce any
demonstrable harmful effects.

"J. Physical Work. We have demonstrated on the other hand
a very marked and significant influence of atmospheric tempera-
ture upon the performance of physical work. An increase of
room temperature from 68 F. to 75 F. caused a decrease of
15 per cent in the physical work performed by men who were not
compelled to maximum effort but were stimulated by a cash
bonus ; and an increase from 68 F. to 86 F. with 80 per cent
relative humidity caused a decrease of 28 per cent in the physical
work performed under conditions of maximal effort:. The fall at
75 F. was most marked in the afternoon hours when fatigue
effects were called into play.

"F. Susceptibility to Disease. Finally, we have found very
definite evidence of the harmful influence of moderately high
atmospheric temperatures, particularly if followed by sudden
exposure to low temperatures, in promoting susceptibility to
bacterial infection. We find that rabbits maintained at a tem-
perature of 86 F. show a distinctly delayed formation of hemo-
lysins and a slightly reduced agglutinative power, as compared
with animals kept at 68 F. ; and that rabbits kept at 76-
70 F. and then chilled to 20-50 or chilled first and kept at
79 F. are much more susceptible to infection than animals kept
at 65-70 F.

"Exhaustive observations of the nasal mucosa of human sub-
jects show that in a warm atmosphere there is an increase, in
a cool atmosphere a decrease in the swelling, moisture, and red-
ness of the nasal mucous membranes. Sudden change from a hot

HEALTH AND THE ATMOSPHERE 169

to a cold atmosphere, particularly when combined with drafts,
produces a moist and distended but anemic condition of the
mucosa, presumably highly favorable to bacterial invasion ; and
workers who have been habitually exposed to high temperatures
show a marked excess of atrophic rhinitis. Hot moist air (as in
the case of laundry workers) seems to be much more harmful
than hot dry air (as in the case of furnace men).

"//. The Effect of Chemically Vitiated Air upon Physiological
State, upon Comfort, and upon Efficiency

"In parallel with our experiments upon the effect of tempera-
ture, we also studied the possible influence of air, chemically
vitiated by human occupancy, but at the same temperature and
humidity as fresh air used as a control. The factor of air move-
ment was excluded by the use of room fans to stir up both fresh
and stale air so that no difference existed save the very slight
changes in oxygen and carbon dioxide and the more obvious
changes in odoriferous organic constituents due to respiration
and effluvia from the body.

"A. Physiological Reactions, Comfort, Mental Efficiency, and
Resistance to Infection. In most of the reactions studied in our
experiments the influence of chemical vitiation of the atmos-
phere appeared to be absolutely nil. Temperature and humidity
being the same, we compared fresh air containing 5 to 11 parts
per 10,000 of carbon dioxide with vitiated air containing 23
to 66 parts and found no difference in body temperature, heart
rate, blood pressure, Crampton index, rate of respiration, dead
space in the lungs, acidosis of the blood, respiratory quotient,
rate of heat production, rate of digestion, and protein metabo-
lism.

"Comfort votes indicated that the subjects were quite unable
to distinguish from the standpoint of sensation between the fresh
and the stale air conditions ; the performance of mental work was
quite unaffected by the chemically vitiated atmosphere.

170 CIVILIZATION AND CLIMATE

"In a special series of animal experiments, guinea pigs ex-
posed to strong fecal odors for considerable periods failed to
exhibit any increased susceptibility to inoculations with foreign
bacteria or to injection of diphtheria toxin.

"B. Physical Work. In regard to the performance of physical
work on the other hand, there appeared to be a distinctly harm-
ful influence of the vitiated air. Temperature and humidity being
the same, our subjects performed 9 per cent less work in stale
than in fresh air, a difference less marked than that produced by
warm as compared with cool air (15 per cent) but apparently
significant. When both unfavorable conditions were combined
(in warm and stale air) only 77 per cent as much physical work
was performed as in cool fresh air.

"C. Appetite and Growth. Finally we found a marked influence
exerted by stale air upon the appetite for food as determined by
serving standard lunches to parallel groups of subjects in stale
and fresh air, respectively, but with the same temperature and
humidity. In the four different scries of experiments which were
successfully completed on this basis without the intrusion of in-
terfering factors, the excess of food consumed under fresh air
conditions was respectively 4.4, 6.8, 8.6, and 13.6 per cent. Since
the probable errors involved in these experiments were relatively
very slight it seems evident that the chemical constituents of
vitiated air may not only diminish the tendency to do physical
work but also the appetite for food.

"This conclusion is strengthened from another direction by
demonstration that exposure to strong fecal odors causes a
restraining influence upon the rate of growth of guinea pigs
during, but not after, the first week of exposure.

'7/7. Practical Conclusions in Regard to Ideal Conditions of

Ventilation

"The experiments of the commission have in general con-
firmed the conclusion of earlier investigators that the first and

HEALTH AND THE ATMOSPHERE 171

foremost condition to be avoided in regulating the atmosphere
of occupied rooms is an excessively high temperature. We have
found that even slight overheating (75 F.) produces the follow-
ing harmful results :

"(1) A burden upon the heat-regulating system of the body
leading to an increased body temperature, an increased heart
rate and a marked decrease in general vaso-motor tone as regis-
tered by a fall in the Crampton index.

"(2) A slight but definite increase in rate of respiration.

"(3) A considerable decrease in the amount of physical work
performed under conditions of equal incentive a decrease
amounting to 15 per cent at 75 F. and to 28 per cent at 86 F.

"(4) A markedly abnormal reaction of the mucous mem-
branes of the nose, leading ultimately to chronic atrophic rhini-
tis and when followed by chill, producing a moist and distended
condition of the membranes calculated to favor bacterial inva-
sion. In animals exposure to high atmospheric temperatures,
particularly when followed by chill, diminishes the protective
power of the blood and markedly increases general susceptibility
to microbic disease.

"For these reasons we believe that the dangers of room over-
heating are far more serious in their effect upon human health
and efficiency than has generally been realized and that every
effort should be made to keep the temperature of the school-
room, the workroom, and the living-room at 68 F. or below.

"With regard to the problem of relative humidity it is obvious
that a high moisture content combined with high temperature
must always be harmful, since the effect of a humid atmosphere
is to decrease the heat loss from the body by evaporation. The
specifically harmful influence of unduly low humidity which has
been postulated by various writers upon ventilation has, on the
other hand, not been apparent in our investigations.

"Our results in regard to the influence of the chemical com-
posHion of vitiated air (temperature and humidity effects being

172 CIVILIZATION AND CLIMATE

excluded) have been generally negative. In two respects, how-
ever, our experiments suggest that some chemical constituents
of the air of an unventilated room may be objectionable. Such
air appears (1) to decrease the appetite of human subjects for
food, and (2) to diminish substantially the amount of physical
work performed under conditions of equivalent stimulation.

"We may conclude then that the primary condition of good
ventilation is the maintenance of a room temperature of 68 F.
or below without the production of chilling drafts ; but that it
is also important, on account of certain subtle but real effects
of vitiated air upon appetite and inclination to work, to pro-
vide for an air change sufficient to avoid a heavy concentration
of effluvia such as was associated in our experiments with a car-
bon dioxide content of 23 to 66 parts per 10,000."

Except in respect to humidity at moderate and low tempera-
tures this long quotation reenforces the conclusions set forth in
this book and in World Power and Evolution. It brings out the
extreme sensitiveness of human health to atmospheric condi-
tions; it shows that temperature is undoubtedly the most im-
portant factor; it adds experimental proof to the widespread
opinion that as soon as the temperature rises much above 70,
man's capacity and inclination for physical work decline and
his susceptibility to disease increases. If high atmospheric hu-
midity is added to high temperature, as in many tropical coun-
tries, the harmful effects are shown to be much accentuated.
The fact that the Ventilation Commission detected no effect of
high temperature and humidity upon mental work seems to mean
merely that when people are subjected to moderately adverse
conditions for short periods amounting to less than twenty per
cent of the time for a few weeks the physical handicaps do not
become pronounced enough to exert a measureable influence
upon the mind. It seems only logical to suppose that if the ad-
verse physical conditions arising from high temperature and

HEALTH AND THE ATMOSPHERE 173

high humidity were to continue and produce their full effect, the
mental powers would ultimately suffer. Thus even this phase of
the Commission's work is not inconsistent with the conclusions
of this book. In this same connection it is interesting to note that
although the Commission made no experiments on the effect of
variability, its report contains in several places the specific sug-
gestion that variability may be an important but unconsidered
factor. The only point wherein the Commission's conclusions
radically differ from my own is in respect to humidity, a subject
which we shall consider in the next chapter. In spite of this dis-
crepancy the report of the Ventilation Commission, in its main
aspects, confirms the general conclusions of this book as to the
relatively depressing effects of tropical climates, regardless of
specific diseases ; and as to the beneficial effects of relatively cool,
variable climates.

CHAPTER VIII

MORTALITY, MOISTURE, AND VARIABILITY

INASMUCH as there is some disagreement as to the effect of
atmospheric moisture upon health, it will be well to examine
the evidence carefully. On the basis of the deaths in Paris from
1904 to 1913, Besson,* the chief of the "Service Physique et
Meteorologique" of that city, has come to the conclusion that
"on the whole, when the humidity increases, the mortality [from
diseases of the respiratory organs] decreases two or three weeks
later. ... If one examines each season separately one finds
that this rule fails only in summer when there is no clear result,
According to a widespread opinion humidity acts in an un-
favorable manner upon human health. The result announced
above cannot fail to surprise many people. One sees that an in-
crease in the number of deaths from diseases of the respiratory
organs is on an average preceded not by an increase but by a
decrease in the humidity." To test this further, Besson divided
the winter months into two groups, one with a mean relative
humidity below 86 per cent and averaging 82.4 and the other
above 86 per cent and averaging 89.2. He found that the deaths
per week when relatively high or low humidity prevailed and in

the succeeeding weeks averaged as follows :

Advantage of

Deaths per week high over low

Low High humidity in per

humidity humidity cent of low

The same week

163

161

1.2%

One week after

177

167

6.0%

Two weeks after

185

175

5.7%

Three weeks after

191

183

4.4%

Four weeks after

195

196

0.5%

* Louis Besson: Relations Entre les Elements M6t6orologiques et la
Mortalite. Annales des Services Techniques d'Hygiene de la Ville de Paris,
1921.

MORTALITY, MOISTURE, AND VARIABILITY 175

It seems clear that during these ten years in Paris the drier
weeks of winter, even though they were quite moist, were accom-
panied by a slightly increased death rate from respiratory dis-
eases and were followed in the next two weeks by a death rate
about 6 per cent higher than that which followed the moist
weeks. Besson ascribes part of this effect to the direction of the
wind, but the effect of the wind must be produced largely
through humidity and temperature. What part is played by
temperature in the results for Paris is not clear.

A study of my own includes temperature as well as humidity,
and gives results almost identical with those of Besson. I used
the deaths from pneumonia in New York City during the year
beginning in April, 1917, and compared them with the weather
on the day of death and on the preceding day. All the days with
any given temperature were divided into two equal groups on the
basis of their relative humidit}'. Here are the results for the
7200 deaths from lobar pneumonia. Those for broncho-pneu-
monia were somewhat similar, but much more irregular, pre-
sumably because the number of deaths was only a third as great.

AVERAGE DEATHS PER DAY FROM LOBAR PNEUMONIA ON SAME

DAY AS GIVEN WEATHER CONDITIONS AND ON SUCCEEDING

DAY. (NEW YORK, APRIL 1, 1917, TO MARCH 31, 1918.)

Mean temperature 20 or less 21-32 38-45 46-55 56-65 G6-70 71-75 76 or

over

Low relative humidity 28.4 26.7 28.7 21.5 19.8 10.5 6.5 6.8
High relative humidity 26.3 25.9 28.0 18.8 15.1 8.2 6.1 6.5

Advantage of moist over

dry 8.0% 3.1% 2.5% 14.3% 31.1% 28.0% 6.6% 4.4%

The obvious fact here is that at all temperatures the moist
days were better than the dry, as appears in the lower line. The
same thing holds true whether we consider the deaths on the

176 CIVILIZATION AND CLIMATE

same day as the humidity or on the succeeding day. While Bes-
son's data seem to show that dryness renders people susceptible
to the initial attack of pneumonia which results in death after
a fortnight more or less, the New York data suggest that when
the disease nears its crisis a dry day may turn the balance
toward death, whereas a moist day turns it toward life. In cold
weather, however, this effect is slight, for when the temperature
is below 45 the moist days in the preceding table reduce the
death rate only 4.5 per cent on an average. At the temperatures
of 56 to 70, on the other hand, an additional grain and a half
of water vapor per cubic foot of space, or a difference of roughly
20 per cent in relative humidity, is associated with a diminution
of about 30 per cent in the death rate. This is especially im-
portant because these are the temperatures at which our houses
are kept, or ought to be kept, most of the year. It adds another
to the bits of evidence which indicate that for respiratory dis-
eases a dry climate is worse than a moist one. The opposite belief
has perhaps become traditional largely because in dry climates
people live out of doors. Other things being equal, it is always
more healthful to live outdoors rather than indoors.

In this connection it might be added that Besson's conclusions
as to the relation of temperature to the death rate from respira-
tory diseases agree with mine as to the similar relation to deaths
from each of the two main types of pneumonia. At temperatures
between freezing and 60 F. there is an almost perfectly regular
decline in the number of deaths as the temperature rises ; then
the decline becomes less and less marked until a minimum is
reached at about 72. At higher temperatures a slight increase
makes itself apparent, but does not go far because there are
only a few days in either Paris or New York when the mean
temperature rises much above 75. Still another investigation,
that of Greenburg,* who studied the monthly deaths from pneu-

* David Greenburg: Relation of Meteorological Conditions to the Preva-
lence of Pneumonia. Journal of American Medical Association, 1919, p. 252.

MORTALITY, MOISTURE, AND VARIABILITY 177

mo ilia in New York, Boston, Newark, and Providence, agrees
with the two already cited as to the effect of both temperature
and humidity upon pneumonia. Such close agreement makes it
practically certain that the humidity of the air, as well as the
temperature, is an important element in determining the death
rate from respiratory diseases.

The need of certainty as to the effect of atmospheric humidity
is so great that I shall sum up a number of other examples in the
form of a long table which the non-scientific reader can skip.
Since the effect of humidity varies according to temperature, we
must carefully distinguish between temperatures above and
below the optimum of 64. At the optimum temperature in every
one of the groups of deaths listed in section A of this table, the
best conditions of health prevailed when the relative humidity
averaged 70 per cent or more for day and night together. At
higher temperatures a relative humidity which averages above
70 per cent (for day and night together) does very decided
harm. At the optimum temperature the effect of humidity seems
to be at a minimum, and a humidity above 70 per cent does little
harm. A lower humidity, although somewhat harmful, has less
effect than at other temperatures, the increase in the death rate
ranging from 5 to 15 per cent, according to the degree of dry-
ness. This may be one reason why the New York Ventilation
Commission found no clear effect of low humidity. Their main
experiments were performed at temperatures close to the opti-
mum. At temperatures below the optimum the effect of humidity
upon the death rate is very clear, much more so than at the opti-
mum. For example, at a temperature of about 40 F. a difference
of only 10 per cent in humidity appears to produce approxi-
mately the effect shown in section A of the following table.

178 CIVILIZATION AND CLIMATE

EFFECT OF HUMIDITY ON THE DEATH RATE

A. Increase in Monthly Death Rate Accompanying a Decrease of 10 Per Cent

in Relative Humidity at a Temperature of 40 F.

(1) Northeastern United States, 1900-1912 0.8%

(2) East central United States, 1900-1912 1.3%

(3) Dry interior of the United States, 1900-1912 2.0%

(4) Boston (deaths after operations), 1906-1915 3.1%

(5) Large cities of United States (whites non-contagious dis-

eases), 1912-1915 3.3%

(6) Large cities of United States (negroes, non-contagious dis-

eases), 1912-1915 4.0%

(7) British Columbia, 1914-1916 7.0%

(8) Large cities of United States (whites, contagious diseases),

1912-1915 9.5%

(9) Southern France, 1901-1910 10.0%

(10) Northern France, 1901-1910 11.0%

(11) Northern Italy, 1899-1913 12.6%

B. Increase in Monthly Death Rate Accompanying a Decrease of 10 Per Cent
in Relative Humidity at All Temperatures, December-March, 1900-1914

(12) St. Louis 0.5%

(13) New York City 1.2%

(14) San Francisco 3.8%

(15) Baltimore 5.0%

(16) Chicago 7.8%

C. Increase in Daily Death Rate Accompanying a Decrease of 10 Per Cent

in Relative Humidity at All Temperatures During the Influenza

Epidemic in New York City (September-December, 1918)

and Boston (October, 1918-April, 1919)

NOTE. The figures in this section indicate percentages of the average daily
change in the death rate instead of percentages of the actual deaths. Hence
the figures are perhaps five times larger than if they were reckoned as in
A and B.

(17) Onset of influenza, Boston 3.0%

(18) Onset of influenza, New York 6.9%

(19) Deaths from influenza, Boston 11.3%

(20) Deaths from pneumonia, New York 21.6%

(21) Deaths from influenza, New York 26.2%

(22) Deaths from pneumonia, Boston 8.3%

(23) Onset of pneumonia, Boston 1.8%

MORTALITY, MOISTURE, AND VARIABILITY 179

This table represents all the available mortality data except
those already mentioned and certain others to be given shortly.
Practically all the evidence seems to point in the same direction.
In 21 out of the 23 sets of data in the table the moister days or
months have an advantage over the drier. The negative figures
for Boston, Nos. 22 and 23, may be accidental, or may mean
that Boston's famous east winds, unlike the moist winds in most
places, are really too damp. Taking the table as a whole, an
increase of 10 per cent in humidity at low temperatures is corre-
lated with an average decrease of not far from 6 per cent in the
death rate.

Let us now turn to quite a different investigation. In Boston
I made a study of the number of deaths following operations
performed in different kinds of weather. On the basis of about
2300 deaths after operations from 1906-1915 the addition of
a grain of moisture per cubic foot of space to the air within
doors would have diminished the death rate as follows, provided
the inside air thereby acquired the qualities pertaining to the
outside air moistened by nature :

APPARENT EFFECT OF ONE GRAIN OF WATER PER CUBIC FOOT

OF AIR IN DIMINISHING THE DEATH RATE AFTER

OPERATIONS IN BOSTON, 1906-1915

Percentage of decrease in
deaths for one additional
A. Mean temperature grain of water per cubic foot

30 or less

9.2%

31-40

. . . . 16.5%

41-50

8.8%

50-60 .... . .

13.7%

61-70

4.6%

Because of the small number of deaths this table is irregular,
but the irregularities have little significance. At temperatures
below the optimum the death rate was higher after operations

180

CIVILIZATION AND CLIMATE

performed in damp weather than after those performed when the
air was dry. This was especially true when the moist weather
continued a day or two after the operations. At high tempera-
tures, however, the effect of humidity is not at all the same as at
low, as appears in Figure 15. Dry conditions are shown toward

100

>0.75

I

C.50

I

& 0.25

a>

Relative Humidiiy at 6 AH

K> 60 SO

Above 70'al 6 AM.

4}-SO'at 8 AM
61-70*ctt 8AM
51-60'ai 8AM

Figure 15. Post-operative Death Rate at Boston in Relation to Humidity

and Temperature

the left, moist toward the right. The height of the lines shows
the number of deaths per day succeeding operations performed
when various conditions of relative humidity prevailed at 8 a.m.
The broken lines, A, B, and C, indicate the number of deaths in
cool weather. Note their regular decline toward the right. High
humidity was evidently an advantage. Now contrast the dotted
lines with the solid line, A, indicating the effect of humidity
when the temperature at 8 a.m. was above 70. Under such con-
ditions, far more than in cooler weather, dry air was bad.
Moderately moist air with a relative humidity of 50 to 60 per
cent was better than even the moistest air at lower temperatures.

MORTALITY, MOISTURE, AND VARIABILITY 181

But note how rapidly the death rate after operations performed
in hot weather rises if high humidity is added to great heat.
Nevertheless the evil effects of very damp hot days in Boston do
not appear to be so bad as those of very dry hot days. This study
of surgical operations seems to afford strong evidence of the
extreme sensitiveness of the human body to variations in hu-
midity as well as temperature. The fact that it gives a different
result at temperatures above and below the optimum tends to
establish confidence, for the results at high temperature and
high humidity are in perfect accord with common experience and
with the experiments carried on by such organizations as the
New York Ventilation Commission and the laboratory of the
American Society of Heating and Ventilating Engineers at
Pittsburgh.

In other countries as well as in France, Italy, and the United
States, whence our data have thus far come, the statistics seem
usually to indicate that extreme dryness is harmful to health.
Other things being equal, the death rates appear generally to be
higher in dry regions than in moist. For example, Lucknow and
Cairo in dry climates have death rates far higher than Madras
and Bombay, which are moist and tropical. Mexico City on its
high, cool, but dry plateau, and Johannesburg and Madrid in
somewhat similar locations, have exceptionally high death rates
in view of their temperature and latitude. In Mexico and India
the dry season of March, April, and May has a decidedly higher
death rate than the succeeding wet season. This happens in spite
of the fact that the temperature in Mexico City during May
averages 65 and is almost ideal, while in July it averages a
degree or two cooler and is almost equally ideal. In India, on
the other hand, both the dry spring and the wet summer are hot,
so that the wet season is very muggy. Yet the people heave a sigh
of relief when the rains come, for it brings hope of a diminution
of disease as well as of good crops.

Let us turn to variability or storminess, another factor which

182 CIVILIZATION AND CLIMATE

seems to cooperate with temperature and humidity in determin-
ing the effect of climate. This factor is especially important be-
cause of its bearing on changes of climate and their effect on
history. In World Power and Evolution the study of climo-
graphs based on eight million deaths in France, Italy, and the
United States suggests that people's sensitiveness to changes of
temperature is almost directly proportional to the uniformity
of their climate. For example, in San Francisco a change of
only 7 in the mean monthly temperature (50 January, 57
July) is associated with a change in the death rate from I6A
per cent above normal to 11.7 per cent below, or about 4 per
cent for every degree of temperature. At St. Paul and Minne-
apolis a change from a mean temperature of 12 in January
to 67 in June is accompanied by a change in the death rate
from 5.5 per cent above normal to 6.9 below, or only 0.23 per
cent for each degree of temperature. In the same way a differ-
ence of about 9 C. between January and May in Naples is
accompanied by a difference of 35.3 per cent in the death rate,
whereas in Milan a difference of 16 C. in temperature is asso-
ciated with a difference of only 26.8 per cent in the death rate.
Put in another way this means that the apparent effect of a
given change of temperature is 2.3 times as great in Naples of
the South as in Milan of the North, and over seventeen times as
great in San Francisco with its remarkably uniform climate as
in St. Paul and Minneapolis with their severe winters and many
storms at all seasons.

In other words, where great changes of weather take place,
people become hardened to them. The reality of this hardening
is demonstrated again and again by the way in which north-
erners who move to the tropics lose their power of resistance to
even the slightest changes of temperature, but regain it after
a few years of renewed residence in a severe climate. The relation
between variability and the power to resist disease seems quite
clear, but as yet there seems to be no conclusive evidence as to

MORTALITY, MOISTURE, AND VARIABILITY 183

the relative importance of variations from day to day such as
accompany storms, and of variations from season to season. The
evidence which will now be set forth pertains to changes of
temperature from day to day.

In World Power and Evolution one of the most important lines
of evidence has to do with variations of temperature from one
day to the next in New York City. A study of about 400,000
deaths during a period of eight years from 1877 to 1884 shows
that when the temperature falls the death rate also falls, while
a rise in temperature is regularly accompanied by a correspond-
ing rise in the death rate. This happens not only in summer,
when one would expect it, but at all seasons, including even the
winter, when one would surely suppose that warm weather would
be beneficial. And so it is, in the long run, but the immediate
change toward warmth is temporarily harmful.

In this investigation, instead of employing the absolute num-
ber of deaths, as in previous cases, the change in the number of
deaths from one day to the next has been used. This is done
partly in order that the greater frequency of days with abrupt
changes of temperature in winter than in summer may not con-
fuse our results. It is likewise because in dealing with changes
of temperature the natural question is whether such changes
have any effect in changing the death rate.

The upper part of Figure 16 illustrates another investigation
of the same kind in respect to daily deaths from pneumonia in
New York. The left-hand side of the diagram indicates the con-
ditions which prevail when the mean temperature of the day of
death is higher than that of the preceding day, while the right-
hand side indicates that the temperature has fallen. The two
solid lines indicate the summer relationships from April to Sep-
tember, A being lobar pneumonia, and B broncho-pneumonia.
The dotted lines show winter conditions, C lobar, and D broncho-
pneumonia. The significant points about these four curves are
as follows :

184

CIVILIZATION AND CLIMATE

(1) They are all essentially alike. This suggests that they all
conform to some definite physiological law.

(2) Every one of them is low at the two ends and high in the

Rise of Temperature Drop of Temperature

10* 5' 5" JO'F

2.0

-ZCfc

-1.0

-zo

205

Deaths from
Pneumonia. In
New York. Apr
1917 lo Mar 1918

Deaths from
Pneunronia and
Inflaenza, com-
bined during
Epidemic of 1918
New York
- -Boston

Figure 16. Relation between Deaths from Pneumonia and Influenza and
Interdiurnal Changes of Temperature

middle. This seems to indicate that under normal conditions such
as prevailed in the year ending in March, 1918, patients suffer-
ing from either form of pneumonia are less likely to die on days
when there has been a marked change of temperature in either

MORTALITY, MOISTURE, AND VARIABILITY 185

direction than on days when there is little or no change. This
conclusion applies to both summer and winter, but is more true
in summer than winter. Hence, for pneumonia patients, a vari-
able climate seems better than one that is uniform. The agree-
ment of the curves with our conclusions derived from factory
work and with those of Campani in Italy is noteworthy.

(3) A drop of temperature, on an average, is decidedly more
effective than a rise in lowering the death rate. This again
agrees with the results of our factory investigation.

The lower part of Figure 16 shows the results of a similar
investigation of the influenza epidemic in the latter part of 1918.
Because of the enormous variations in the death rate I have here
used the percentage rather than the actual number of deaths
by which each day's death toll differed from that of the pre-
ceding day. Both New York (solid lines) and Boston (dotted
lines) have been included. The investigation has been broadened
to include (1) the number of cases in which influenza (a) and
pneumonia (b) attacked a patient, or at least, in the case of
Boston, were reported to have done so, and (2) the number of
deaths from influenza (c) and pneumonia (d). The significant
points in Figure 16 are as follows :

(1) All the lines except that for deaths from pneumonia in
Boston (c') slope in the same way, thus indicating that a rise of
temperature is worse than no change, and that a fall is much
better than either. This disagrees with the preceding investiga-
tion of normal pneumonia, but that may be simply because the
lower part of Figure 16 includes only the winter months and
not those of the spring when the advent of warm days would be
beneficial.

(2) The average changes in the rates for all the conditions
shown in the lower part of Figure 16 amount to an increase of
14.1 per cent in the deaths and illnesses on days with a strong
rise of temperature and 3.4 per cent on days with little or no
change, whereas on days with a strong drop of temperature the

186 CIVILIZATION AND CLIMATE

change takes the form of a drop of 18.4 per cent. This means
that for all the days when the temperature was markedly dif-
ferent from that of the preceding days, no matter whether it was
higher or lower, there was an average drop of 2.2 per cent in the
rates of disease and death against a rise of 3.4 per cent on the
days of comparatively uniform weather. In other words, here,
just as in normal pneumonia, the variable weather had an ad-
vantage over the uniform weather. This conclusion is quite con-
trary to the usual ideas, but it is supported by many other bits
of evidence.

One such bit is found in the following data as to the average
number of deaths per day in Boston hospitals after surgical
operations, when specified changes of temperature took place
between the day of an operation and the succeeding day.

Interdiurnal change of temperature

Daily deaths
December to
February

Daily deaths
March to
November

Changes of 9 F. or more in either direction
Change of 4 to 8 in either direction .
Change of 3 or less in either direction .

0.340
0.336
0.326

0.235
0.288
0.343

The first column suggests that in winter the changes of tem-
perature are too severe in Boston, so that days with only a little
change have an advantage of about 4 per cent over the d&ys
with a violent change. From March to November, however, the
conditions are quite different: operations performed on days
with the smallest changes are followed by 41 per cent more
deaths than are those performed when the temperature changed
most strongly from the day of the operation to the next day.
Thus for the year as a whole the variable weather displays a
distinct advantage just as in New York.

The net result of our studies thus far is the conclusion that
temperature, humidity, and variability all play important parts

MORTALITY, MOISTURE, AND VARIABILITY 187

in determining variations in health and mortality from day to
day, month to month, season to season, and year to year. For
each climatic factor there appears to be a distinct optimum or
most favorable condition, but this varies considerably in re-
sponse to differences in the other factors. Thus the optimum
temperature in dry weather is not the same as in wet, and pre-
sumably is not the same in variable weather as in that which is
uniform. The high level of the optimum temperature which we
found in California may possibly be due in part to the uniformity
of the weather. Again, a degree of humidity which is highly
favorable at a temperature of 50 F. may work severe harm at
80, just as a degree of variability which is highly beneficial in
warm weather may be too extreme in the midst of a cold winter.
In the rest of this chapter I shall describe certain investiga-
tions in which the three climatic factors of temperature, hu-
midity, and variability are analyzed in reference to the influenza
epidemic in 1918. The investigation of influenza was carried out
by a Committee on the Atmosphere and Man appointed by the
National Research Council of the United States. A mathe-
matical method known as partial correlation coefficients was em-
ployed. This method has the remarkable quality of picking out
and isolating the effect of any one among a number of factors
as in an experiment. In the present case the Committee set itself
the completion, or at least the extension, of a task already begun
by Professor Pearl of Johns Hopkins University. The task was
to ascertain whether any environmental conditions were re-
sponsible for the great differences in the mortality of the influ-
enza epidemic of 1918 from one city to another. During the ten
weeks of the main epidemic Philadelphia, for example, had a
death rate from influenza and the resultant pneumonia four times
as great as that of Milwaukee, while the rate in Pittsburgh was
twice as great as in the neighboring and similar city of Cleve-
land. Before the committee's investigation was finished the fol-
lowing twenty-two factors had been examined :

188 CIVILIZATION AND CLIMATE

A. Factors of human environment (demography).

1. Age (proportion of inhabitants of various ages).

2. Sex (number of females per hundred males).

3. Density of the population (persons per acre within city).

4. Rate of growth from 1900 to 1910.

B. Factors of geographical position.

5. Distance from Boston, where the epidemic began.

6. Longitude.

7. Latitude.

C. Physiological factors; normal death rates in 1916, 1916, and 1917 from:

8. All causes.

9. Pulmonary tuberculosis.

10. Organic diseases of the heart.

11. Nephritis and acute B right's disease.

12. Typhoid fever.

13. Cancer and other malignant tumors.

D. Racial factors.

14. Percentage of negroes, 1920.

15. Percentage of foreign born, 1920

E. Industrial factor.

16. Percentage of population engaged in manufacturing, 1919.

F. Climatic factors.

17. Mean temperature for day and night.

18. Change of mean temperature from one day to the next.

19. Absolute humidity (weight of water vapor per cubic foot of space).

20. Relative humidity, or percentage of possible water vapor.

21. Weather a combination of Nos. 17-20.

22. Climatic energy as denned in this book.

Directly or indirectly these twenty-two factors embrace most
of the conditions which may have been effective in causing
people's power of resistance to the epidemic to vary from city to
city. Sanitation and medical practice fail to appear in the list
because their degree of excellence cannot easily be expressed in
figures. But the death rate from typhoid fever is generally sup-
posed to be an unusually good measure of sanitary efficiency,
while other death rates are in most places a fairly good index
of the excellence of the medical service. Almost the only im-

MORTALITY, MOISTURE, AND VARIABILITY 189

portant field which the factors do not cover is that of variations
in the disease-bringing bacteria so far as such variations are
due to causes not included in our table. When all these various
factors are investigated by means of the most exact and delicate
mathematical method yet known, the only one which shows any
conclusive causal relation to the destructiveness of this particu-
lar epidemic is the weather.

In the work which ultimately led to this conclusion, the Com-
mittee on the Atmosphere and Man took the death rate from
influenza and pneumonia during the ten weeks succeeding the
outbreak of the epidemic in each of thirty-six large cities in the
United States. These ten weeks cover the first and, in most
places, much the more important outbreak. The committee also
obtained data as to the temperature, relative humidity, absolute
humidity, and change of temperature from one day to the next.
The weather data were tabulated for periods of ten days be-
ginning seventy days before the onset of the epidemic and con-
tinuing fifty days thereafter. Previous to the thirtieth day
before the epidemic there is evidence of no real relationship be-
tween any weather condition and the destructiveness of the in-
fluenza. During the thirty days just before the onset of the
epidemic, however, the temperature and especially the absolute,
as distinguished from the relative humidity show a distinct rela-
tion to the succeeding death rate. This means that if the weather
was warm during the month before the influenza reached a city,
the death rate was high; if the amount of moisture in the air
was great, the conditions were still worse. At Boston, for ex-
ample, from the twentieth to the eleventh day before the epi-
demic the temperature was higher than during the corresponding
period in any other cities except New Orleans, New York, and
Los Angeles. This was natural, for the epidemic broke out in Bos-
ton earlier than elsewhere. In places like St. Paul, Toledo, and
Grand Rapids, the cool and fairly dry autumn weather which
prevailed for a month before the epidemic apparently gave peo-

190 CIVILIZATION AND CLIMATE

pie a certain degree of stored-up vigor which stood them in good
stead and lessened the ravages of the disease. If the tempera-
ture was variable, as in Cleveland, Columbus, and Richmond,
and especially if it fell during the ten days after the onset of
the epidemic, the death rate was lower than where the contrary
conditions prevailed. On the other hand, high relative humidity
during the ten days before the onset was associated with a
relatively high death rate. Cambridge, New Haven, and New
Orleans suffered most in this respect. The dampness perhaps
made it easy for the bacteria to be transmitted. Droplets of
water in the air may act as carriers of the bacteria, or may pre-
serve their virility.

From the tenth to the thirtieth days after the onset of the
epidemic the virulence of the bacteria was apparently so great
that the state of the weather made no difference in the death
rate. At any rate there is no evidence that the immediate weather
conditions had any effect in overcoming the sudden and sweeping
character of the infection. After the thirtieth day, however,
there came another change, and the apparent effects of tempera-
ture and absolute humidity again rose high. This was the time
when in most places the disease reached its maximum and began
to decline. At that time cool and moderately dry weather once
more was associated with a low death rate. This does not neces-
sarily mean that cold weather is favorable at the time of an
epidemic. In fact, quite the contrary may be the case, for very
low temperature may be as bad as high. Labrador suffered
greatly in the epidemic of 1918.

Having reached the conclusion that atmospheric conditions
influenced the severity of the epidemic, the next step was to find
a numerical expression for the weather by combining the tem-
perature, humidity, and variability according to their apparent
importance. When this had been done, the method of partial
correlations was used to compare the weather with the severity
of the epidemic and with all the other factors which showed any

MORTALITY, MOISTURE, AND VARIABILITY 191

sign of being important, namely, deaths from tuberculosis,
deaths from all causes, deaths from heart diseases, and climatic
energy. The weather proved to be the only one whose correlation
coefficient was more than four times the probable error, and
hence large enough to be significant. The final partial correla-
tion coefficient, when the four other factors named above were
held constant, that is eliminated, amounts to 0.57. This is 7.6
times the probable error, which means that there is only one
chance in hundreds of millions that we are being misled by acci-
dental agreements between the weather and the death rate.

"Thus," to quote the report of the committee, "the statistical
fact is clear. The weather, which means primarily the weather
just before the onset of the epidemic and at or just after the
climax, is the one factor thus far investigated which shows a
clear, pronounced, and persistent relation to the destructiveness
of the epidemic. This does not mean that the weather was in
any sense a cause of the epidemic. It is even possible that the
weather may be related to the epidemic only indirectly, as is the
case with the death rate from heart disease, although no factor
capable of producing this result has yet been suggested. Even
if the weather is a causal factor in producing variations in the
virulence of the epidemic, there is no reason to think that it is
the only factor. If the degree of relationship between two vari-
ables is proportional to the square of the correlation coefficient,
as is sometimes held, the weather, even if it is a direct a;^ent, may
be responsible for no more than a third of the variations from
city to city. Nor do our high correlation coefficients mean that
the weather had anything to do with setting the date of the epi-
demic or with determining the severity of the 1918 epidemic
compared with other epidemics. Neither do they prove anything
as to the effect of the weather in other countries, although else-
where a relationship similar to that found in the United States
seems probable. For instance, the British government estimates
that in India the death rate from the epidemic was about six

192 CIVILIZATION AND CLIMATE

times as great as in the United States, while scanty reports from
other tropical countries indicate a similar excessive mortality.
The one thing which seems clear from the present investigation
is that the weather is the one factor whose apparent relation-
ship to the epidemic is not seriously reduced or modified when
other conditions are held constant.

"The results of this investigation should be qualified in still
another respect. It is not necessary to suppose that other epi-
demics will show exactly the same relationships as the epidemic
of 1918, even though they may be strongly influenced by the
weather. In the first place, the epidemic of 1918 was so peculiar
in its virulence, its rapid dissemination, its fatality for persons
in the prime of life, and in other respects, that it may well have
been peculiar in its climatic relationships. In the second place,
the epidemic occurred at a season when the approach of cold
weather normally exerts a strong stimulating effect in the
United States. It is well known that from August to October or
even November the death rate normally declines. The epidemic
seems to have reflected this condition. Just so far as the weather
approached the conditions which prevail at the time when the
autumn mortality is lowest, the ravages of the epidemic were
checked. At some other, colder season, relatively low tempera-
ture and low humidity might be as harmful as high temperature
and high humidity appear to have been in September, October,
and November, 1918. As a matter of fact, the epidemic of Feb-
ruary and March, 1919, shows only a small positive correlation
between the monthly death rate and the temperature. Other con-
ditions, perhaps other conditions of weather, were then domi-
nant; or possibly some cities were too cold while others were
too warm, a condition which would make the use of correlation
coefficients impracticable.

"Finally, even if the weather should prove to be an important
factor in causing variations in the virulence of influenza, we still
have little evidence as to how its effects are produced. Presum-

MORTALITY, MOISTURE, AND VARIABILITY 193

ably the weather gives to the human being more or less power of
resistance to disease. But it is not improbable that the weather
also has an important effect upon the vigor, reproductive rate,
or transmission of the disease-bringing bacteria."

CHAPTER IX
HEALTH AND WEATHER

THIS chapter deals with two investigations of the relation
between health and the weather. They seem to me the most
conclusive evidence yet available along this line because the
various weather elements are more clearly separated than in
most cases, and because there is no danger of confusing the
effects of different seasons. The first investigation pertains to
the weather day by day, and is by far the most extensive in
which daily data have been employed. It represents a coopera-
tive effort carried on by the Committee on the Atmosphere and
Man of the National Research Council of the United States and
the Metropolitan and New York Life Insurance companies. In
addition to the author those most closely concerned in planning
the work were Dr. J. Arthur Harris of the Carnegie Institution's
laboratory at Cold Spring Harbor and Dr. L. I. Dublin of the
Metropolitan Life Insurance Company, but many suggestions
were received from others. The results appear to be fairly con-
clusive as to mean temperature and changes of temperature,
but are inconclusive as to relative humidity.

In order to obtain an adequate series of daily mortality data
the Committee was obliged to go back to the years 1882 to 1888
in New York City. At that time and for a few years previous
the actual day of death was recorded and the facts were sum-
marized by days and published in the annual reports of the New
York Board of Health. This highly valuable record has not
since been equalled either in New York or elsewhere, so far as
I am aware. The Committee used three sets of mortality data:

HEALTH AND WEATHER 195

(1) deaths of children under five years of age; (2) deaths of
persons over five years of age; (3) deaths from pneumonia. In
order to avoid errors due to the growth of population and the
improvement in medical practice, each year was treated as a
separate unit, and each category of deaths was reduced to per-
centages of the average number of deaths per day in that par-
ticular year.

For the phase of the investigation here under discussion, the
method of partial correlation coefficients was employed. For the
layman it may be well to repeat that by this method the
effects of different factors can be separated as in an experiment.
In order to avoid periods when the temperature is sometimes
above and sometimes below the optimum, the months of De-
cember to March were chosen. This is also advisable because
the effect of the various climatic elements in winter is less under-
stood than their effect in summer. One aim of the work was to
eliminate the effect of the seasons and determine whether a
mere departure of the weather elements from the normal for any
special month has any effect. Accordingly, in preparing the data
for the correlation coefficients each month was treated as a sepa-
rate unit, and the departures of all kinds were reckoned from the
averages for December, January, and so forth. This made it
possible to ascertain almost beyond question that the weather
day by day causes small variations in health which are super-
posed upon the large seasonal variations.

The Committee used three elements of the weather: (1) the
mean daily temperature; (2) the interdiurnal change of mean
temperature from one day to the next; and (3) the mean daily
relative humidity. In nature the effects of these three elements
are inextricably mixed, but the method of partial correlation
sorts them out. The results of this sorting appear in Figure 17.
The day marked zero is the day on which a given condition of
weather occurs. The weather elements on each such day have
been compared with the deaths on that day and on each of the

Mean Temp&raiurc l,,k r-diun?al Change of

Cha/jyes of Tempemlure iu..{ TciY)peTCllUTe

Rdalive Hiunidity hdil totwl^ Temp and Retatwe uniid% Comtant

Relative Humidity

Temperature and Changes of
Temperature hriUl Ccmalank

C <a

3'

X

Iq

I>e

Fx

B
odhs over
ve Ycur5

G R,

f

UJb

-.15
: 17.

Correlation between Weather Elements and Daily Deaths in New York City,
December to March, 1882-1886

HEALTH AND WEATHER 197

fourteen succeeding days. The lengths of the bars in Figure 17
indicate the size of the partial correlation coefficients. The three
diagrams on the left show the relation between mean temperature
and the deaths in our three categories when the interdiurnal
change of temperature and the relative humidity are held con-
stant and thus eliminated. The three central diagrams show
similar coefficients for changes of temperature when mean tem-
perature and relative humidity are held constant ; while the right-
hand set apply to relative humidity when mean temperature and
changes of temperature are held constant.

Where the bars of Figure 17 lie above the central line, the
coefficients are positive ; that is, a high condition of the weather,
such as high humidity or high temperature, is associated with
a high death rate. Where the bars are below this central line the
reverse is the case, high temperature, for example, being asso-
ciated with few deaths. But note that in studying changes of
temperature our purpose is not to discover the effect of a small
change compared with that of a large change regardless of
whether the temperature rises or falls. It is to discover whether
the effects of a rise and of a fall are the same or different. Ac-
cordingly, the average condition has been counted as that in
which there is no change of temperature from one day to the
next ; a rise of temperature has been given a plus sign and a fall
a negative sign. Hence in the central column of Figure 17 a bar
above the line means that a high death rate is associated with a
rise of temperature, while a bar below the line means that the
death rate is high when the temperature falls.

The degree of significance of the bars in Figure 17 may be
judged from the shading. Where the bars are lightly shaded
their length is less than three times the probable error. This
means that they have little or no significance. When the light
shading reaches its greatest length, that is, when the coefficient
is three times the probable error, there is one chance in twenty-
two that a bar of this length would be produced accidentally

198 CIVILIZATION AND CLIMATE

even if our two sets of figures have no real relation. Suppose for
a moment that the death rate and the relative humidity have no
real relationship. Nevertheless, with figures the size of those
here used we should accidentally get a correlation of 0.075 (three
times the probable error) once in every twenty-two correlations.
The three diagrams on the right of Figure 17 depict forty-five
correlation coefficients. Hence mere chance would be likely to give
us two that rise as high as 0.075. What we actually find is two
which rise a little above that level.

The areas marked by diagonals indicate values between three
and four times the probable error. Now, the likelihood that a
coefficient of any particular size will be produced accidentally
decreases very rapidly as the coefficients become larger. Thus
while there is one chance in twenty-two that a coefficient will be
three times the probable error, there is only one chance in one
hundred forty-two that it will be four times the probable error.
Among the ninety coefficients in the first and second columns of
Figure 17 mere chance would not be likely to give more than one
coefficient rising to the outer limit of the diagonals, but as a
matter of fact we have eleven. If the coefficient is five times the
probable error there is only one chance in 1341 that it is due
to accident arid not to a real relationship ; if six times, only one
in 19,300; seven, 427,000; eight, 14,700,000. For all practical
purposes a coefficient eight times the probable error gives full
certainty of a relationship of some sort.

In general we may say that in Figure 17, or any similar dia-
gram, the conditions that suggest a relationship are (1) coeffi-
cients more than three times the probable error; (2) a con-
siderable series of coefficients all having the same sign and hence
all falling either above or below the central line; (3) a series of
coefficients which systematically change from high values to low
or from positive to negative, or vice versa. The conditions which
are generally agreed to amount to practical proof of a relation-
ship are ( 1 ) individual coefficients which rise to at least six times

HEALTH AND WEATHER 199

the probable error; (2) several successive coefficients rising to
at least four times the probable error; (3) a considerable series
of coefficients which systematically and persistently change their
values in some orderly sequence such as from three or four times
the probable error on the plus side to an equal value on the
negative side. All three types are represented in Figure 17.

We are now ready to interpret that Figure. Bear in mind that
the length of the columns indicates the degree of relationship
between deaths and the weather elements, and has nothing to do
with the actual number of deaths. Remember also that when we
speak of temperature in what follows, we mean temperature after
the effects of relative humidity and interdiurnal changes of tem-
perature have been eliminated by means of partial correlations.
In similar fashion relative humidity and changes of temperature
mean those two factors individually after the other two have
been eliminated. To begin in the upper left-hand corner, the posi-
tion of the bar for day above the central line in diagram A
indicates that during the 726 days from December to March in
the six years under discussion, high temperature on any par-
ticular day tended to be accompanied by a large number of
deaths of children under five years of age on that same day. The
small size of the bar, however, indicates that this relationship
is too small to be considered seriously. On the other hand, the
temperature on any given day had a pronounced relation to the
deaths during the next three days, as appears in the heavy black
shading of days 1 to 3. High temperature was systematically
followed by a low death rate, and low temperature by a high
death rate. Inasmuch as we are dealing with only a single day's
weather, and inasmuch as the largest coefficient ( 0.167) is
seven times the probable error, the total effect of the mean tem-
perature for all days on the deaths among children must be
great. On the fourth day after a given temperature, however,
it has practically disappeared.

Diagram B suggests that the reaction of older persons to the

200 CIVILIZATION AND CLIMATE

outside temperature is similar to that of little children. Curi-
ously enough, however, on the day when a given temperature
occurs the effect is stronger than among the children, while on
succeeding days the opposite effect is weaker and is somewhat
more delayed. Probably the immediate harmful effect of high
winter temperature arises from the fact that when the outer air
is unusually warm for the season, our houses are likely to be
kept too warm, especially on the first day of such warmth. On
the other hand, when unusually cold days arrive, many houses
which have been too warm become cooler and that is helpful, but
soon the fires are pushed and the old condition of hot, stuffy
rooms returns.

In diagram C, showing the relationship between the mean tem-
perature and the deaths from pneumonia, the most important
feature is the fairly regular decline from a moderately high
level on the left to a low level on the right. This apparently
means that temperatures which are high for the season tend to
cause death among pneumonia patients, but have a good effect
in preventing other people from contracting the disease, so that
the death rate from pneumonia falls off after about two weeks.
Further comment on these first three diagrams is unnecessary.
They confirm the results obtained in other ways, and show that
the temperature of even a single day plays an appreciable and
measurable part in determining the general health of the com-
munity.

Look now at the middle column of Figure 17. This depicts the
relationship between the death rate and the change of tempera-
ture from one day to the next when the mean temperature and
the relative humidity are both eliminated by means of partial cor-
relations. Among children less than five years old, as appears at
the top, a rise of temperature tends strongly to cause many
deaths on the day when it occurs and on the succeeding day,
while a drop acts in the opposite fashion. So strong is this effect
that the largest partial correlation coefficient (0.202) is 8.4

HEALTH AND WEATHER 201

times the probable error. Inasmuch as there is scarcely one
chance in one hundred million that so large a coefficient should
be accidentally obtained, we may be practically certain that
changes of temperature from one day to the next (regardless of
the mean temperature) exert an important effect upon the health
of young children. The suddenness with which this effect comes
to an end is noteworthy. The portion of diagram D from the
third to the fourteenth day is typical of the coefficients obtained
when there is no relationship between two sets of phenomena.

Diagram E indicates that among older people changes of
temperature from day to day have almost the same effect as
among children, a rise being harmful and a drop beneficial. In
this case, however, the relationship is not so marked as among
the children, the delay is greater, and there is a reaction on the
fifth day. Thus the harm done by a rise of temperature, or the
good done by a fall, is partly neutralized by effects of the oppo-
site kind a few days later, but the neutralization is only partial,
as appears from the greater size of the shaded areas above than
below the zero line.

Pneumonia patients (diagram F) present another case where
effects of opposite types occur at different times. People suffer-
ing from this disease are probably harmed somewhat by a rise
of temperature on the very day when it happens, and are simi-
larly helped by a fall, but these effects are too slight to be sig-
nificant. Two days after a given change of temperature, on the
contrary, the pneumonia patients show a distinct benefit if the
change has been toward warmer conditions ; they are harmed by
a change in the opposite direction. This occurs regardless of
whether the actual mean temperature is high or low, for that
factor has been eliminated by our partial correlations, as has
relative humidity. The changes themselves appear to be the
effective agent. But how about the relatively high positive corre-
lation on the eleventh day in diagram F? There is about one
chance in one hundred and fifty that this is due to accident,

202 CIVILIZATION AND CLIMATE

whereas there is only one in about fifty thousand that the larger
coefficient of the second day is accidental. Nevertheless, the posi-
tive correlation on the eleventh day may be significant. If so, it
presumably means that a rise of temperature is accompanied by
conditions favorable to the development of pneumonia, so that
an unusually large number of people die about eleven days after-
ward. Here again we appear to have a curious contradiction
between the effect of relatively high temperature and of the
change toward such a temperature. This contrast appears so
constantly and consistently that its reality can scarcely be
doubted.

One of the clearest and most convincing features of this in-
vestigation of daily changes of temperature is its unequivocal
character. In all three diagrams (D, E, and F) the high coeffi-
cients are either higher or more numerous than in the corre-
sponding diagrams for mean temperature (A, B, and C). This
agrees with several other lines of evidence, such as the sensitive-
ness of people in monotonous climates, in suggesting that varia-
bility of temperature not only from season to season but from
day to day may be almost as important as the mean temperature
itself. Such slight evidence as is yet available also suggests that
variability in other respects such as sunshine, rainfall, moisture,
and wind may have an appreciable effect upon health. The em-
phasis thus given to variability as a distinct factor, apart from
the conditions which vary, is of much significance in connection
with changes of climate and the relation of climate to the dis-
tribution of civilization. It confirms the conclusions derived from
our study of factories, general death rates, influenza, pneu-
monia, and operations, and is itself confirmed by strong evidence
which is yet to come.

Turning now to relative humidity, we find that diagram G is
completely negative. During the years in question the relative
humidity of the air had no appreciable effect upon the health
of children under five years old in New York. This is true even

HEALTH AND WEATHER 203

when the effects of mean temperature and changes of tempera-
ture are eliminated. Among older people (diagram H) the same
is true so far as any immediate effect is concerned, for days
to 6 have irregular and insignificant coefficients. From the sixth
to the fourteenth day after any given condition of relative hu-
midity, however, there is a slight but persistent positive correla-
tion every day. On two days this rises to almost four times the
probable error. This suggests that in winter high humidity may
possibly be favorable to the contraction of diseases from which
people die a week or two later. A similar suggestion in respect
to influenza has already been discussed. The pneumonia diagram
(I), however, has a different aspect. None of its coefficients are
large enough to be significant, but the fact that the first seven
are all positive gives a hint that high relative humidity may
have a slightly unfavorable effect upon pneumonia patients.

It is most perplexing to find that different sets of data give
different indications as to the relation between atmospheric
humidity and health. The present investigation with its almost
negative results, but with a slight suggestion that damp air
facilitates the transmission of harmful micro-organisms, agrees
with the experiments of the New York Ventilation Commission,
and with the results obtained by the Committee on the Atmos-
phere and Man in its work on influenza. Opposed to this are the
results of what seem to be equally reliable investigations per-
taining to deaths after operations, to the death rate from pneu-
monia by months as set forth by Greenburg, to Besson's inquiry
into the weekly death rate in Paris, and to the monthly death
rate where millions of people were studied as described in World
Power and Evolution. Moreover, a new investigation, shortly to
be described, points even more strongly toward atmospheric
humidity as an important agent in promoting health. A possible
explanation of this apparent contradiction may be that humid-
ity affects people in two ways, directly through the skin, nerves,
and lungs ; and indirectly through minute organisms that bear

204 CIVILIZATION AND CLIMATE

disease. The disease-bearing organisms being very short lived,
are quickly influenced by variations in atmospheric moisture.
Hence a day or two of unusually high relative humidity may be
enough to give them an opportunity to produce disease. Man, on
the other hand, may be influenced more slowly so that the bene-
ficial effect of moisture upon him becomes apparent only when
he is subjected to moderately moist conditions for some time.
Thus relatively long oscillations in health may arise through the
effect of atmospheric moisture upon man, and short oscillation*
through the effect upon the bringers of disease, and the two may
easily be of opposite character.

We now come to what seems to me a most conclusive study as
to the general effect of the weather upon health. In order to gain
a comprehensive view of the variations in this effect from place
to place and likewise from season to season, I have made a fresh
investigation of the deaths each month from 1900 to 1915 in
thirty-three cities of the United States. Every city with over
100,000 population in 1910 has been used so far as mortality
data are available. Each city and each month of the year has
been treated as a separate unit, the Januaries, Februaries, and
so on being divided into two equal groups on the basis of each of
the following climatic factors :

(1) Mean daily temperature.

(2) Mean daily relative humidity (average of 8 a.m. and
8p.m.).

(3) Variability or storminess (number of storms whose
centers passed within two hundred miles of the given city, allow-
ing double weight to those within one hundred miles).

(4) Wind (total number of miles per month). Five cities
only.

Data for the wind were investigated only at New York, Balti-
more, Chicago, St. Louis, and San Francisco. The Chicago data
are doubtful because the growth of the city appears gradually

HEALTH AND WEATHER 205

to have cut off the wind from the Weather Bureau Station and
caused an apparent decline in windiness.

An example will show how the data were used. The eight warm-
est Januaries in New York averaged 6.0 F. warmer than the
eight coldest, and had fewer deaths by 0.6 per cent. In February
the excess of temperature in the eight warmest months amounted
to 6.5 and their death rate was 4.1 per cent less than that of
the cooler months. In March the corresponding figures were
6.4 and 9.7 per cent; in April 3.8 and 4.5 per cent; in May,
on the contrary, an excess of 3.5 in temperature was accom-
panied by a death rate 1.5 per cent greater in the warm months
than in the cool months, while in July, although the eight warm
months averaged only 2.8 above the eight cooler months, the
excess in their death rate rose to 14.2 per cent.

The use of this method brings out many interesting facts on
which we cannot now dwell. It eliminates entirely all complica-
tions due to the seasons, for each month stands by itself and can
be compared with any other month. It likewise shows nothing as
to the general effect of the seasonal variations of either weather
or mortality. It merely shows how the departures of any par-
ticular climatic factor from the normal for that particular
month affect the death rate. Boston, for example, is thus found
to be benefited in summer by the coolness of its damp east winds ;
dampness without coolness is the bane of New Haven in summer ;
New York, by reason partly of its great size and partly of its
fine climate, is unusually regular in its responses to the weather,
but for some unexplained and possibly accidental reason is
averse to storms in midsummer as well as in midwinter. Although
Baltimore is hot in summer, it suffers little harm from humidity
even when that factor runs high. Baltimore is likewise benefited
at all seasons if it has more than its usual allowance of storms,
while Boston, being far north, normally has so many storms that
it is better off during the less stormy months except in summer.
Chicago, on the other hand, is benefited by storms except during

o

CO

A

.

Q

*H

O

4)

3

Q

H
O

208 CIVILIZATION AND CLIMATE

the coldest months ; it is likewise benefited by high relative hu-
midity except in the autumn, but curiously enough it is not bene-
fited by high temperature in winter, perhaps because the warm
months experience dry southwest winds, bare ground, and dust.
Somewhat the same is true in Pittsburgh and Denver. Cleveland
and San Francisco have climates of such a type that departures
from the normal produce relatively little effect, whereas in cities
like St. Paul and Minneapolis in the north and St. Louis and
Cincinnati farther south the effects are much greater. In
southerly places like Nashville and Memphis the effect of de-
partures from the normal weather is especially great. Presum-
ably it is still greater in the far south, but Atlanta is the only
other southern city for which data are available.

In Figure 18 (a and b) the thirty-three cities have been com-
bined into geographical groups. Each city is weighted according
to its population as follows : 100,000 to 200,000 = 1 ; 200,000
to 400,000^2; 400,000 to 1,000,000 = 3; 1,000,000 to
2,000,000 = 4 ; over 2,000,000 = 5. The curved lines indicate
the extent to which the death rates in the eight warmer, moister,
or stormier Januaries, Februaries, and so forth, from 1900 to
1915 differed from the corresponding death rates in the eight
cooler, drier, and less stormy Januaries and other months. The
figures in the scales beside the diagrams indicate departures in
percentages of the normals. The normals are the estimated num-
bers of deaths that each place would have experienced per month
in any given year if the number of deaths changed regularly in
response to the growth of the city and the improvements in medi-
cal practice without regard to weather, seasons, epidemics, and
so forth. The method of getting the normals is explained in
World Power and Evolution. In Figure 18 and the other figures
of the same kind all the curves have been smoothed by the
formula

a + 2b + c
b,

HEALTH AND WEATHER 209

which is a common way of eliminating the confusing minor ir-
regularities which arise because of the small number of years
for which data are available. In the figure the heavily shaded
areas mean that the months with relatively high temperature,
high humidity, high storminess, or high winds had lower death
rates than the months in which the weather factors stood lower.
For example New York, Philadelphia, and New Haven form a
group of cities lying within a distance of about one hundred and
fifty miles and having similar climates in spite of individual
idiosyncrasies. In the first column of Figure 18, diagram B
shows that in these cities the months of January, February, and
especially March are too cold, for in each month the death rate
in the eight warmest years was from 3 to 5 per cent lower than
in the eight coldest, as appears from the dark shading. In April
the average temperature was about right, for the curve crosses
the zero line ; in that month the bad effect produced by weather
that is a little too cool is balanced by the corresponding effect
of high temperature during the same month. As summer ad-
vances the warm months begin to have a disadvantage, as is
indicated by the dotted shading. The eight warmer Julies had an
average death rate about 6 per cent greater than that of the
cooler Julies even in the smoothed curve ; in the unsmoothed
curve this rises to 10.4. Inasmuch as the eight warmest Julies
averaged only about 3 warmer than the coolest eight, each
degree of excessive temperature raised the death rate more than
3 per cent.

Passing on to the second column of Figure 18 it appears that
high relative humidity is beneficial to New York and its neigh-
bors throughout the winter, and especially in April when an
excess of 6 per cent in relative humidity is accompanied by a
diminution of 6 per cent in the death rate. During the three
summer months the highest humidities do harm, the smoothed
maximum excess of deaths being 4 per cent while the unsmoothed
figures are 5.8 per cent accompanying an excess of 6.4 per cent

210 CIVILIZATION AND CLIMATE

in the relative humidity. The duration of the period when hu-
midity is harmful is short. During the autumn and early winter
the nearness of the curve to the zero line indicates that it makes
little difference whether those months are relatively dry or
moist, perhaps because the general conditions of health are so
good that people can resist extremes which might harm them
at less favorable seasons. Nevertheless, the damper rather than
the drier months were the best.

In the third column of Figure 18 the storms of the New York
group of cities appear to have had less influence than either the
temperature or the humidity. In the late winter and spring the
more stormy months were the most healthful; in summer the
less stormy ones had a very slight advantage, too small to be
significant ; the autumn again was like the late winter, while
November and December were like June and July. This particu-
lar curve happens to be one where the effect of storms is at a
minimum. The most important thing about it is that the dark
shading is much more extensive than the light, which means
that on the whole the stormiest months were times of better
health than those that were less stormy.

In the right-hand column of Figure 18, where the effect of the
wind is shown, the upper diagram is based only on New York.
It is very symmetrical, and indicates that in winter high winds
are accompanied by a high death rate, while in summer they are
accompanied by a low death rate.

The question at once arises whether the four types of curves
in Figure 18 really represent the effect of the individual climatic
factors or whether each curve is compounded of the effects of
all four factors. For example, are high winds really favorable
in summer, or do they merely appear to be so because they are
accompanied by low temperature or low humidity or some other
favorable condition? The answer is found in Figure 19. The solid
curves there are the same as those which we have just been
examining for the New York group of cities in Figure 18, but

HEALTH AND WEATHER

211

are plotted on a larger vertical scale. The dotted lines are the
same curves corrected to allow for the other climatic factors.
In making the corrections it was assumed that temperature is
the most important climatic factor, and is followed by humidity,

JFMAMJJyASOVDJ

5 Temperature

Relative Humidity

-2
-3
-f

Storms

3

z

1

o Wind New York

-1

-2

Figure 19. Correction for Effect of Other Climatic
Factors

storms, and wind in this order, for this is what is commonly
supposed, and it is supported by the investigation here de-
scribed. The method of making the corrections was as follows :
Knowing how much the temperature and the death rate in the
moister months differed from those in the drier months, it was

212 CIVILIZATION AND CLIMATE

easy to determine how much effect a given difference of tempera-
ture produces. Since we know the difference in temperature be-
tween the moister and drier months, it is possible to make
allowance for this difference month by month and thereby elimi-
nate from the humidity curve the effect of temperature. When
the humidity curve had thus been corrected, the storm curve was
corrected in the same way on the basis not only of the tempera-
ture curve but of the corrected humidity curve. Next the humid-
ity curve was corrected to allow for the effect of storms, and the
temperature curve to allow for the effect of both storms and
humidity. Finally, the wind curve was corrected to allow for
variations in all three of the other factors. The result was the
dotted lines in Figure 19. At the top of the diagram the cor-
rected and uncorrected lines showing the effect of temperature
upon the death rate are practically alike. The next pair of lines,
those for humidity, are almost alike, but allowance for the other
factors seems slightly to reduce the importance of humidity in
the spring and raise it in summer. In the third pair of lines the
corrections seem to lower the general level a trifle, thus making
it appear that storms are a bit more beneficial than appeared at
first sight. Finally, the corrected and uncorrected curves for the
winds are completely different. The good effect in summer has
practically disappeared in the corrected curve. If the data for
the wind had been available from New Haven and Philadelphia,
as well as New York, the smoothing of the wind curve might pos-
sibly have been still more complete. As things now stand, the cor-
rections seem to indicate that in Figure 18, in spite of minor
details due to other factors, the apparent effects of temperature,
humidity, and storms represent approximately the real effects,
and would represent them still more closely if a larger number
of cities were averaged together as will be done in Figure 20.
As for the wind, it may be that high winds in winter have some
direct effect in raising the death rate, but in summer practically
all of their effect appears to arise from the conditions of tern-

HEALTH AND WEATHER 213

perature, humidity, and storms, or variability which accompany
them.

Let us return now to Figure 18. In the left-hand column the
groups of cities east of the Mississippi behave almost as one
would expect. Except for Rochester and Buffalo (D), which
appear to be practically never too hot, all the diagrams are
heavily shaded in winter and lightly shaded in summer, thus
indicating that the winters from Tennessee and northern
Georgia to Minneapolis and Boston are too cold and the sum-
mers too hot for the best health. In the center of this area, to be
sure, group E (Cleveland, Toledo, and Detroit), group F
(Pittsburgh, Columbus, and Indianapolis), and group G (At-
lanta, Nashville, and Memphis), show a curious depression in
summer, as if the harmful effect of hot weather was somehow
inhibited, perhaps because the hot winds are dry so that the bad
effects of high humidity are mitigated when the temperature
rises. West of the Mississippi all of the groups (K to N) suggest
that while hot summer weather is generally bad except on the
cool Pacific coast, unusually warm winter weather is also often
harmful. This is presumably because the warm months are gen-
erally dry ; and our monthly data seem to show that dryness
is almost always harmful in cold weather.

That this last statement is true seems to be abundantly veri-
fied by the second column in Figure 18. Here there is no such
equal distribution of heavy and light shading as in the diagrams
showing the effect of temperature. On the contrary, almost every
individual diagram displays a greater area of heavy shading
than of light, and some of the diagrams such as C, G, and K have
practically no light shading. In general the amount of heavy
shading, that is, the good effect of atmospheric moisture, in-
creases as one goes from the moister and cooler parts of the
country to those that are warmer and drier. It reaches a maxi-
mum in diagram L for Denver and Spokane, the two cities in
our list where the atmospheric moisture is least. This clearly

214 CIVILIZATION AND CLIMATE

means that in practically all parts of the United States, so far
as data are available, and especially in the drier parts, the health
of the inhabitants would be materially improved if there were
more atmospheric moisture. This clear-cut and apparently un-
equivocal result agrees with the study of deaths set forth in
World Power and Evolution, and with the study of the death
rate after operations. The contrast between these three lines of
evidence, on the one hand, and the results of our investigation of
daily deaths in New York, together with the work of the Ventila-
tion Commission, on the other, is the reason for our suggestion
that humidity has two diverse and opposed effects. It seems to
be beneficial in its direct effect, except at temperatures above
the optimum, and harmful in its indirect effect through bacteria.

Much of what has been said of atmospheric moisture is like-
wise true of storms, as appears in the third column of Figure 18.
Notice how largely the heavy shading predominates. Note also
that it is scarce in northern groups of cities such as B, H,
and M, but increases as one goes southward until in groups G,
J, and especially C practically every month of the year shows
a lower death rate when storms are relatively abundant than
when they are few. Here, just as in the case of humidity, the
regions which have few storms, like those which have little at-
mospheric moisture, give heavily shaded diagrams because an
increase in the number of storms is an advantage to health at
practically all seasons. But places like Boston, which are exposed
not only to many storms but to strong oceanic winds, may get
too many storms in the winter.

The question of the effect of storminess is so important that
I have prepared Figure 20 to show what happens when stormy
periods last several months. The upper diagram in each case
shows the conditions when the storms of a given month are com-
pared with the deaths of that month. In the second diagram the
months of the sixteen years used in our study have been grouped
into halves according to the number of storms not only in the

HEALTH AND WEATHER 215

month when deaths occurred, but in the preceding month. In the
lower diagram the storminess of three consecutive months has
been compared with the deaths in the third month. To begin
with Boston, a relatively high degree of storminess lasting only
a single month is slightly beneficial in summer and again, curi-
ously enough, in winter, but this may be a mere accident due to
the shortness of our record. If the stormy period lasts two months
the good effect of storms is much increased. In fact, Boston's
health would apparently be distinctly improved if the city could
have frequent periods of relatively high storminess lasting two
months during the summer, but not during the late winter. The
lightly shaded area in the autumn is so small that it may be acci-
dental. If the periods of storminess last three months, however,
Boston gets too much of them and the death rate rises markedly.
In other words, Boston seems to lie close to the fortunate level
where it gets neither too many nor too few storms in the long
run, although in the more extreme periods it gets too many, just
as in milder periods it does not get enough.

Contrast Boston with Chicago in Figure 20. A single month
of more than the average storminess helps Chicago a good deal
during all seasons except midwinter. But two successive months
of unusual storminess, and especially three, do harm at prac-
tically all seasons. In other words, an increase in storminess
hurts Chicago more than Boston. Nevertheless, both cities evi-
dently profit greatly by the fact that they have many storms,
as appears when they are compared with cities farther south.
In New York, for example, Figure 20 shows that increased
storminess during periods of more than one month is beneficial in
summer but not in winter, while Seattle is benefited by increased
and prolonged storminess at practically all seasons. This simply
means that New York, with less severe storms than Chicago or
Boston, would profit by a mild increase in storminess. On the
other hand, Seattle, with far less storminess than the other cities
would be better off to have decidedly more.

Boston
AMJJyASOND,!

CJiieogo
JFM A M J Jy A 6 O N D

New Yorlt

Figure 20. Excess or Deficiency of Deaths in Relation to
Stormy Periods Lasting One Month (Upper Diagrams),
Two Months (Middle), and Three Months (Lower),
1900-1915

HEALTH AND WEATHER 217

At the bottom of Figure 20 Baltimore and Washington, which
are treated as a single unit, and Memphis on the Mississippi
River in Tennessee present still a third type. At all seasons
their stormiest months are the most healthful, for these cities
lie toward the southern edge of the storm belt. They have far
less stormy weather than New York, Chicago, and Boston, but
more than Seattle. During the years under discussion all these
places were especially stormy in March. They are benefited by
prolonged periods of storminess in summer and autumn, but
cannot stand such periods in the winter and spring. Too much
emphasis must not be placed on the minor details of any of the
diagrams in Figure 20, especially that of a small city like Mem-
phis, for the number of years included in our data is small.
Nevertheless, there seems to be little question that storminess
has an important effect upon health. In a belt of country ex-
tending from New York and Boston westward to Chicago, the
beneficial effects of storminess are greatest. To the north of that
belt increased storminess appears to have a harmful effect upon
health; to the south the present degree of storminess is not
enough on an average, and a higher degree regularly causes an
improvement in health at most seasons.

In concluding this chapter let us turn to Figure 21. Here the
data for all of our thirty-three American cities have been com-
bined into a single diagram. This shows the effect produced by
an unusually high condition of any one of the four climatic fac-
tors upon the death rate in the whole northern United States
together with the Pacific coast. In using such a large area there
is great opportunity for opposed conditions in different regions
to cancel one another, but this cancellation is far from complete.
Before we interpret Figure 21 let us consider for a moment the
degree to which the diagrams may be the result of chance. In
most such cases it is the custom for mathematicians to compute
the probable error by means of a formula. Our method, however,
whereby cities are weighted according to their population, and

218

CIVILIZATION AND CLIMATE

the departures are reckoned from normals which pertain to the
year instead of the individual months, would cause such compu-
tations to take an excessive amount of time. Accordingly I have
calculated the data for four months exactly as in the diagrams,
except that pure chance has been allowed to control the choice
of months for each of the two groups of eight years into which
the sixteen years have been divided. The smoothed results for

JFMAMJJyASONDJ

Temperature

Relative Humidity

Storms

Wind

Figure 21. Net Effect of Weather in the United States

comparison with Figure 21 are January 0.47, April 0.40, July
0.72, October 0.79. In each of the four diagrams the extremes
are four to seven times as great as the accidental variations thus
obtained. This fact, together with the systematic character of
the results, makes it practically certain that we are dealing with
real relationships and not with accidental coincidences.

If this be accepted, our diagrams in Figure 21 show that in
the United States as a whole, excluding the South except in

HEALTH AND WEATHER 219

California, a certain amount of harm is done to health in winter
by low temperature, but not so much as one would expect. The
high temperatures of summer do much more harm. The effects
of the wind seem to be very clear, but are probably due largely
to the conditions of temperature, humidity, and storminess
which accompany the winds. Now for the most surprising fact ;
lack of moisture does almost as much harm to the United States
in winter as does low temperature, while in the spring it does
only a little less than does high temperature in summer. At all
seasons the United States as a whole, omitting the parts for
which we have no data, has poorer health in unusually dry
weather than in that which is unusually moist. Almost the same
thing is true of storms. The country is better off when stormi-
ness is unusually abundant, except in the late fall. In Figure 21
the average departures from the zero line, regardless of whether
the departures are positive or negative, are as follows : tempera-
ture 4.09 per cent, relative humidity 3.96, and storminess 3.60.
These figures seem to represent approximately the relative im-
portance of the three great climatic factors. Thus in spite of
certain puzzling facts as to humidity, the general result of our
study of health in relation to the weather is to confirm the re-
sults of our previous study of efficiency in factories. The same
conditions of temperature, humidity, and variability which cause
people to work quickly or slowly in the ordinary affairs of life
seem also to cause their health to be good or poor.

CHAPTER X
THE IDEAL CLIMATE

WE are frequently told that the Riviera or southern Cali-
fornia has an ideal climate. Florida lays claim to it m
winter, the Alps in summer. Two of the few regions which rarely
assert their preeminence in this respect are Boston with its east
winds and London with its fogs. Yet in many ways they have a
strong claim to high rank. It all depends upon what we mean by
"ideal." For rest and recreation a warm, equable climate is
doubtless most delightful; for a fishing or climbing trip some-
thing quite different is desirable. For most people the really
essential thing in life is the ordinary work of every day. Hence,
the climate which is best for work may in the long run claim to
be the most nearly ideal. But such climates are also the ones
that are best for health. Hence they are the ones which people
will ultimately choose in the largest numbers. The few disagree-
able features at certain seasons are no worse than the shiver
at the beginning of a cold plunge.

On the basis of both work and health, the best climate would
apparently be one in which the mean temperature rarely falls
below the mental optimum of perhaps 38, or rises above the
physical optimum of about 64. From this point of view the
most ideal conditions would seem to be found where the tempera-
ture for the year as a whole averages not far from 51, as at
London, Paris, New York, and Peking. In four chief portions
of the globe, the winter temperature averages not far from 38,
and that of summer not far from 64. The first of these is Eng-
land. At London the thermometer averages 38 in January and

THE IDEAL CLIMATE 221

63 in July, while at Liverpool the figures are 39 and 60. If
an average of 51 at all seasons were ideal, southwestern Ireland,
with a range of from 45 to 59, and the Hebrides, from 42
to 55, would be more ideal than London. On the continent,
where the seasonal variation is greater than in Britain, the
length of the relatively unfavorable periods with temperatures
above 65 and below 38 also increases.

A second region where the temperature conditions approach
the ideal is the Pacific coast of the northern United States and
southern British Columbia. Seattle, averaging 39 in January
and 64 in July has practically the same temperature as Lon-
don. Southward the seasonal range decreases. San Francisco,
averaging 49 in January and 59 in September after the cool
summer fogs have passed away, may claim in many ways to be
ideal. Still farther south the temperatures of Los Angeles and
San Diego, 53 or 54 in January and 69 in August, fluctuate
about the physical optimum and would be ideal for physical ac-
tivity if mean temperature were the only criterion. The mental
optimum, however, is lower than the temperature of all except
the unusually cold days, and variations from day to day are
rare. A short distance inland the Californian climate becomes
less favorable than on the coast, for the average in summer at
Fresno, for example, is 82. Even though the heat is mitigated
by low humidity, the continuance of such high temperature
causes people to feel indisposed to activity.

England and the Pacific coast owe their climatic excellence
largely to the fact that ocean winds from the west blow freely
over them. Two regions in the southern hemisphere enjoy the
same advantage, namely, New Zealand and part of South Amer-
ica, including southern Chili and portions of Patagonia. We are
apt to think of these South American regions as sparsely settled
places of little importance. This is true, for the present, but it
is not because the climate prevents activity. The climate, to be
sure, is a drawback, but the harmful feature is not the tempera-

222 CIVILIZATION AND CLIMATE

ture, but the rainfall. Plants, not man, are the chief sufferers.
Unlike our other three regions, this part of the world has a
deficient rainfall except in the west, and there high mountains
hinder settlement. In spite of this, the few small portions of
Patagonia that permit profitable agriculture are making prog-
ress and would doubtless do so rapidly if not hampered by re-
moteness, by the absence of railroads, and by the social handi-
cap of their aborigines.

It must not be inferred that the climates of Patagonia, New
Zealand, England, and the Pacific coast of the United States
are necessarily ideal. Mean temperature is not the only impor-
tant factor. Among other things the relative humidity must be
considered. The deficiency of moisture in Patagonia not only is
disastrous economically, but, to judge from the preceding
studies, it lessens man's energy. A similar effect is produced by
excess of moisture, and thus harm is done in Ireland and western
Scotland, which would otherwise be almost as fortunate as Eng-
land. New Zealand, the central Pacific coast of North America,
and England itself are sometimes unduly damp for long periods,
but nevertheless enjoy a relative humidity of not far from 70 per
cent much of the time. Other regions, however, such as the
eastern United States and central Europe, seem to be more
favored in this respect.

The change of temperature from day to day, as we have seen,
seems to be as important as relative humidity, and must accord-
ingly be considered more fully. Its effect on human activity seems
to be second only to that of the mean temperature of the seasons.
The intensity and number of daily changes depend upon two
chief factors, first, the range of temperature from the warmest
to the coldest part of the year, and second, the number of
cyclonic storms. Where the winters are cold and the summers
hot, the changes from day to day are also extreme. For instance,
in the Dakotas, where the mean temperatures of January and
July differ by 60 F., a change of equal magnitude may take

THE IDEAL CLIMATE 223

place in twenty-four hours. On the other hand, in a place like
the Congo, where the difference between the coldest and warmest
months is only three or four degrees, the days are correspond-
ingly uniform. The whole matter is illustrated by maps in many
physical geographies and in such publications as Bartholomew's
Meteorological A tlas. The parts of the world where the change
of seasons favors a highly advantageous degree of change from
one day to the next include most of North America, but omit
Florida, the Pacific coast of the United States, and the regions
from the Mexican border southward. All of central and eastern
Europe is also included except parts of Italy and Greece. A large
area in North Africa and a small area in the south of that conti-
nent also rank high, as do central Australia and the part of
South America which includes central Argentina. Finally, all
except the southern parts of Asia lie within the high area. Thus
this particular favorable condition occurs not only in many
regions whose climate is also good from other standpoints, but
in a much larger number whose general climatic conditions are
decidedly unfavorable. This is not surprising, for the beneficial
effect of pronounced changes of temperature from day to day
is often nullified by great heat in summer or extreme cold in
winter. Moreover, the seasonal range of temperature forms only
one of the two factors which determine the amount of stimulation
derived from changes of temperature from day to day.

The other factor is the number of cyclonic storms. By this,
as has already been explained, we mean the ordinary storms
which produce our changes of weather from day to day in the
United States and Europe. Probably the storms are more im-
portant than the range of temperature from season to season,
for they bring rain, humidity, changes in the winds, and all sorts
of stimulating variations. The world's stormiest region, so far
as known, includes the Great Lakes region of the United States
and southern Canada. Around this center there is an area of
great storminess extending southward approximately to Mary-

224 CIVILIZATION AND CLIMATE

land and Kansas, and northwestward through the Dakotas to
Alberta. Eastward it includes New England and the Maritime
provinces, while northward it quickly disappears. To the south
the storminess diminishes gradually, so that Florida has a
moderate degree of variability in winter but not in summer.
Southern California is the least stormy part of the United
States. In Europe the very stormy regions include Britain,
most of France, Germany, parts of Scandinavia, and the north-
ern part of Italy, together with western Austria and the Baltic
region. In Asia, Japan is the only place where cyclonic storms
are at all numerous. The lands of the southern hemisphere gen-
erally have few storms. New Zealand is the chief exception, al-
though there they do not cause such great changes of tempera-
ture as in America and Europe. The extreme southern tip of
South America is likewise stormy, but its storms do not cause
much variability. On the contrary, they give rise to a monotony
of wind and clouds which is extremely deadening, according to
the testimony of those who have lived in such places as Tierra
del Fuego or the Falkland Islands. Farther north, in central
Argentina, there is a moderate number of storms, comparable
to those in the southern United States, and their effect is dis-
tinctly favorable.

We are now prepared to estimate the relative stimulating
power of the various climates of the world. In England, for ex-
ample, the mean temperature of the seasons and the degree of
storminess are both highly favorable, while the seasonal changes
are only moderate. Germany is above medium in temperature,
and high in seasonal changes and storminess. In this respect, it
resembles the northeastern United States and southern Canada.
Japan is similar except that it is somewhat too warm and damp.
The coast of British Columbia and of the neighboring states is
highly favorable in mean temperature, and medium in storminess
and seasonal changes. Around San Francisco, the mean tempera-
ture is still better, but both seasonal changes and storms are

THE IDEAL CLIMATE 225

mild. In compensation for this, however, there are frequent
changes of temperature because fogs blow in from the ocean,
and are quickly succeeded by the warm, bright weather which
generally characterizes the interior. Farther south where the
fogs cease, the conditions become less favorable from the point
of view of the changes from one day to another, although the
mean temperature of the seasons still remains advantageous.

The chief defect of the climate of the California coast is that
it is too uniformly stimulating. Perhaps the constant activity
which it incites may be a factor in causing nervous disorders.
When allowance is made for the fact that California's urban
population is relatively smaller than that of states like Massa-
chusetts and New York, insanity appears to be even more
prevalent than in those states. Moreover, the cities of the Cali-
fornia coast have the highest rate of suicide. In 1922 four
California cities led the list in suicides, the number per 100,000
population being : San Diego 47.8, Sacramento 37.9, San Fran-
cisco 30.4, and Los Angeles 30.3, against about 15 in the eastern
cities. Possibly these facts may be connected with the constant
stimulation of the favorable temperature and the lack of relaxa-
tion through variations from season to season and from day
to day, although other factors must also play a part. The people
of California may perhaps be likened to horses which are urged
to the limit so that some of them become unduly tired and break
down.

In the same way the people of the eastern and central United
States are more nervous and active than those of Europe but
not necessarily more efficient because of still different climatic
handicaps. They are alternately stimulated and relaxed by
frequent changes from day to day, and in this are like horses
that are well driven. In the spring and autumn, however, the
combined effect of ideal temperature and highly invigorating
daily changes spurs them to an astonishing degree of effort.
Then comes the hot summer or the cold winter, either of which is
debilitating. People do not diminish their activity at once, espe-

226 CIVILIZATION AND CLIMATE

ciallj in the winter. They draw on their nervous energy, and thus
exhaust themselves. They are like horses which pull on the bit,
and when urged a little break into a run, straining themselves by
their extreme speed. Then they are pulled up so suddenly that
they are thrown back on their haunches and injured. In Ger-
many somewhat the same conditions prevail, although not to so
great an extent. England apparently comes nearer to the ideal
than almost any other place. The climate is stimulating at all
times, both by reason of abundant storms and because of a
moderate seasonal range. It never, however, reaches such ex-
tremes as to induce the nervous tension which prevails so largely
in the United States.

In strong contrast to these highly favored regions are such
places as the center of Asia, where the winters are depressingly
cold and the summers unduly hot. The range from season to
season is apparently helpful, but its good effects are largely
nullified by the infrequency of storms. Day succeeds day with
no apparent change. In the desert of Takla-Makan in Chinese
Turkestan in the fall of 1905, 1 found that one of the most sur-
prising features was the way in which winter came upon us
unawares. Each morning the thermometer stood a trifle lower
than the preceding morning, but there was never any change
such as that which we so often experience in America when the
first severe frost suddenly comes after a series of days as warm
as summer. Frost at last began to prevail at night, but not until
we found the water frozen hard in the morning did we realize that
winter was upon us. So it goes, month after month, with deaden-
ing monotony. Yet when a storm does come the change is often
much more extreme than in more oceanic regions. It is frequently
so great that its value as a stimulus is much diminished.

Tropical regions suffer even greater disadvantages than do
places like the center of Asia. Not only is the temperature un-
favorably high, but there are practically no cyclonic storms
except in portions where a few hurricanes occur each year.

THE IDEAL CLIMATE 227

Thunder storms, to be sure, are abundant, but they rarely bring
any important change of temperature. Moreover, the seasonal
range from the warmest to the coldest month is generally less
than the difference between day and night. Day after day dis-
plays no appreciable variation from its predecessor. The uni-
formity of the climate seems to be more deadly than its heat.
Such uniformity, perhaps as much as the high temperature and
high humidity, may be one of the most potent causes of the
physical debility which affects so many white men within the
tropics, and which manifests itself in weaknesses such as drunk-
enness, immorality, anger, and laziness. Even in tropical high-
lands the same deadening monotony prevails, although to a less
degree than in the lowlands. Such monotony is perhaps the
condition which will do most to prevent the white man from living
there permanently for generation after generation. His general
health may not seem to suffer, but if he works hard he is in great
danger of breaking down nervously. The temperature of the
highlands may be highly stimulating. There are many places
where the mean temperature during every month in the year is
within a few degrees of either the physical or mental optimum,
or of their average. At Quito in Ecuador, for example, the cold-
est month, November, averages 54.3 F., and the warmest
months, February and September, 55. Nowhere within the
tropics, however, are there any regions which enjoy the physical
optimum at one season and the mental optimum at another. If
we are justified in associating a high rate of insanity on both
the Atlantic and Pacific coasts of the United States with the
peculiar climatic conditions, we should expect that white men
in tropical regions at high altitudes would suffer still more in
the same way, or else would become inert, but no figures seem to
be available to determine this point.

We might proceed to discuss scores of ways in which a knowl-
edge of the exact effects of climate may assist in the under-
standing of historic events, or help in guiding future develop-

228 CIVILIZATION AND CLIMATE

ment. As a step in this direction let us construct a map of the
world showing the degree of energy which we should expect
among normal Europeans in various regions on the basis of cli-
mate. This map will be based entirely on our studies of work
among factory operatives and students, and I shall leave the
description of it unchanged from the first edition of this book
except for the correction of an error. Later we shall test this
map of "climatic energy" by one of "climatic health" based on
our studies of mortality and disease. Both maps are determined
by (1) the mean temperature month by month, (2) the amount
of change from one day to another, and (3) the relative humid-
ity. The conditions which prevail at various seasons in the east-
ern United States duplicate those of almost every portion of
the globe. There are hot, dry days like those of the Sahara ; hot,
damp days like those in the Amazon forests ; cold days like those
on the great ice-sheet of Greenland, and days of almost every
other description. At this point we must make an assumption
which cannot be tested until vastly more data have been col-
lected. Let us assume that the continuance of a given condition
produces the same effect as its temporary occurrence. For ex-
ample, in Connecticut our measurements of the effect of days
having a mean temperature of 75 are based on occasional days
scattered through the summers of several years. Only in rare
cases do four or five days of such extreme temperature follow in
succession without interruption by more moderate weather. The
actual figures show that the first hot day does not greatly
diminish people's energy, for the human body is able to resist
for a while and to carry the impetus of previous good conditions
into the first part of a bad period. After two or three days, how-
ever, the heat takes hold on people and makes them inefficient,
or even causes some to collapse. If such weather continued for
months we should become somewhat accustomed to it, and the
period of collapse would be past. Just what the rate of work
would then be we cannot yet determine. It would almost certainly

THE IDEAL CLIMATE 229

be slower than on the first hot day, but it would probably be
faster than on the third or fourth. Because of this uncer-
tainty we are obliged for the present to assume that it would
be equal to the average of a number of first days and a much
smaller number of second, third, fourth, and so on. Having
made this assumption, but recognizing that it needs testing, we
may go on to construct our map. We must remember that it is
not supposed to be a map of the actual energy displayed by the
people of various places, but of the energy that we should expect
among Europeans if they lived in these places and were influ-
enced as are the people of the eastern United States.

In making such a map it is fortunate that the most important
factor is also the one most carefully tabulated by climatologists,
and for which our investigations of energy give the most un-
equivocal results. The mean temperature for every month in
the year is given for about 1100 stations in all parts of the
world in Hann's Klimatologie. The third curve from the top in
Figure 10, it will be remembered, shows the average efficiency
which would be expected at any given temperature on the basis
of the work of factory operatives in Connecticut and New York.
A table inserted as an appendix to this book shows the value for
each degree of temperature according to the centigrade scale,
the maximum being reckoned as 100 at a temperature of 15 C.
or 59 F. To determine the effect of mean temperature upon
human activity we simply take from Hann the mean temperature
of each month, and then from the table in the Appendix, or from
the curve in Figure 10, find the corresponding relative efficiency.
Then we add the values for all the months. If every month had
an average temperature of 59, with a corresponding relative
efficiency of 100, the efficiency for the place in question would
be 1200. As a matter of fact this is never reached, but London
stands at 1196.6, San Francisco at 1198.6, and Quito in Ecua-
dor at 1198.9. The worst place is Massaua near the southern
end of the Red Sea, where the figure is 1070.

230 CIVILIZATION AND CLIMATE

The next process in constructing a map of climatic energy is
to determine the effect of changes of temperature from day to
day. Unfortunately exact statistics are not available in most
regions, and we are obliged to employ an approximation. Since
changes from day to day depend chiefly upon the seasonal range
of temperature and the number of storms, I have combined the
two, giving equal weight to each, and giving the two together
approximately one half the weight assigned to seasonal tempera-
ture. That is, the difference between Quito and Massaua, as
stated in the last paragraph, is 128.9. This represents the maxi-
mum effect of the seasons, so far as the average temperature is
concerned. The maximum effect of the seasons as far as changes
from day to day are concerned is reckoned as 30, and the maxi-
mum effect of storminess on the same basis is also reckoned as
30. Since highly extreme conditions are not favorable, I have
assumed that no seasonal change beyond 30 C., or 54 F., is of
value, and also that changes below 7 C. (19.4 F.) or above
23 C. (73.4 F.) are of no value. In other words, if the range
from the mean temperature of the coldest to the warmest month
is from below 7 C. to above 23 C., it is reckoned as having a
value of 30, just as it would be if 7 were the lowest point
and 23 the highest. If the range should be from 4 C. to 16 C.,
it would be reckoned as having a value of 12, while if it were
from 20 C. to 28 C., the value would be only 3, because the
extremely hot weather above 23 would scarcely be stimulating
even if there were slight changes from day to day. In the same
wav extreme storminess does not produce an effect in propor-
tion to the number of storms. One storm may succeed another
so rapidly that the weather ceases to have sufficient variety, and
becomes dull and lowering all the time. This is the case at Cape
Horn, and also in certain parts of the North American Great
Lakes region in winter. Accordingly a storminess of 20 centers
per year according to Kullmer's scale which means far more
than 20 storms is reckoned as the optimum. Greater stormi-

THE IDEAL CLIMATE 231

ness is held to have the same stimulating value as 20 centers,
while everything lower is counted proportionally. The whole
matter is so technical that it cannot be understood without de-
tailed explanations. These are now unnecessary, since the very
simple method to be described shortly gives an almost identical
but more accurate map.

Humidity has not been considered, because the necessary
figures are not available. In most of the cooler parts of the
world it would make little difference, although a few unduly
damp places like Ireland, or excessively dry regions like Chinese
Turkestan would be lower than now appears. The chief differ-
ence would be in the warm portions of the world. Agra in north-
ern India, for instance, now has a lower rank than Bombay and
Calcutta, but if allowance were made for humidity this would
probably be reversed, for Agra is pleasantly dry much of the
year. The same reversal would probably occur between dry
Khartum and wet Equatorville on the Congo. Arizona and other
desert portions of the United States would also make a better
appearance than on the present map. It must not be forgotten,
however, that our data for New England show that extreme
dryness does more harm than extreme humidity. This, however,
does not apply to high temperatures. Under such conditions
great humidity is undoubtedly most debilitating. Yet even when
the air is hot, it may be too dry. In such a place as Death Valley,
in summer with the thermometer at 100 to 135 in the shade,
it is almost impossible to drink enough water to preserve normal
physiological conditions. Even a brief period of physical ac-
tivity gives rise to much discomfort, and people who stay
through the summer are in danger of suffering permanent injury
to health.

Our knowledge of the effect of both extreme humidity and
extreme dryness is unfortunately still qualitative rather than
quantitative. Some day, however, exact figures for all the vari-
ous climatic elements will be obtainable, and we shall construct

232 CIVILIZATION AND CLIMATE

a map showing the actual efficiency to be expected in every part
of the world. It will be so accurate that the manufacturer, for
example, who contemplates establishing a factory, will be able
to determine the precise efficiency of labor in the different places
which he has in mind, and can put the matter into dollars and
cents for comparison with the cost of transportation, raw ma-
terials, and other factors.

Meanwhile, our map makes no claim to be more than a first
approximation to the truth. Therefore no maps of individual
continents are now presented, but merely a map of the world,
Figure 22, and of the United States, Figure 34. In preparing
these the figures for Hann's stations have been placed on the
maps. Then a line has been drawn to include all places falling
not more than 25 points below the possible maximum. These
are ranked as "very high" and are shaded black. A second line
includes places falling from 25 to 50 below the maximum, and
the area thus delimited is ranked as "high," and shaded with
heavy black lines. The next division, indicated by light lines,
is ranked as "medium," and the values range from 50 to 75 below
the maximum. The area shaded with thickly scattered spots
includes places ranging from 75 to 115 below maximum, and is
counted as low. The fifth division, shaded with widely scattered
spots, is "very low," and ranges from 115 to 155. Finally, the
hot desert areas which fall below 155 are left unshaded, but if
humidity were considered they would probably rank as high as
the wet parts of the tropics.

Let us now make a similar map on the basis of health. For this
purpose I have used the climograph for the eastern United
States as given in World Power and Evolution. Each column and
each horizontal line of the original climograph was plotted,
smoothed, and prolonged to the necessary limits of temperature
or humidity where these exceeded those actually occurring in the
eastern United States. It was assumed that a temperature of
100 F. and a relative humidity of 100 per cent would speedily

THE IDEAL CLIMATE 233

cause death, and that a mean temperature of 120 for day and
night, even with the lowest relative humidity, would ultimately be
fatal. From the smoothed data thus prepared the departures of
the death rate from the normal under any combination of tem-
perature and humidity can be determined. On the basis of the
data previously given it was assumed that the effect of variability
is approximately equal to that of temperature or humidity. Since
the best degree of variability seems to be about twenty-one
storm centers per year according to Kullmer's scale, this number
was counted as the optimum. It was assumed that places having
this degree of storminess or more have an advantage in health
equal to one half of that due to temperature and humidity com-
bined. This combined effect was measured by the difference be-
tween the two extremes of Yuma, 7.8 per cent above normal, and
San Diego, 6.3 per cent below normal. Although equal impor-
tance is thus assigned to temperature, humidity, and storminess
in the United States, the importance of storms for the world
as a whole amounts to only about one fourth that of the other
two combined.

Unfortunately the data as to relative humidity are scanty and
lack uniformity. Excellent and uniform figures, however, are
available for the United States and India, and less satisfactory
figures for Russia and Siberia. These indicate that if complete
data were available a map of climatic energy on the basis of the
effect of the weather on health in the United States would
closely resemble the one based on factory work, as may be seen
by comparing Figures 34 and 36. Hence in the absence of good
data as to humidity we may for the present use the map of
climatic energy based on factory work. (Figure 22.) The simi-
larity of this map to the Indian, Siberian, and American portions
of the map based on our data for health and the weather in the
United States goes far toward proving its general reliability.

The most noticeable feature of Figure 22 is two large black
areas of "very high" energy in the United States and southern

234

CIVILIZATION AND CLIMATE

Jf9 MO KO 100 %0 60 40 JtO

AO 40 60 So 200 1W

Figure 22. The Effect of Climate on Human Energy as Inferred from
Work in Factories

Canada on the one hand, and in western Europe on the other.
Each is surrounded by a heavily shaded "high" area of large
extent. The remaining high areas, four in number, are surpris-
ingly limited. One lies chiefly in Japan. It is shown as extending
into Korea, but the correctness of this is doubtful, for the
records of storms in that region are imperfect. The second lies
chiefly in New Zealand, but extends into Australia. The records
of storms in this region have been published less fully than in
Europe and America, but the general appearance of the map
seems to be approximately correct. The third of the minor high
areas is located in the southern part of South America. The
records here are very imperfect, and the extent of the high area
is doubtful. The reason for this uncertainty is not only that

THE IDEAL CLIMATE 235

reliable records of storms are not abundant, but that the avail-
able data do not enable us to determine how much change from
day to day is caused by the average storm. The amount of
change must be slight compared with that experienced under
similar circumstances in North America and Eurasia, because
no part of the southern end of South America is far from the
ocean. The fourth of the minor high areas lies along the Pacific
coast of North America. As already stated, its southern portion
owes its character not to storms or seasonal changes, but to
frequent breezes blowing in from the ocean. It extends only a
short distance inland, and is too narrow to be prominent on the
small-scale maps of this volume.

In the far North human energy appears to decline more than
would be expected. We know that population is scanty, and
civilization low, but we commonly ascribe this to the difficulties
of agriculture. Little can be demanded of people who must get
a living by hunting and fishing. From the map, however, it ap-
pears that even if other circumstances were favorable we should
not look for any great achievements. This accords with the slow,
inefficient character of the Eskimos, and of the Ostiaks and other
inhabitants of northern Siberia. Grenfell in his book on Labora-
dor says that the Eskimo "cannot compare with the Newfound-
land white fisherman for perseverance and 'snap.' An Eskimo
does not get one fish for the other's ten." This happens even
when the Eskimo is in his native habitat, and is doing work to
which he has been trained from childhood. Racial inheritance
may have much to do with this, but the testimony of white men
is that after a long stay in the Arctic they themselves lose ambi-
tion and energy.

At the other extreme of climate the regions within thirty de-
grees of the equator seem to be characterized by essentially the
conditions that we should expect. The status of the highlands is
striking. A high degree of energy among white men would not
be expected permanently in any of them. We are often told that

236 CIVILIZATION AND CLIMATE

the climate of tropical highlands is as fine as any in the world.
Not infrequently people are urged to colonize such regions. In
book after book we read that there is not the slightest reason
why the white man should not live there as well as at home. I do
not assert that this may not be possible. In fact, I strongly hope
that some day it will come to pass. Nevertheless, our map seems
to indicate that previous to any such desirable consummation
we must greatly increase our knowledge of how to adapt our-
selves to nature, and especially of how to select physical types
which are capable of preserving their own health and raising
children in spite of the monotonous climate. At present, while
the white man may learn to preserve his health in tropical
regions, he can scarcely expect to retain the vigor which he dis-
plays in the more favored parts of Europe.

The most unexpected feature of the map is the diminution of
energy as one proceeds eastward from western Europe to central
Asia. In the deserts of Turkestan and Mongolia, and especially
in the Tibetan highland, the map should probably show lower
conditions than are actually depicted, but as records are not
available, the medium shading has been extended across the whole
of the unknown area. Before making the studies here described
I should have said that a man in Siberia could be as efficient as
in far western Russia in the same latitude. Yet the Baltic Prov-
inces are very high in climatic energy, while eastward there is a
steady decline until only medium conditions are reached. The
reason is readily apparent. In the first place, the Siberian winter
is colder and longer than that of the region near the Baltic Sea.
More important than this, however, is the decline in storminess
as one passes eastward across Russia into Siberia. The cyclonic
centers of low pressure, which constitute storms, are either
broken up when they approach Asia in winter, or else swing out
toward the sea to avoid the great area of high barometric pres-
sure which lies over the continent during the cold season. Hence,
during midwinter the far interior is characterized by clear and

THE IDEAL CLIMATE 287

extremely cold weather, not hard to bear, but steadily benumb-
ing. In the spring and autumn, on the other hand, storms are
fairly frequent, and are often of most terrific intensity. The
burans, as they are called, are even worse than our western
blizzards, which are the same thing under another name. They
destroy cattle and horses by the thousand, and human beings
often perish within a hundred yards of their houses. Only when
the burans are at an end and the milder storms of the late spring
and summer prevail does Siberia enjoy a highly stimulating
climate.

The conditions just described afford an interesting commen-
tary on the common idea that the plains of Siberia are to be the
scene of a wonderful development of European civilization
during the next few centuries. I formerly shared this opinion,
but have now been obliged to modify it. While this chapter was
being written I spoke of this change of opinion to a Russian
friend who has come to America for the sake of greater freedom.
"Yes," he said, "that is just what the exiles say. I have many
friends who are exiles. When they are sent to Siberia they take
books with them and expect to do much work in writing and
along other lines. Some plan to carry on linguistic studies, and
some to make various other kinds of scientific investigations, but
they almost never do it. They say that at first they begin to
work with great vigor, but after a year or two their energy de-
clines. They have the desire to work, but do not seem able to do
so. They attribute this to being so far from home, and to the
lack of stimulating contact with civilization. I think there may
be more to it than that, for they seem to lose their energy."

Nansen, in his fine book, Through Siberia: The Land of the
Future, emphasizes this point. He frequently speaks of the slow-
ness and inertia which he encountered. Here per acre the Sibe-
rians raise far smaller crops than the Norwegians, and the main
reason assigned by Nansen is lack of care, forethought, and
energy in cultivating and fertilizing the soil. "There is no hurry

238 CIVILIZATION AND CLIMATE

here," he says, "Siberia is still a country that has a super-
abundance of time, as of everything else ; they may think them-
selves lucky for having so far escaped the nervous stress that
we know too well in Europe." He also quotes Rodishev, one of
the most enlightened members of the Duma, who sums up his
impressions of a journey in Siberia with the statement that the
Siberians are "a people without enterprise or initiative." Yet
the vast body of Russians in Siberia consists of pioneers who
voluntarily went to the new country. The mere fact that they
broke old ties and made the hard journey into the wilderness
bespeaks more than the usual degree of energy and initiative.
Similar people in the United States and Canada display great
activity ; in Siberia the climate seems to damp their vigor.

All this suggests that the old Russian autocracy accomplished
its purpose more fully than it realized. It not only exiled many
of its most thoughtful and active people, but sent them to a
place where not only do the isolation and hardships diminish
their power, but where nature insidiously accomplishes exactly
the kind of repression that the authorities desire. From the
standpoint of climate,, without respect to the many other factors
which may cause quite other results, the relative positions of
Russia and Siberia do not seem likely to change. Both, we may
rightly hope, are destined to advance far beyond their present
position, but while there is reason to think that western Russia
may approach the standard of western Europe, Siberia suffers
from a handicap which may never let her overtake the Baltic
regions on the west of the great northern republic.

Turning to China, we find that the summers are often debili-
tatingly hot, with a steady, damp heat that is apt to be trying.
The winters, on the other hand, are by no means so long as in
Siberia, nor so severe. Yet they are far worse than in western
Europe, and as bad as in any part of the United States. Cold
waves often sweep down from the north, and are so severe that
instead of being stimulants, they are depressing in regions like

THE IDEAL CLIMATE 239

Peking. In the south, however, they are beneficial. Everywhere
cyclonic storms are rare, so that there is no stimulus of great
importance from that source. This is one of the chief reasons
why China does not stand high on the energy map. The northern
parts of the country are more favored than those in the south or
in the far interior, but the difference is not great. Indeed, the
uniformity of all parts is surprising. The disadvantages of high
temperature in the south are balanced by those of low in the
north. If China were part of a smaller continent her nearness to
the moderating influence of the sea would help her much more
than is now the case. All through the winter she is under the
benumbing control of the vast continent to the west, which not
only sends out severe cold waves, but prevents the passage of
storms. Japan, on the contrary, does not suffer so much in this
way. Extremes of temperature are 'milder than in China, and
stimulating storms are frequent. Her greatest drawback is the
long period of hot, damp weather in summer. Nevertheless she
stands high. Here we must bring our review of the map of cli-
matic energy to a close. We shall come back to it again when
we have studied the distribution of civilization.

CHAPTER XI
THE DISTRIBUTION OF CIVILIZATION

DOUBTLESS the reader has already noticed the striking
resemblance between the distribution of climatic energy
and of civilization. Look again at Figure 22 and see how the
black areas agree with the places of highest culture. In view of
this it seems advisable to construct a map of civilization to serve
as a standard of reference. Only two methods appear feasible.
One is by statistics ; the other on the basis of opinion. Both pre-
sent grave difficulties. The statistical method will ultimately
prove far the better, but it may not be practicable for centuries.
For a fair estimate of the position of a country we need accurate
statistics of education, morality, industry, inventions, scientific
and artistic skill, wealth, pauperism, charity, crime, and many
other aspects of human life which will readily suggest them-
selves. No reliable figures for many of these things have ever
been gathered in any part of the world ; no statistics for any of
them have ever been gathered in many countries. Statistically
it is almost impossible to compare Afghanistan, for example,
with Kamchatka. Even where accurate statistics are available,
the methods of compiling them are often so diverse as to make
comparisons misleading. We may know exactly how many
people are arrested and convicted for theft in half a dozen
countries, but in one country the police may be so inefficient that
few criminals are apprehended, while in another practically
every thief may be caught. Thus the better may easily appear
the worse. The only way to use statistics at present seems to be
as a check upon the other method. We can select some country so

DISTRIBUTION OF CIVILIZATION

extensive that its various parts differ decidedly, but sufficiently
homogeneous so that the figures for all portions are comparable.
Since the United States meets these conditions better than any
other country we shall examine its statistics in several cases, and
shall use them as a test of a map prepared on the basis of opinion.

For a map of the civilization of the entire world we must rely
on the opinion of well-informed persons, but we shall find that
this agrees closely with the indications of statistics. The value
of a map based on personal opinion depends partly on our defi-
nition of civilization and partly on our confidence in the judg-
ment of the persons in question. Even the best and broadest ex-
perience does not eliminate personal or racial bias. Therefore,
the only safe course is to obtain the opinions of many people
belonging to different races and ruled by different ideals. Ac-
cordingly, in the autumn of 1913, I asked over two hundred
people in twenty-seven countries to help in preparing a map.
Most fortunately this was before the great war broke out. Good
feeling prevailed everywhere, and among men of sound judg-
ment there was perhaps as little racial prejudice as at any time
during the course of history. This is especially important
because similar conditions may not prevail again for years.

The persons whose assistance was asked were selected for
various reasons. The larger number were geographers whose
first duty is to know all parts of the world. Ethnologists in con-
siderable numbers were included for the same reason, but they
responded less freely than the geographers. Historians, diplo-
mats, colonial officials, travelers, missionaries, editors, educa-
tors, and business men were all included. The only criterion was
that each person should possess an extensive knowledge of the
world through personal knowledge, or, in a few cases, through
reading. Some were selected because of knowledge of special
regions not well known to most people and only reached by ex-
tensive travel. To all these many kinds of people, numbering 213
in all, I sent the following letter :

242 CIVILIZATION AND CLIMATE

"May I ask your cooperation in the preparation of a map
showing the distribution of the higher elements of civilization
throughout the world? My purpose is to prepare a map which
shall show the distribution of those characteristics which are
generally recognized as of the highest value. I mean by this the
power of initiative, the capacity for formulating new ideas and
for carrying them into effect, the power of self-control, high
standards of honesty and morality, the power to lead and to
control other races, the capacity for disseminating ideas, and
other similar qualities which will readily suggest themselves.
These qualities find expression in high ideals, respect for law,
inventiveness, ability to develop philosophical systems, stability
and honesty of government, a highly developed system of educa-
tion, the capacity to dominate the less civilized parts of the
world, and the ability to carry out far-reaching enterprises
covering long periods of time and great areas of the earth's
surface.

"In preparing such a map it is evident that statistics may
afford much assistance, but they need to be supplemented. They
touch only upon material things in most cases, and none are
available for a large part of the world. Therefore, our best
resource is the personal opinion of competent judges. Accord-
ingly, I am asking a hundred geographers, anthropologists, and
other persons of wide knowledge, whether they are willing to
take the time to divide the regions indicated in the accompanying
list into ten groups according to the criteria mentioned above.
Group 10 will include regions of the very highest character,
that is, those where the greatest number of valuable qualities
are found in high degree. Group 1 will include those which are
lowest in these respects. On the basis of this grouping I shall
determine the average position of each region and shall prepare
a map accordingly.

"The purpose of such a map is threefold. In the first place,
it will prove intrinsically interesting to a large number of

DISTRIBUTION OF CIVILIZATION 243

people, and is likely to arouse considerable discussion. In the
second place, in all geographical, historical, sociological, and
economic discussions it seems to me that we need a clearer,
stronger emphasis upon human character, that is, upon the
mental and moral qualities which dominate the civilization of
the various nations. If this is so, it is highly important, in the
third place, that we should determine much more fully than has
yet been the case how far various moral and mental qualities are
influenced by physical environment, race, historical develop-
ment, biological variations, and other causes. In order to deter-
mine these things we need a map which shall serve at least ap-
proximately as a standard of reference. In discussing the
influence of such things as racial character, differences of reli-
gion, social institutions, modern means of communication, the
form of the land, the relation of land and sea, variations of
climate and the like, we shall be able to gain much light by com-
paring their distribution with that of human character as it
now exists according to a consensus of expert opinion.

"The matter can best be illustrated by outlining a specific
purpose to which I mean at once to apply the proposed map.
[Here follows a brief description of the map of human energy
on the basis of factory work.]

"I recognize that those to whom this letter is sent will say at
once that they are not sufficiently familiar with all parts of the
world, and that they have no means of distinguishing between
different parts of China, for example, or between the different
portions of equatorial Africa. This is certainly true, but it must
be remembered that the classification is very rough. It is only
desired that the one hundred and eighty-five names on the en-
closed list be divided into ten groups, no group to contain less
than fifteen names or more than twenty-one, and each preferably
to contain eighteen or nineteen. It may not be easy to determine
whether all of the divisions of France, for example, fall in the
first group, but it is perfectly evident that none of them will

244 CIVILIZATION AND CLIMATE

fall in the fifth or lower. The chief thing is to place them as
nearly as possible in their proper group according to one's own
personal opinion. A given region may properly fall in the fifth
group, but the purpose of this classification will not be defeated
if it is placed in either the fourth, fifth, or sixth, for when the
opinions of one hundred persons are averaged, individual idio-
syncrasies will disappear. In view of the varying degrees to
which each individual is familiar with the different regions of
the world, I should be glad if each contributor would underline
the names of regions with which he is well acquainted either by
travel or reading, and would place question marks after the
names of regions as to which his knowledge is especially deficient.
Names not underlined or questioned will be considered as inter-
mediate. The three grades of familiarity thus indicated will be
weighted in the ratios of 3, 2, 1. [This has not been done, partly
because only about half of the contributors made the division
into three grades, and partly because the final results are not
appreciably changed by using the unweighted values.] The
grade of the various regions should be indicated by underlining
or questioning the names upon the small slips mentioned below,
but may be done upon the accompanying list if that is more con-
venient, but in that case please be sure to return the list. For
convenience of classification I enclose slips containing the names
of the different divisions. These may be spread out upon a table
and arranged in ten columns and shifted from column to column
until an approximately satisfactory arrangement is reached.
When thus arranged those of each column should be placed in
the corresponding envelope of the ten here enclosed, and all may
be mailed in the large addressed envelope. Envelope ten is for
the highest group, and one for the lowest.

"In making the classification, one or two points need to be
borne in mind. In the first place, the past should not be con-
sidered: Greece, for example, should be placed in the group
where its condition during the past one hundred years would

DISTRIBUTION OF CIVILIZATION 245

place it without reference to its ancient greatness. In the second
place, if two races inhabit a given region, both must be con-
sidered, and the rank of the region must depend upon the aver-
age of the two, giving each one a weight proportional to the
number of people. For instance, in a state such as Georgia where
nearly half the people are negroes, they must receive half the
weight. Still another point is that the rank of a country can
often be determined by considering the position which its people
take when they migrate elsewhere. For instance, the position of
Syrians as compared with Germans when they migrate to Eng-
land or the United States is a fair criterion as to the relative
merits of the two races. After the first generation, however, this
should not be applied, for the younger generation owes much of
what it is to the new country. A final point concerns countries
which are poor in natural resources, or which are not located in
the main centers of the world's activity, but which are neverthe-
less of high character. For example, so far as importance in
the affairs of the world is concerned, England vastly outranks
Scotland. Nevertheless, our estimate of the greatness of Eng-
land owes much to the large number of Scotchmen who have
gone out to build up the British Empire. Therefore, in estimat-
ing the relative merits of Scotland and England, the matter of
size or even of commercial importance should receive relatively
little consideration, whereas the character and ability of the
people as rulers, merchants, scientists, writers, and men of all
sorts should have a predominating weight.

"In publishing the final results I should be glad if I might
print the names of those who have contributed, but of course this
must be as each individual may choose. The individual lists will
not be published, and will be treated as confidential. I judge
that other contributors will feel, as I do, that their classifica-
tions are of necessity so imperfect that they do not care to
distribute them to the world at large. Hence, while the list of con-

246 CIVILIZATION AND CLIMATE

tributors will be published unless the contributors prefer other-
wise, their individual opinions will be withheld. I hope, however,
to publish a list showing the average rank of each country and
the range of opinion between those who put it highest and those
who put it lowest. Inasmuch as the plan here outlined depends
upon the cooperation of many contributors, no single individual
can in any respect be held responsible for features of the final
map which do not meet his approval.

*"In addition to the general list of divisions, I enclose a set
of cards bearing the names of the states of the United States,
and of the provinces of Canada. Would you be willing to arrange
these in groups and place them in the proper envelopes, employ-
ing the same method as for the larger divisions? Group 1 will
be the states or provinces which are least progressive, or least
influential, so far as the general character of their citizens is
concerned, and Group 6 the highest. Each group should contain
about ten names. The object of this you see is to make a map of
the United States and southern Canada on the same basis as
that of the world, but on a more minute scale.

"The rough grouping here suggested ought not to take more
than a few hours' time. Many days, to be sure, might be devoted
to it, but the added accuracy thus gained would not be sufficient
to make it worth while. If you can give the necessary time at
your earliest convenience I shall be most grateful. If you cannot,
would you be willing to return the list, the slips, and the en-
velopes in order that I may ask someone else to do it in your
place? Whether you contribute or not, I shall take pleasure in
sending you copies of the final results. Trusting that I may hear
from you soon, I am

"Very truly yours,

"ELLSWORTH HUNTINGTON."

* This paragraph was included only in the letters sent to Americans and
to one or two Europeans especially familiar with America.

DISTRIBUTION OF CIVILIZATION 247

Replies were received from 137 persons, while others sent
copies of their publications, so that an answer of some sort came
from about three fourths of those addressed. The majority of
the remaining fourth were foreigners to whom a six-page letter
in English might appear formidable. About 90 per cent of the
English and Americans sent replies, which is a very large pro-
portion as such things go. I am convinced that the rest failed
to answer chiefly because the classification required more time
and was more difficult than I at first realized. The fact that
classifications continued to be received for an entire year indi-
cates that many meant to answer, but put my letter aside for a
more convenient season which never arrived. A third of those who
replied, fifty-four to be exact, actually made classifications, and
all but two or three conformed so closely to the general plan
that it has been possible to use them. The names of the con-
tributors are given in the Appendix. I take this opportunity to
express the warmest appreciation of their kindness in cooperat-
ing so cordially. Not only their classifications, but their letters
were of the highest value and in many cases contained sugges-
tions which have been of great assistance in preparing this vol-
ume. The same is true of many letters from persons who did not
contribute, but who took pains to explain their reasons and to
suggest ways in which my plan might be improved. Except where
direct quotations are employed I have not attempted to acknowl-
edge my indebtedness for various ideas which distinctly modify
the tenor of these pages. This is partly because the same thought
was often expressed by several persons, and partly because in
many cases I cannot tell from which of several letters an idea
was derived. Except in two instances I have also refrained from
mentioning names, because where so many have contributed
materials of great value, it might seem invidious to mention some
and not others. Therefore, I can merely express my gratitude
to all concerned. The net result of this attempt at scientific
cooperation among men of many races and tongues leaves a

248 CIVILIZATION AND CLIMATE

strong impression of the spirit of fellowship and friendly help-
fulness among men of wide interests in all portions of the world.

The countries represented in the final classification and the
number of contributors are as follows: Australia 1, Canada 1,
Norway 1, Sweden 1, Netherlands 1, Russia 1, Spain 1, Portu-
gal 1, France 2, Italy 2, Japan 3, China 3, Germany 5, Great
Britain 7, and the United States 24. The number of Chinese and
Japanese is particularly gratifying. The ratio between the num-
ber of contributors and the number to whom letters were sent is
higher among them than among any other main group except
the Americans, as may be seen in the Appendix. It is to be re-
gretted that no one from India or South America cooperated,
and only one, a Russian, from the European countries east of
Germany.

The difficulty of making the classification is considerable.
Several contributors spoke of spending an entire day upon it,
or of taking out the slips time after time to arrange them more
satisfactorily. Some said that they spent two entire days upon
it. All seemed impressed by the way in which a systematic classi-
fication of this kind brings out the weak spots in a man's knowl-
edge. For instance, here is the way in which one contributor
expressed himself :

"One appreciates what a big world this is and how little one
knows about it when he attempts such a task as you have set.
It is a most excellent means of taking the conceit out of one."

Another puts it in this way :

"I must confess that it is the most difficult and one of the most
humiliating games I have ever tried to play ! I always knew I
was a fraud as a President of a Geographical Society, but I
never knew before how great was my deception! The greatest
difficulty I found lay through my ignorance of the proportion
of the different races inhabiting a district."

DISTRIBUTION OF CIVILIZATION 249

An interesting feature of the letters was the diversity of
opinion as to the advisability of any such classification of coun-
tries. To take the adverse opinions first, one of my best friends,
an American geographer, put the matter very strongly :

"I am complying with your request for a sorting of the slips
you sent me. It is a very bad plan, and not, I think, of value.
Indeed I am not sure that I would have done it for anyone else
than you."

I am glad to say that later he expressed a much less severe
opinion. Another geographer, a Teutonic European, speaks
most cordially in part of his letter, but comes out bluntly in
opposition to this particular plan :

"I am wholly unable to take part in this work. I take your
scheme as a failure ... I guess you are here, like some other
Americans, under the influence of a too systematizing spirit.
It seems to me impossible to classify mankind by this simple
method."

Still a third, an American anthropologist, is equally uncom-
promising :

"Speaking frankly I do not conceive that the method you
suggest is possible of scientific results. One must choose between
statistics which are definite and mere judgments which are gen-
eral. To apply the geographic method to a compound of sta-
tistics and loose generalization may be productive of grave
error."

And a fourth, also an American anthropologist, expresses
himself as follows :

"It has been my endeavor, in my anthropological studies, to
follow the same principles that are laid down for natural
sciences ; and the first condition of progress is therefore to elimi-
nate the element of subjective value; not that I wish to deny

250 CIVILIZATION AND CLIMATE

that there are values, but it seems to me necessary to eliminate
the peculiar combination of the development of cultural forms
and the intrusion of the idea of our estimate of their value,
which has nothing to do with these forms. It seems to my mind
that in doing so these obtain subjective values, which in them-
selves may be the subject of interesting studies, but which do
not give any answer to the question that you are trying to
solve."

Another anthropologist, this time a European, at the end of
a particularly long and suggestive letter, expresses himself thus :

"Taking all that I have written here into consideration, I
think that if we were going to grind all the different regions of
your long list in the same statistical mill, and to try to compute
an average, a highly improbable and fantastic result would be
obtained. For my own edification I put some of your criteria
to the test, though in a different way. I drew up a list of twelve
characteristics of the 'highest value, 5 in which I included sense
for beauty in literature (belles lettres) and a few others, and
then distributed them to eleven different regions of the globe.
My familiarity with those regions by a long sojourn or travel
and reading, covers a lifetime. To each characteristic for each
region I assigned a number, from 1 to 1 0. I then added the differ-
ent values or points to try to find a ratio, which might be called
'index of civilization.' I give it valeat quantum valere potest."

The table possesses so many points of interest that it is in-
serted below. At the end I have added a column showing the rank
of the various regions according to all the contributors, as com-
puted according to the system presently to be outlined. If the
plan embodied in this table could be carried out on a large scale,
it would undoubtedly be better than mine. The difficulty is that
it requires a vast amount of work and a degree of familiarity
with the various peoples of the earth which is found only among

DISTRIBUTION OF CIVILIZATION 251

a few exceptional students who can almost be counted on the
fingers. In course of time we may perhaps hope for a map
based on some such minute study of human nature. Yet when it
is before us, there is every reason to think that its general fea-
tures, with which alone we are concerned in this book, will be
almost identical with those of the map which we shall shortly
consider. The reasons for this will be given later.

One important point stands out in this table. I have given too
little weight to the aesthetic side of human nature. In framing
a definition of civilization I consciously thought of art in all its
forms, but it seemed as if this were included in "the capacity
for formulating new ideas and for carrying them into effect,"
just as science is meant to be included. Moreover, the course of
history seems to show that every nation which rises high in other
respects sooner or later experiences a period of high develop-
ment in art, architecture, literature, and science. Nevertheless,
these things should have received more specific recognition in my
definition.

To turn now to the other side of the question, those who be-
lieve in the utility of the plan presented in my letter naturally
do not feel the necessity of stating their reasons. Nevertheless,
a considerable number take pains to express approval. For in-
stance, a widely traveled Englishman thinks that "there are
tremendous possibilities in all such attempts." An American,
who is familiar with most of the countries of the world, says :
"Permit me to say how heartily I thank you for engaging in
this enterprise. Despite any and all sympathetic or hostile criti-
cism of such a work, or the cheap and hasty or really valuable
appraisal likely to be made, such a scheme will be invaluable to
all students of human progress."

Another American, who has spent a large part of his life in
the Orient, is of the opinion that: "It is a fascinating and very
significant set of standards by which you have asked us to group

zss

8gggSS8S8SS

jo xapin

auiAvoqs

(a} a 'sjaMog

-u ut A^tiBaq

f-H r- 1

joj asuas

(ea^s^)

aOOOOit-0^DG*0

ui A^Ti^aq

i 1 4

joj asuag

(papnir>ui

<<

^JB ui A^nBoq

^0005G*OC<5*-.Q*^i^

M

joj asuag

s

A^ajBS |BUOS

^^OOSOJ^^^^O^.^

o

-jod jo aajSaa

I

A^n^owi

33

2

JO pJBpUBJS

ffi

auaiSAii jo

h

sa[dK>uud jo

t^^Ot-XODOOQCQ^'*^^.

W

H

00 <

uoqtjonpa jo

v.

raa^sAs pado

*Ct-OiODX^t-^-iG*r-t^

3 w

-pAap AtqSiH

LH

O

[oj^uoo PUB

OjCO^^-l^M-tr-H^r-,

pB9\ o^ aaAvo j

5 *

c

3 "

sreaiB

R

w

PUB spouad

a

<g

amqSi"!

O^.^W-WO^-M^OO

v

x:

fc

ino Ajn?3

a

o^ A^ijRV

.s

siua^sAs

"c

w

dojaAap

COXCCOi^OT^^r-lr-r^

8

g

<n ^?nqv

8

^Bapi

.2

-^Binuuoj joj

^CC^XC?^Oi^H-.rHr-,

i

A^lDBdBD PUB

5

T?

aAnniui

O?COXCOCOO.H<^W

%

JO JOA\OJ

a

M

g * -0

!

* a o 3 ^

g

I

S^ g ^ -^ "0 8- -^ -a *
JS? -S? a * -S3 *"

|2 S^|2 &2|^ -s$ -e|S fr S ^

*- 8 ^ d | 8 g 8 Sd ' S T8 i . S a & - c -

g z I 2 ~* I 2 --g 55 ^ -s 52 S 2 &* | z &*
ol^o^S^s^cS,?

*Thenumb<

DISTRIBUTION OF CIVILIZATION 253

the regions. From the time I started after lunch until now, late
in the evening, I have done nothing else. But there is more than
fascination. There is a very deep issue involved, and I am glad
to share in the construction of the map and the charts you are
aiming to devise."

One of the fairest estimates of both the advantages and dis-
advantages of the plan is contained in a letter from "Ambassa-
dor" Bryce, as we Americans still like to call him. With his per-
mission I quote it in full.

"Your idea is ingenious and interesting, and I should much
like to see how it works out, though it seems to me, on first
impressions, that the various factors involved are so many and
so complex that the visual presentation you contemplate would
need an amount of comment and explanation which would re-
quire something like a treatise to accompany the map. You give
one instance in the case of the state of Georgia. Another might
be found in the other Georgia, south of the Caucasus, where
besides the native Georgians (Karthli) there are Armenians,
Tartars, and Russians, not to speak of minor races. Or take
Japan. If one were to think of the educated and ruling Japanese
and the sort of civilization they have created one would make
a rating quite inapplicable to the ordinary Japanese. Without
therefore doubting that by means of a map presentation very
interesting results may be obtained, I should think that some
modifications would be needed. 'Efficiency' is a very complex
conception.

"If I had time I should like to try to think further about the
scheme, and lend what aid I could to it ; but unfortunately I am
so fully occupied by trying to finish a book for the sake of which
I retired from the Embassy at Washington that I am, to my
regret, unable to cooperate with you. I shall be grateful for any
information as to the results you can send me if you have
leisure."

254 CIVILIZATION AND CLIMATE

"P. S. [dated two weeks after the letter].

"I have kept the papers some weeks in order to see if I could
help you. I have dealt with the states and provinces of the
United States and Canada, though roughly. But I have found
the difficulty of any adequate arrangement of the extra-Euro-
pean world in point of efficiency impossible in respect of the
extraordinary mixture of races. Turkey and India are nuts too
hard to crack; ... In South America it would be easier, be-
cause the races are really so mixed that a new race results. But
in India they live side by side and are quite different. [Here
follows a classification of South America.] It is quite interesting
and makes one think. But after all it is the natural race divisions
rather than territorial divisions that count everywhere outside
western Europe. Even in Russia and the Balkans, race doesn't
correspond to territory."

These various letters give a good idea of the general opinion
of the attempted classification. As an illustration of the diffi-
culties to be overcome, lot me quote from an Italian contributor,
an anthropologist: "What is the standard of these various
civilizations? Yours, as it seems to me, is an European one, and
this, I think, is a very limited one and cannot solve the prob-
lems." A similar objection is expressed in a letter from an Eng-
lish anthropologist. Quoting from my letter, he says :

" 'The power of self-control, high standards of honesty and
morality, . . . high ideals, respect for law' are eminently char-
acteristic of many savage and barbarian peoples, notably the
North American Indians, and in my opinion these latter stand
very much higher than the average American citizen, but the
latter lead in 'the power to lead and control other races, and
capacity for disseminating ideas . . . inventiveness, highly
developed systems, etc.' "

I recognize the force of these comments. Races certainly differ
greatly, even though they happen to dwell in the same physical

DISTRIBUTION OF CIVILIZATION 255

environment. Moreover, people of a given race who live under
the same environment may differ widely because of diversity in
religion, government, or institutions. Furthermore, the defini-
tion of civilization here presented does not pretend to be perfect.
It is a European definition. Yet it is also a world-wide definition.
The contemplative Hindu may perhaps be a higher type than
the aggressive citizen of western Europe, but the contemplative
type has made relatively little impression upon the world as a
whole. If we turn to antiquity, the people who have left their
impress are those who had this so-called European activity. The
Greeks and Romans had it to a marked degree. The people of
India had something of it when they wrote the Vedas. Gautama
had it when he founded the Buddhist religion. He was contem-
plative, but yet he had the qualities expressed in our definition.
He was preeminently possessed of "high ideals, respect for law,
inventiveness [in the broad, non-technical sense], ability to
develop philosophical systems, and the capacity to dominate
the less civilized parts of the world." He dominated through
ideas, not force. The Jews had this same power. Such men as
St. Paul, although not aggressive in the ruder European sense,
were unsurpassed in "the power of initiative, the capacity for
formulating new ideas and for carrying them into effect, the
power of self-control, high standards of honesty and morality,
the power to lead and control," and "the capacity for dissemi-
nating ideas." The reason for regarding the standard here set
forth as European is that Europe is today its great exponent.
In the past, however, not only did Rome, Greece, Palestine, and
even northern India possess it, but Egypt, Mesopotamia, and
Carthage all displayed it. China, too, in her days of early great-
ness, and the wonderful Maya people of Yucatan, the only ones
to develop the art of writing in America, were animated by the
same active, stirring, "European" qualities. That is why we
remember them, but have forgotten most of their contempo-
raries.

256 CIVILIZATION AND CLIMATE

Granting that our definition of civilization is imperfect, but
admitting that it includes the qualities which are of greatest
importance in causing a nation to impress itself upon the world,
let us now proceed to ascertain how these qualities seem to be
distributed. The fifty contributors whose classifications could
be used have been divided into five divisions as follows : ( 1 ) 25
Americans, 1 of whom is a Canadian; (2) 7 British, 1 of whom
is an Australian; (3) 6 Germanic Europeans, 4 of whom are
Germans, 1 a Swede, and 1 a Norwegian; (4) 6 Latin Euro-
peans, namely, 2 Frenchmen, 2 Italians, 1 Spaniard, and 1 Por-
tuguese, with whom has been included 1 Russian because there is
no other group in which he fits more appropriately; and (5)
5 Asiatics, 2 of whom are Chinese and 3 Japanese. A third
Chinese contribution was most unfortunately lost in the mail.
The average opinion of each of these five groups is given in the
tables in the Appendix. To obtain the final rank of each coun-
try the averages for the five groups have again been averaged.
Thus each race, or at least each of our five divisions, has equal
weight in determining the figures on which will be based the
map used in later discussions. The opinion of twenty-five
Americans, for example, has no more weight than that of five
Asiatics. This may seem unfair, but on the whole it seems to
be the method best calculated to give a reliable result. All of
us are inevitably prejudiced. The Americans put America, espe-
cially its more backward parts, higher than is correct. The
Asiatics put their own countries too high. By giving America
and Asia equal weight and by dividing Europeans into three
groups animated by different ideals and different sympathies,
we are able largely to eliminate the effect of racial prejudice.
The final results are summed up in Figures 23 to 28, but I shall
defer comment upon these for the present.

In the tables in the Appendix the countries of Europe, North
America, and Asia have been divided into groups corresponding
as nearly as possible to race, while those of Australia, Africa,

DISTRIBUTION OF CIVILIZATION

257

and South America have each been put in a single group because
the racial differences are either not strongly marked, as in
Australia and South America, or are highly complicated as in
Africa. Under each group the country whose final rank is highest
has been placed first and the rest in consecutive order. To begin

Figure 23. The Distribution of Civilization in Europe

with Europe, England heads the list. It is the only region placed
in the first rank by every contributor. Northwestern Germany,
which includes Berlin, comes next. The Germanic Europeans
and the Asiatics all place it in the first rank. The British almost
invariably do so, and their opinion, 9.9, averages the same as
that of all divisions. Among the Americans and Latins several
place this part of Germany in the ninth class instead of the
tenth. Hence, its average position according to them is about

258 CIVILIZATION AND CLIMATE

9.8. Such slight differences have little significance, for 98 per
cent is almost as good as 100 per cent. Yet they suggest that in
1913 people who live in other parts of the world were on the
whole not quite so sure of the Germans as of the English.

Following northwestern Germany we find central Germany,
then Scotland, Denmark, Holland, and so on to northeastern
Germany. Here for the first time we come to a region which is
placed by one group, the Latins, in a rank lower than 9.0. Yet
even here the Latins do not assign a rank lower than 8.5 and
everything above this ranks as "very high." Nowhere in these
first fourteen regions does the greatest diversity amount to more
than one degree on our scale of ten, or 10 per cent on the scale
of 100. This is comparatively insignificant, for it means that
while the difference between the highest and the lowest may be
10 per cent, each of them usually differs from the average by
only about 5 per cent. Thus so far as numbers 1 to 14 are con-
cerned the difference of opinion among Americans, British, Ger-
mans, Latins, and Asiatics is almost negligible. All alike rank
these regions very high.

Coming to Ireland, a country which, for lack of any other
suitable group, is placed with the Teutonic regions, we find
much more diversity of opinion. The English, presumably be-
cause it is part of their own empire, and the Asiatics, perhaps
because ifc seems to them like a part of England, place it very
high, with a rank close to 9.0. The Germans, on the other hand,
place it at 7.0, scarcely above the medium grade, while the
Americans and Latins place it near 8.0, which means that com-
pared with the world as a whole they think that Ireland stands
high, but not very high. The fact that all the Teutonic regions
except Ireland and the Austrian Alps rank above 9.0, and that
these two, which are partly Teutonic, stand in the high group
near 8.0, suggests that race is a dominant factor in determining
the status of civilization. The same suggestion is enforced when
we note that among the Romance nations the most Teutonized

DISTRIBUTION OF CIVILIZATION

259

portions stand highest. Yet the fact that innate racial differ-
ences may be of great and even overwhelming importance, as I
have shown in The Character of Races, by no means alters the
fact that geographical conditions also play a highly important
part.

Figure 24. The Distribution of Civilization in Asia

The Romance nations of Europe seem to possess much more
diversity of civilization than do the Teutons. Even if Albania
and Montenegro be omitted as not being truly Romance, the
range is from medium in Corsica and Sardinia to high in many
regions, and very high in parts of France and northern Italy.
Some places, such as southeastern Spain and southern Italy,
are rated exceptionally high by the Asiatics, although the other
contributors agree quite closely.

260

CIVILIZATION AND CLIMATE

Among the Slavic nations central Russia stands at only 6.2
in American opinion, possibly because Jewish atrocities were
freshly in mind when the classification was made. In general there
can be no question that the Baltic Provinces and Bohemia stand
at the top, while southeastern and northeastern Russia are at

Figure 25. The Distribution of Civilization in Australia

the bottom. The European region whose rank is most doubtful
is southern Finland. The Asiatics reduce it to about 6.6, while
the Germans rate it at 9.4. Even in so extreme a case the average
opinion is not open to much criticism. The final rank is about
8.0, which puts Finland at nearly the same level as Bohemia, the
Baltic Provinces of Russia, Ireland, and the Austrian Alps.

It would be interesting to go through all the tables and point
out their special features, but this must be left for the reader

DISTRIBUTION OF CIVILIZATION

261

to do by himself. Only one or two additional points can be
indicated. In Asia the first thing that strikes one is the great
diversity of opinion as to Japan and China. This is due to the

Figure 26. The Distribution of Civilization in Africa

fact that the Japanese and Chinese place their own countries
much higher than do the people of other races. This is natural.
The surprising thing is rather that these people, with their

Figure 27. The Distribution of Civilization in South America

DISTRIBUTION OF CIVILIZATION 263

justifiable pride in a great past, do not place their own countries
at the very top. They recognize that Europe and North Amer-
ica have in certain ways surpassed them. Aside from Japan and
China the agreement of the different groups as to the position
of Asiatic countries is on the whole surprising. Only in rare
cases does the opinion of any one group depart from the average
opinion by more than a single degree on the scale of ten. Two of
the few exceptions are llajputana and the Syrian Desert, both
of which are placed exceptionally low by the Asiatics.

In Australia and North America the diversity of opinion is
much greater than in the older continents. It reaches a maximum
in Canada, especially in Alberta, and in southern Greenland,
and Iceland. In such places opinion ceases to be of any special
value. People simply do not know what sort of conditions
actually prevail. In Australia, likewise, the comparative newness
of the country causes some people to overrate it and others to
depreciate it. It has not yet reached the stable equilibrium
which gives the world as a whole a well-defined opinion. In the
United States there is no more diversity of opinion than in
Europe, for that country has taken its true place among the
nations. In Australia and Canada, on the other hand, outsiders
who live far away and who are not racially connected with the
inhabitants are still sceptical as to whether those countries are
actually destined to rise as high as their people claim.

Thus far we have drawn attention to differences of opinion
more than to agreements. Let us now examine the matter in
another way, which brings out the essential agreements. We will
take North America because this is the continent as to which
there is the greatest diversity of opinion. Figures 28 to 33 pre-
sent a series of maps based on the average opinion of all divisions
of contributors and on the individual opinion of each division
respectively. Here, as in Figures 23 to 27, the rank of each
region has been written in its proper place on the map. Then
lines have been drawn in such a way as to separate the areas

Figure 28. Civilization in North America, According to all Contributors

Figure 29. Civilization in North America, According to Twenty-five Americans

266 CIVILIZATION AND CLIMATE

above and below 9, above and below 8, and so forth. The areas
ranking above 9.0 have been heavily shaded, those from 8.0 to
9.0 less heavily, and so on until places below 2.0 are almost un-
shaded. This gives a map of civilization whose main features can
readily be grasped. In Figure 28, which represents the average
opinion of all groups of contributors, certain main features
stand out prominently. They are ( 1 ) an area where civilization
declines rapidly northward in British America; (2) a high area
extending from east to west across the northern half of the
United States and a narrow southern band of Canada; (3) an
exceptionally high area in the northeastern United States from
the Mississippi to the Atlantic Ocean; (4) a similar small, high
area on the Pacific coast; (5) a bight of lower, but not of low,
territory extending into Nevada; (6) a rather rapid decline
south of the United States, which is interrupted by (7) an area
of slight improvement in central Mexico.

Let us now turn to Figure 29 and see how the opinion of
Americans differs from that of the world as a whole. The general
aspect of the two maps is the same, for every one of the seven
features just mentioned can be detected. The older states along
the Atlantic from New England to the northern Gulf States
have almost identically the same rank in both maps. The chief
difference lies in this : the Americans have a higher idea of the
new parts of their country and of the new parts of Canada than
have the people of other places. The3 r have a correspondingly
low idea of the countries immediately to the south of them. In
Central America their opinion does not differ greatly from that
of the rest of the world, but they place the West Indies and
Mexico relatively low. In the West Indies this is probably due in
large measure to the fact that in order to prevent internal con-
vulsion the United States has repeatedly intervened in Cuba
during the brief period of self-government enjoyed by that
country. The low estimate of Mexico probably arises from re-
cent revolutions which at the time when the classification was

DISTRIBUTION OF CIVILIZATION 267

made were producing a most unpleasant impression in the United
States. In Europe and Asia the disorders of Cuba and Mexico
attract relatively little attention, and the contributors to our
classification probably thought of these countries as they were
during their periods of long peace.

Now let us study British opinion as expressed in Figure 30.
One of the most noticeable things is the English opinion of the
southern United States and especially of the southwestern
deserts. No patriotic feelings, either conscious or unconscious,
lead them to believe that high conditions extend to the Mexican
boundary beyond which Americans believe that there is a great
and sudden change. Their opinion of the Atlantic portions of
the United States is almost identical with that of the Americans
and of the rest of the world, and they agree with other races in
their opinion of the regions south of the United States. They
believe, as does almost everyone, that the Pacific coast of the
United States stands higher than the western part of the in-
terior, but they do not think quite so well of it as do the Ameri-
cans, although the difference is slight. The places where their
national pride comes into prominence are in Newfoundland and
the Canadian Northwest. It is interesting to notice that while
they place Newfoundland and British Columbia higher than do
the Americans, they do not have so high an opinion of the Mari-
time Provinces nor of Alberta, Saskatchewan, and Winnipeg.
In the latter regions this is easily explicable. A great number of
Americans have gone to Winnipeg, Alberta, and Saskatchewan
during recent years, and their glowing reports have made upon
Americans a great impression which has not yet reached Eng-
land so strongly, and has scarcely touched the rest of the world.
Americans perhaps think more highly of the Maritime Provinces
than do the English because many of the most energetic Nova
Scotians and other provincials come to the States and display
marked ability. To cite a case which has come under my own ob-
servation, the students from Acadia College in Nova Scotia,

Figure 80. Distribution of Civilization in North America, According to
Seven British Contributors

Figure 81. Civilization in North America, According to Six Germanic Europeans

270 CIVILIZATION AND CLIMATE

who carry on work in the Graduate School of Yale University,
for some years maintained a higher average grade in their
studies than did those from any other institution.

In the German map, Figure 31, the seven general features
already enumerated can readily be detected, just as in the
American and English maps. The differences are largely matters
of detail. Such a feature as the assignment of a rank of 9.0 to
the northern Rocky Mountain states of Idaho, Wyoming, and
Montana, and of only 8.0 to the Dakotas, Minnesota, Nebraska,
and Iowa is probably due to lack of familiarity with the interior
of the continent. Why Newfoundland should stand so high is not
evident, unless it be because the name has long been known and
is very familiar. In the high position of Iceland racial pride is
again evident, for the Icelanders are close akin to the Germans.

The Romance peoples present a map (Figure 32) which again
shows all the main features. They know, however, even less of
the recent development of the western United States and of the
Canadian Northwest than do the Germans. To them Alberta and
Saskatchewan are apparently undeveloped tracts where trap-
pers and Indians still roam, for otherwise they surely would not
assign so low a rank as 4.4. Newfoundland for some reason is
also placed low, only 5.2, and Greenland still lower, 3.4. On the
other hand, just as the presence of Germanic people in Iceland
and Greenland draws the high lines far to the northeast in the
Germanic map, and as pride of country bulges the high lines
to include the whole of the United States in the American map,
so a similar racial pride causes Cuba and the Creole region of
Louisiana to be much higher on the Latin map than on any other.

Last come the Asiatics with a map, Figure 33, which is sur-
prisingly like the other maps. They do not yet realize what has
happened in the Canadian Northwest, but their opinion of
eastern Canada and Newfoundland is close to the average. The
place where they differ chiefly from others is in their relative
ideas of the northern Rocky Mountain states to which they

DISTRIBUTION OF CIVILIZATION 271

assign a rank of only 7.6, and of the southern states which they
place all the way from 8.2 to 9.4. Apparently the eastern United
States seems to them highly progressive in all parts, and they
do not attach much importance to the presence of the negroes,
a disadvantage which Americans themselves feel strongly.

Taking these maps as a whole we see that they all give the
same general impression. Where one group goes to a great
extreme, as in Newfoundland or Greenland, some other is likely
to go to the opposite extreme. Thus the eccentricities of judg-
ment displayed by one race are largely counteracted by those of
another. When all are averaged, the number of inconsistencies
is greatly diminished. To Americans and English it ms,y seem
that Alberta and Saskatchewan, for example, should have a rank
higher than is here given them, but we are forced to admit that
final judgment is not possible until these regions have been well
populated for a generation or two. Taken as a whole Figure 28
seems to contain no important inconsistencies, even though one
might wish to change the rank of certain places in which he is
particularly interested. The map represents the judgment of
thoughtful people in many countries. If fifty other equally well-
informed people representing an equally large number of coun-
tries in three separate continents were chosen to make another
classification, it is highly improbable that their map would differ
from Figure 28 so much as this map differs from the others.
The chances are that the two would be so nearly alike as to be
indistinguishable. Inasmuch as thei ^e is more diversity of opinion
in regard to North America than to any other continent, the
maps of the others approach still more closely to a true repre-
sentation of the opinion of thoughtful people all over the world.
The maps are by no means perfect, for human judgment is
fallible. Yet they at least present so close an approximation to
the truth that we may use their general features without danger
of error.

During the World War and the succeeding period of hot

Figure 32. Civilization in North America, According to Six Latin Europeans
and One Russian

Pigttte aa. OfttatSott i Katth toetka, According to Five Aaiati<ss

274 CIVILIZATION AND CLIMATE

prejudice it seemed to many people that the truth of this last
statement was doubtful. But soberer thought today shows that
this is by no means the case. During the war the steadfastness
and ability with which the various races fought was almost
directly in harmony with their position in our map of civiliza-
tion. The Portuguese, with a rank of 70, had to be taken out of
the main line of battle, for they could not withstand Germans,
who rank above 95. The Austrians, Poles, Italians, and Rus-
sians, not to mention the Asiatics and Africans came out of the
war with less military prestige than the British, French, and
Germans. Moreover, since the war the European countries
shaded black in Figure 23, that is, those ranking highest in civili-
zation, have recovered more rapidly than their neighbors of lower
rank. France and Belgium in spite of being ravished, and Ger-
many, in spite of her financial debacle, have strengthened their
positions in many respects, and are pushing ahead with remark-
able energy. Russia, on the other hand, remains in a curious
state of suspended animation, perplexing the world by her com-
bination of inertia and radicalism. The Soviet system is perhaps
only nominally at the root of her troubles : the wholesale exodus
and slaughter of the old leading classes have probably done ir-
reparable damage whose full results may not appear for many
years. But back of both these facts lies something in Russian
character, something which before the war held that country
down to a rank of from 46 to 73 in Figure 23. Spain, Italy,
Hungary, Greece, and the rest of southern and eastern Europe
have likewise behaved as one would expect of countries on the
level of progress indicated by their respective positions on our
maps of civilization. Thus, on the whole, the World War and
its aftermath seem to confirm the value of our method of esti-
mating human progress.

CHAPTER XII

VITALITY AND EDUCATION IN THE
UNITED STATES

WE now have before us maps of the world depicting the dis-
tribution of climatic energy and of civilization. Our ulti-
mate object is to compare them, and see whether they are caus-
ally related. It will add to the certainty of our results, however,
if we first apply an independent, statistical test to each of
our two maps. Since the methods of compiling statistics vary
greatly, our results will be most reliable if we confine ourselves
to a single country. The United States is easily the best for this
purpose. It possesses a large and highly varied area which tends
to produce diversity. Its census and such organizations as life
insurance companies furnish data compiled according to the
same method in all parts. It is homogeneous in government and
institutions ; equal opportunities are open to all ; and the same
ideals and methods prevail almost everywhere. Moreover, no
part has been devastated by war or any other great disturbance
for nearly two generations ; the people have moved freely from
place to place ; and there has been a constant tendency to foster
a single type of culture. From all these points of view no other
equal area is so nearly uniform. Racially the country is of course
complex, but until the last few decades the great majority of the
people have been Nordics from northwestern Europe. The vari-
ous types have mixed to such a degree and have become so
strongly Americanized that the native white population is every-
where similar. The only large elements which tend strongly
toward diversity are the negroes and the recent immigrants from

Figure 34. Climatic Energy in the United States on the Basis of Factory-
Work

m 122 f~\ 117 112 107 102 97 92 87 TO T

in m

Figure 35. Vitality in the United States, According to Life Insurance

Statistics

Figure 36. Distribution of Climatic Health on Basis of Seasonal Variations in

Cities, 1900-1915

Figure 37. Civilization in the United States. The Numbers Indicate the
Relative Rank on a Scale of 6.0 According to 23 Contributors. The
Highest Possible Rank is 6.0 and the Lowest 1.0. This Scale is Entirely
Independent of the One Used for the World as a Whole

278 CIVILIZATION AND CLIMATE

non-Teutonic countries. If these are eliminated, as they are to
a large degree in the maps that we shall consider, essentially the
same degree of energy and civilization ought to prevail through-
out the country except where geographic surroundings, the
presence of some special institution, or some other disturbing
factor makes itself felt.

One of the best tests of energy is vitality. We have already
seen that the death rate varies from month to month in close
harmony with variations in the strength of factory operatives.
Let us now see whether there is similar geographic harmony.
Figure 34 shows the distribution of climatic energy in the United
States. It is a revision of Figure 22, but on a larger scale. Figure
36 is a new map of climatic energy based on seasonal variations
in the death rate as described in Chapters VII to IX. The general
similarity of these two maps compiled in absolutely different
ways is one of the strong reasons for believing that they give a
true idea of the distribution of climatic energy. Where either one
is heavily shaded we should expect people to be strong, vigor-
ous, and long-lived. We might test this expectation by ordinary
mortality data, but such data make no allowance for age nor
for such conditions as the preponderance of unhealthful cities
and manufacturing in the Northeast, and the natural selection of
more vigorous types as migrants to the West. A more reliable
test is furnished by insurance companies, for their policy holders
are much more homogeneous than the population as a whole, and
full allowance is made for age. A comparison between the actual
deaths among policy holders and the deaths expected on the
basis of standard life tables gives a good idea of the vitality of
a community. Figure 35 shows such a comparison. It is based
on the combined experience of three prominent life insurance
companies whose officials have kindly placed in my hands the
figures for hundreds of thousands of people arranged by states.

On the map the figures indicate the actual number of deaths
compared with the average for the whole country, which is taken

VITALITY AND EDUCATION 279

as 100. The country has been divided into five grades corre-
sponding to those of the energy map. The two heaviest shadings
include regions where the mortality is distinctly less than would
be expected from the actuarial tables. The third degree of shad-
ing indicates conditions a little better than the average; the
fourth, high mortality ; and the lightest, very high. The indi-
vidual maps for each of the three companies are closely similar,
which indicates that the general features are not due to any
special policy of one company. Indeed, so far as the policy of
the companies is concerned, the southern states ought to present
a better record than the northern, for the restrictions upon the
issuance of policies are there more rigid. For instance, one
company entirely refuses to issue any policies in certain un-
healthful sections of the South. Elsewhere only people engaged
in special kinds of healthful occupations are accepted. More-
over, in the places where risks are accepted upon the same terms
as in the North, the medical examination is often more strict,
especially in respect to preventable diseases such as tuberculosis.
Various other considerations differentiate one region from
another. For instance, in cities, especially in manufacturing
cities, the death rate is higher than in country districts, and this
causes New York State to be unduly high. Miners are an espe-
cially precarious class from the standpoint of the insurance
companies, and, therefore, are excluded by at least one of our
three companies in some of the more remote western states, and
are accepted only in small numbers by the others. They may
account for the poor position of Montana, but the neighboring
mining state of Idaho ought in that case to be equally bad. In
all the more important respects the tendency would be to cause
the death rate in the South to be lower than in the North, were it
not for the disturbing element of physical weakness due to cli-
mate. In tropical countries the figures are far worse than in the
southern states, which is what would naturally be expected. The
fact that sick people often go west for their health does not

280 CIVILIZATION AND CLIMATE

enter into the problem. Such people cannot obtain insurance.
If they have been insured before they become sick, they are
reckoned as belonging to the place where they lived when the
policies were issued, and not to the place where they die.

It would be interesting to enter into further details, but space
does not permit. The outstanding feature of the vitality map is
its agreement with the maps of climatic energy. As people die in
greatest numbers during months when their energy is low, so
they die in parts of the country where their energy would be
expected to be diminished on account of the climate. In Figures
34, 35, and 36 conditions are best in the North from New Eng-
land to Kansas. Westward and southward they become less
favorable. The decline in health is probably due in part to the
direct physiological effects of climate, and in part to its indirect
effects upon sanitation and other methods for the promotion of
health. The direct and indirect effects of climate almost in-
evitably go together, for where people's energy is great, they
are quick to adopt new means for the prevention of disease and
the improvement of health. Moreover, in dry, and still more
in warm, regions it is easy to tolerate unsanitary methods of
disposing of sewage, which thus pollutes the water supply. Yet
all these things would apparently lose part of their importance
were it not for the weakening effect which certain climatic condi-
tions unquestionably produce, as is so well proved by the varia-
tion of the death rate from month to month even in places where
the conditions of health are most carefully looked after. Here
we must leave the matter. The general agreement of the vitality
map with the map of climatic energy affords strong evidence
that the climatic map is correct.

We are now ready to test our map of civilization in the same
way that we have tested the map of climatic energy, that is, by
comparing it with a map based on statistics. Strength and
energy of character cannot easily be reduced to statistics, for
most of the conditions and activities for which we possess exact

VITALITY AND EDUCATION 281

data depend too much upon outside circumstances and not suf-
ficiently upon the actual qualities of the people. Prof. Mark
Jefferson has studied the matter carefully, and I have borrowed
freely both from his published and unpublished work. It might
seem as if such things as railroads, the number of letters, the
amount of manufactures, or other similar matters might furnish
a good clue to the intellectual capacity and cultural development
of a people, but unfortunately this is not so. Take the case of
railroads. Nevada has more miles of railroad in proportion both
to the number of inhabitants and the inhabited area than any
other state in the Union. In 1920 it had 198 miles for every
10,000 inhabitants, while Rhode Island had only 3.3, and Massa-
chusetts 5.5. This does not mean that Nevada is more progres-
sive than New England. The case is like that of a desert through
which a cowboy was riding when he met a friend.

"What you doing here?" asked the friend.

"Nothing," was the answer. "I'm just crossing this here desert
because it's here."

In the same way the railroads cross Nevada because it hap-
pens to lie between the East and the West. If it were uninhabited,
or peopled by savages, it would still have many railroads.

The letters sent out by a community furnish a criterion of its
state of civilization, but even this must be used with much cau-
tion. A hundred letters sent by a Chicago mail-order house in
response to orders averaging two dollars apiece are no more sig-
nificant than a single letter from Detroit in answer to an order
for an automobile. In this case, as in many others, a concentra-
tion of activity in certain regions may occur without any corre-
spondingly high ability or culture. The same is true of manu-
factures. Doubtless, manufactures generally develop wherever a
community rises to a high state of civilization, and the manu-
facturing processes and all that goes with them are in turn a
help in the development of a still higher civilization. Neverthe-
less, the accidents of position, or the presence of natural re-

282 CIVILIZATION AND CLIMATE

sources are commonly supposed to cause equally progressive and
competent communities to differ enormously in the number of
factories. We shall examine this supposition later.

We are forced, therefore, to turn to something more personal.
Illiteracy and education are fairly good tests, for they depend
largely on the immediate surroundings of each individual. Illit-
eracy would answer the purpose excellently were it not that
people have moved about so much in recent years. Education,
however, is still a local matter, especially in a country like the
United States. Each state, and often each county or town, de-
cides for itself how much it will spend for schools, how long they
shall be open, and how stringently attendance shall be enforced.
Hence, the schools form an unusually delicate test of the real
character of a community. The quality of an educational sys-
tem cannot, indeed, be measured exactly by statistics. Never-
theless it is fairly well indicated by certain conditions for which
accurate statistics are available. For this purpose I have selected
the following data for the year 1920 from the reports of the
United States Commissioner of Education and from the census.

1 . Percentage of young people seven to fourteen years of age
who attend school. Colored children, those born in foreign coun-
tries, and the children of foreign-born parents are excluded in
order to make our data as homogeneous as possible. The per-
centages range from 82.5 in Utah and 79.6 in Idaho to 66.9 in
Maryland and Georgia and 65.4 in Louisiana.

2. Percentage of young people eighteen years of age who
graduate from a four-year course in either a public or private
high school. Since colored and foreign-born persons are not
distinguished from native whites in the data for high school
graduates I have assumed that all the graduates were whites,
and have calculated the percentage of high school graduates
accordingly. This of course gives the South a certain advantage
and thus compensates for the fact that in using the next two

VITALITY AND EDUCATION 283

criteria it has been necessary to include colored pupils in one
case and colored teachers in the other. It must be remembered,
however, that the foreign-born population of the Northeast
tends to lower the standards there just as do the negroes in the
South, although not so much. The maximum percentage of high
school graduates is reached in Maine 28.6, New Hampshire
27.8, and Oregon 26.2. The minimum is found in Georgia 3.9
and South Carolina 2.2, but when the allowance described above
is made for colored people the figures become 6.6 and 4.5.

3. Average number of days per year that each pupil in the
public schools was actually in attendance. Good schools as a
rule are in session fairly long, and always insist on a high degree
of regularity in attendance. The range here is from 151.3 in
New Jersey and 149.6 in Massachusetts, to 76.0 in South Caro-
lina and 76.9 in Mississippi.

4. Average salary of teachers. Unless teachers are well paid,
the more capable young people are likely to give up teaching
for some more lucrative and less nerve-racking occupation. Con-
trary to general belief, teachers are paid at about the same rate
in the Northeast and the far West, the maximum averages being
$1282 in New Jersey and $1262 in Massachusetts compared
with $1375 in Oregon and $1281 in Washington. In the inter-
mediate regions the general scale of recompense falls lower,
dropping to $1081 in Illinois, which outranks all its neighbors
except Ohio, and to $696 in South Dakota. In the South the level
is very low with a minimum of $426 in Georgia and only $291
in Mississippi.

5. The excess of young men over young women among stu-
dents eighteen to twenty years of age. This last item may at
first sight seem unimportant, but its significance grows as one
studies it. In regions where the standards are low there is a
tendency for the older boys to drop out of school, while the girls
keep on in order to become stenographers, bookkeepers, or, espe-
cially, teachers. Maryland and Utah have the greatest excess of

Figure 38. Distribution of Education in the United States in 1920 on Basis of Five
Factors Equally Weighted

Figure 39. Percentage of Gainfully Employed Persons in the United
States Engaged in Manufacturing, 1919

VITALITY AND EDUCATION 285

boys over girls, and Wyoming, Montana, and the Dakotas the
greatest excess of girls.

In using these five criteria each one was given exactly the same
weight, that is, the census figures for any given criterion were all
lowered or raised proportionally, so that for each criterion the
difference between the lowest and highest figures was the same.
As the number of criteria was increased it was noticeable that
the resultant map, Figure 38, became more regular. In its final
form the map shows that in a relatively narrow belt extending
from southern New England through New York, but not Penn-
sylvania, and continuing to Illinois, the general conditions of
education are excellent. The same is true in Utah and on the
entire Pacific coast. In those places, on the whole, a large per-
centage of the children are in school, a large proportion gradu-
ate from the high school, the school year is long and the pupils
are regular in attendance, the teachers are well paid, and many
boys as well as girls continue to study after the age of eighteen.
One or another of these conditions may break down in any indi-
vidual state, but when all are taken together the heavily shaded
areas of Figure 38 rank high.

One of the most noteworthy features of Figure 38 is the rela-
tively low position of the Dakotas and the other states in the
western part of the Great Plains, and also the similar position
of all the Rocky Mountain states with the exception of Utah.
The comparatively recent settlement of some of these states
explains the situation in part, but Washington is almost as new
as the Dakotas or New Mexico and yet stands in the first rank.
The sparsity of population is another factor which hinders
education. Utah, however, has a population decidedly more
sparse than that of the Dakotas and Wyoming, but ranks 126
where they rank from 113 to 119. Its density of population is
scarcely greater than that of New Mexico, whose rank is 110.
The relatively poor conditions in the Dakotas and Wyoming
find a partial explanation in the fact that the population is more

286 CIVILIZATION AND CLIMATE

completely agricultural than in almost any other part of the
United States. There are no large cities and even villages are
comparatively scarce or small. The great majority of the people
live scattered over the vast plain, each family on its own quarter
section. Hence, the children are obliged to go long distances to
school. Muddy roads often make this difficult in the spring,
while the severe storms of winter are an even greater obstacle.
In spite of this, the Dakotas stand at the top in the literacy of
their people, so that the school system must be fairly efficient.

In New Mexico quite the contrary conditions prevail. A part
of the people, to be sure, live on widely scattered ranches where
the children cannot go to school. By far the larger number, how-
ever, live in compact settlements where the houses are grouped
in a comparatively small area because of the necessity for using
a common water supply for irrigation, or else because of mining
industries. Under such conditions, schools can be maintained
more easily than on huge, townless plains like those of the Dako-
tas. The same conditions prevail in Utah and Nevada, and to
a less extent in Idaho, Wyoming, and Montana. Nevada's rank
of 117 may be less creditable than South Dakota's of 114. Yet
the fact remains that the Dakotas are lower than would be ex-
pected, while Utah is surprisingly high.

The proud position of Utah is presumably the result of Mor-
monism. The leaders of that faith have had the wisdom to insist
on a thorough system of schools, and have obliged the children
to attend them. The "Gentiles" have in self-defense been forced
to do equally well, and the result has been admirable. Whatever
one may think of Mormonism as a religious belief, it must be
credited with having accomplished a remarkable work in spread-
ing a moderate degree of education almost universally among
the people of Utah. Without its influence, the rank of Utah
would probably be about 118, that is, between Colorado (119),
on one side, and Nevada (117), on the other. I emphasize this
because it shows how clearly our maps reflect the influence of

VITALITY AND EDUCATION 287

any peculiar condition. Manifest^, the distribution of education
throughout most of the United States does not depend upon the
influence of any particular institution, for essentially the same
institutions prevail everywhere. Yet in the map of education,
Utah is conspicuous because it is strongly influenced by a unique
American institution which is limited to one small area.

Another factor which would be expected to bear an important
part in determining the distribution of education is the presence
of the negro. Doubtless this has a pronounced effect, for an in-
ferior race inevitably retards a higher. Yet the map indicates
that other factors are equally important. Not only is education
at a low ebb in New Mexico which has few negroes, but also in
West Virginia which is comparatively free from both negroes
and Mexicans. Moreover, Texas, where the colored people form
only a fifth of the population, stands lower than South Caro-
lina and Mississippi, where half the people are colored. Another
significant fact is that the education of negroes varies from state
to state almost as does that of whites. Although a smaller pro-
portion of negroes than of whites go to school, the two races
are well or poorly educated in the same places. In Figure 38 the
figures for states with a moderate number of negroes, such as
Virginia with 80 per cent, give a fairly correct impression, but
the impression is not so accurate in those states where there are
many negroes. The reason is that our index figures for states
with many negroes are too high by reason of our assumption that
all the high school graduates are white. Disregarding these minor
discrepancies it seems quite clear that the degree of education is
not proportional to the number of colored people. The southern
states differ among themselves because of special circumstances
such as good or bad laws, but all stand low because of more gen-
eral factors. These factors give to Figure 38 its general char-
acter, and seem to make it as good an epitome of the general dis-
tribution of culture in the United States as we are yet able to
obtain on the basis of the statistics of a single activity. Neverthe-

288 CIVILIZATION AND CLIMATE

less, it can scarcely be doubted that the general level of the
South is much lower than it would be if there never had been any
colored people there, but that would not change the general
aspect of Figure 38.

Let us now compare our educational map with Figure 37,
which represents the distribution of civilization in the United
States according to the opinion of 23 people. All of these except
Ambassador Bryce were Americans. They grouped the states
and provinces of the United States and Canada into six classes,
number 6 being the highest. Massachusetts is the only state
invariably placed in the highest class. New Mexico and Arizona,
which stand lowest, have an average rank of 1.6. In order to
judge how much reliance to place on the classifications, I took
the first ten that were received and averaged them, and when ten
more had come to hand averaged them also. Somewhat to my
surprise, and much to my pleasure, the two sets of averages were
practically identical. How much they differed may be seen in
the Appendix. The average difference is only 0.2, and the maxi-
mum 0.6. The agreement of the two sets probably indicates that
any other group of equally well-informed persons would have
made essentially the same classification. To be sure, in spite of
several attempts, I was unable to obtain any contributor in the
states west of Minnesota or south of the Ohio River. Local
prejudices, however, have probably not exerted much effect on
the final results, for California stands in the highest class, with
practically the same grade as Minnesota and Iowa.

A comparison of the maps of civilization (Figure 37) and
education (Figure 38) is interesting. In general aspect the two
are similar. Both have two high areas, one in the northeast and
center, and the other on the Pacific coast. In both there is a
decline from north to south. Another common feature is a tongue
of high conditions jutting out toward Kansas. This eastern high
area and its western counterpart are almost identical with those
of the map of civilization. In both maps Massachusetts takes

VITALITY AND EDUCATION 289

first place. In the map of progress Connecticut, New York, and
Ohio come next, all three being equal: in the educational map
the rank is Connecticut, New Jersey, Washington, California,
Ohio, Utah, Oregon, and then New York. Judging by this our
eastern contributors to the map of progress did not quite do
justice to the Pacific coast. Nevertheless, the differences among
the states just mentioned are too slight to be important.

The only real discrepancies between Figures 37 and 38 are
the high position of Utah in the educational map, which has
already been explained, and the absence of the low tongue jutting
into Nevada. This is perhaps due in part to an undue lowering
of the general educational level of the Southeast by reason of
the negroes. Also the contributors to the map of progress prob-
ably paid too much attention to aridity, sparseness of popula-
tion, and the ephemeral character of mining towns, and nob
enough to the fact that the inherent quality of the people of
newly settled regions is almost invariably high. The process of
selection incident to migration, as I have shown in The Char-
acter of Races, goes far to insure such quality, provided the
migration is unassisted and is beset with sufficient difficulty.

In concluding this chapter, let us look at another criterion
which is often supposed to be closely associated with civiliza-
tion. This is the percentage of the population engaged in manu-
facturing. It is often asserted that manufacturing is most pre-
dominant in places where supplies of coal and iron are present,
but this is a mistake. A mathematical analysis of the states of
the United States by means of correlation coefficients shows that
there is no relation whatever between the amount of coal mined
per inhabitant and the percentage of the population engaged in
manufacturing. In the world as a whole the same is true; we
have been misled by the accidental circumstance that England,
western Europe, and the United States happen to be places
where coal is very abundant. But, relatively speaking, those
regions stood just as high in civilization before the age of ma-

290 CIVILIZATION AND CLIMATE

chinery as since. They were leaders then as now. The important
factor in determining the presence of manufacturing is the char-
acter and energy of the people. If anyone doubts this, let him
study the progress of manufacturing in Switzerland, Sweden,
New England, Japan, or on our own Pacific coast, in spite of
small supplies of coal. Hence Figure 39 (page 284), showing the
percentage of the population engaged in manufacturing, is in
many ways a map of the distribution of progress. It also reflects
the location of certain natural resources such as cotton in the
South and Cuban tobacco which is manufactured on a large scale
in Florida at Tampa and Jacksonville. Nevertheless, the main
aspects of Figure 39 are surprisingly like those of Figures 37
and 38. The similarity of our maps of education and manu-
facturing to one another and to the map of progress based on
the opinions of our twenty-three contributors, is too close to be
accidental. It seems to show that all three present a fairly accu-
rate portrayal of the actual distribution of human progress.
Since our two maps of climatic energy likewise not only resemble
one another but are closely similar to the map of mortality, we
may feel fairly sure that in all parts of the world the same rela-
tionship holds true.

CHAPTER XIII
THE CONDITIONS OF CIVILIZATION

WE have now reached a crucial point in our investigation.
We must compare the distribution of civilization and
of climatic energy. The reader has doubtless noticed that our
maps of climatic energy in the United States, whether based on
factory work or health, display an unmistakable similarity to
the maps of human progress based on education, manufacturing,
and general opinion. In spite of minor discrepancies, all the maps
show the following features: (1) a high area elongated east and
west from New England and New Jersey to a little beyond the
Mississippi, (2) another and much narrower high area elongated
north and south along the immediate coast of the Pacific Ocean,
(8) a decline from north to south in the east, and (4) a low bay
running across the country in the Rocky Mountain region, with
a pronounced accentuation in the dry Southwest.

While these details are fresh in mind let us turn to three maps
of Europe showing the relation of climatic energy (Figure 40),
health (Figure 41), and progress (Figure 42). The map of
climatic energy is based on our factory data but with some modi-
fication, because we have found that the optimum temperature
for health is 64 or 65 rather than 60 as was inferred from
factory work. The second map shows the distribution of health.
It is based on the official mortality statistics of the European
countries for the years 1909-1913. Infants under one year of
age and old people over seventy-five have been omitted because
the figures for those groups, especially the infants, are unre-
liable. The data for other ages have been reduced to what is

292

CIVILIZATION AND CLIMATE

called a standard population, so that differences in age and in
the proportion of children from country to country have been
eliminated. The map shows the actual distribution of health in
Europe under normal conditions. The third map illustrates the
distribution of civilization. It is the same as Figure 23 in the
chapter on civilization, but is reproduced here in another form
in order that it may easily be compared with the other two.

From Huntington and Williams' Business Geography, by permission of
John Wiley <& Sons, Inc.

Figure 40. Distribution of Climatic Energy in Europe

These three maps are so much alike that if the titles were
removed, most people would not be able to tell which is which.
In each map an area of heavy shading surrounds the North
Sea, and shades off gradually in every direction. In each there
are three projections of heavy shading, one toward Italy, a

From Huntin'jton and Williams' Business Geography, by permission of

John Wiley c Sons, Inc .
Figure 41. Distribution of Health in Europe

80 10 10 20 80 40 6O 60 TO

From Huntington and Williams 1 Business Geography, by pvi mission of

John Wiley <& Sons, Inc,
Figure 42. Distribution of Civilization in Europe

294 CIVILIZATION AND CLIMATE

second toward the Black Sea, and a third along the Baltic. The
maps are so much alike that there can scarcely be any question
as to the reality of their relationship. And that relationship can
be of only one kind. Civilization and progress undoubtedly influ-
ence health, and health in turn has an effect on progress. But
neither civilization nor health can have any appreciable effect
upon the distribution of climate. The only way in which the three
maps can be so alike, unless by sheer accident, which is practi-
cally impossible, is for climate to exert a direct and dominating
influence upon the distribution of health and an indirect and
perhaps less dominating influence upon civilization through
other agencies, such as agriculture.

Now turn to the world as a whole. Unfortunately the mor-
tality records of many backward countries are highly inaccu-
rate, while for vast areas they are wholly lacking. Hence it is
impossible to construct a map of health. We can, however, con-
struct maps showing the distribution of climatic energy and of
civilization. A map of climatic energy based on factory work
has already been given in Figure 22. This is repeated in Figure
43. The map of civilization on the same page, Figure 44, sums
up the data already shown in Figures 23 to 28. All regions to
which our fifty contributors assign a rank of 8.5 or higher are
rated as "very high," and are shaded in solid black. Those from
7 to 8.5 are rated as "high," and arc shaded in heavy lines ; those
from 5 to 7 are "medium," and are indicated by light lines ; 3 to
5, "low," shaded with abundant dots ; and under 3, "very low,"
and dotted only lightly.

The first thing that attracts attention is the general resem-
blance between the maps of energy and of civilization. Both, for
example, show a high area in northwestern Europe. A tongue
extends into Italy, another toward Roumania, and a third to
the Baltic. Another projection runs out into western Siberia.
Here the high area of the map of civilization extends about as
far as the medium area of the map of energy. This is not surpris-

(66 W MO K9 100 HO 60 49 O AO 40 60 80 309 MO 240

Figure 48. The Distribution of Human Health and Energy on the

Basis of Climate

Figure 44. The Distribution of Civilization

296 CIVILIZATION AND CLIMATE

ing, for even if the people of Siberia have the energy indicated in
Figure 43, they are hampered by the remoteness and newness
of their country, not to mention other conditions. In central and
northern Siberia, the difference between the two maps is slight.
The significant thing is that in both there is the same falling off
toward the center of Asia. Still farther east in China and Japan
conditions are once more alike, China being medium and Japan
high.

In Indo-China and especially in India, the maps differ. Appar-
ently, this arises largely from European domination, and is due
to the constant addition of strength from that continent. This
does not apply to Siam, however, which has worked out its own
salvation. It ranks as very low on the energy map, and only as
low on the other. This may have no significance, for our maps
are still in their early stages. Further knowledge may change
such slight disagreements into agreements. On the other hand,
it may increase the disagreement. In that case we may discover
that by long residence within the tropics, the races of Indo-China
and India have become differentiated from Europeans and are
less susceptible to the influence of steady heat. Again, race
differs from race in its inheritance, and the Siamese may inherit
stronger traits than are possessed by their neighbors. Finally,
the level of Siamese civilization may have been raised by contact
with other races, by the adoption of particular institutions of
government, forms of religion, or social organization, or by the
inspiration and energy of a few men of unusual gifts. I mention
these possibilities not because they are of special importance in
Siam, but because they illustrate the many and varied influences
which cooperate to determine the position of a country in the
scale of civilization.

In comparing the maps of energy and civilization, one of the
clearest features is the effect of a strong race upon regions
which it rules or colonizes. Again and again the presence of such
a race causes a region to be higher in civilization than would be

CONDITIONS OF CIVILIZATION 297

expected on the basis of climatic energy. Java, the Philippines,
and India are examples. It is especially noticeable in regions
controlled by Great Britain. In Australia, for instance, the
general decrease in both civilization and energy from southeast
to northwest is the same in both maps, but the presence of the
English raises the places of "very low" energy to "low" in
civilization, and so on, each grade being raised one degree, so to
speak, until the map of civilization shows a large high area in
the southeast. In South Africa and Egypt, British influence
is displayed in the same way. In the Canadian Northwest, on
the other hand, it is not apparent. The northern parts of Alberta
and Saskatchewan appear higher in energy than in civilization.
We have already seen that according to American and, to a less
extent, British opinion, this is not the case, for recent settlement
has raised these regions to a comparatively high degree of
culture.

In the United States the energy map shows a strip of medium
conditions along the southern frontier, but this is rated as high
on the other map. Such a condition illustrates how a high type
of government causes efficient people to settle in unfavorable
regions, and how it also adds to the effectiveness of less efficient
people such as Mexicans and negroes, thus in part overcoming
the handicap of climate. In the central states, on the other hand,
civilization is not rated so high as one would expect on the cli-
matic basis. Probably this is because the country is so new that
our Chinese, Russian, Spanish, and other foreign contributors,
though they have traveled and studied extensively, do not realize
how great is the progress of recent times. California, like the
southern states, is higher on the map of civilization than on the
other. As already explained, this may in part be due to the im-
possibility of making a wholly accurate map of climatic energy.
It may also arise from the location of California on the sea-
board, and from its early development as contrasted with the
newer states of the interior. A comparison of the United States

298 CIVILIZATION AND CLIMATE

as it appears on the world-maps and as it appears on the maps
of that country alone is important. Where the country stands
by itself, and its parts are classified by people who live in it
and are thoroughly familiar with it, the resemblance between
climatic energy and civilization is greater than where the classi-
fication is on a rougher scale and is made by people less familiar
with it. Another reason for the difference is that a classification
of places where a uniform standard of culture prevails and where
the same race is everywhere dominant is much easier than where
many types of culture and highly diverse races are considered.
The maps of the United States represent the kind which must
be made for each country. The difference between the features of
the United States on the world-map and on the other represents
the extent to which our general map of civilization is in error.
In spite of this, however, the general features of the country are
unmistakably the same on both maps. So far as the conclusions
of this volume are concerned, it makes no difference which we use.
In this lies the importance of our various tests of the United
States. They show that although much remains to be done before
we can construct a map which is approximately perfect, the
most important features are reasonably distinct and unmis-
takable.

Turning to Latin America, we find about what would be ex-
pected in Mexico and Central America. The highlands are me-
dium and the lowlands low. South America, on the contrary,
presents some unexpected features. The Andean highlands,
including Venezuela, Colombia, Ecuador, Peru, and Bolivia, are
all ranked as low in civilization, whereas the climatic map would
indicate medium energy. In the belt of highlands on the east
side of Africa the same phenomenon is observable. Perhaps an
equatorial climate is more debilitating than would be expected
from the work of factory operatives in summer. In South
America the presence of an ancient race whose vigor was already
waning at the time of the discovery of the New World has doubt-

CONDITIONS OF CIVILIZATION 299

less hindered Spanish immigrants in accomplishing what might
otherwise have been looked for. This, however, does not alter the
case, for the original inhabitants in the Andean countries, just
as in the African highlands, stand lower than would be expected.
Argentina, on the contrary, goes to the opposite extreme, and
is higher on the map of civilization than on that of energy. The
importance of this must not be overrated, for the climatic data
are somewhat doubtful because of the paucity of statistics as to
changes of temperature from day to day. As the maps now
stand, however, they are encouraging, for they suggest that
even with a moderately favorable climate, the Latin race in
America is competent to rise to a high level.

Let us turn now from these details, and look once more at the
general aspect of the two maps. In spite of minor disagreements,
the main features are essentially alike. There are, in each case,
the same two great high areas in western Europe and the United
States. The decline from western Russia eastward to the center
of Asia, and the rise to high conditions on the eastern edge of
Asia in Japan are equally apparent. Likewise, the maps are
strikingly alike in the shape of the very low areas in Africa and
South America. South of latitude 30 S. each of the southern
continents begins to rise in energy and in civilization, and the
rise is more pronounced on the eastern side than on the western.
Even where the maps disagree, the explanation of the disagree-
ment is often obvious from a consideration of the recent move-
ments of European peoples. Some of the remaining discrep-
ancies are explicable on well-known grounds, such as the
impossibility of agriculture, which hinders civilization in the far
northern parts of America and Asia. In addition to all this many
differences in the degree of progress among people in similar
climates are due to racial inheritance. A conspicuous example of
this sort, as I have shown in The Character of Races, is the
contrast between the progressive Icelanders and the backward
people who live in a similar climate in the southern tip of South

800 CIVILIZATION AND CLIMATE

America. A still more conspicuous example is the difference
between the primitive hill tribes of India and the highly compe-
tent Parsis whose ancestors came to India more than a thousand
years ago because of their loyalty to their old Zoroastrian reli-
gion.

When allowance is made for obvious facts like these, the re-
semblance between the two maps becomes more striking. Call to
mind the method of their construction. Neither represents the
personal opinion or bias of any one man. Any other person with
the same data before him would have obtained similar results.
The maps simply give expression to two distinct sets of facts.
The first is that the opinion of men of many races agrees as to
the general distribution of civilization. The second is that if
the various conditions of climate produced the same effect upon
all the people of the world as upon students and factory opera-
tives in the eastern United States, the amount of work accom-
plished in different countries would be closely proportional to
the status of civilization.

Aside from the map of climatic energy, it is hard to think
of any other which would so closely reproduce the features of
the map of civilization. Suppose that race were made the cri-
terion, and that a map were shaded in proportion to the number
of Teutons. We should find that in Europe such a map would
closely resemble the map of civilization except that places like
Finland, southern France, central Italy, Hungary, Bohemia,
Servia, and others are relatively high in civilization even though
none are more than half Teutonic, and some only very slightly,
or almost unappreciably so. In Asia, on the other hand, there
is much more Teutonic blood in Syria and Asia Minor than in
Japan, yet Japan ranks far higher. The Japanese might claim
racial superiority almost as fairly as the Teutons, and both the
Latins and Slavs may justly point to the fact that they pre-
dominate in some of the most advanced portions of the globe.
When we look at the low places, we find that Teutonic areas,

CONDITIONS OF CIVILIZATION 301

such as the Transvaal, Alaska, southern Greenland, and parts
of Australia make a poor showing; the Latins in parts of Latin
America are even worse ; the Slavs at their worst fall no lower
than the Teutons ; while the Japanese nowhere fall so low.

A map of religion does not resemble a map of civilization, no
matter which religion is employed. Protestant Christianity,
indeed, prevails chiefly in regions which are either high or very
high; Iceland is by no means an exception, for its civilization
is much higher than would appear from our table, where it is
grouped with Greenland. Roman Catholic Christianity, on the
other hand, prevails in locations which range from very high
to very low ; and Greek Christianity from high to low. Buddhism,
likewise, ranges from high in Japan to low in Tibet, while Mo-
hammedanism never rises above medium, and in some places falls
very low. That religion raises or lowers the tone of a country
I do not for a moment question, but if a people are physically
weak and are lacking in self-control because of something in
their surroundings, the history of the world as exemplified by
groups of people who have long been nominally Christians in
Abyssinia, India, Latin America, and elsewhere seems to show
that they debase even the finest religion. The higher the form of
religion and the more self-sacrifice and devotion it requires, the
more difficult it becomes to keep any but the most energetic and
determined races even approximately true to it. The only cases
where people of low efficiency seem to remain true to a high
religion are where they are continually stimulated by the pres-
ence of a stronger race.

As a third criterion, suppose that we take form of govern-
ment, and inquire whether a map of governments would resem-
ble one of civilization. Of course, the excellence of the govern-
ment is closely related to the degree of civilization, but not so
with the form. Republics range from very high in Switzerland
and France to very low in Venezuela. Limited, but autocratic
monarchies existed in high countries like Germany, at least,

802 CIVILIZATION AND CLIMATE

before the great war, and also in low countries like Turkey and
Persia. Thus we might go on to consider one after another of
the great factors which cooperate in giving form to modern
civilization. The nature of a nation's religious faith, its form of
government, its social organization, its ease of intercourse with
other nations, and various other conditions play a fundamental
part in the distribution of civilization. Yet each is conditioned
by the degree of energy possessed by a people, for if a race lacks
energy, no amount of excellence along other lines will place it in
the first rank. Energy, in turn, is greatly influenced by climate,
and thus climate becomes an essential element in determining the
status of civilization. We may well reverse our statement, how-
ever, and say that no amount of energy will make a nation great
if none of its people are gifted with genius, or if it never evolves
an orderly form of government or a moral code which allows a
man to enjoy life, property, and home without constant fear of
outsiders Thus, the material and immaterial elements of civiliza-
tion play into each other in such a way that either seems the
more important according to the angle from which we view it.
The interplay of diverse factors is so important that it is
worth while to examine it in a concrete case.* Dr. Scott Nearing
in the Popular Science Monthly for 1914 published an article
entitled, "The Geographical Distribution of American Genius.'
While it is impossible to measure genius, it is possible to ascer-
tain how many people of unusual ability are born in a given
region. That useful publication, Who's Who in America, though
not infallible, forms a good summary of about twenty thousand
people who have either achieved "special prominence in credit-
able lines of effort, making them the subjects of extensive inter-
est, inquiry, or discussion in this country," or who occupy
positions which could scarcely be attained except by persons of

* The following discussion is reproduced unchanged from the first edition
of this book. In The Character of Races I have considered the changes in the
past ten years, but they do not alter the conclusions set forth below.

CONDITIONS OF CIVILIZATION 303

unusual ability. Taking Who's Who for 1912-1913 as a basis,
Nearing has tabulated the birthplaces of the first 10,000 names
according to states. He took only 10,000 because that number
seemed enough to give reliable results. His tabulation strikingly
reenforces the common opinion that New England, especially
Massachusetts, has produced far more than its proportionate
share of persons of unusual ability. The utility of his investiga-
tion seems so great and the method so reliable that I have asked
Dr. Nearing for permission to make use of fuller data than were
contained in his article, and he has kindly supplied me with the
figures for each state. In order to determine the relative status
of the various parts of the country it is not fair to compare the
number of eminent persons who were born in a given area with
the present population. At the time when the men who are now
prominent were born many of the western states contained only
a handful of settlers. It is equally unfair to compare the number
of such persons who live in a given region with the present popu-
lation, for many persons who have achieved prominence owe it
to the place where they grew up and not to that where they now
live. The only fair way seems to be to ascertain the relation
between the number of eminent persons born in a given region and
the population of the region at the time of their birth. Accord-
ingly, the first thing to do is to find when the people in Who's
Who were born. Nearing gives the following table :

NUMBER OF EMINENT PERSONS IN THE UNITED STATES IN 1912
WHO WERE BORN AT CERTAIN TIMES

TIME OF BIRTH HUMBER

Before 1850 2,818

1850-1859 2,715

1860-1869 2,717

1870-1879 1,304

1880-1889 95

1890-1899 2

Unknown 349

Total 10,000

804

CIVILIZATION AND CLIMATE

The number who were born before 1840 is not given, but it
must be considerable, for the people who attain eminence are
among the most long-lived portions of the community. On the
other hand, the number who were born after 1880 is too small to
be considered. People rarely become eminent before they are at
least thirty-five years of age. The forty years from 1835 to
1875 cover the births of practically all who had attained suffi-
cient distinction to be included in Who's Who for 1912. Accord-
ingly, we must find the average population of each state accord-
ing to the censuses from 1840 to 1870, but inasmuch as the
number who were born previous to 1840 is less than in later
decades, we shall come nearer to the truth if we give that census
only half as much weight as the others. In all cases we employ
the figures for the entire white population, whether native or
immigrant, but omit the negroes, Chinese, and Indians. The way
in which the matter works out is illustrated in the following
table, where the population is given in thousands :

White population in thousands

Eminent
persons

Eminent
persons
per
100,000

1840

1850

I860

1870

Average
1840-70

Massachusetts
South Carolina
Nebraska

729
259

985

275

62

1221
291
29
83

1443

290
122
90

1147

282
43
67

1123
111
24
1

98.0
39.4
55.8
1.5

New Mexico ....

The figures in the last column show the relative rank of these
states in the production of persons of unusual ability from 1835
to 1875. Similar figures for each state are given in Figure 45.
Since Nearing used only the first 10,000 names of American-
born persons in Who's Who, or only about 60 per cent of the
total, the index figures really mean the number of eminent
persons for every 60,000 people instead of 100,000. In a few

CONDITIONS OF CIVILIZATION 305

cases where the average population previous to 1875 was less
than 10,000, two or more adjacent states have been combined
so as to give a total large enough to be significant. The numbers
thus obtained have been enclosed in parentheses. On the map the
United States has been divided into four grades, much as in
Figures 34 to 39. Thus all these maps are comparable. The only
essential difference is that Figure 45 belongs to a period averag-
ing more than half a century earlier than the others. It presents
the most accurate picture now available of the distribution of
ability at that time. New states are at no disadvantage compared
with the old, for if a region had no population previous to 1860,
for example, and only a few thousand in 1870, full allowance is
made for this. Many of our 10,000 eminent people have moved
away from their early homes, but the great majority did not
go until they had at least reached an age approaching twenty
and the main elements of their character were already formed.
Thus the peculiarities of the map depend not only on whether the
population was of such a caliber that children of high ability
were produced, but also on the conditions which molded the early
life of such children.

Aside from accidents three chief conditions determine the num-
ber of eminent persons in a community. The first is inherited
ability. Unless a man is born with more than the average mental
capacity, the chances of his inclusion in Who's Who are slight.
The second condition is opportunity in the broadest sense of the
word. A bright child born on a remote farm in Maine, on a ranch
in Arizona, or in a clearing among the Tennessee mountains may
be so hampered by lack of education and of the stimulus derived
from contact with people outside his own little circle that he
never accomplishes anything that attracts attention. The third
condition is energy. Many a man of high ability, who is also
blessed with the best education and with all sorts of opportu-
nities to develop his talents, fails to make any impression on the
world because he is indolent. Frequently, a man of less ability

306 CIVILIZATION AND CLIMATE

but endowed with energy achieves much more. Energy depends
partly on inheritance, but also on climate. So far as it depends
on inheritance it should be included under the first of our three
conditions. Thus the three may be briefly defined as (1) inherited
qualities of all kinds, (2) opportunities, which include educa-
tion, the degree of culture in a community, and the freedom with
which a person can find scope for his particular talents, and (3)
energy so far as this depends upon physical circumstances not
connected with either heredity or opportunity.

Let us now inspect Figure 45 to see how far our three con-
ditions make themselves evident. Each gives rise to certain
features which stand out unmistakably. To begin with inherit-
ance, Massachusetts gave birth to 98 eminent persons for every
60,000 of its white population during the specified period. That
is, 1 white child out of every 600 born at that time has dis-
tinguished himself. The figures for the surrounding New Eng-
land States and New York range from 50 to 78. Such a striking
difference is certainly not due to climate. It is equally certain
that it is not due to opportunity. The average child in New
York has as good a chance to go to school and enter any sort of
occupation as has the child in Massachusetts. Yet the rank of
New York is only half as high as that of Massachusetts. In
Maine, Vermont, and New Hampshire the opportunities are dis-
tinctly less than in New York. There is much less wealth, the
people are more isolated, the number of cities is proportionately
smaller, the common school system is no better developed, and
the facilities for sending children to college are not so great.
Yet even Maine outranks New York, for she produced 54 emi-
nent persons per 60,000 while New York produced only 50. Yet
New York itself stands very high. Aside from the New England
States only Nebraska exceeds it, while Oregon and Delaware
rival it. South Carolina is another state which stands far higher
than its neighbors, for although 39 is low compared with the
78 of Connecticut, for example, it is high compared with the 24

CONDITIONS OF CIVILIZATION

307

of North Carolina and the 25 of Georgia. Probably, heredity
plays an important part here, as in other cases ; although, as we
shall shortly see, the matter is complicated by other conditions.
Oregon and especially Nebraska, however, are unmistakable.
Proportionately, they stand as high above their neighbors as

Figure 45. Birthplaces of Persons of Unusual Ability in the United States. The Nu-
merals Indicate the Number of Eminent Persons Born in Each State per 60,000 of
the Average White Population from 1835 to 1875

Massachusetts, Connecticut, Rhode Island, and Vermont above
theirs. The case of these two states is most suggestive. So far
as energy is concerned, there is nothing in the climate of either
Oregon or Nebraska to give them a special advantage. Previous
to 1890, by which time the education of four fifths of the people
in Who's Who was completed, these two states did not offer
their children especially great opportunities. In fact, the oppor-

308 CIVILIZATION AND CLIMATE

tunities were much less than in Oregon's next neighbor, Cali-
fornia, or in Iowa, Illinois, Indiana, Ohio, and Pennsylvania, the
states directly east of Nebraska. Yet Oregon ranks 30 per cent
higher than California, and Nebraska exceeds Illinois by 60 per
cent, and Indiana by over 90 per cent. In striking contrast to
Nebraska we find New Mexico with a rank of only 1.5, which
appears as 2 on the map because we have avoided the use of
fractions. Here again neither climate nor opportunities explain
why this state falls so far behind its neighbors. The only reason-
able explanation is that until 1870 or later its "white" popula-
tion consisted almost wholly of Mexicans.

In reading the preceding pages it may have occurred to the
reader that the preeminence of New England is only apparent,
not real. It may be due largely to the local prejudices or limited
viewpoint of the compilers of Who's Who. This is not the case,
however. The book is edited and published in Chicago. Yet
Illinois and the neighboring states all receive a relatively low
rank.

The facts just stated are of profound significance. Massachu-
setts, because she was settled by the strong-willed Pilgrim
Fathers and by other Puritans who fled to the wilderness to
maintain their high ideals, has produced vastly more than her
proportion of the men who have made America what it is. Con-
necticut and Rhode Island for similar reasons have followed
closely on her heels, while the northern New England States
have much more than held their own compared with the rest of
the country. It has been a fad to decry puritanism, but people
of puritan descent have taken the foremost place. They have
done so because they inherit the strength of mind which made
it possible for the Puritan Fathers to develop their stern con-
scientious system and carry out their noble purposes in the face
of temptation and opposition. When that old stock has been
transported to places such as Nebraska and Oregon, where for
a while it was dominant before the great tide of later immigra-

CONDITIONS OF CIVILIZATION 309

tion, it raised the average ability to a level reached nowhere else
except in New England. In New Mexico, on the contrary, we
harbor a group of people, fortunately small, who are even more
conspicuous by their lack of ability than the New Englanders
are in the opposite way. We may excuse the Mexicans by saying
that they do not learn our language and do not merge them-
selves in our civilization. The competent Mexicans, however,
usually those who possess the greatest proportion of Spanish
blood, do learn English and make themselves felt among us. The
others, perhaps because they inherit an inert disposition from
their Indian ancestors, are content to remain backward.

This brings up the great question of immigration and racial
character. In the earliest days of colonization we received only
the stronger elements of the various European populations. The
North had its Pilgrims, Puritans, Quakers, and others, while in
the southern states a part of the settlers were people who as
Huguenots or other religious refugees were notable for tenacity
of purpose and high ideals. The rest of the settlers were in large
measure people of unusual courage and initiative, for others were
not brave enough to come. For this reason, apparently, the
states of the Atlantic coast from Georgia northward stand
higher than those west of them. After America had been settled
so long that migration thither was easy, we began to get immi-
grants of medium grade, not the best nor the worst, but from
advanced countries and from the substantial middle classes.
These are what predominate from Pennsylvania to Iowa. They
are good material, but not so good as the old. Otherwise why
should so fine a state as Wisconsin have produced only half as
many eminent men per 100,000 as has Connecticut, and no more
than South Carolina, which labors under far greater disad-
vantages? In these last decades we are taking into our midst
many people scarcely better than the Mexicans. We may say
what we choose about absorbing them and making them good
Americans. It is our duty to do so as far as we can, but why

310 CIVILIZATION AND CLIMATE

blind ourselves to the facts of biology? Plough horses cannot
race like thoroughbreds. Do men gather grapes of thorns or
figs of thistles ?

Today Massachusetts and New England seem to be losing
their supremacy in the production of men of special ability.
A study of Who's Who for 1922 shows that the conditions of
ten years earlier still prevail, although less marked. But where
the old New England families have sent their sons out over the
wide expanse of our land, the loss to the mother states is more
than compensated by the gain to the rest of the country. Unfor-
tunately, the change means more than that. It means, first, that
we are steadily diluting our strength. We are acting as would
a dairyman who thought that by adding a dozen low-grade ani-
mals to his herd of a hundred prize-winners and letting them
breed together he was going to increase the value of his stock.
In addition to this we are losing in another and more dangerous
way. It is as if the dairyman should not only add poor animals,
but should also prevent his best animals from bearing young. No
amount of care would make the low-grade animals give as much
milk or be of as much value as the prize-winners. Man is subject
to the same biological laws as animals. High mental ability and
strength of purpose are his most valuable qualities. Yet we act
as if we thought that though these are not reproduced, our coun-
try can continue to advance. Our unwillingness to live simply
either prevents a large proportion of our most competent men
and women from marrying, or causes many of those who marry
to have few children. All men are not created equal biologically,
and it is the best who are dying out. We must recognize that fact,
and act upon it before we have worked irreparable injury. All
this has been said many times by eugenists, but it must be
repeated again and again until it is not only believed but acted
upon. Biology teaches it; common sense insists upon it; and
now our purely geographical studies enforce the same con-
clusion.

CONDITIONS OF CIVILIZATION 311

The second condition which controls the distribution of people
who attain eminence is opportunity. This appears unmistakably
in only one portion of Figure 4<5, but there it stands out sharply.
Notice how West Virginia with 19 eminent persons per 100,000,
Tennessee with 18, and Arkansas with 11 fall below the sur-
rounding states. This is apparently because these are the por-
tions of the South where mountains and other physiographic
disadvantages cause the people to degenerate into "poor whites"
and "crackers" in spite of a good inheritance. What these back-
ward communities need is a "chance." They need the opportu-
nities that are brought by schools, railroads, factories, and the
other appurtenances of civilization. They need also the oppor-
tunity brought by freedom from such bodily afflictions as the
hookworm disease. Kentucky, which now has a rank of 23, would
probably stand much higher were not a large part of the state
peopled by mountain whites. The same is true of North Caro-
lina. Perhaps this state would not equal South Carolina, which
had a large number of old families of unusual ability, but the
two would be much nearer than now. A large fraction of North
Carolina consists either of mountains or of swampy, unhealthy
tracts along the coast, while South Carolina is almost free from
such disadvantages. In another portion of the country it may
be that Maine lags behind New Hampshire and Vermont in part
because of her relative remoteness and lack of opportunity.
Doubtless other places show the same conditions, but the matter
is not certain. For example, Nevada's low position with a rank
of only 15 is probably due in part to this cause, but it is doubtful
whether she has been much worse off than Utah, which has the
respectable rank of 36. Idaho, Wyoming, Montana, and the
Dakotas certainly had no more opportunities than Nevada
previous to 1890, for they were supplied with fewer railroads,
and were much less easily in touch with the rest of the world.
Yet their rank is 33, or more than twice that of Nevada. Taking

312 CIVILIZATION AND CLIMATE

the map as a whole it seems that although opportunity is highly
important, it is less important than heredity.

The preceding paragraph stands as it was written ten years
ago. Today I am inclined to think that I have not given sufficient
weight to inheritance in estimating the position of the mountain
whites. Note first, however, that in Figure 36, where climatic
energy is estimated on the basis of the effect of climate on health,
the mountain white region, by reason of its altitude, forms a
peninsula of more favorable conditions jutting southward. Some
non-climatic factor seems completely to overshadow this climatic
condition. That this other factor is in part the isolation of the
mountain valleys I do not doubt. But this is probably supple-
mented to a large degree by inheritance. When people are able
to move freely from one region to another, there is a strong
tendency for those who are most competent to gain possession of
the best places. The rich cotton soil of the Black Belt of Georgia
and Alabama, as R. M. Harper has well shown, has attracted
and held a large number of unusually competent families the
kind that produce leaders. In such a region the lands of the less
competent people are gradually bought by the competent. The
less competent often move into the mountains, where land is
cheap. There is likewise a backward movement whereby the most
competent mountaineers drift to the lowlands. Thus in course of
time there arises a genuine inherent difference of ability, and
that may be one cause of the backwardness of the mountain
whites.

Turning now to our last factor, that is, energy as determined
by climate, we see that, in general, the outlines of Figure 45 are
like those of the climatic map in Figure 36. To be sure there
are important differences. For instance, the very high area
which covers all the northeast and center of the country in the
energy map is split into a New England and a Nebraska por-
tion in the map of ability. Yet the Nebraska area of many emi-
nent people displays an interesting resemblance to the tongue

CONDITIONS OF CIVILIZATION 313

which projects out in the same direction on the energy map.
The Pacific coast is likewise high on both maps, although there
are differences of detail. Nevada, too, is at the head of a tongue
of low conditions in both cases. On the Atlantic coast both maps
rise from Maine to Massachusetts, and decline from New York
to Florida. In the map of ability, however, high conditions go
somewhat farther south than in the other map, and South Caro-
lina, presumably because of heredity, rises unexpectedly. The
other southern states from Georgia to Louisiana are also a little
higher on the ability map than on its companion, probably
because of the wealth and opportunities which prevailed in them
previous to the Civil War, or else because of the abundance of
old families with high ideals and strong minds. Yet even these
states are lower than the tier of northern states from Pennsyl-
vania to Iowa, where the average inheritance is probably no
higher, if as high, but where the climate gives energy.

Taken as a whole the map of ability is an admirable example
of the way in which a variety of factors cooperate in determining
the status of civilization. Climate, as it were, paints a broad
background, shading gradually from very high in certain areas
to lower in others. Then the other factors come into play. They
paint fresh colors which may or may not resemble those of cli-
mate. In some cases, such as Massachusetts, the same color is
laid on by climate, heredity, and opportunity, not to mention
proximity to the sea and to Europe, facilities for manufactur-
ing, and various other factors which perhaps may be considered
as opportunities. Where that happens, high civilization is sure
to prevail. In other cases, such as South Carolina, the climate
paints only a moderately high color, inheritance paints a higher
one, education a low, the presence of the negroes a still lower,
and so on indefinitely.

Such, then, is the meaning of our maps. They do not indicate
that climate is the only factor in determining the condition of
civilization, or even the main one. Far from it. Yet they indicate

314 CIVILIZATION AND CLIMATE

that it is as essential as any other. Today civilization seems to
make great progress only where a stimulating climate exists. A
high civilization may be carried from such places to others, but
it makes a vigorous growth and is fruitful in new ideas only
where the climate gives men energy. Elsewhere it lags, or is kept
at a high pitch only by constant reinforcements from more
favored regions. In the past men have perceived that climate is
apparently one of the most important conditions which favor
or retard the growth of civilization. They have been greatly
impressed not only by its effects upon their own bodies and
minds, but by the fact that in warm countries the amount of
progress is closely in harmony with what would be expected on
the basis of one's own feelings. At the same time they have
realized that among countries located in the same latitude there
are differences of culture almost as great as between temperate
and tropical countries. This has seemed to indicate that climate
is not so important as the tropical regions would suggest. Now,
however, we see that when people's actual achievements under
various climatic conditions are measured, we must revise our
opinion. Variations of temperature from day to day are much
more important than has been realized. Therefore, in the same
latitude the stimulating effect of the climate may differ greatly.
The civilization of the world varies almost precisely as we should
expect if human energy were one of the essential conditions, and
if energy were in large measure dependent upon climate.

CHAPTER XIV
THE SHIFTING OF CLIMATIC ZONES

WE now have before us the main hypothesis of this book,
as stated at the end of the last chapter. But even if
many facts suggest that civilization at present varies from place
to place almost precisely as we should expect if it were dependent
upon climatic energy and health, is there not abundant evidence
to the contrary? Three objections at once present themselves:
(1) The great nations of antiquity developed their culture in
regions where the climate is now relatively unstimulating. (2)
The American Indians, even when they lived in some of the
world's best climates, failed to evolve a high civilization. (3)
People of European races are today maintaining and developing
a high civilization in relatively unstimulating climates such as
that of northern Australia. Each of these objections is highly
significant and must receive careful attention. The first will
form the subject of the present chapter.

In a series of books, among which The Pulse of Asia, Pales-
tine and Its Transformation, and Climatic Changes are the
most important, I have set forth the hypothesis that during
historic times the earth's climate has been subject to pulsatory
changes. It cannot be too clearly understood that this hypothe-
sis was not framed with reference to the hypothesis of civiliza-
tion and climate presented in this book. It was developed inde-
pendently before I realized how closely the distribution of civi-
lization is bound up with that of climate. In fact, the hypothesis
of climatic changes was what led originally to the studies de-
scribed in this book.

316 CIVILIZATION AND CLIMATE

The steps which have led to this hypothesis may be summed
up as follows : In many parts of Asia, Africa, and America ruins
of towns and cities are located where now the supply of water
seems utterly inadequate. Old strands surround lakes that are
now dry ; old alluvial terraces in which lie remains of human
occupation show that the rivers have changed their habits of
erosion and deposition since men began to become civilized. Old
roads traverse deserts where caravans cannot now travel ; traces
of dry springs are seen; bridges span channels which carry no
water for years at a time ; old fields are walled and terraced in
places where now the rainfall is too scanty to permit agriculture
and where no water can be brought for irrigation. Elsewhere old
irrigation canals abound in districts where today there is little
or no water supply, or where what water there may be in the
streams is so salty that it kills the crops instead of invigorating
them. These things and many others, which almost every traveler
in semiarid or desert countries has seen for himself, seem to be
almost irrefutable evidence that at some time the climate was
moister than now.

Such evidence long ago gave rise to two hypotheses, which are
now almost abandoned, those of deforestation and of progressive
desiccation. According to supporters of the first hypothesis, the
reckless cutting of forests has not only allowed the rains to
denude the mountain sides of soil, but has caused an actual
diminution in rainfall. This view once had a considerable popu-
lar vogue, but for various reasons it has now practically ceased
to be considered among scientists. In the first place, modern
measurements of rainfall before and after the deforestation of
large tracts are contradictory. At best they show only slight
differences, too small to have any appreciable effect, and in prac-
tically every case so doubtful that they may be due merely to
the accident of an especially dry or rainy period of a few years
coming not long before or after the forest was cut. Moreover,
many of the strongest evidences of desiccation are found in

SHIFTING OF CLIMATIC ZONES 317

places such as southeastern Syria where there is no reason to
think that the country has ever been forested since it was first
occupied by civilized man. Finally, there are thousands of square
miles in Chinese Turkestan where the forests themselves have
died because of lack of water, and are still standing as gaunt
skeletons preserved for a thousand years or more because of the
extreme dryness of the air.

The hypothesis of progressive desiccation assumes that dur-
ing historic times the earth has steadily been growing drier.
This was my own view when I published Explorations in Turkes-
tan in 1905. Many careful students still uphold it. The majority
of its supporters, however, apparently think that it needs modi-
fication along lines which will shortly appear.

While evidence of more water in the past than at present is
prominent in many places, there is also much of the contrary
nature, less noticeable, but no less convincing. For example,
ruins are located on the floor of lakes which must have been
partially dry when the structures were erected. Elsewhere one
finds irrigation canals in places now so damp that their con-
struction would seem to be a waste of energy. In north Africa
and Syria huge irrigation works are located in regions of
another kind, which not only are dry now, but must have been
dry in the past. Otherwise the Romans would not have expended
such enormous labor to get water. These things and others
furnish almost irrefutable evidence that at certain periods the
water supply of many semiarid regions was no greater than at
present. Because such evidence is less abundant and noticeable
than the other kind, the believers in progressive desiccation
have overlooked it. On the other hand, other students have been
so impressed by it that they have held that there have been no
changes of climate during historic times, and that the fluctua-
tions which followed the last glacial epoch came to an end before
the beginning of history.

There seems only one way to reconcile these two opposing

SHIFTING OF CLIMATIC ZONES 319

views, each of which is based on unassailable evidence. That
way, as I first showed in The Pulse of Asia (1907), is to group
the evidence according to its date, and see how far the indica-
tions of moisture and aridity come at different times. For ex-
ample, several lines of evidence, such as the location of dwellings
in tracts that are now great swamps in Ireland, unusually pro-
longed famines in China and western Asia, and vegetation of
warmth-loving types in Europe suggest a dry period perhaps
1400 to 1200 years before Christ. At the time of Herodotus, be-
tween 400 and 500 B.C., all the evidence points to moist condi-
tions in western Asia and northern Africa ; about 200 B.C. a
somewhat drier climate apparently prevailed, although not so
dry as now ; then at the time of Christ conditions were once more
favorable. After about six centuries of gradually increasing
aridity, a dry period more severe than that of 1200 B.C. reached
its climax about 650 A.D. An improvement followed which cul-
minated about 1000 A.D., then came another bad time, reaching
its worst in the thirteenth century. It was followed by a rapid
recovery, which did not last long enough to be of great value.
Since the fourteenth century minor fluctuations have continued
to take place. The whole matter is summed up in the dotted line
of Figure 46. There the high parts of the curve represent moist
conditions and the low dry. The curve is only approximate, and
does not represent all the data now available, but it seems wise
to reproduce it in the original form which it had when the "pulsa-
tory hypothesis" of climatic changes was first formulated.

From what has been said, it appears that during historic
times climatic pulsations have taken place. They seem to be of
essentially the same nature as glacial epochs and post-glacial
stages, the difference being only in degree. Apparently, the con-
ditions of the geological past merge without break into those of
the present. This fact is rarely appreciated by historians and
archaeologists, who naturally have little occasion to come in
contact with it. Students of glaciation, however, have carried

820 CIVILIZATION AND CLIMATE

the matter so far that we now have abundant evidence of a great
succession of climatic variations covering the entire period from
the present time back to the date when ice covered much of the
northern United States and Europe. They find no evidence of
any sudden break between the past and the present. On the con-
trary, by means of "varves" or layers of clay deposited each
year in lakes that disappeared after the retreat of the ice, De
Geer and Antevs, for example, find a constant series of pulsa-
tions. In studying the retreat of the ice in North America, F. B.
Taylor and others have discovered a series of some fifty small
moraines indicating stages of retreat. These point to climatic
cycles having an average duration of somewhere around five
hundred years, although varying considerably. Most students
of post-glacial climates believe that such cycles, on a diminish-
ing scale, have continued into the historic period.

In order to test the pulsatory hypothesis, some means of
actually measuring the climate of the past seems necessary. In
the southwestern United States there appear to have been
changes like those in western Asia. In that region, Prof. A. E.
Douglass has found that the thickness of the annual rings of
trees furnishes a reliable indication of variations in the water
supply from year to year. In California among the big trees I
found that correlation coefficients for a period of about fifty
years show clearly that the rainfall is a main determinant of
the rate at which the trees form their rings of growth. Other
factors of course enter into the matter, as Antevs has pointed
out in a forthcoming publication of the Carnegie Institution of
Washington. But these likewise are climatic. Corrections must
be made to eliminate the effects of age, but this can be done by
mathematical methods of considerable accuracy. It is difficult
to determine whether the climate at the beginning and end of a
tree's life was the same, but it is easy to determine whether there
have been pulsations while the tree was making its growth. If
the trees from various parts of a given district form thick rings

SHIFTING OF CLIMATIC ZONES

321

for a century, then thin ones for another hundred years, and
again thick ones, we may be almost sure that they have lived
through a long period of unfavorable climate.

During the years 1911 and 1912, under the auspices of the
Carnegie Institution of Washington, I measured the thickness
of the rings of growth on the stumps of about 450 Sequoia
trees which had been cut for fence posts, shingles, and pencil
wood in California. The trees varied from 250 to nearly 3250
years of age. The great majority were over 1000 years old, 79
over 2000, and 8 over 3000. Even where only a few trees are
available, the record indicates the main fluctuations, although

Figure 47. Changes of Climate in California for 2000 Years. Adjusted to
Levels of Owens Lake. Dotted Lines Indicate Main Respects in Which
Antevs' Method of Correction Gives a Different Result from That of
the Author

not the details. Where the number approximates 100, acci-
dental variations are largely eliminated. Accordingly, in Cali-
fornia we have a climatic record which is fairly accurate for
2000 years and approximate for 1000 years more. This is
expressed graphically in the solid line of Figure 46. In order to
bring out the details the more reliable portion from 100 B.C.
to the present time has been reproduced in Figure 47. This
resembles the corresponding part of Figure 46, except that the
vertical scale is three times as great, and corrections have been
made on the basis of Owens Lake.

In general, the tree curve resembles that of changes of cli-
mate in Asia, although there are differences in detail. Beginning

322 CIVILIZATION AND CLIMATE

with 1000 B.C., both curves have a maximum. They dip down
about 800 B.C., and rise high not far from 700. About 400 B.C.
they disagree, but this is probably due largely to the absence of
reliable data for the Asiatic curve. In the second century before
Christ both are low, but not so low as at present; at the time
of Christ they rise high, and continue to fluctuate together till
300 A.D. At that time they show a difference which may have
arisen because I was unduly impressed by the abandonment of
many ruins in Chinese Turkestan at the end of the third century.
The dash line is probably more accurate than the dotted and
should be used. In the seventh century both curves reach their
lowest point. Then, from 650 A.D. onward, their general course
is closely similar, especially about 1000 A.D. In the tree curve
the evidence of pulsations is even clearer than in the Asiatic
curve. The general agreement between the two curves seems to
indicate that the main climatic variations of western Asia and
the region of similar climate in the United States are approxi-
mately the same. Moreover, the fluctuations of both tree growth
and rainfall among the Sierras during the last half century show
a strong resemblance to those of rainfall in Palestine, as is set
forth in Climatic Changes and in Post-glacial Climatic Cycles,
a publication of the Carnegie Institution of Washington. Since
the present growth of trees in California shows so close an agree-
ment with the rainfall in western Asia, it is almost certain that
a similar agreement existed in the past. The growth of the trees,
with its pronounced variations lasting hundreds of years, is
generally accepted as the strongest single piece of evidence that
pronounced climatic pulsations have taken place in historic
times. The agreement between the curve obtained from exact
measurements in California and that obtained from physio-
graphic and archaeological evidence in western Asia unites with
present climatic variations in proving that these two regions of
similar climatic type have experienced a similar series of climatic
pulsations.

SHIFTING OF CLIMATIC ZONES 323

Another line of evidence in the western United States is pecu-
liarly important because it employs a method absolutely differ-
ent from those already mentioned, but reaches the same con-
clusion. Every river holds in solution a certain amount of
sodium, chlorine, calcium, carbon dioxide, and various other
materials. Under ordinary circumstances this cannot be detected
except by chemical analysis. If the river flows into a lake which
has no outlet, however, the water is evaporated, but the dissolved
constituents remain, and gradually increase until a brine is
formed. Certain materials, such as the calcite, which forms tufa
or other kinds of limestone, are removed by algae or bacteria,
and certain others, such as potassium, seem to be absorbed by
the clays of the lake bottoms. Sodium, and especially chlorine,
however, do not appear to be removed until the brine becomes
completely saturated so that crystals of common salt are formed.
Hence, if we know the amount of sodium and chlorine brought in
by the rivers each year and the amount dissolved in an unsatu-
rated lake, we can easily calculate the time that has elapsed
either since the lake was first formed, or since it last overflowed,
A body of water that overflows, as everyone knows, soon becomes
essentially as fresh as the rivers that supply it.

On the eastern side of the Sierra Nevada mountains Owens
Lake, a body of salt water in southern California, is peculiarly
well adapted to our present purpose. Owens Lake receives most
of its water from the river of the same name. Both have been
measured and analyzed with unusual thoroughness because part
of the river is carried to Los Angeles in a remarkable aqueduct
nearly 250 miles long. H. S. Gale of the United States Geological
Survey has carefully gathered all the available data. He con-
cludes that, according to the figures of the aqueduct engineers,
the accumulation of the chlorine of Owens Lake would require
4200 years, and the sodium 3500, the average being 3850. A
series of fresh strands and an old outlet channel show clearly
that the lake overflowed not long ago. Hence, Gale concludes

324 CIVILIZATION AND CLIMATE

that "4000 years or considerably less" is the length of time
since the lake stood at the outlet level, 180 or 190 feet higher
than at present. These figures, however, require modification.
The reason why Gale gives the period as "4000 years or con-
siderably less" is that the figures of the engineers omit the lower
third of the drainage area of Owens Lake, and this is the part
where the waters flow most slowly and where the clays and other
deposits which surround them are most saline. Hence, more salt
proportionately should be accumulated here than higher up.
Moreover, as he carefully points out, no allowance is made for
the well-ascertained fact that when the rivers are more abun-
dantly supplied with water, as must have been the case when
the lake was full, the amount of dissolved salt is also greater
although not in direct ratio to the water supply. When due
allowance is made for these conditions and for others of minor
importance, the time since the last overflow is reduced to 2500
or 2000 years. In other words, at the time of Herodotus and
perhaps at the beginning of the Christian era, the climate of the
Owens Lake region was so moist that the lake expanded to two
and one half times its present size and sent a stream down the
outlet channel.

Owens Lake does more than indicate a change of climate
within two or three thousand years. It also shows that the change
has been highly irregular. This is proved by a large number of
strands lying below the level of the outlets, and by the way in
which these vary in character and in the extent to which they
have been covered by fresh detritus washed down from the moun-
tains. At Owens Lake there are four series of strands. These
apparently correspond to the four chief periods when the cli-
mate has grown moist as shown by the growth of the big trees
in Figure 47. Fortunately, Owens Lake lies only fifty miles east
of the region where the trees were measured. The general cli-
matic fluctuations of both districts are the same. The uppermost
strand, the huge gravel beach at the level of the outlet, was pre-

SHIFTING OF CLIMATIC ZONES 325

suniably last reached by the water about the time of Christ, or
possibly in the days of Herodotus, for both the chemical evi-
dence and the trees point to this conclusion. A series of similar,
but much smaller beaches at lower levels record the approach of
a dry period during which the lake fell to a low level whose exact
position cannot be determined. Judging by the trees this must
have culminated about 650 A.D. During this period gravels were
washed in by mountain streams and formed what are known
as fans, or low, flattened cones, which may be several miles long.
These covered the old strands in many places, and extended far
below their level to the diminished lake.

Next the waters rose again, but not halfway to their former
level. They formed two small strands, not gravelly like their
predecessors, but faint and sandy as if the winds were weak.
They must date from about 1000 A.D., when the trees indicate
a wet period, for they are younger than the gravel fans of the
preceding dry time. The next phase of the lake was a dry period,
which was most extreme about 1250 A.D. More gravels were
then deposited, and the fact that they cover the preceding
strands and extend to a much lower level shows that the lake
then stood low, as would be expected from the trees.

The next high period of the lake, about 1350 A.D. according
to the trees, is unusually interesting. The water did not reach
so high a level as formerly, perhaps because the rainy period was
short, but it formed a large beach of gravel quite different from
the preceding beaches. This seems to indicate great storminess,
a condition which is also suggested by the fact that the growth
of the trees at this time increased more rapidly than at any other
period for nearly 3000 years. In Europe during the same cen-
tury, unprecedented storms caused great floods in France, while
the severity of the waves was so intense as to break through
beaches and sand dunes, and convert large marshy areas into
portions of the sea along the coasts of Holland and Lincoln-
shire. During the winters the rivers froze to an unheard-of

326 CIVILIZATION AND CLIMATE

degree, and three or four times men and animals passed from
Germany to Sweden on the solid ice of the Baltic Sea, an occur-
rence unknown in our day. In England the summers were so
rainy that the average yield of grain diminished disastrously.
In self-defense many landowners gave up grain-raising, and
turned their attention to sheep and cattle. Distress and discon-
tent were the inevitable result among the peasants. Far away in
central Asia the Caspian Sea and the lake of Lop Nor both
rose with great rapidity between 1300 and 1350 A.D. Thus from
California to China evidence of various kinds unites to indicate
that during the fourteenth century there occurred a short
period of unusual storminess. Such conditions, if intensified and
prolonged, would probably cause the accumulation of enormous
glaciers.

To return to Owens Lake, the lowest series of strands is sandy
and small compared with the large gravel bar of 1350, and was
evidently formed under different conditions. Presumably the
lake fell to a low level about 1500 A.D., and rose during the next
century or more to form the highest strand of the latest series.
The evidence of Owens Lake is much strengthened by that of its
neighbors. At Mono Lake farther north in California, the more
recent strands are almost identical with those at Owens. Since
1900 Mono Lake appears to have risen higher than at any time
since at least 1775, as we judge from the rings of growth of a
tree killed by the rising salt water. Thus on all sides there is
the strongest evidence not only that the climate of the past
differed from that of the present, but that many minor pulsa-
tions have occurred, and have grown less intense during the
historic period.

Up to about 1912 I supposed, as did practically all students
of the subject, that the same kind of climatic changes have
taken place in all parts of the world. At that time Penck, a
leading student of the glacial period, came to the conclusion
that this is by no means the case. He showed that on the north-

SHIFTING OF CLIMATIC ZONES 327

ern side of arid or desert areas we find densely saline lakes, like
Great Salt Lake in Utah, surrounded by old strands. These
indicate that the lakes have long been contracting, so that they
have abandoned first one strand and then another, while at the
same time their water has become more and more highly concen-
trated. On the equatorial side of the desert belt, on the contrary,
we have such lakes as Chad in Africa, a shallow sheet of water,
only slightly salty and not surrounded by any great series of
strands. It has the appearance of being a new lake formed by a
recent increase in rainfall. Penck points out a similar contrast
between the sand dunes on the poleward and equatorward sides
of the desert belt in both hemispheres. On the poleward side, the
dunes consist of loose, moving sand appropriate to places that
are growing drier. On the other side there are plenty of dunes,
but they are covered with a sparse vegetation which is sufficient
to keep them from moving and to prevent the formation of new
dunes. Since their formation, the rainfall has evidently in-
creased. From the dunes and lakes, as well as from other evi-
dence, such as the snowline, Penck concludes that changes of
climate consist of an alternate shifting of the climatic zones.
During a glacial period he holds that the northern storm belt is
shifted southward so that the storminess of Germany and the
northern United States is pushed into Italy or the southern
United States. In the same way the desert belt is displaced
toward the equator. Thus the polar side of the desert has
more storminess and moisture than formerly, while on the equa-
torial side the desert is shoved into an area where equatorial
rains formerly supported abundant vegetation. During a time
such as the present, on the contrary, the desert expands on its
northern border. Its lakes diminish, leaving strands behind them
and becoming very saline ; vegetation dies ; and the wind is free
to pile up sand dunes. On the other, or equatorial side of the
desert, the amount of rain is greater than before. Hence, basins
which formerly contained no water are now filled with shallow

828 CIVILIZATION AND CLIMATE

lakes, such as Chad, which have not yet had time to become
highly saline. Vegetation also spreads into the desert and sand
dunes become covered with it and cease to be moved by the wind.
Few people have studied glacial problems more carefully than
Penck. One result of the minuteness and care of his studies is
that he is one of the foremost advocates of great climatic com-
plexity. He distinguishes several post-glacial stages, and in
addition brings his conclusions so far toward our own time that
he states that a peculiarly dry period prevailed in central Asia
as late as the early part of our era.

While Penck was formulating his hypothesis of a shifting
of climatic zones as an explanation of the glacial period, I was
working out the same hypothesis in respect to historic times.
Our work was wholly independent, being based on different lines
of evidence, and, so far as I am aware, neither knew what the
other was doing. The shifting of climatic zones is by no means
a new idea, for it has been vaguely suggested many times. The
new thing is to find direct evidence of it, such as Penck presents
in his discussion of lakes and dunes. As to historic times, the
ruins of Guatemala and Yucatan furnish perhaps the best avail-
able test, for those regions are the most notable example of a
civilization which developed within the torrid zone. The ruins of
that civilization are the most remarkable archaeological remains
in the western hemisphere. Part lie near the northern coast of
Yucatan in a relatively dry region inhabited today by a fairly
prosperous agricultural population. Many of the finest ruins,
however, and most of the more ancient ones, lie back from the
coast in a wilderness of dense forest and jungle. There agri-
culture is almost impossible. The difficulty of clearing the rank
vegetation and getting it dry enough to burn before a new crop
of lusty bushes grows up is enormous. Fevers, too, prevail
most of the year. They include the worst types of tropical
malaria, as well as many other kinds. Foreigners are quickly
attacked, and it would be dangerous for a white man to attempt

SHIFTING OF CLIMATIC ZONES 329

to live there permanently, even with all the appliances of modern
medical science. The natives also suffer terribly. Many of the
children are apparently killed by malaria and other tropical dis-
eases in infancy. Those who grow up carry the effects with them
through life. Seventy-five per cent of the tropical workers at
the Panama Canal are thought to have had malaria germs in
their systems, even though they did not show outward signs of
the disease, and the case is probably even worse among the
natives of the lowland forests of Quintana Roo, Campeche, and
Peten, where most of the early Maya ruins are located. One has
only to look at their swollen paunches to see that something is
the matter ; and their dull, apathetic manner, both in work and
play, is another sign of some deep-seated physical ailment.

Here, where today the Indians are so diseased and agricul-
ture is so difficult, there dwelt an ancient race characterized by
the qualities which we have defined as most essential to a high
civilization. The Mayas of Yucatan and Guatemala, alone
among the aborigines of America, carried to high perfection
the arts of sculpture and architecture. The Incas of Peru, to
be sure, made striking buildings, but they had no idea how to
adorn their simple structures with columns, rosettes, arches,
gargoyles, pediments, and many other architectural devices
which were universal among the Mayas. These clever people
were also adepts in astronomy, for they framed the most exact
calendar ever known except our own. Theirs was better than
that which eastern Europe used up to the time of the Great War.
Most remarkable of all, the Mayas developed the art of writing
and carried it to a pitch higher than that reached by the Chinese,
for they apparently began to use signs to denote sounds instead
of having a sign for each individual word. Their temples are
great structures which sometimes rise three stories in height and
have a length of three or four hundred feet. A single city often
contained a score of noble buildings and extended over many
square miles of territory. The Mayas need not shrink from com-

330 CIVILIZATION AND CLIMATE

parison with any people who had no greater opportunities. They
were terribly hampered in many ways. They were not blessed
with progressive neighbors to spur them on and teach them new
ways. They had no beasts of burden. They could not plough the
fields, nor transport loads except by their own labor. Every one
of the great stones which form their temples and palaces must
have been brought by human effort, a task which none but a
most energetic race would undertake. Another disadvantage was
the absence of metal tools. The Mayas perhaps had a little
copper, but not a trace has been found of any metal tool that
would be of service in carving their intricate f a9ades and deli-
cately wrought statues. Flint or obsidian was all that they had,
and yet with such poor tools they created works of art which
command sincere admiration. To do all this with such small
opportunities was surely one of the greatest feats ever achieved
by any race.

The most surprising thing about the Mayas is that they
developed their high civilization in what are now the hot, damp,
malarial lowlands where agriculture is practically impossible.
A hundred miles away on the coasts of Yucatan or in the Guate-
malan highlands far more favorable conditions now prevail.
There agriculture is comparatively easy, the climate, while not
bracing, is at least good for the torrid zone, and malarial fevers
are rare. Today the main cities lie in these more favorable re-
gions ; the energetic part of the population is there, and the in-
terior lowlands are hated and shunned by all except a degraded
handful. In the past the more favorable localities were occupied
by people close akin to the Mayas, yet civilization never rose to
any great height. Ruins are found there, but they are as far
behind those of the lowlands as the cities of Yucatan are today
behind those of the United States.

In explanation of these peculiar conditions several possi-
bilities suggest themselves. First, we may suppose that the
Mayas were the most remarkable people who ever lived. They

SHIFTING OF CLIMATIC ZONES 331

were able to carry on agriculture under conditions with which
no modern people, not even those of European race, have ever
succeeded in coping. They chose the worst place they could
find, almost the worst in any part of America, even though far
better places lay close at hand and were occupied by an allied
people relatively few in numbers and backward in civilization.
They were able to preserve their energy for a thousand years or
more under the most debilitating climatic conditions ; and, lastly,
they were immune to the many fevers which today weaken the
dwellers in their old habitat. It is possible that a process of
natural selection had given these people an extraordinary degree
of genius, energy, and ability to combat disease. Indeed, it seems
to me that this must have been the case. Nevertheless, I doubt
whether it fully explains why the Mayas settled and thrived in
what now would generally be accounted the worst parts of their
general region. A second possibility is that in the time of the
Mayas tropical diseases were less harmful than at present. We
have no shred of evidence one way or the other. Such a thing is
possible, but in view of the fact that there are several kinds of
malaria, not to mention other diseases, any one of which greatly
weakens a race, there is little chance that the Mayas were free
from disease unless something else was also different. Even if
diseases were not so prevalent as now, this does not explain the
other apparent anomalies of the Maya situation. It may be,
however, that a fairly satisfactory explanation will be found if
the two preceding possibilities are joined with a third, namely,
a climatic change such that the dry conditions which prevail a
little farther north prevailed in the Maya region when these
people attained eminence. Such a shifting of zones would increase
the length of the dry season which now comes in February and
March. This would diminish the amount of vegetation and cause
scrub to take the place of dense forest. Under such conditions
agriculture would become comparatively easy. Fevers would
also greatly diminish, for in the drier parts of Yucatan they

332 CIVILIZATION AND CLIMATE

are today relatively mild, and the lowland plain would be the
natural site of the chief development of civilization just as is
the case in other countries.

When the dates of Maya history are compared with the curve
of tree growth in California, they seem to agree with the hy-
pothesis of a shifting of climatic zones. The early chronology
is so doubtful that we may pass it by. The seventh century,
however, is known to have been a time when Maya civilization
sank to a very low ebb, for scarcely a building dates from that
time, and the traditions become most vague. The great fact of
this period is that for some reason the Mayas completely aban-
doned their old homes in what is now the forested south of
Mayaland. They migrated north to the drier, less forested, and
more healthful region of northern Yucatan. This would be what
we should expect, for when the California trees grew slowly, as
in the seventh century, the desert zone that almost girdles the
globe would lie far to the north. At the same time the equatorial
zone of rains would expand northward over Guatemala and
Yucatan; the rainfall would be abundant and the dry season
short ; the forests would become rank, agriculture would be
difficult; disease would be rife; and the vitality of the Mayas
would be sapped. From about 900 A.D. to 1100 A.D., on the other
hand, the California trees grew rapidly. At such a time the desert
belt would be pushed south, and favorable conditions would pre-
vail in the home of the Mayas. At that time occurred the last
great revival of architecture and the construction of the great
buildings whose ruins now adorn Yucatan. Whatever the cause,
there is no doubt that there was a marked outburst of energy.
The later history of the Mayas has no great abundance of archi-
tectural monuments to serve as landmarks, and hence cannot be
correlated with the fluctuations of the California curve. Yet,
so far as the indications enable us to judge, the two sets of
phenomena are in harmony.

In addition to the Maya ruins, the conditions of Palestine

SHIFTING OF CLIMATIC ZONES 383

and the growth of the California trees furnish independent evi-
dence that the zone of cyclonic storms has at certain periods
suffered a shift equatorward. In Palestine, the rainy zone ap-
parently once extended at least fifty miles south of its present
limit, in spite of the fact that the mountains there die out and
thus render the conditions doubly unfavorable to rainfall. Today
Hebron is the last large town and Beersheba the last village. In
the old days, Aujeh, far to the south, was comparable to Hebron,
and other ruins indicate that settlements were located still
farther southward. The California trees indicate a shifting of
zones, because they grow most rapidly during years when storms
continue late into the spring. At such times the winter zone of
storms is pushed equatorward and continues to give moisture
late in the spring.

Further consideration of changes of climate is not here pos-
sible. For that the reader must turn to Climatic Changes and
Earth and Sun. Here our main interest centers in the results of
such changes, and in the degree of certainty with which we can
make use of them in the interpretation of history. For our pres-
ent purpose the essential point is that a series of climatic pulsa-
tions has apparently continued from a maximum phase in the
last glacial epoch down to a minimum phase today. Nowhere do
we find any marked break between the past and the present. On
the whole, the early milleniums of the historic period appear to
have been drier than the present in such regions as Syria and
California, but this general condition of more abundant rain
appears to have been broken by frequent dry periods which
may have lasted hundreds of years. Some of these, as in the
seventh century of the Christian era, may have been even drier
than the present. Moreover, the effect of any given phase of a
climatic pulsation is not the same in all parts of the world. On
the contrary, the effects differ from zone to zone, from the coast
of a continent to the interior, and from the east side to the west.
Except in magnitude the climatic cycles of the past appear to

334 CIVILIZATION AND CLIMATE

have been like the small cycles of the present. California, for
example, tends to be rainy when Florida is dry.

It must be clearly understood that the chief characteristic of
the climatic pulsations of the historic period has not been
changes of temperature. Failure to understand this point has
led to many mistakes. The chief variations have apparently oc-
curred in the number, intensity, and paths of the kind of
cyclonic storms which bring ordinary changes of weather in
the progressive regions of the United States and Europe. Such
storms are by far the most variable climatic factor. The extent
to which the mean annual temperature or relative humidity of a
region may vary from one year to another is very limited, but
the rainfall, and still more the number of storm tracks whose
centers pass near a given place, may vary greatly. Some years
storm after storm sweeps across the country at intervals of
only a few days, while in others a real storm occurs scarcely
more than once a month. Even in a place like New York City
where the climate is very regular, the number of storm tracks
whose centers passed within 200 miles of the city in the short
period from 1900 to 1915 varied from 21 in 1903 to 48 in 1915,
while from July to December the variation was from 7 to 24. In
regions having the type of climate now found in the eastern
Mediterranean region and in California, such variations are
far greater. At San Francisco, for example, the number of storm
centers passing within 200 miles of the city during this same
period varied from one in 1900, 1904, and 1914, or none in the
twelve months from December, 1913, to November, 1914, to
13 in 1903. In the past, according to our hypothesis, the number
of storm tracks for century after century approached this
higher figure, while at other periods it fell toward the lower
figures. In other latitudes and in parts of the continents where
other types of climate prevailed, an opposite type of change took
place, while in intermediate regions no distinct alternation was
evident.

CHAPTER XV
THE PULSATORY HYPOTHESIS AND ITS CRITICS

CHANGES of climate are so vital an element in the main
conclusions of this book, that I shall devote a chapter to
criticisms of climatic hypotheses. I shall begin by summarizing
once more the changes in my own opinions, for these changes
have had much to do with the nature of the criticisms.

1. When I first became convinced that climatic changes had
taken place in historic times in western Asia I accepted without
question an opinion then somewhat widely held that the earth
as a whole was growing drier and warmer.

2. Further study, as already explained, soon led me to aban-
don this idea of progressive desiccation in favor of the hypothe-
sis of pulsations whereby the climate first became warmer and
drier, and then moister and cooler.

3. The next step was the conclusion that variations in tem-
perature are of negligible importance, and that the main change
has been in the amount of rainfall.

4. Investigations in regions other than deserts advanced the
hypothesis another step by leading to the conclusion that
changes of opposite types have taken place in different latitudes,
and even on opposite sides of a continent in the same latitude.

5. The last step was the conviction that the evidence points
clearly to changes in the location, number, and intensity of
cyclonic storms as the main factor in climatic pulsations.

Bearing in mind, then, that the hypothesis of climatic pulsa-
tions as presented in this book has become quite different from
what it was when first published in 1905, let us consider some
of the main criticisms.

836 CIVILIZATION AND CLIMATE

1. Many authors have endeavored to disprove all theories of
climatic change during historic times by suggesting that the
supposed evidences of moisture in the past are purely accidental.
In one place, so they say, a river may have changed its course ;
in another an earthquake may have diverted an underground
water supply ; famous marches like that of Alexander with his
elephants in Persia may have been possible not because the cli-
mate was moister than that of today, but because the march
occurred in an unusually wet year. In the same way the thou-
sands of seemingly waterless ruins in the southwestern United
States may all have depended upon hidden springs, one or two
of which have been found. Such arguments are of value merely
as guides in determining the kind of investigations that shall
be made in the field. They break down when it is realized that
although many conscientious investigators have had them in
mind, the evidence that would give them weight is not forth-
coming. It must be remembered that the evidence on which the
hypothesis of climatic pulsations is founded is of many types ;
it is widely spread in at least four continents ; and it depends
upon literally thousands of ruined towns, waterless roads, aban-
doned irrigation systems, old unirrigated fields now too dry for
cultivation, old records of lake levels and the like. It is statisti-
cally impossible that so many accidents should all point in the
same direction. Accidents by their very nature tend some-
times in one direction and sometimes in another. The evidence
of climatic changes, however, is not only widespread, but ex-
tremely systematic. At certain periods and in certain types
of climate it all, so far as I can see, tends to indicate aridity.
In other types of climate and at other periods it seems to point
with equal unanimity in the other direction. If the evidence
were merely the result of accidents, such a separation on the
basis of both time and place would be impossible.

2. A number of persons have criticized the pulsatory hy-
pothesis under a misapprehension, or have directed their criti-

PULSATORY HYPOTHESIS AND ITS CRITICS 337

cisms at earlier statements of the hypothesis without knowing of
later modifications. A good instance of this is Woeikof's book
on Turkestan. Apparently that Russian author had not read
or had failed to understand The Pulse of Asia, but had become
thoroughly familiar with a book written by his countryman,
Berg, for the express purpose of combating what he believed to
be the false conclusions of The Pulse of Asia. At any rate,
Woeikof pronounced the conclusions of that book "inane," and
then proceeds to prove the very thing that led to the choice of
its title, namely, that during historical times the Caspian Sea
and other lakes in central Asia indicate alternating periods of
more and of less rainfall than at present. Herbette, a French
critic, also aims his main arguments at progressive climatic
changes, but agrees that the Caspian Sea affords unmistakable
evidence of climatic pulsations. In general, the critics have
tended to attack not only the idea of a progressive climatic
change in one direction during historic times, but the idea of
climatic uniformity. The alternative to which almost all authori-
ties turn is climatic pulsations having a greater intensity in the
past than at present. Over nine tenths of the geographers of
America, if we may judge from the fifty whose opinions have
been expressed in writing, hold this view, although they differ
as to the degree to which climatic pulsations have influenced
history.

3. A certain type of critic has asserted again and again that
in the southwestern United States, and to a less degree in other
places, the population in regions that are now too dry to sup-
port many people was never so dense as the ruins seem to indi-
cate. What really happened, they say, is that the same people
built village after village and house after house. Superstitious
fears of death or disease, or a tendency to wander may have
caused a house to be inhabited for a few decades and then aban-
doned. The old house never was reoccupied, but a new one was
built by its side. Thus ruins which appear to be large enough

838 CIVILIZATION AND CLIMATE

for a thousand people may never have been occupied by more
than fifty or one hundred at any one time. Wherever the history
of a region is well known this explanation is wholly out of the
question, as in Syria, the Libyan desert, Persia, and many other
parts of the Old World. It has flourished in America simply
because the history of pre-Columbian America is unwritten, and
archaeologists have been free to exercise their imaginations.
Douglass, however, who perfected the methods of measuring
climate by means of tree growth, has most effectively proved
that in at least one case a relatively large set of ruins was all
built at practically the same time. By comparing the rings of
different trees he is able to tell just which ring in one tree is
synchronous with a given ring in another. This is possible be-
cause the rings vary greatly from year to year, and the same
succession of rings of various thicknesses is practically never re-
peated. By this method Douglass finds that in one of the large
ruins at Chaco Canyon, New Mexico, the houses were all con-
structed within about a decade. Thus the whole village must have
been occupied at one time and there was no successive addition of
new rooms to replace those no longer occupied. Of course this is
only one instance, but a single concrete, well-established fact like
this is worth more than any amount of supposition.

4. In many quarters there has been an assumption that cli-
matic uniformity is a normal condition. Meteorologists do in-
deed find that so far as records are yet available, which means
for scarcely a century, there are no certain indications of
progressive climatic changes. But there is overwhelming evidence
of pulsations which increase in magnitude and duration as the
records become longer. In the same way the geological record
continually discloses new evidence of climatic pulsations. Where
one glacial period consisting of a single epoch was generally
recognized half a century ago, at least six are now recognized.
These date from all parts of geological time, and some have
several epochs and subepochs. Moreover, the studies of Bar-

PULSATORY HYPOTHESIS AND ITS CRITICS 339

rell, De Geer, Sayles, and many others are rapidly showing that
during practically the whole of the earth's known history smaller
climatic variations have occurred. Thus pulsations of climate
are now almost universally regarded as characteristic of the
vast periods covered by geology as well as of the short period
covered by exact climatic records. The presumption is that
similar pulsations of intermediate magnitude were characteristic
of the intervening historical period.

One of the most interesting facts in this connection is the
change of attitude among meteorologists. For example, Ward
of Harvard, who was for many years the only professor of
climatology in an American university, has long been one of the
strongest champions of climatic uniformity. In his later works,
however, and especially in his forthcoming Climate of the United
States, his attitude toward the hypothesis of pulsatory climatic
changes has altered from pronounced opposition to benevolent
neutrality. His position is that the kind of physiographic and
archaeological evidence which must be used in solving the problem
of climatic changes during the period prior to exact records is
utterly different from the statistical data to which climatolo-
gists and meteorologists are accustomed. Therefore specialists
of those types cannot properly evaluate it, but must wait for the
final decision of specialists in this particular line. As a matter
of fact, the great majority of specialists in post-glacial clima-
tology now accept the idea that climatic pulsations have con-
tinued into historic times.

5. The most important criticism of my climatic hypotheses
that has yet appeared was written by J. W. Gregory,* a British
geographer who has traveled very widely. He makes two im-
portant points, but both pertain to matters where I had already
come to the selfsame conclusion which he advocates. His first
point is that various types of vegetation, such as the palm tree,
grew in places like Palestine in the past and grow there today.

* "Is the Earth Drying Up?" Geographical Journal, vol. 43, 1914.

340 CIVILIZATION AND CLIMATE

This, he says, precludes a change of climate, for if Palestine
were colder two thousand years ago than now, the palm could
not have flourished. The truth of this statement is undeniable
if we are talking about pronounced changes in the mean tem-
perature. As I have already shown, however, the evidence points
toward changes of precipitation much more than of tempera-
ture. The best students of glaciation agree that during the
glacial period the mean temperature of the earth as a whole
was probably not more than 10 or 15 F. cooler than at pres-
ent. If the climate at the time of Christ differed from that of
the present time by one tenth of the difference between our
modern climate and that of the glacial period, the temperature
of Palestine would only have to be about 1 F. cooler than now.
That would scarcely produce any appreciable effect upon vegeta-
tion. The actual figures show that Palestine might be at least 2
or 3 F. colder than now without preventing the growth of the
palm tree. Moreover, the extremes may have been greater than
at present without much change in the average temperature. If
the winter storms were colder than now by a few degrees, that
would not hurt the palm tree. In Seistan in eastern Persia I
have camped in six or eight inches of snow almost under the
shade of a palm grove whose dates, packed in a sheepskin, were
uncommonly delicious. The thermometer fell below 20 F., and
a strong north wind overturned the tent, but the palm trees did
not suffer, for in two or three days the air was again as balmy
as spring. So far as has yet been shown, the conditions of vege-
tation nowhere seem to be out of harmony with our hypothesis.
Gregory's other point, like the one already discussed, agrees
completely with a conclusion to which I had been led inde-
pendently. He shows that so far as there is evidence of climatic
changes, it points in one direction in some regions and in the op-
posite direction elsewhere. He even published a map showing the
distribution of positive and negative changes. Inasmuch as this
conclusion agrees with that to which both Penck and myself

PULSATORY HYPOTHESIS AND ITS CRITICS 341

had been led on entirely different evidence at essentially the
same time, I see no reason to question it. In a word, the article
by Gregory, which some historians have regarded as demon-
strating that there have been no changes of climate, goes far
toward substantiating the hypothesis of this book, for it helps
to clear up some of the main difficulties. It does not touch at all
upon the problem of pulsations. Like practically every other
important criticism of my conclusions it was published while
the pulsatory hypothesis was still being developed and before
the most convincing evidence, that of the Big Trees, had been
published in The Climatic Factor. Since the publication of that
book, so far as I am aware, there has been no serious quantita-
tive attempt to disapprove the general hypothesis of climatic
pulsations during historic times.

6. The contrast in the types of evidence which are conclusive
to workers in diverse lines of research is illustrated not only
by the fact that meteorologists are slow to accept the evidence
of historic changes of climate, while geologists are quite ready
to accept it, but by the fact that historians seem to think that
if changes of climate have occurred there ought to be written
records to this effect. This may be illustrated by two eminent
geographers who follow the historical method. Brunhes and
Seniple both reject the idea that climatic changes have had an
important effect upon history, but neither gives reasons for
such rejection. Miss Semple, for example., in an admirable paper
on the Mediterranean lumber trade merely states that, having
studied the evidence, she does nob think it convincing. Her work,
with its remarkably complete references even when minor matters
are discussed, shows that to her the real evidence on this point
is the written word. With her usual fairness, however, she points
out that the accounts of widespread forests in ancient Cyprus
suggest a different climate. She also gives many other facts
which seem much more consistent with climatic changes than
with uniformity. At least, they seem so to persons with a geo-

342 CIVILIZATION AND CLIMATE

logical training. Such persons have become more and more accus-
tomed not only to the idea of repeated climatic fluctuations of
all grades throughout geological times, but to the idea that
such fluctuations have caused the destruction of many types of
plants and animals, and have hastened the development and mi-
gration of new types. The historian, on the contrary, is accus-
tomed to attribute almost everything to purely human action.
Hence, in the absence of records to the contrary, he assumes
that human causes are mainly responsible for such facts as de-
forestation, changes in trade routes, the apparent drying up
of sources of water, the abandonment of settlements in the drier
parts of the world, and the migrations which so largely originate
in dry regions.

7. Closely connected with the preceding attitude of mind is
a reason for the rejection of the hypothesis of climatic changes
which has been advanced by some historians. They say that
such changes are not needed in order to explain the historic
facts. But this has nothing to do with the matter. The primary
question is whether climatic changes have taken place during
historic times. If they have taken place, the historian must per-
force take note of them and inquire into their effect, just as he
inquires into the effect of barriers like the Alps or of great
men like Socrates. This wholesome attitude is beginning to
prevail in many quarters and will doubtless grow.

A good example of the way in which a knowledge of climatic
changes throws light on history is set forth by A. T. Clay in an
article on "The So-called Fertile Crescent and Desert Bay."*
From a study of Babylonian tablets Clay discovered that a
Semitic people known as the Amaru or Amorites had already
risen to a position of relatively high civilization in the fourth
millennium B.C. Their capital was at Mari on the Euphrates
River not far from 37 N. in the very heart of what the oriental
historian Breasted has described as a "desert bay." The tablets

* Journal American Oriental Society, 1924.

PULSATORY HYPOTHESIS AND ITS CRITICS 348

seemed to show that this capital was powerful enough to rule
Babylonia, but Breasted maintained that this was impossible
because there was no sufficient "material basis" for an empire
such as that of the Amorites. The arable land along the Eu-
phrates is too limited to support an empire, and the surround-
ing land is desert. Nevertheless Legrain of the University of
Pennsylvania found a fragment of a dynastic list which showed
beyond question that the Amorite city of Mari on the Euphrates
ruled Babylonia in the fourth millennium B.C. Thus the historic
facts and the present condition of climate are completely out of
harmony. When Clay investigated the matter on the spot, he
found about seven hundred square miles of arable land within
the alluvial tract which might be watered from the Euphrates.
Much of this, however, could not now be irrigated under the
present conditions of floods and erosion. Even if it could all be
irrigated it is scattered in so many small bits that it would form
a very inadequate basis for an empire which could rule Baby-
lonia. The region round about, however, is full of evidence that
the country was by no means a desert in the past. As Clay puts
it : "To include the Euphrates valley in the so-called 'desert
bay' would be equivalent to including the Nile Valley in the
Sahara desert, the difference, however, being that unlike the
Sahara, Mesopotamia was not a desert in ancient times.

"It is not improbable, since we know that the climate has
changed within the past two thousand years, that practically
the entire area of many thousand square miles of Mesopotamia
north of the [Euphrates] river was fertile, or at least was a
great pastoral territory.

"The Khabur and the Balikh rivers, which flowed through
this region southward to the Euphrates, were augmented by
numerous streams, some of these at present containing water,
while others are dry. Today the Balikh river at the end of
summer is almost dry at its mouth. In commenting upon the
Khabur and its tributaries an Arab writer says it is 'such as

344 CIVILIZATION AND CLIMATE

is not to be found in all the land of the Moslems for there are
more than three hundred pure running fountains.' Certainly
the Hebrew writers and redactors of the old Testament [would
not have made] themselves ridiculous in the eyes of their in-
telligent contemporaries by placing the Garden of Eden in this
region, if it were a 'desert bay.'

"And what is true of the region north of the Euphrates, is
true of the land lying to the west of the river. There are dry
beds of rivers and streams with sand and pebble bottoms in
which, at present, water is not seen from one end of the year
to the other; even many of these streams are spanned by well-
constructed bridges. There are well-heads, spring houses, in
which water formerly gushed from the earth, some even con-
taining inscriptions; but where neither wells nor springs exist
today. Even in the flat and once fertile plateau, which was
thickly inhabited, there are no signs of irrigation having been
practiced, showing that there was once sufficient rain to make
the country habitable.

"There are reasons for believing that forests existed in cer-
tain regions, where today the tree and the vine could not secure
a footing, for the hills are denuded of their soil. The land which
the Greeks and the Romans found so profitable to develop, is
now largely a waste ; and it is difficult to appreciate, from what
we see at present, what certain ancient writers tell us about the
land ; for example Cicero, who said that 'the country is so rich
and so productive that in the fertility of its soil, and in the
variety of its fruits, and in the vastness of its pasture lands, and
in the multitude of all things which are matters of exploitation
it is greatly superior to all other countries.' " (Manilian Law
VI.)

A little to the west of this upper Mesopotamian kingdom of
Amaru, Butler points out that a pronounced change of climate
is necessary in order to explain the history of Syria where an
area of 20,000 square miles which is now practically "desert

PULSATORY HYPOTHESIS AND ITS CRITICS 345

and deserted was more thickly populated than any area of
similar dimensions in England or the United States is today if
one excludes the immediate vicinity of the large modern cities.
It has also been discovered that an enormous desert tract lying
to the east of Palestine, stretching eastward and southward into
the country which we know as Arabia, was also a densely popu-
lated country. How far these settled regions extended in antiq-
uity is still unknown, but the most distant explorations in these
directions have failed to reach the end of ruins and other signs
of former occupation."

It would be easy to cite many other instances where a knowl-
edge of climatic changes helps greatly in understanding history.
In World Power and Evolution I have shown this in relation to
Rome, while The Character of Races gives similar illustrations
for China, Iceland, and other regions. But regardless of whether
the historian feels the need of changes of climate to explain the
historic record, no scientifically-minded historian can fail to
take account of such changes if the specialists in such matters
agree that they have taken place. The general attitude of such
specialists, even in 1916, and still more at present, seems to be
well summed up in the words of D. W. Johnson, professor of
physiography in Columbia University. "The Climatic Factor,"
he says in a review of that book, "does not solve all the problems
presented by a theory of climatic pulsations. It does not pretend
to do so. It does aim to show that the evidence thus far available
strongly supports the belief that there have been within historic
time climatic changes of a pulsatory nature. In this the author
has, in the opinion of the reviewer, been successful."

One of the latest expressions of opinion is that of Aiitevs in
Post-glacial Climatic Cycles (1924). "The view of distinct cli-
matic fluctuations, advanced especially by Huntington in a
number of papers, is being more and more accepted. The very
best evidence of the occurrence of climatic fluctuations of short
as well as long duration, up to thousands of years, is probably

346 CIVILIZATION AND CLIMATE

to be found in the varying rate of growth of the big trees and in
the periodic retreat and halt of the edge of the Pleistocene ice-
sheets. Huntington shows that there have been during historic
times, within limited regions, climatic fluctuations of different
length and character, but particularly in precipitation. The
most noteworthy fact is that, especially in western and central
Asia, the climate 2000 or 3000 years ago appears to have been
distinctly moister than today."

CHAPTER XVI
THE SHIFTING CENTERS OF CIVILIZATION

THE final stage of our investigation is now before us. We
have compared the present distribution of civilization and
of climatic energy, and have seen their remarkable similarity.
We have found that during historic times a pulsatory shifting
of climatic zones appears to have taken place. Let us now
examine the shiftings of civilization, and determine how far they
are in harmony with those of climatic energy.

The black area of the map of energy, Figure 43, embraces
about 1,100,000 square miles, or 2 per cent of the total land
surface of the earth. Its population is about 260,000,000 or 17
per cent of the world's total. Consider what would have hap-
pened if all except this small 2 per cent of the lands had ceased
to exist seventeen or eighteen centuries ago. Civilization would
probably have been altered only slightly, although history
would have been very different. If the rest of Europe had ceased
to exist, Byzantine art, the works of the early Christian Fathers,
Moorish architecture, and Spanish explorations, conquests, and
colonization would be lacking. Yet this would not seriously
change the conditions under which we now live.

Suppose that not only the more backward European coun-
tries, but all the rest of the world except the very high Euro-
pean area had been blotted out in 200 A.D. The influence of
Greece and Rome would have persisted in Italy. The Roman
Church, the Renaissance, and the Reformation would all have
played their part. The absence of new lands for discovery in
the Middle Ages would doubtless have retarded the development

848 CIVILIZATION AND CLIMATE

of certain ideas. Otherwise the status of civilization would have
been changed only a little. The United States is the only non-
European country which has succeeded in touching the daily
life of the world as a whole. American inventions have gone
everywhere. The denizen of Turkestan who discusses news re-
ceived by telegraph and whose wife makes his clothes with an
American sewing machine is being genuinely touched by the
New World. Yet America must not boast too loudly, for without
her aid, Britain, France, Germany, Italy, and their small neigh-
bors would have made most of the world's great inventions. To
an even greater degree they would have been able to develop the
arts and sciences.

The energetic area of western and central Europe has been the
great center of civilization for a thousand years. From it have
gone forth more new ideas than from all the rest of the world
combined. Its manufactures have flooded all parts of the earth.
Politically it dominates about 60 per cent of the earth's surface.
Socially its domination is still greater. The United States has
long been chagrined because South America has looked to
Europe not only commercially but for her inspiration in litera-
ture, art, education, and almost every other phase of activity.
Politically the United States helped greatly in inspiring the
South American republics, and was at the front in stirring
Japan to the new life of the last half -century. Yet those coun-
tries have turned their faces Europeward, because that is where
ideas have been most numerous and have been most fully em-
bodied in the concrete form of institutions, inventions, or scien-
tific principles.

Wherever one turns, he feels the tentacles of the great Euro-
pean center of civilization reaching out and vivifying the life of
the whole world. To tell why this is so would require a volume.
The position of western Europe, the strong racial inheritance
of its people, their legacy from the civilizations of the past,
their inspiring religion, their political freedom, and their many

SHIFTING CENTERS OF CIVILIZATION 349

powerful institutions, all play an essential part. With these,
however, and as one of the conditions which make them possible,
stands the fact that western and central Europe is the only
large region which for many centuries has had these advantages
and at the same time has enjoyed a highly stimulating climate.

The main American high area differs somewhat in shape on
the maps of civilization and climate, but not essentially. In
general, it occupies the portion of the United States lying
north of about latitude 38 and east of the Rocky Moun-
tains, together with the adjacent strip of southern Canada. Its
area, taking the average of the black portions of the two maps,
is approximately the same as that of the similar European
region, that is, about 1,100,000 square miles. The population,
however, is only 70 or 80 million. Here, then, we have another 2
per cent of the earth's land surface, containing between 4 and 5
per cent of the earth's total population. Consider how dispro-
portionately great an influence it exercises. To be sure, it falls
far behind Europe, but that is partly because it is newer and
less densely populated. Yet it differs from Europe only in the
degree, not the kind of influence. We have already referred to
some of its inventions. All the world recognizes that the tele-
phone and telegraph and many other forms of applying electri-
cal energy are primarily American. In architecture we have
evolved such wholly new types as the sky scraper. In art and
literature we follow far in the wake of Europe. Yet we have
done some things which are widely known in other countries,
and which are steadily molding the world's opinion. The same
is true in science, for the day is past when Europeans can ignore
the work done on the western side of the Atlantic. In the future
the disparity between Europe and America along this line bids
fair to decrease rapidly, for great institutions devoted to pure
science are growing on a scale unknown in other lands.

This is in part the result of another line of effort in which
Americans have achieved unusual success. By this I mean what is

350 CIVILIZATION AND CLIMATE

sometimes called "big business," the amalgamation of many
minor industries or branches of an industry into one huge
whole. It is illustrated in such organizations as our great rail-
roads, and the United States Steel and Standard Oil companies.
These represent in part the effect of unusual opportunities, but
they also indicate a great expenditure of energy in planning
and directing such widely ramifying activities. Lastly, and
perhaps most important of all, the United States has touched
the world upon the political and social side. In spite of political
sins, which are many, this country has stood for freedom and
justice. Its example has inspired other nations, including parts
of western Europe, and has given them a stronger desire for
equality of opportunity and for the breaking down of unfair
discriminations between man and man. In more backward places
a similar influence is felt. Good judges say that in Turkey, for
example, one of the most potent causes of the revolution of 1908
and the establishment of representative government was the
teaching of American missionaries. The same is true in China,
except that there the Americans have been associated with mis-
sionaries of many other nations, while in Turkey they have
been almost alone. In many other ways the ideas of this small
area radiate in all directions. In that lies the essential point.
Because this 4 or 5 per cent of the world's population are en-
dowed with unusual energy, they are rapidly obtaining an in-
fluence wholly disproportionate to their numbers.

The third important region of high civilization and energy is
Japan. Although the area is only about one tenth that of the
corresponding portion of Europe and America, the population
is over 50,000,000. In proportion to population, its influence
upon the world is less than that of the European countries and
the United States, which perhaps corresponds with an appar-
ently lower degree of climatic energy. Yet we all recognize that
Japan is a factor to be reckoned with. We admire its art and the
skill and certainty with which the country has made use of our

SHIFTING CENTERS OF CIVILIZATION 351

western civilization. We see students flocking thither from China.
In India the news of the progress of Japan causes a stirring of
dead leaves as by an autumn wind. As Japan has done, so India
would do. In India, however, the agitators chiefly talk and com-
mit violence, whereas in Japan they go to work quietly without
much talk or violence and make themselves so efficient that they
do not need the help of Europeans. When Japan sees an oppor-
tunity, such as was afforded by the war in 1914, she takes it,
and thus advances another step in her role as the most capable
nation in Asia. Like Europe and America she makes mistakes
sometimes, but they are the mistakes of strength more than of
weakness.

Next to Japan among our areas of high civilization comes
JJew Zealand, together with the adjacent corner of Australia.
Here we have chiefly a reflection of British civilization. Most
of the five or six million people are grouped in an area some-
what more than twice the size of Japan. The question is not
whether they show evidences of a high civilization, which they
certainly do, but whether they have added to the endowment of
ideas and institutions with which they came from England.
They have clearly done so, for when we think of Australia there
come to mind such things as the Australian ballot, whereby a
man is enabled to vote freely without undue influence from
bystanders. We think also of progressive labor legislation, and
of a pioneer attempt at old age pensions, especially in New
Zealand. We are made conscious that these people are not
mere followers, but are blessed with ideas which they themselves
originate, and which spread out from them to other parts of the
world.

The last area which rises to the very high grade on the map
of civilization is California and the coast farther north. The
population is still relatively small, and the majority grew up
in other regions. Hence, it is too soon to judge of the ultimate
status of this area, It is so bound up with the rest of the United

352 CIVILIZATION AND CLIMATE

States and Canada that its particular contribution to civiliza-
tion cannot be clearly differentiated. Yet, through their uni-
versities and their readiness to attempt great things, the people
of the Pacific coast impress the world as unusually capable and
energetic. All together these five regions comprise less than one
twentieth of the lands. Though they are densely populated
because of the skill of their people in agriculture and manu-
facture, they contain only about one fourth of the world's
population. In influence, however, they many times outweigh the
remaining three fourths. If they decide on anything the rest of
the world must submit. They never care to imitate the more
backward nations, whereas such nations show a growing desire
to imitate Europe, the United States, or Japan. The reason
seems to be that the favored one twentieth of the land is inhabited
by people of great energy ; and this energy appears to be due,
in part at least, to the frequency of cyclonic storms with their
stimulating changes of weather from day to day.

This leads to the question whether similar conditions pre-
vailed in the past. Throughout the course of history there have
been centers of civilization like those which exist today. One of
the most important was the plain of Mesopotamia. Here, in the
earliest days, the people of Sumer and Accad invented hiero-
glyphs, founded great cities, built mighty irrigation works,
drained the marshes, and laid the foundations of a complex
system of law and religion. They clearly possessed the power of
originating ideas and of putting them into effect in a way that
is now characteristic of the modern centers of civilization. Where
they originally came from we do not know, but apparently they
were not Semitic. In course of time, they were overwhelmed by
Semites who presumably came in as rude invaders. Then there
happened one of the most characteristic events in the history of
centers of civilization. The Semites dropped their old culture and
adopted that of the people whom they had conquered. They
began also to be inventive like their predecessors, and thus de-

SHIFTING CENTERS OF CIVILIZATION 353

veloped cuneiform writing out of the hieroglyphics of the Su-
merians. In later days, other Semitic invasions took place. In
each case the invaders had done no great things in their original
homes so far as we know, but when they came to Babylonia they
blossomed into people of inventive minds full of new ideas which
caused the arts to rise to higher and higher levels. Then the
Semites were conquered by a non-Semitic race, the Scythians,
who were also called Medes, and they in turn were stirred to great
achievements. They were conquered by still another people, the
Persians under Cyrus, and once more the old process of stimula-
tion was repeated.

One of the most significant things in Mesopotamian history
is the way in which each invading race seems suddenly to have
risen in civilization as soon as it reached the new country. How
much of this was due to natural selection, whereby the weaker
people among both the conquerors and the conquered were
weeded out during the migrations, wars, and pestilences which
preceded the final settlement; how much to contact with a
higher civilization; how much to more abundant opportunities
afforded by the new environment; and how much to a direct
physical invigoration? I shall not try to answer these questions.
That some of the invading races achieved great things because
they possessed innate ability seems certain. That their minds
were rendered more alert by contact with the achievements of
the races which they conquered is likewise beyond doubt. That
the wealth and agricultural possibilities of Mesopotamia and the
abundance of labor stimulated the invaders to construct palaces
and temples, build irrigation works, gather stores of treasure,
carve huge monuments, spend money in supporting libraries
and students, and carry on other great enterprises also seems
certain. These principles are so firmly established that there is
no need to discuss them. The question before us is simply
whether there may also have been a climatic stimulus compa-

354 CIVILIZATION AND CLIMATE

rable to that which now seems so important according to our
maps of civilization and energy.

I do not purpose to discuss the empires of antiquity. I merely
wish to point out some of the features of their location and his-
tory which pertain to our main hypothesis. While Babylonia
was flourishing, Egypt in the same latitude was also developing
a great civilization. The two were rivals. Which rose higher it
would be hard to say. They moved along parallel lines, and at
last both fell during the same general period, although Egypt
retained a moderate importance for some centuries after Meso-
potamia had lapsed into insignificance. Far to the south the
Sabseans in southwestern Arabia also developed a fairly high
civilization, but they formed in no sense a great center. We know
little about them, but apparently they were behind Mesopotamia
and Egypt in the way that modern Bulgaria is behind England.
Therefore, we may pass them by for the present although they
will come up again in another connection. East of Mesopotamia
the highlands of Persia were the seat of a comparatively high
culture, but it scarcely rivaled that of the two great countries
in the river valleys. Going on still farther toward the rising sun,
we find high civilization in northern India, at a very early time.
That region never rose to such a pitch as did the countries
farther west, and it decayed somewhat earlier. Beyond India
the chain of ancient empires is broken until we come to China.
In Indo-China, to be sure, traces of a state of culture higher
than that of today are found, but they belong to a compara-
tively late date, and it is doubtful whether they were indigenous.
They are important, but do not apply to the period now under
discussion. China, however, stands in the same class with Egypt
and Mesopotamia. It reached a high state of development about
as long ago as they did. Here, too, the art of writing developed,
inventions of many kinds were made, strong governments were
organized, other powerful institutions were evolved, and in gen-
eral there was intense mental activity.

SHIFTING CENTERS OF CIVILIZATION 355

At this point the reader is advised to consult a map of Asia
and Europe. Notice how the great countries of the past lie
along an arc. They begin on the east with China, whose main
center was north of the Yangtse basin. The civilization which
flourished there possessed qualities like those of modern Europe
and America. The Chinese were an aggressive, active people, able
to fight or to evolve a new religious system as the case might be.
The sword was in their hands not merely because those were
ancient days, but because the people or their rulers wanted to
make conquests, or because they were quick to resent infringe-
ments of their rights. They were not like their modern successors
who see themselves defrauded because they have not the power to
cope with their despoilers. Follow the map westward along a
band eight or ten degrees wide with the thirtieth parallel a little
south of its center. At first we find ourselves among lofty moun-
tains which merge into the great Tibetan plateau. Here we
should not expect any great development because of the un-
favorable topography. At length, however, our band descends to
the plain of northern India. On the flank of the Himalayas lived
Gautama, the great thinker who founded Buddhism. Out on the
plains around Delhi and Agra and in surrounding regions was
the home of the early Aryans, who developed the Sanscrit lan-
guage, wrote the Vedas, and made a real contribution to human
progress. Here, just as in China, a broad expanse of arable land
was favorable to the development of a great people. Next our
band strikes the high mountain rim of Baluchistan and Afghani-
stan, and then passes over the plateaus of those countries, high
and rugged in the north, lower and intensely desert in the south.
In these regions the ancient civilization was higher than that
of today, but not until we reach the western side of the plateau
do we come upon the really advanced country of ancient Persia.

As our band leaves the Persian highland, let us change its
direction a trifle so that its central line will point toward the
center of the Mediterranean Sea. First it lies over Mesopo-

356 CIVILIZATION AND CLIMATE

tamia. There in the highly fruitful plain we find the two great
empires of Babylonia and Assyria. They were like their modern
successors in the Europe of our own day. When they were weak,
they were invaded by the more backward people from the moun-
tains on the north and east or from the desert on the south and
west. When they were strong, they waxed ambitious ; they strove
to extend their borders ; they desired to rule the world ; and
they were jealous of rival nations. Thus, at their strongest, like
the great European nations which were locked in a death
struggle only a few years ago, they fought tooth and nail for
supremacy. The fighting in itself is a proof of superabundant
energy. A weak people may fight in self-defense, or to gain new
lands when sorely pinched with want at home, but only a people
who have a surplus of both strength and wealth can stand the
strain of great wars whose purpose is primarily national ag-
grandizement.

On the flanks of our band of culture, as we call it, lay the
Hittites on the borders of Mesopotamia and Anatolia. Beyond
the Syrian desert the Phoenicians, Syrians, and Jews lived in
the center of the band. Their degree of culture was equal to
that of the Mesopotamians, but displayed itself in different
ways, partly because of the accidents of mountain chains and
seas, and partly because of peculiar circumstances of their
racial character or history. Next comes Egypt, still within the
belt. Here still another fertile plain gave opportunity for the
accumulation of great wealth and for the leisure which is essen-
tial to a part of the population if new ideas are to be developed.
They fought the same kind of wars as did their neighbors on
the east. When they were most powerful, their strength and
energy were such that the mere conquest of their immediate sur-
roundings did not satisfy them. Taking Palestine, they pene-
trated Syria and even advanced to the Euphrates to fight with
Assyria, their great rival. And the Assyrians in turn came down
through Palestine to wrest the sovereignty from Egypt.

SHIFTING CENTERS OF CIVILIZATION 357

Now we pass on, curving our band a little more, so that its
center shall pass along the main axis of the Mediterranean Sea.
Rhodes, the JEgean Sea, Greece, Italy, and Carthage fall within
its limits. Here, too, we see the same phenomena of the rise and
fall of great nations. The rise did not begin quite so early in
the west as in the east, and the fall did not come so soon. In
Greece, as in Mesopotamia, a growth in strength brought with
it a great struggle for supremacy, which did not end till Athens
had subdued her rivals. The wars of Rome and Carthage were
of this same kind in which great nations, jealous of the great-
ness of a rival, well-nigh destroy themselves.

In this rapid sketch we have mentioned each of the chief
civilizations of Asia and Europe up to the time of Christ. All
were in their day centers comparable to those which we have
seen to exist at present. One main center lay in China, and cor-
responded to that of modern Japan. A second, less important
and less enduring, was located in northern India. It has no mod-
ern representative, unless we hold that the vogue of an esoteric
Hinduism among a few people in America and Europe is a sign
of the existence of a widely pervasive influence derived from
an Indian center. The third center extended widely. Its eastern
limit was on the western border of Persia. It embraced Meso-
potamia, Syria, Egypt, and Greece, and on its flanks included
surrounding countries such as Asia Minor. Ultimately it moved
westward, so that Mesopotamia and then Egypt ceased to belong
to it, while Italy was included. Finally, it migrated still farther
westward and also northward until now Italy is on its south-
eastern border, and its main area is the great European center
of our own time.

In the western hemisphere only two centers of civilization seem
to have existed in ancient times. One was in Peru. We shall not
consider it here, partly because it will come up in another con-
nection, and partly because it failed to rise to a great height,
and never succeeded in spreading its ideas very widely. The other

Figure 48. The Shifting of the Storm Belt

1 Regions of Great Storminess (Over 20 Centers per Year According: to Kullmer's Scale).

2 Regions of Moderate Storminess (10-20 Centers).
8. Regions of Slight Storminess (1-10 Centers).

4. Regions Without Cyclonic Storms. .

The Median Lines of the Main Storm Area and its Branches are Indicated by Heavy Solid Linen.
The Hypothetical Median Lines of the Ancient Storm Belts are Shown by Dotted Lines.

SHIFTING CENTERS OF CIVILIZATION 359

ancient American civilization, that of the Mayas, is much more
important. It falls into the same class with those of Eurasia.
Its fertility in ideas has already been discussed. Its influence
spread to other lands, and may be seen in various ruins in
Mexico, and in the calendars which were employed by several
races in that country. It must have lasted many centuries, for
we find splendid buildings whose dates are at least seven hundred
years apart and possibly more. Moreover, before the first of
these was constructed there must have been a long period of
development during which the arts of architecture and sculpture
were perfected. At that same period the Mayas must have been
so far advanced that they kept most accurate records of astro-
nomical phenomena, for otherwise they could not have deter-
mined the length of the year with the extraordinary accuracy
displayed in their calendar. In all these respects the Maya center
resembled those of Eurasia. It differed from most of them in not
leaving a successor.

So much for the ancient distribution of civilization. Now for
that of climatic energy. Storms, as we have seen are the most
variable of the climatic elements and also one of the most im-
portant in their effect on human energy. Figure 48 shows the
approximate distribution of storminess in the northern hemi-
sphere at the present time. It is constructed with the north pole
in the center to show the complete storm belt encircling the north
temperate zone. The lands arc shaded to represent three grades
of storminess. Heavy lines indicate the centers of the main storm
belt and its branches. The region of maximum storminess lies in
southern Canada north of Lake Ontario. Thence the main belt
extends eastward along the St. Lawrence River to the southern
edge of Newfoundland, where it is joined by a branch coining up
the Atlantic coast. Then it crosses the Atlantic with a slight
inclination toward the north, and before reaching Ireland splits
into three branches. One passes along the northern edge of Scot-
land, through northern Scandinavia and thence along the Arctic

360 CIVILIZATION AND CLIMATE

coast of Europe toward Asia where it almost dies out and
probably merges with the second or main branch. The latter
splits into two parts which unite again. One passes through
Scotland and southern Scandinavia. The other goes through
southern England, follows the Baltic Sea and then, after join-
ing its companion, extends eastward in the latitude of Moscow
for an indefinite distance into Siberia. It apparently passes
entirely across Asia, but in an attenuated form. The third
branch swings from the central Atlantic toward the Bay of
Biscay. It becomes weakened over southern France, strengthens
once more in the stormy Gulf of Lions, continues with great in-
tensity in northern Italy, then loses intensity as it crosses the
Bosnian highland, and finally passes through southern Russia
where it rapidly dies out. Returning to the main belt of stormi-
ness we find that after it has crossed Asia it is joined in the
Pacific Ocean by a branch which has been highly developed in
Japan, but which weakens northward. The combined branches
cross the Pacific Ocean to the American shore. Then in southern
Canada and the northern United States, after crossing to the
east side of the Rocky Mountains, the belt once more becomes
extremely stormy. It is joined by a branch from Colorado and
so completes its course back to the point of greatest intensity.

Several features of Figure 48 deserve careful attention. In
the first place, wherever the storms in their eastward course
approach an extensive highland they tend to be weakened or
deflected, after which they may or may not recover their former
intensity. This may be seen on the western side of North Amer-
ica where storms are relatively rare until the Rocky Mountains
have been crossed. The highlands of western Europe are not great
enough to destroy the storms, but the main belt is broken into
several branches, each of which seeks to avoid the high country.
The northern branch would apparently skirt the whole of
Scandinavia were it not that some unknown force prevents it
from going much to the north of the Arctic circle. The next

SHIFTING CENTERS OF CIVILIZATION 361

avoids the Welsh and English highlands, crosses into the North
Sea over the lowest part of England, receives an accession of
intensity, and follows the line of least resistance along the Baltic
and across the plains of Russia. Ultimately, however, the vast
expanse of Eurasia, especially when it becomes a high pressure
center in winter, largely overcomes the tendency of the air to
move in the cyclonic fashion, and storms become rare. The third
branch, which swings still farther south, is also greatly weak-
ened when it reaches the land of southern France, but becomes
very pronounced in the northern Mediterranean, only to be
weakened once more by the Balkan highland.

Another important feature of Figure 48 is the peculiar de-
tached areas of high storminess in northern Italy and Japan.
In the United States two similar areas are visible on a large
scale map of storms, although they are masked in Figure 48.
In both cases they are apparently remnants of a southern, or
subtropical belt whose existence at times of many sunspots is
described in Climatic Changes. Another belt where storminess
increases greatly at times of sunspot maxima is found on the
north side of the present belt. It tends to reduce the habitability
of the northern parts of Europe and North America, but with
that we are not now concerned. In the southern belt of increased
storminess, when sunspots are numerous, the American areas
move somewhat southward, become intensified, and lengthen
east and west until they coalesce, forming a continuous, although
very faint, belt. In Eurasia the same thing seems to happen,
although the meteorological records in Asia are not yet suffi-
cient to permit us to test it. If the American subtropical belt
were intensified in the way that appears to have occurred in the
past, its place of maximum storminess would probably be in
the Gulf of Mexico. There the obstacles presented by the high-
lands of the southwestern United States and Mexico give place
to the open sea and allow free play for the development of
storms. A development of this subtropical storm belt seems to

362 CIVILIZATION AND CLIMATE

be a pronounced feature of the changes which now take place
from sunspot minima to maxima, and would presumably be
magnified if the changes were on a greater scale than those of
the 11-year cycle. Hence, we should infer that when the Mayas
made their greatest progress, their country was blessed with
a stormy area like that of the Gulf of Lions which now does
much to make northern Italy far more progressive than the
southern part of the country. Under such conditions the climate
of Yucatan would have been comparatively stimulating in spite
of its warmth, for during much of the year cool waves from
the north would have been frequent.

In Eurasia the same principles apply with appropriate modi-
fications because of the topography and size of the continents.
If the subtropical storm belt should be magnified in the way
that seems characteristic of times of many sunspots, it would
apparently reach its chief development in the places where the
great civilizations of the past were located. The present Italian
center would presumably move south and then expand eastward
along the open Mediterranean. Leaving the highlands of Asia
Minor on its northern flank, it would pass over the Syrian moun-
tains because it could not spread eastward in any other way.
Possibly it would be forced a little southward so that northern
Egypt and Palestine would get its full force. In the lee of the
mountains the storminess of the Syrian desert would be less
than that of Syria, but by the time the storms reached Mesopo-
tamia they would have had an opportunity to form themselves
anew, and we should expect a well-developed center. In Persia,
judging by what happens today, this would be weakened but
not destroyed. Then the great plains of northern India would
permit the storms to gather once more in a center corresponding
to the faint one of today. That would probably be the end of
any distinct belt for a thousand miles, for the eastern Himalayas
and the mountains of western China interpose so tremendous a
barrier that we should not expect the Japanese center, even

SHIFTING CENTERS OF CIVILIZATION 363

though it moved southward and westward, to coalesce with the
Italian. The plains of central China, especially the Yangtse
valley, would be the place where the present Japanese center
would be most highly developed, provided it suffered changes
corresponding to those indicated in the corresponding American
area. Japan, on the other hand, would presumably be less stormy
than now, for the storm tracks would be pushed southward and
oceanward.

If these things actually occurred at the times indicated by
our California curve, the countries which now stand highest in
civilization would then have had a long, cold winter, ending in
a very stormy spring and followed by a cool damp summer. The
stimulating qualities of the climate would have been less than
now, and the possibilities of agriculture, especially in such
places as Germany, would have been much diminished. In the
countries like Greece where civilization was then at its highest,
on the contrary, the number of storms and the duration of the
stormy period would have been decidedly greater than now.
The rainfall also would have been greater so that agriculture
would have been favored. More important than this, however,
would have been the high degree of climatic stimulus because
of frequent changes of temperature. To apply the matter con-
cretely, lower Egypt has an average temperature of about 80
during July, and Mesopotamia and northern India are even
hotter. Three thousand years ago the heat was probably almost
equally intense at certain periods, although somewhat less ex-
treme on an average, and by no means so uniform. All the
summers were probably hotter than was the summer of 1911
in the northeastern United States. That summer, in some lo-
calities, was the most severe for a century. Yet its effect on
work was no worse than that of an ordinary winter. In Meso-
potamia and Egypt, summers of that kind doubtless made
people slow and inert. They were probably not so bad as the
present summers, however, for although no appreciable quan-

364 CIVILIZATION AND CLIMATE

tity of rain may have fallen, a large number of storms must
have passed over regions not far to the north. Such storms
would stir the air and bring fresh breezes. Everyone knows that
in hot weather a change in the wind even without rain is most
refreshing. When the summer was over, the storm belt, according
to our hypothesis, would migrate southward in its normal fash-
ion, and soon Egypt and Mesopotamia would be swept by storm
after storm. During the early fall and late spring conditions
would be about as they are in the homes of our Connecticut
factory operatives during cool summers such as 1912 or 1913.
For the intervening five or six months the average temperature
would range from about 55 to 65 F., there would be a con-
stant succession of cool waves, and the conditions would be
almost ideal for great physical activity. Thus, even though the
summers were distinctly bad, the total debilitating effect would
be little greater than that of summer and winter combined in
Connecticut. In Greece and Italy, with their more favorable
mean temperature, conditions would be still better than in Egypt
or Mesopotamia. In the same way favorable conditions appear
to have prevailed in each of the great countries of the past at
the time when it made its most rapid progress.

The two phases of our climatic hypothesis are now before
us. In point of time, though not of presentation in this book,
the first step was a study of the climate of the past. Ten years
of work along this line had led to the hypothesis of pulsatory
changes, and finally to the idea that the changes consist pri-
marily of a shifting of the belt of storms. After this conclusion
had been reached, a wholly independent investigation of the
effect of present climatic conditions upon human activity led to
two conclusions, neither of which was anticipated. One was
that in the eastern United States the cold weather of winter is
as bad for work as for health, while only the warmest summers
cause any serious curtailment of work. The other was that
storminess and variability from day to day are of great im-

SHIFTING CENTERS OF CIVILIZATION 365

portance. On the basis of these two conclusions it at once be-
comes evident that the stimulating effect of climates in the same
latitude and having the same kind of seasonal changes may be
very different. It also becomes clear that the distribution of
civilization at the present time closely resembles that of climatic
energy. From this the next step is naturally back to our previous
conclusion that changes of climate in the past have consisted
largely of variations in the location of the storm belt. If this is
so, evidently the amount of climatic stimulus must have varied
correspondingly. Thus we are led to the final conclusion that,
not only at present, but also in the past, no nation has risen to
the highest grade of civilization except in regions where the
climatic stimulus is great. This statement sums up our entire
hypothesis. It seems to be the inevitable result of the facts that
are before us. Other factors, to be sure, are also highly im-
portant. One of them is natural selection, a subject so much
neglected in its bearing on human history, that I made it the
theme of a book on The Character of Races, which might almost
be considered the continuation of the present book. The de-
velopment of human culture and its spread from land to land
is, I believe, a third factor quite as important as either of the
others. Yet, unless we have gone wholly astray, the surprising
way in which independent lines of investigation dovetail into one
another seems to indicate that a favorable climate is one of the
essential conditions of high civilization.

CHAPTER XVII
ABORIGINAL AMERICA AND MODERN AUSTRALIA

HAVING seen that the objections to climatic pulsations
are largely based upon misapprehensions, let us now take
up certain other objections to our hypothesis of the relation
between civilization and climate. "How about the high civiliza-
tions of the past in places outside the storm belt?" the objector
may say. "Yucatan and Guatemala may possibly have been
stormy, but surely not Peru, southern Arabia, Rhodesia, Ceylon,
Java, and Indo-China. Yet in the past these have boasted a
civilization much higher than now prevails there."

The force of this objection must frankly be admitted. The
ruins of Indo-Chino and northern Ceylon, and any other traces
of high civilization located in tropical lowlands, with the excep-
tion of Yucatan and Guatemala, appear to have little relation
to our hypothesis. Nevertheless, they do not require a modifica-
tion of it. More probably they represent the triumph of other
factors over the climatic factor. They apparently indicate one or
both of two things. First, they appear to represent a temporary
wave of progress due to the incursion of higher cultures from
more favored regions such as India or the neighboring highlands.
If that is so, they possess no more significance for our climatic
hypothesis than do the railroads and other innovations intro-
duced by Europeans within the past few decades. In addition to
this they probably indicate that prolonged migration or other
difficult circumstances gave rise to pronounced natural selection.
Thus the people who built the wonderful ruins of Java and Indo-
China, as I have shown in The Character of Races, and perhaps

AMERICA AND AUSTRALIA 367

those of Ceylon, Peru, southern Arabia, and Rhodesia, had pre-
sumably been gradually sifted so that only a very competent
remnant was left. Under such circumstances the effects of cli-
mate seem to be overcome for a while, but reassert themselves in
due time as the strong inheritance becomes diluted or weakened.

Another possibility is that the small and insignificant belt of
storms which now traverses northern India may have been in-
tensified in the past. In eastern India because of the barrier of
great mountains, a branch may have swung southeast around
the lofty Burmese mountains to the neck of the Malay Peninsula
and then northeast to join the storm belt which now traverses
Japan. In crossing the plains of Cambodia, where the most re-
markable ruins of southeastern Asia are located, it may have
developed a distinct storm center like that of northern Italy.
The conditions would resemble those inferred in Yucatan and
Guatemala, although perhaps less favorable because cold waves
would be checked by the great mountains. Indo-China and Cen-
tral America lie in the same relative position in .their respective
continents, a fact which may have considerable significance. I do
not, however, attach much importance to what has just been said
about storms in southeastern Asia, for it is pure supposition.
Other conditions, such as racial selection and the spread of the
highly developed culture of India seem competent to explain a
large part of the sudden rise of these regions, while the tropical
climate largely explains their rapid decline.

Yemen, Rhodesia, and Peru furnish admirable examples of
comparatively high civilizations which developed in tropical
highlands and lasted for a considerable period without much
intercourse with regions of higher culture. They lie at such an
altitude that the mean temperature never departs far from the
optimum. For instance, at Fontein, near the Rhodesian ruins of
Zimbabwe, the figures are 54 F. in June and 71 in November.
For Yemen none are available, but judging by the altitude its
temperature is probably slightly lower than that of Rhodesia.

368 CIVILIZATION AND CLIMATE

Even at Cuzco the elevated capital of old Peru, although the
monthly mean temperature according to Hann ranges from 46
in July to 52 in November, the conditions are very stimulating
compared with tropical lowlands. The temperature is too low
for comfort, to be sure, but it is of the type that appears to
stimulate mental activity. If immigrants possessed of an un-
usually high inheritance, either by reason of strenuous natural
selection or of actual biological mutations, should come to such
a region, the relatively stimulating quality of the climate would
combine with their innate ability to enable them speedily to
dominate the indigenous population, and to develop many new
ideas. Yet we should not expect such a civilization to endure so
long or rise so high as those in more favored regions. The people
would tend to exhaust themselves for they would never experi-
ence any restful changes of seasons, and would be stimulated at
all times. To revert to an earlier illustration, their condition
would be like that of a horse which is always driven at full speed.
Such a horse might go rapidly for a while, but would wear him-
self out at an age when carefully driven animals were in their
prime. The constant nervous excitation produced by such a
climate in immigrants from a less favorable region would induce
both progress and decay. The most nervous people would die
out partly because they would exert themselves too strenuously,
and partly because nervousness is a potent agency in reducing
the birthrate. Self-control would also be weakened, thus leading
to vice and excesses of various kinds. The chewing of the coca
leaf, a narcotic stimulant which is thought by many people to
be one great cause of the backwardness of Peru today, would
harm the high-strung, competent parts of the community more
than the dull, apathetic ones. In various other ways such a
climate as that of Peru would be stimulating for a while, but
would lead to exhaustion sooner than would one where greater
variability prevails.

We now come to perhaps the strongest objection aside from

AMERICA AND AUSTRALIA 369

the question of the reality of climatic changes. A large part of
this book has been concerned with the United States. The present
status of that country is one of the main foundations which
originally led to the framing of our hypothesis. Yet before the
coming of the white man savagery prevailed where civilization
is now highest. In the sixteenth century the climate must have
been approximately the same as that of today, so that no expla-
nation can be sought along that line. It is true, to be sure, that
before the coming of the white man the most energetic and
capable of the Indian tribes lived in what is now the north-
eastern United States. Dixon, for example, in The Racial His-
tory of Man, says that the "Iroquois and southern Algonkian
tribes were among the first of those north of Mexico in ability
and prowess." Many other authorities speak in similar fashion.
Wissler in Man and Culture reproduces a map by Moorhead
showing that the region from Illinois to New York was an im-
portant center of culture from which many new types of stone
ornaments spread in all directions.

In spite of this we are still confronted by the stubborn fact
that before the coming of the white man no great civilization
had ever developed in the northern United States, while Mexico,
although far inferior climatically, harbored a relatively civi-
lized population. At first this seems to prove that our hypothesis
cannot be correct. Yet more careful examination leads to a dif-
ferent conclusion. The distribution of civilization in pre-Colum-
bian America merely brings out a fact which I have again and
again tried to emphasize : although climate is highly important,
there are other factors whose weight is equally great. Even if
our hypothesis be fully accepted, no less importance will thereby
attach to the other great factors which condition the events of
history. Because a man dies for lack of air, we do not think that
air is more essential than food, drink, warmth, the circulation of
the blood, the reproduction of the species, the ravages of viru-
lent germs, and many other conditions. So it is in history. Even

370 CIVILIZATION AND CLIMATE

if our climatic ideas arc correct, it will still be true that the
ordinary events of the historical record are due to the differing
traits of races, the force of economic pressure, the ambition of
kings, the intrigues of statesmen, the zeal of religion, the
jealousy of races, the rise of men of genius, the evolution of new
political or social institutions, and other similar circumstances.
Yet a comprehension of the part played by the climatic factor
will enable us to explain some of the many events which hitherto
have puzzled us. Not all will be thus explained, for others must
wait until the action of some other set of as yet unknown condi-
tions is understood.

In the present case I believe that the explanation of the fail-
ure of the Indians to rise to a level equal to that of the people
of the Old World is found partly in inheritance, and partly in
the history of their culture. I shall not dwell on inheritance,
although it may be of great importance. In The Red Man's
Continent and more fully in The Character of Races I have set
forth the hypothesis that the aboriginal Americans during their
migration from Asia by way of Siberia and Alaska suffered an
adverse natural selection due to the severe arctic environment.
This may have given them a permanent handicap compared with
the people of Europe and western Asia among whom the selec-
tion was peculiarly favorable. Among the Americans, so it seems,
there was a premium on powers of passive resistance and en-
durance. Moreover, among all hunting tribes there is a premium
upon the ability to make sudden but not prolonged efforts such
as are involved in hunting and raiding. Only when agriculture is
introduced does a new type of natural selection put a premium
upon such qualities as steady industry. Among hunting tribes,
moreover, the acquisitive temperament which stores up food or
other property for periods of want, is not nearly so great an
advantage and hence so strong a selective factor as among those
who practice agriculture. Because the wandering hunter cannot
transport much property. J^ loses the biological advantage

AMERICA AND AUSTRALIA 371

which causes the people who have a careful, thrifty spirit to have
a greater chance of survival than have the careless and un-
thrifty.

Turning now to the cultural side of early America, at least
two great factors appear to have imposed a tremendous handi-
cap upon the Indians of the northeastern United States. One is
the absence of tools of iron, and the other the lack of beasts of
burden. These, in conjunction with the type of vegetation, offer
such hindrance that the Indians could probably never have
developed a very high civilization, even if they had been as com-
petent as the races of Europe. The discovery of the use of iron
depends on the coincidence of three conditions whose occurrence
is almost accidental. One is the birth of a man with sufficient
inventive genius to devise a way of obtaining the metal from
rocky ores. The second is the occurrence of ores and other neces-
sary materials within easy reach of such a man. The third is
that the genius must be so relieved from the fear of enemies, the
danger of starvation, and the ravages of disease that he has both
time and strength to elaborate his idea. The combination of
these three fortunate circumstances never occurred in America.
Although a little native copper was used, we have no assurance
that the ore was ever smelted in any large amounts. It certainly
never was used to any great extent. Iron, except a few bits from
meteorites, was absolutely unknown. Hence, through no fault of
their own, the original Americans were dependent upon such
tools as they could fashion from flint, obsidian, and other stones,
or from bones, shells, wood, and similar materials.

We are so accustomed to iron tools that we scarcely realize
their immense importance. Consider their effect upon agricul-
ture. Go into a virgin forest with its labyrinth of trees. Imagine
the task of cutting them down with a stone hatchet. The mere
physical labor is such that none but people of high energy
would ever attempt it. An easier way is to girdle the trees,
cutting off the bark in a circle, and then leaving them to dry

372 CIVILIZATION AND CLIMATE

until they can be burned. Even that, however, is a long process.
In the moister parts of the world its difficulty is greatly in-
creased by the fact that while the trees are becoming dry enough
to burn, new vegetation is rapidly growing. In many tropical
regions, as we have already seen, the clearing of the forests is
scarcely feasible even today. In temperate forests the difficulties
are not so great, but they are practically insurmountable from
the standpoint of a savage who has never known the meaning
of hard, steady work. In the hunting stag a man may follow
the trail until he falls from exhaustion, but that does not give
him the power to wield a stone axe day after day upon stubborn
trees. Hence, if agriculture is to be practiced by primitive people,
it must originate in regions where there are no forests. The
vegetation must be such that the primitive savage with his hands
or with a stick or stone can easily grub it up. Or else it must be
so scanty that it will not interfere with his crops. Anyone who
has seen the agriculture practiced by people who are just emerg-
ing from some other mode of making a living knows that the
fields are sown in the spring and then generally left untended
until the time of harvest. Since the people are not yet able to
get their living entirely from the crops, which is always the case
during the transition period, they must perforce carry on the
old occupations while the crops are growing. Hence, the weeds
are allowed to grow as freely as the grain. In a dry climate where
the natural growth consists only of bunchy grasses or low bushy
weeds which grow apart from one another and can easily be
pulled up with the hands before the next sowing, this does little
harm. In wetter regions, such as the forested areas, the weeds
arc much more difficult to eradicate.

The reader has probably noticed that I have said nothing
about grass-lands. The reason is that primitive people never
attempt to cultivate them. If grass covers the entire ground and
forms what we commonly call sod, crops cannot be raised in it.
The dry grass may be burned off in the early spring, and seeds

AMERICA AND AUSTRALIA 373

may be dropped in holes punched with a sharp stick, but no crop
will be reaped. The roots of the grass are not killed. The new
blades shoot up at once, and by the time the seeds begin to sprout
are so high as to strangle them. The primitive savage who has
no iron tools cannot possibly dig up the sod. Even if he has such
tools, the laboriousness of the task bans it as a practical method
of making a living. If anyone doubts this, let him spade up a
turfy spot ten feet in diameter. Then let him calculate how many
days he would have to dig in order to make a living for himself
and his family by sowing grain. Let him also determine how a
primitive savage who had to do this task would support himself
and his children while he was digging before he could get a first
crop. Finally, remember that the savage had no good spade to
help him, but only a clumsy, heavy stone set in a handle, a
brittle bone, or a piece of wood rudely shaped into a flat imple-
ment.

I have emphasized the difficulties of grass-land because they
apply directly to the problem of civilization among the Indians.
The part of North America which is highly stimulating cli-
matically is covered either with grass of the prairie and great
plains type, or with temperate forest. It is universally recog-
nized that a high state of civilization is impossible unless it is
based on agriculture. Otherwise people must wander all the
time. They cannot accumulate the appliances which are essen-
tial even to the lowest real civilization. They are not attached
to the soil, and thus have no incentive to improve a particular
tract. If they possess domestic animals, the case is better, for
these foster the sense of ownership and various other feelings
which serve as uplifting influences. In North America no do-
mestic animals except the dog and the turkey were known to the
aborigines. This was not the f $ult of the Indians. The horse and
camel families were then extinct in North America. The bison
was the only animal of the ox family with which the Indians
came in contact. It was too large and fierce to be domesticated.

374f CIVILIZATION AND CLIMATE

The mountain sheep might possibly have been tamed, but it is
very wild, its habitat is the barren mountains, and it is too small
to plough sod. Even if the Indians had been familiar with the
idea of taming beasts of burden, they could not have achieved
any important success.

Without beasts of burden ploughing is impossible. Unless the
ground is ploughed, grass-lands, as we have seen, cannot be
cultivated. Therefore, civilization could not flourish in the
grassy plains of North America before the coming of the white
man. A few highly favored spots such as river flood plains
where the grass grows in bunches and does not form turf were
cultivated, but the areas of that kind were too small and scat-
tered to give rise to any widespread progress. They were ex-
posed to the ravages of Indians who had not yet learned the
value of permanent abodes and individual property, two pri-
mary requisites of civilization. The people who lived upon them
were swamped in the flood of surrounding savagery. Moreover,
the places which they were able to cultivate were subject to
floods and other disasters, so that the cultivators were often
forced to rely on the chase. Then again, without iron tools or
beasts of burden, a man cannot cultivate a large area because
the task is so great. Therefore, the crops were likely to be
insufficient to support the family throughout the year even
when the harvest was good. The only recourse was the chase.
In Asia people who were in such circumstances might rely on
domestic animals, but not in America. Hence, it was practically
impossible for the Indians of the grassy plains to get away from
the savage life of the hunter.

In the forests which covered all the eastern states conditions
were scarcely better. A man might girdle enough trees and cut
enough brush to make a small field. When the wood was dry,
he might burn it and obtain a crop of corn. While it was ma-
turing he and his famity would make a living by hunting and
fishing. The crop would support them through the winter, but

AMERICA AND AUSTRALIA 375

it is extremely doubtful whether they would ever cease to regard
hunting as a main source of livelihood. The reason is this. The
first crop on a burned piece of forest land in the eastern United
States meets with little difficult} 7 . For the second crop it is
necessary to cut down and burn the weeds and bushes that have
grown up during the preceding summer. Remember that the
savage has no iron hoe, his seeds are planted haphazard here
and there among the stumps, and he cannot tend his garden
while it is growing, but must go off and hunt. When he cuts
the bushes and weeds the second spring he does not root up the
grass. He has no spade with which to dig it up, and no ox or
donkey to draw a plough. If he takes good care of his field he
soon finds that he has a meadow of tough sod on his hands. If
he neglects it, he has a few acres of bushes. In either case if
he would make a living from agriculture, he must go through
the labor of clearing a new field. Meanwhile he still must be a
nomadic hunter. He takes no great pride in his field ; he does not
care to improve it ; he rarely builds a permanent house beside
it. Why should he? Next year, or at most in two or three years,
lie must build another. In a few specially favored spots the con-
ditions may be such that agriculture is not quite so difficult. In
the North, where the stimulating climate is found, such spots
are few and small. The chief reason for wonder is that the
Indians had energy to cultivate any crops whatever. In the
southern states the grass does not form so dense a sod as in
the North, but even there the difficulties are great. Yet a begin-
ning had been made, and the civilized nations, including the
Cherokees and others, lived in permanent villages and practiced
agriculture as their main means of getting a living. If they had
had iron tools and oxen, America today might be filled with
highly civilized Indians instead of Anglo-Saxons and other
Europeans. Thus, although an appropriate physical environ-
ment and a strong mental inheritance may be essential to the
rise of a great civilization, these two conditions may be at their

376 CIVILIZATION AND CLIMATE

best, and yet savagery may prevail because certain cultural
essentials such as iron or beasts of burden are lacking.

From this example of people who lived in one of the finest
climates but failed to make progress, let us turn to a case of
quite the opposite type. The white settlers in northern Aus-
tralia live in an almost tropical climate, which is poor as climates
are rated in this book. Nevertheless, they maintain a high civi-
lization and believe that they can raise that civilization still
higher. They are sometimes cited as proof that climate in itself
has little or no effect on health and that high civilizations may
arise regardless of climate. Nowhere else in all the world does so
large a body of white people, chiefly of British descent, live
within the tropics. Some two hundred thousand upon whom the
sun shines vertically at certain seasons are doing all the work
of life with their own hands. The men work in the fields, on the
docks, and among the cattle and sheep ; they cut sugar cane
under the tropical sun ; but their health docs not appear to
suffer. The women work in intolerably hot kitchens, often among
swarms of pestiferous flies, and in rooms where the blazing sun
on the unceiled roof joins with the kitchen fire to produce an
appalling temperature. Yet their death rate is phenomenally
low, and they give birth to unusually healthy children. In fact
the death rate in tropical Australia among all classes of the
population is one of the lowest in the world.

In order to make fair comparisons between regions where the
proportion of young people is unusually large and those where
the proportion is smaller, let us use what is known as a "standard
population." On this basis, which is universally accepted as the
fairest yet devised, the death rate in Queensland, the most
northerly state of Australia was 12.1 during the normal years
before the World War. Compare this with 10.5 in New Zealand
which is the healthiest country in the world, 12.4 in Denmark the
healthiest country in Europe, 15.2 in England, and 16.1 in the
part of the United States then included in the registration area.

AMERICA AND AUSTRALIA 377

In addition to this the medical examinations of young men who
enlisted during the World War showed that the Queenslanders,
even the tropical Queenslanders, were certainly not less robust
and vigorous than those from other parts of Australia and much
more so than were the similar men from England. Moreover,
children of the second and third generation have been born in
tropical Australia and are fine, active specimens. It seems to be
proved beyond question that the white man can live in tropical
Australia, that he can enjoy good health, and that white children
of high vitality can be born there, live there all their lives, and
become the parents of other children who seem to be equally
healthy.

These facts are highly important and encouraging. They hold
out a definite and not unreasonable hope that some day a branch
of the white race may live permanently within the tropics and
carry on all the work of life without the help of any colored race.
The fond hope of the Australians that there may permanently
be a "White Australia" is by no means without foundation. Does
this invalidate our hypothesis of the relation between climate
and civilization? Let us see what the facts really are. The Aus-
tralians are so much interested in the problem of a White Aus-
tralia that their Commonwealth Statistician has issued a special
bulletin on tropical Australia. He has also kindly placed at
my disposal certain other unpublished data. I am afraid that I
shall seem to the Australians to use these data in a wa}^ that they
will not like. But after all, what they really want is the truth. I
am perhaps as keen as they for a White Australia ; not only do I
want to see my own race show its ability to conquer every kind of
environment, but I am extremely anxious to see a conclusive
tropical experiment on a large scale. I want to know beyond ques-
tion whether white men can live permanently in a tropical coun-
try, conquer the diseases, and make progress without the help or
hindrance of any colored race. Tropical Australia affords by
far the greatest opportunity for such an experiment. It is

378 CIVILIZATION AND CLIMATE

peculiarly free from tropical diseases, and there seems to be no
good reason why it should fail to be kept free ; its colored in-
habitants are few in number and most of them belong to the
aboriginal Australian race which seems to be dying out and
which is quite easily kept separate from the whites ; and it is
under a government which is disposed to do everything possible
to make the experiment a success. I would give a great deal to be
able to see what kind of people live in tropical Australia one or
two hundred years hence. But if the tropical Australians of the
next few generations are to succeed indeed, if they are to avoid
great misery perhaps they must know exactly what they are
facing. To let the dream of a White Australia and the hope of
developing a great country blind ourselves to the facts is sure
to produce evil.

We have already seen how extraordinarily low is the death
rate of Queensland. If we take only the part of Australia that
lies within the tropics we find that during the years 1920, 1921,
and 1922 the following remarkable conditions prevailed.

Non-

Australia

tropical

Tropical

as a

Queens-

Queens-

whole

land

land

Birth rate per 1000 women aged 15-44

107

114

141

Infant death rate per 1000 births

63

56

55

Female death rate per 1000

8.66

8.03

7.90

Male death rate per 1000 persons

11.03

10.61

14.00

In the first three of these conditions non-tropical Queensland
is more healthful and vigorous than Australia as a whole, while
tropical Queensland is still more healthful and vigorous. In
other words the white women in the tropical part of Queensland
enjoy remarkably good health; they give birth to a relatively
large number of children ; and those children are uncommonly
healthy. What more could one ask as a recommendation of the
climate of a country ? The men, to be sure, have a much higher

AMERICA AND AUSTRALIA 379

death rate than those of other parts of Australia, but from the
standpoint of the future of the race they are much less im-
portant than the women and children. The Census Bulletin on
Tropical Australia attempts to explain the differences between
the death rates of men and women as follows: "As both sexes
should be affected by tropical conditions to something approach-
ing an equal degree, it is apparent that there must be some cause,
apart from purely tropical attributes, to account for the un-
favorable male death rate. . . . The great excess of men over
women in the tropical population suggests that there are many
men living under primitive conditions. Under such circumstances
there is too frequently a disregard for precautionary sanitary
measures, and a reckless neglect of the first symptoms of disease.
There are other cases, too, where it is impossible by reason of
great distance to procure medical or surgical assistance. It
would appear to be mainly to such circumstances as these that
the high male death rate in the tropical parts of Queensland is
due. Otherwise, the great difference between the mortality of the
sexes is not readily explained."

There may be some truth in this, for in tropical Queensland
there were in 1921 about 133 men for every 100 women. Never-
theless it must be a minor factor. Not only is it generally be-
lieved that frontier life is far more healthful than city life, but
much more than half the men of tropical Australia, as well as
a still larger proportion of the women, live in the towns where
there are physicians and often hospitals. Moreover, as appears
in the last sentence of the quotation from the census bulletin,
the isolation of the *nen on the frontier is appealed to as an
explanation of their high death rate only in the absence of any
other reasonable cause.

There is, however, another reasonable cause of the low death
rate among the women, a cause which the Census recognizes as
applying to the population as a whole, but does not recognize as
applying to women more than to men. That cause is natural

380 CIVILIZATION AND CLIMATE

selection. As the bulletin on tropical Australia puts it : "Per-
haps the most potent influence [in causing the low death rate of
tropical Australia] has been that sort of natural selection which
operates in the settlement of all new territories. It may be stated,
as a rule, that only the bolder and more virile of any community
will venture on the role of pioneer in new unsettled country, and,
when such a new country is a tropical one, to which popular
opinion generally ascribes more than the usual discomfort, the
physical standard of the settlers is probably more than ordi-
narily high."

The logical corollary of this statement is that among women
the selection of those who are physically fit is much stronger
than among men. Young men rarely think of their health ; young
women think of theirs a great deal, and between men and women
30 to 50 years of age the contrast is still greater. That the
women distribute themselves over tropical Australia in a dif-
ferent way from the men is obvious from the fact that in the
tropical part of western Australia, which is the most inacces-
sible and undeveloped, there were in 1921 about 524 men for
every hundred women ; in the tropical part of the Northern
Territory, which is not quite so remote, the number of men per
hundred women was 271 ; in tropical Queensland, which is still
better, 133 ; and in Towns ville, one of the two largest cities, 108.
On the other hand, in Rockhampton, which is the largest city
of tropical Australia and also the one located farthest south
and hence in the least enervating climate so far as the coast is
concerned, the number of men per 100 women was only 94. In
other words the women become relatively more and more numer-
ous as one gets away from two conditions, namely the frontier
and a truly tropical climate.

But this does not end the matter. The women tend not only to
concentrate in the best parts of tropical Queensland, but to
remain out of tropical Australia altogether, or else to go away
after once getting there. This is evident from the fact that

AMERICA AND AUSTRALIA 381

among children up to the age of about fifteen years the relative
numbers of boys and girls are normal. From that age onward
the number of men in proportion to women steadily increases.
For example, among young people 15 to 19 years of age there
are 110 boys for 100 girls, but at the age of 30 to 34* years this
proportion has increased to 155, while between the ages of 65 to
69 it rises to 24*9. Yet in a normal population at this latter age
there are only about 89 men for 100 women. Where such con-
ditions prevail it needs no demonstration to prove that the
women either have never come to tropical Australia or have
come and gone away. In either case it is certain that a large
number of the women are missing because of their health. One
has only to talk with those who are still there, or with the
Australian physicians, to learn how often the men like the cli-
mate while the women dislike it, and how often the health of a
wife or daughter causes a family either to move away, or to be
temporarily separated. The women frequently go to the cooler
South, while the men stay in the tropical North. Often the whole
family moves away from tropical Australia because the women
do not like the climate. The selective process which causes the
men in a new tropical country to be especially strong and vigor-
ous works still more effectively in causing the women and hence
the children to be even more markedly of the same type. The low
death rate in tropical Australia does not indicate that the
climate is favorable, but merely that natural selection has been
unusually vigorous.

Now look at the matter in still another way. Here is a little
table showing the death rate among persons from 15 to 49 years
of age in the three Australian states of Victoria, which is the
most southerly and hence the coolest, New South Wales which
also lies quite far to the south, and Queensland which lies so far
to the north that a quarter of its population is within the
tropics. The death rate has been calculated according to a
"standard population" so as to eliminate any differences arising

382 CIVILIZATION AND CLIMATE

from the fact that there are more young people in one section
than in another. The residents of the three states have been
divided according to their birthplace, the great majority of
Australians being natives of one of the three states, or else of
England or Scotland.

DEATH RATES PER "STANDARD POPULATION" AT AGE 15-49

YEARS IN VICTORIA, NEW SOUTH WALES, AND QUEENSLAND

ACCORDING TO PLACES OF BIRTH AND RESIDENCE,

1920, 1921, 1922.

Birthplace

Victoria
Male Female

Residence
New
South Wales

Male Female

Queensland
Male Female

Aver.
M. F.

Grand
Av.
both
sexes

Victoria
England
Scotland
New S. Wales
Queensland
Average

4.38
3.65

4.83

4.47

4.51

4.01

3.87
4.13
4.27
5.01*
4.26

3.60
3.78
3.94
3.76
4.42*
3.90

3.38
3.39
3.48
3.57
4.13*
3.59

4.28
4.66
4.85
4.84
4.87*
4.70

3.72
3.99
3.47
4.03

4.27*
4.00

4.
4.
4.
4.

4,

.09
03
54
36

.84*

3.70
3.75
3.69
3.96

4.47*

3.90
3.89
4.08
4.16
4.66*

Grand average for

the two sexes 4.39 3.75 4.35

The significant fact about the table is this : no matter whether
they reside in Victoria, New South Wales, or Queensland, the
people who were born in Queensland have a higher death rate
than those born in the other two Australian states or in Eng-
land or Scotland. If we combine the figures in the last column
it appears that the residents of Australia who were born in
Victoria, England, Scotland, and New South Wales have an
average death rate of 4.01. On the other hand the death rate
among the people born in Queensland, no matter where they
reside, is 4.66 or approximately 15 per cent greater. The high
death rate of Queenslanders in Victoria, or New South Wales,

*Maximura.

AMERICA AND AUSTRALIA 383

may be due to the fact that Queenslanders whose health is im-
paired migrate to those cooler states. But in that case the born
Queenslanders who remain in Queensland ought to be stronger
than the average and their death rate should be correspondingly
low. On the contrary, their death rate exceeds that of the people
born in any of the other four regions. In other words, although
the people whose residence is in Queensland have a very low
death rate, those who were born there have a high death rate.
Those who reside there were largely born elsewhere, and have
had the advantage of a strenuous though unconscious process of
natural selection. Those who were born there are more nearly
like the average of their race than were their selected parents,
for that is the universal biological rule, and they show what
seems to be the effect of the tropical climate.

Another significant fact in this connection is the number of
children per family. We have already seen that in proportion
to the number of women between 15 and 44 years of age the
birth rate in Queensland is high compared with the rest of Aus-
tralia, and is still higher in the tropical sections. That appar-
ently is due to the fact that the weaker women either leave their
husbands in the North arid go to the cooler South, or else per-
suade their husbands to move away. Those who remain are
strong and vigorous and have many children. But when we
examine the number of children born to persons who are Queens-
landers by birth, regardless of what part of Australia they later
reside in, we find quite a different condition. The following
figures show the number of children born to persons whose own
birthplace was in various regions and who died in Australia
during the year 1921.

Germany 6.3 Tasmania 5.0

Ireland 5.8 South Australia .... 4.5

Scotland 5.7 Victoria 4.1

England 5.4 Queensland 3.8

New South Wales . ... 5.1

384 CIVILIZATION AND CLIMATE

The high birth rate among the foreign-born Australians is
normal. There is nothing surprising about the differences among
the birth rates of people born in New South Wales, Tasmania,
and South Australia. The low rate among natives of Victoria,
however, is not easily explained unless it be that Victoria, more
than any of the other states, has lost the pioneer, frontier
quality, and hence has the relatively low birth rate which is
normal in old regions. For Queensland, no such explanation can
be advanced, for that state has more of the pioneer quality than
any other in our table. The most probable explanation is that
the birth rate is low for the same reason that the death rate is
high. Both conditions indicate that in spite of the strong selec-
tive process to which their parents were subjected, the native-
born whites of Queensland display qualities which suggest a
lower degree of physical stamina than prevails among the other
Australians, and which seem to be the result of the tropical
climate. If this is the case it strongly bears out the general
thesis of this book.

Even the part of Australia where the tropical Queenslanders
live is not tropical in the ordinary sense. The climate is indeed
quite trving by reason of the length and monotony of the summer
and because of the dampness near the coast and the great heat
in the interior. There is, however, a real winter in the regions
where most of the tropical Queenslanders actually live. For two
or three months the temperature almost everywhere falls to
between 40 and 50 during the night, while on the highlands
there is often frost. If a climate so mildly tropical can produce
such clear results, it seems probable that a thoroughly tropical
climate may handicap the white man to a far greater extent.
Nevertheless, I do not despair of the ultimate triumph of the
white man over the tropics, and much less do I despair of a
White Australia. The process of selection for climate has never
been really tested. All that has happened thus far has been
purely fortuitous. The hope of the future, I believe, lies in an

AMERICA AND AUSTRALIA 385

orderly and far-sighted selection of the right types of people, as
well as in the further development of tropical medicine and
hygiene.

But suppose the matter of health should be taken care of to
such a degree that the white race could live, thrive, and per-
manently reproduce itself within the tropics, would civilization
there advance as in cooler regions? I doubt it. In the first place
the most casual observation in any tropical region shows that
the white man as well as the colored does not act with so much
energy as in cooler climates. No matter how good his health, he
is forced to work comparatively slowly or for short hours, or
else he exhausts himself. It was a Queensland sheep "selector,"
the owner of twenty or thirty thousand animals, who first called
my attention to the "Queensland walk." We were watching a
man who slowly sauntered from the hotel to the railroad station.
"That's the way we all walk after we've been here in tropical
Australia a while." Not only do people almost of necessity walk
that way, but they often work that way. In tropical Australia,
just as in other tropical countries, all sorts of things are allowed
to lie at loose ends much more than among the same kind of
people in more stimulating climates.

The new point which I wish to make in regard to Australia
is that natural selection works against a tropical country, and
is working that way even more now than in the past. This sounds
like a contradiction to what has just been said about the strong
physique of the Queenslanders and the possibility of building
up a genuine tropical race of white men. But there is no con-
tradiction. I am talking now about the qualities which lead to
progress in civilization. As one goes about in tropical countries
and especially in Australia, one soon finds that the people who
complain most about the climate are, first, the women, and then
the professional men. In other words it is the people whose work
keeps them indoors. The rancher, planter, promoter, real estate
agent, cane cutter, and dock laborer do not seem to feel the heat

386 CIVILIZATION AND CLIMATE

half as much as do the minister, lawyer, doctor, engineer, and
teacher. These intellectual workers may and do adapt them-
selves to the climate and achieve results of high importance.
Nevertheless, almost without exception, they are on the watch
for a chance to get away to a pleasanter and more stimulating
climate. Many of them are ready to accept smaller salaries for
the sake of being in an atmosphere which is physically and
mentally more stimulating. In this fact, perhaps, lies one of the
greatest hindrances to the development of a high civilization
in countries with unfavorable climates. We have already seen
this selective process at work in the Bahamas, it is very active
in tropical Australia, and it is energetically at work in the
southern United States. The greater the wealth of a country,
the greater its intellectual and physical activity, the better its
transportation system and the more fully it picks out and edu-
cates all its bright sons and daughters, the more likely those
same sons and daughters are to move away from the unstimu-
lating environments and into those which most fully give scope
to all their faculties. In this respect, as in so many others, each
new advance of knowledge, each new step in the mastery of
nature, gives greater power to the regions whose climatic ad-
vantages already help them to be leaders. To him that hath
shall be given, and from him that hath not shall be taken away
even that which he seemeth to have.

CHAPTER XVIII
THE CLIMATIC HYPOTHESIS OF CIVILIZATION

WE come now to the final step in our study of the hy-
pothesis that climate ranks with racial inheritance and
cultural development as one of the three great factors in deter-
mining the conditions of civilization. As to which of the three is
most important it is impossible to say. The absence of good
conditions in any one respect may hold a country back, while
all three must rise to a high level if a race is to reach the highest
plane of civilization. In this last chapter I propose to consider
one final objection to the general hypothesis that climate has
been a determining factor in the geographical distribution of
human progress. That objection is that it seems impossible that
changes of climate can have exerted so great an influence as is
claimed in this book.

The first and most obvious effects of climatic changes are
economic. We have said little about them in this book because
they have been fully discussed in the publications numbered 2,
3, 7, 8, 10, 12, 14, 15, 17, and 22 in the list at the end of the
preface. A single concrete case will illustrate the matter. In
1909 I visited Palestine. That happened to be an unusually dry
year. During the months of April and May I rode scores of
miles on horseback amid fields of wheat that showed merely a
few scanty stalks three to six inches high. In Beersheba I talked
with a number of men who had tried to raise grain during a few
preceding years of good rainfall, but were financially ruined
by the drought of 1909. In Moab many villages which had been
recently reoccupied under the stimulus of the building of the

388 CIVILIZATION AND CLIMATE

Mecca railroad were being abandoned. The inhabitants were
moving back across the Jordan to the better watered parts of
the country. Coincident with this migration of the agricultural
people away from the drier areas along the borders of the desert,
there was a corresponding outward movement of the Arabs from
the desert. Because of the drought the great tribes from the
interior of the Arabian desert swarmed over the grain fields
on the eastern and southern border of Palestine, and let their
camels eat up the few scattered stalks of wheat that had sur-
vived. The poor villagers shot at the Arabs when they first
appeared ; then ran in terror to the villages. Soldiers were sent
to stop the Arabs, but in vain. The quarrels between the Arabs
and the peasants resulted in so many gunshot wounds that Dr.
Patterson of the Presbyterian Hospital at Hebron was swamped
with patients. I myself was in an Arab encampment when it was
raided by other Arabs. I was held up and my money demanded
at another time, and I came in personal contact with raiding
parties on three other occasions. Thus a single year of drought
brought distress upon all kinds of people ; it caused migrations
among those of the settled population who occupied the least
favored locations ; it set the desert tribes in motion ; and it
spread poverty, hunger, raids, war, wounds, and death on every
side.

Consider now what must have happened when Palestine
changed from a condition of relatively abundant rainfall in the
early part of the Christian era to a condition drier than that of
today in the seventh century. Remember that at certain periods,
such as the third century B.C. and the beginning of the seventh
century A.D., a relatively rapid change took place. Remember
ioo that Professor Butler of Princeton has been quoted above
as saying that an area of "20,000 square miles of land on the
(astern border of Syria is now too dry for permanent human
occupation, but was once more thickly populated than any area
of similar dimensions in England or in the United States outside

CLIMATIC HYPOTHESIS OF CIVILIZATION 389

the immediate vicinity of the large modern cities." This must
mean a population of at least 100 per square mile or 2,000,000
people. South of the region described by Butler there are addi-
tional thousands of square miles of similar dry land full of
ruins, while further east, in northern Mesopotamia, in the
region described above by Clay, the same is true. Suppose for
the sake of argument that 3,000,000 people lived in these areas
which are now too dry to be occupied.

When a diminution of rainfall gradually rendered this great
multitude homeless, a corresponding reduction must have taken
place in the productivity of the areas which are still inhabited.
Palestine today contains about 650,000 people and Syria about
3,000,000. There is abundant evidence that when the outlying
districts were habitable the population of the still habitable
areas was much greater than now. It seems, then, that at the
time of Christ, or at least some 400 years earlier, there may
have been over 6,000,000 people in these areas of Palestine and
Syria. Then what happened? Regions which had been sup-
porting perhaps 9,000,000 people in comparative comfort were
so stricken by drought that the homes of about three million
became almost uninhabitable, while the productivity of the
remainder was reduced nearly half. Thus only about three and a
half million people could live where 9,000,000 had lived before.
Of course so great a decline did not occur in a single generation
or even a single century. But a reduction of the capacity of the
country from 9,000,000 to 8,000,000 persons would produce
misery, famine, death, migration, raids, wars, and misgovern-
rnent to an almost incredible degree. Then would follow a few
decades or scores of years of recovery and again a reduction
in the capacity of the country to support people. Each such
blow must have dragged civilization downward, and the result-
ant chaos of the Dark Ages is by no means surprising.

The misery and discontent due to prolonged poor crops tend
to make people unstable not only politically, but in other ways.

390 CIVILIZATION AND CLIMATE

Religious bitterness is almost sure to increase under such con-
ditions. A portion of the community attributes its poverty to
the fact that there is something wrong with the present form of
religion. The rest are inclined to attribute their distress to the
wickedness of their neighbors who decry the old religion. Thus
bitterness and persecution are engendered. Those who become
discontented with the old religion are prone to accept the new
ideas propounded by religious enthusiasts. This seems to have
been the case when Mohammed made his appeal to the Arabs
after the prolonged period of increasing aridity which cul-
minated with a sudden access of dryness in the first half of the
seventh century. Without the genius of Mohammed that long
period of adversity might have come to an end without any
serious upsetting of the old conditions ; but without the dis-
content and unrest fostered by years of distress Mohammed
might have appealed in vain, for he would have had to speak to
men who did not desire change, instead of to those who ardently
longed for it.

The people of the desert may perhaps occasionally pour forth
from their homes without any special stimulation, but this is
doubtful. Anyone who has had much to do with the Arabs and
other desert nomads knows that when there is plenty of water and
grass there is very little raiding and fighting. On the other hand,
a single dry year causes raids in the fashion described above;
and prolonged dry periods appear to lead to great outbursts
of desert people like that which reached its crest under the influ-
ence of Mohammedanism, although it began before Mohammed
came into prominence. The Mohammedan migration was by no
means unique. Many of the barbarian migrations of earlier times
seem to have been impelled by similar dry periods.

At a later time two similar invasions took place. First, from
about 1000 A.D. to 1200 A.D. the climate of central Asia seems
to have grown decidedly drier, and distress and discontent
reigned among the people of the tents. At this time, as in the

CLIMATIC HYPOTHESIS OF CIVILIZATION 391

days of Mohammed, no great concerted movement might have
arisen, had it not been for the ambitions of one man. Jenghiz
Khan may have been no more ambitious and no abler than other
gifted men of his race, but he happened to live at a period when
natural calamities had brought his people to a condition of dis-
content favorable to his aspirations. This condition, it would
seem, was an important help in enabling him in a few years to
arouse all the tribes of the steppes and deserts, and sweep over
Asia, bringing almost unparalleled devastation. Three centuries
later, at the beginning of the sixteenth century, another ambi-
tious Asiatic, Baber by name, arose in Turkestan and emulated
his great ancestor Jenghiz Khan. In Baber's case, also, physical
conditions seem to have favored his projects, for after a period
of improved climate, a rather rapid decrease in rainfall cul-
minated at the time of his conquests. How much this had to do
with the fact that he was able to conquer India and establish
there the Mogul dynasty has not yet been thoroughly investi-
gated. It seems clear, however, that this dry period should at
least be carefully considered before any conclusions are drawn
as to Baber and the Mogul conquest of India.

From great wars and movements of the nations let us turn
back to the individual people, and see how increasing aridity
may effect a race physiologically. The chief effect is probably
produced by the selective action of disease. Insidious diseases
such as malaria, consumption, neurasthenia, and the like are
presumably among the most important sifters of the wheat from
the chaff in the physical make-up of a nation. Although malaria,
for example, docs not kill people in any such spectacular fashion
as do the great epidemics, it may be far more dangerous in its
ultimate effects. The plague passes over the land and is gone;
the dead are dead, and the living have suffered no serious injury.
Malaria on the contrary, hangs on year after year, not killing
its victims, but sapping their energy and vitality. The presence
and the abundance of malaria are closely associated with cli-

392 CIVILIZATION AND CLIMATE

mate and topography. Without entering into any discussion of
the origin of malaria, let me point out how a change toward
aridity in a country like Greece and, to a less extent, Italy,
would probably foster the disease.

Malaria is preeminently a disease of tropical and subtropical
countries whose climate is characterized by alternate wet and
dry seasons. Except in the perennially moist portions of the
tropics, the streams of such regions are subject to seasonal
floods which spread over wide areas for a short period and then
disappear, leaving innumerable stagnant pools and swamps,
ideal breeding places for the anopheles mosquito. Permanent
bodies of water usually contain fish which eat the mosquito
larvae and reduce their numbers, or else the water moves suffi-
ciently to carry away most of the eggs that are laid in it. When
the climate of a subtropical country becomes drier, the condi-
tions which favor the mosquito arc intensified. This is due in
large measure to the fact that a diminution of the rainfall les-
sens the amount of vegetation upon the slopes and thus allows
the soil to be washed away rapidly. The streams are thereby
overloaded and begin to fill the valleys with sand and gravel.
This causes the flowing water to wander hither and thither over
broad flood plains in innumerable channels, which form pools
when the floods assuage. Or it may be that the water loses
itself in marginal swamps. The streams also become intermittent,
and no longer contain large quantities of fish. Thus everything
cooperates to reduce the number of streams which flow steadily
throughout the year, and to increase the number of bodies of
stagnant water in which the mosquitoes may live. This in itself
may produce most widespread effects. How great they are may
be judged from the success of the United States government in
eradicating malaria at Panama by the opposite process of
reducing the number of places where mosquitoes can breed.

At the present time malaria is endemic in Greece and Rome.
That is, it is always there, and is looked upon as one of the

CLIMATIC HYPOTHESIS OF CIVILIZATION 393

necessary diseases of childhood, much as we look upon measles.
Sir Ronald Ross of the Liverpool School of Tropical Medicine
is responsible for the statement that nearly half the people of
Greece have suffered genuine injury from malaria, and in Italy
the case is scarcely better. Up to the age of puberty children
are attacked by it every autumn. They grow weak and sallow,
their spleens are permanently enlarged, and their vitality is
lowered for life. No one who has known much of malaria will
question the severity of its results and the length of time which
elapses before they are eradicated even in the case of adults.
In spite of quinine, which has come to our aid in modern days,
malaria is one of the most insidious of diseases. Every traveler
who is really familiar with the Orient knows how the sufferers
from malaria lie and groan for days, and may have little energy
for months. They go languidly to the necessary tasks, and as
soon as possible sit down to rest with open, stupid mouths. Phy-
sicians agree that it is impossible to expect much initiative or
energy from a nation in which for centuries almost half of every
generation has been devitalized by this baneful disease.

From a .painstaking study of classical authors W. H. S.
Jones* has concluded that up to about 400 B.C. in Greece and
200 B.C. in Rome, malaria was almost unknown. Then it ap-
peared, and during the succeeding century or two became com-
mon. At first it attacked adults, which shows that it was a rela-
tively new disease, which was still epidemic and not endemic,
or else, we would add, that Greece was on the very border of its
habitat. Later it became permanently located in the respective
countries and attacked chiefly children, the older people having
become immune after suffering in childhood. It is noticeable that
the introduction of malaria coincides with the beginning of the
weakening of Greece and Rome, and the time when it became

* Malaria: A Neglected Factor in the History of Greece and Rome.
Cambridge, England, 1907.

394 CIVILIZATION AND CLIMATE

endemic, in Greece at least, is synchronous with the epoch when
the luster of the ancient names became irretrievably dimmed.

Ross and Jones are of the opinion that, along with various
other factors, malaria was one of the important causes of the
fall of Greece and Rome. The growing effeminacy and lightness
of the Greeks, and the brutality of the Romans, are just the
effects which they think would be produced upon people of the
respective temperaments of the two races. The case is so strong
that one can scarcely resist the conclusion that this pathological
factor may have played an important part in the psychological
changes which appear to have accompanied the decline of civili-
zation and of population in both Greece and Rome. It would
be unwarranted to assert that the increase in the amount and
severity of malaria was due wholly to climatic changes. Other
influences, such as contact with Egypt and the introduction of
slaves, may have been equally effective. Nevertheless, it seems
probable that the spread of the disease in both Greece and Rome
proceeded most rapidly when a change of climate not only ren-
dered the topography of the valleys and the behavior of the
streams more favorable than hitherto to the propagation of
the anopheles mosquito, but likewise weakened the physique of
the people so that they readily succumbed to disease.

Natural selection presents a still more insidious way in which
the change from relatively moist, stormy, cool conditions to
those of aridity may have affected the Greek, Roman, and other
races. In the opinion of many scholars one of the most important
factors in the greatness of these powers was the presence of a
race of blond northern invaders. Take the case of Greece. These
northern Achseans apparently came into the country in the
thirteenth century B.C. Their coming may have been influenced
by the dry period of which we find some evidence at that time
both in America and Asia. After their arrival the climate on
the whole, although with many fluctuations, appears to have
become more propitious. Up to the third century it continued

CLIMATIC HYPOTHESIS OF CIVILIZATION 395

to be favorable. Then it became more arid. We have seen how
sensitive people are to climatic environment. The negro would
apparently disappear in the northern United States were he not
replenished from the South. The Scandinavian does not seen)
to prosper greatly in the dry, sunny portions of the United
States ; he is there subject to diseases of the skin and nerves
which appear seriously to deplete his numbers in a few genera-
tions ; whereas in the rainy Northwest, which resembles his
native habitat, he thrives greatly both in body and estate. It
was probably the same with the northern invaders in Greece,
So long as the climate was propitious they flourished and lent
strength to the country. Then, when conditions became less
favorable, the unseen ravages of malaria and other diseases
presumably attacked them with especial severity, and in the
course of centuries they gradually disappeared. Today blond
Greeks are almost unknown, although classical literature and
many fair-haired old statues demonstrate their presence for-
merly in considerable numbers.

Thus far we have been discussing changes of climate, and
have overlooked the fact that even though the climate remains
constant, man's cultural progress may alter the location of
the most favorable climate and hence of the centers of civiliza-
tion. This depends upon the fact that the climate which is most
favorable to a race in a primitive stage of development is not
the most favorable for the same race in a higher stage of culture.
GilFillan, in an article on "The Coldward Course of Progress,"*
has presented an interesting study of the way in which man's
cultural heritage, in the form of a constantly growing command
over nature, has enabled mankind to advance farther and farther
into regions of low winter temperature. In a savage state, with-
out fire, clothing, or shelter other than caves, man is not likely
to thrive where the winter temperature remains for any length

* Political Science Quarterly, 1920. The same subject is discussed from
another standpoint in Stefansson's The Northward Course of Empire.

396 CIVILIZATION AND CLIMATE

of time as low as freezing, and his main development is almost
sure to be where the coldest months are not much below the
physical optimum of 64 or 65 F. With each step of progress
he is able to endure greater extremes of climate. The open fire,
the grate, the stove, the furnace, and the central heating plant
are all steps toward a condition where man is able to resist the
cold, no matter how severe it may be. The use of the skins of
animals for clothing ; the invention of the arts of spinning and
weaving; the domestication of the sheep, goat, and camel; the
cultivation of flax, and especially cotton ; the invention of power
looms and of the steam engine ; all these are important steps
which have tended to make man independent of climate. So too
are the inventions which enabled mankind to close the mouths of
the caves in which he lived, and then to build huts, thatch them,
cover them with close-fitting shingles, construct them of thick
bricks, or otherwise make them warm and weatherproof. In like
manner man's power to resist unfavorable climates has been
increased by every invention that has made it possible to trans-
port food cheaply and quickly, and to preserve it for long
periods. Today it is not difficult for large communities to live
in health and comfort in the cold parts of the earth where their
naked ancestors would have perished in a few days. Human
energy may perhaps ultimately rise higher in those regions than
in others, for many of the diseases of warmer regions do not
flourish there, and the constant changes from the warmth within-
doors to the cold without seem to be highly stimulating and
healthful, provided they are properly guarded against. More-
over, in cold regions man can be free from the effects of undue
heat against which he has thus far shown little ability to pro-
tect himself. Thus it appears that even without changes of
climate the highest civilization would have tended gradually to
migrate farther and farther north. How far this tendency will
go is not yet clear, but the far northern location of such pro-
gressive people as the Scotch, Canadians, Swedes, Norwegians,

CLIMATIC HYPOTHESIS OF CIVILIZATION 397

Icelanders, and Finns seems to suggest that the limit has not
yet been reached.

One other phase of GilFillan's work deserves careful notice.
He points out that when civilization takes a backward step it
also tends to move equatorward. For instance, about two hun-
dred years before Christ the leadership of the world had passed
from Greece, while Rome was falling into a state of confusion
and retrogression from which its recovery was never complete,
even in the days of Augustus. At that time the intellectual
leadership went back toward the south, and the Alexandrines
held aloft the torch of learning. They did not, to be sure, make
any great intellectual or material innovations, but they at least
kept the spirit of progress alive. In the same way in the Dark
Ages the intellectual leadership of the world went back to North
Africa where lived such men as Saint Augustine and his con-
freres.

GilFillan suggests an interesting explanation of what happens
when the most progressive nations of any given period are
suddenly thrown backward by barbarian invasions, climatic
changes, or similar occurrences. At such times regions which
have formerly ranked high stand out once more as leaders.
When the cooler countries fall to a low estate, the eclipsed
warmer countries come into prominence. They still represent the
highest type of civilization that is compatible with their climate,
even though they have not been able to make that type appear
great or important because the cooler and more energetic
countries have not given them a chance. Thus in the Alexandrine
period, when Greece had fallen and even Rome was at a low ebb,
northern Egypt took the lead because the stage of progress to
which the world had regressed was suited to Egypt's climate.
How true this hypothesis may be, it is hard to say. The migra-
tion of the more competent Greeks to Egypt may explain the
Alexandrine period. Greece was becoming uninhabitable and
poverty stricken by reason of an adverse climatic change, but

398 CIVILIZATION AND CLIMATE

Egypt with its perennial river was still prosperous and attrac-
tive. Nevertheless, there seems to be much truth in the idea that
man's social progress constantly alters his relation to climate.
In the past great inventions have helped chiefly in enabling man
to overcome low temperature; in the future, perhaps, they will
help him in equal measure to overcome high temperature, dry-
ness, and monotony.

The last matter to engage our attention is the effect of
changes in storminess. In reading what has been said about
Egypt, for example, the reader has probably said to himself:
"This book claims that the people there were once energetic and
healthy because of frequent storms. If this is so, the rainfall
must have been much more abundant than now. For the sake
of argument we grant that at certain periods the precipitation
was greater than at present, but we cannot possibly believe
that it increased to the extent demanded by this hypothesis."

This objection is so important that I have made a special
investigation to determine the exact relation between stormi-
ness and rainfall. The word storm, as already explained, does
not primarily mean rain. To the meteorologists a storm is an
area of low barometric pressure which is always accompanied
by inblowing winds, and usually, but not invariably, by rain,
Even if there is not a drop of rain, a storm may otherwise be
fully developed and may cause strong winds which give rise to
changes of temperature and humidity. The matter is well illus-
trated by comparing Colorado and Georgia. According to Kull-
mer's maps, Colorado is one of the stormy parts of the United
States. It is crossed by the centers of about three times as many
storms as Georgia. Yet its rainfah 1 is only about one third as
great. In other words, Georgia has nine times as much rain in
proportion to its storms. The reason is simple. Colorado lies
far from the ocean, and the air which rises in the centers of
its areas of low pressure has already lost most of its moisture.
Georgia, on the other hand, lies so near the broad Atlantic

CLIMATIC HYPOTHESIS OF CIVILIZATION 399

and the warm Gulf of Mexico that storms draw in great quanti-
ties of moisture. The lands around the eastern Mediterranean
and in western Asia have a climate far more like that of Colo-
rado than of Georgia. Part, to be sure, are near the Mediter-
ranean Sea, but that furnishes only a little water compared
with the great oceans.

Another comparison shows that in the entire United States
if the number of storms increases 20 per cent during a given
series of years, the rainfall increases scarcely 10 per cent. In
the arid southwest, however, the ratio is larger, the increase in
storminess for all available stations being three times as great
as the increase in rainfall. These figures apply to present con-
ditions where the variations are slight and of short duration.
If larger, more permanent changes took place, the rapidity of
the atmospheric circulation would probably be so much in-
creased that the ratio between increase in storminess and in rain-
fall would be greater than at present, and might be four, or
even six to one. The drier the region, the greater would probably
be the ratio.

Suppose that the ratio were four to one in Syria. A change
which would increase the storminess by 200 per cent would
involve a change of rainfall amounting to 50 per cent. That is,
the number of stimulating changes would be three times as
great as now, while the rainfall would only increase by one half.
Such a change would render the Phoenician coast much more
stimulating than at present. It would also increase the agri-
cultural wealth, and would cause the limits of permanent habi-
tation to advance some miles into the desert. There the stimu-
lating influence might be less than on the coast, because the
storms might be somewhat interrupted by crossing the moun-
tains. Nevertheless, it would be important. The effect of in-
creased storminess upon habitability, however, would be much
less noticeable than upon energy or than in better watered
regions. An increase of 50 per cent in a rainfall of twenty

400 CIVILIZATION AND CLIMATE

inches, such as is now enjoyed by many parts of Syria, would
raise the precipitation to thirty inches, a figure which permits
great prosperity. A corresponding change in Egypt would in-
crease the rainfall of Alexandria from 8.8 inches to 13.2, that
of Port Said from 3.3 to 5, and that of Cairo from 1.3 to 2.
Farther east in Mesopotamia, Bagdad would change from 9 to
13.5, while in Persia, Teheran would rise from 10 to 15 and
Ispahan from 5.2 to 7.8. Deserts would still be deserts. They
would be easier to cross than at present, and the number of
inhabitants might be greater, for there would be more pasturage
for camels and sheep. The springs would also be larger and more
permanent than now, and some new ones would appear. Yet the
predominant feature would still be great wastes of blowing sand
and barren gravel ; the people would have to be nomads ; and
those who entered the desert would frequently encounter rigors
like those of today. Thus a large change in the stimulating quali-
ties of a subarid or desert climate is possible without a change
of rainfall greater than that for which there seems to be good
evidence.

Consider more specifically exactly what happens when stormi-
ness increases in regions such as western Asia and eastern Cali-
fornia. The general conditions at the time when Greece was in
its prime, for example, appear to have been approximately as
follows: (1) The average temperature for the year as a whole
was probably a little lower than at present, but it is doubtful
whether the difference amounted to more than two or three
degrees at most. (2) The temperature of the seasons may have
varied somewhat more than that of the year as a whole. The
greater mixing of the air from wide areas by means of the winds
that accompany storms presumably lowered the summer tem-
perature a little over the lands and raised the winter tempera-
ture. (3) The amount of rain was apparently considerably
greater than now. Doubtless the rain then, as now, came chiefly
in winter, while the summers were very dry just as at present,

CLIMATIC HYPOTHESIS OF CIVILIZATION 401

but the length of the rainy season appears to have been in-
creased. That is, the rains began earlier in the autumn and
lasted longer in the spring than at present. (4) The amount of
atmospheric moisture over the lands was presumably greater
than now. This would arise partly from the increased cloudiness
and rainfall and partly from the fact that the winds were pre-
sumably more efficacious in bringing moisture from the neighbor-
ing seas. (5) The winds were apparently stronger than at pres-
ent, arid varied in direction much more than now because of the
frequent storms. (6) The greater storminess and the more fre-
quent changes in the winds, as well as their greater strength,
must have caused great or at least frequent variability of tem-
perature as well as variations in other respects. Such variability
would occur even in summer when the storms passed to the north
of the lands that were most progressive two or three thousand
years ago; it would be still more pronounced in the rainy
season.

Let us apply these generalizations to a specific case. Suppose
that from 500 to 400 B.C. the climate of Athens differed from
that of the present in the following respects :

1. Temperature of July 77 F. instead of 81 F.; January
48 instead of 46. Mean temperature of the year 62.0 instead
of 63.1.

2. Relative humidity at all seasons 10 per cent higher than
now, January the moistest month, 84 per cent instead of 74 per
cent ; July the driest month, 58 per cent instead of 48 per cent.
As a matter of fact the change was probably not the same at
all seasons, but I wish to keep our example simple.

3. Annual rainfall 22 inches instead of 15, ranging perhaps
from 3.3 inches in November to 1.0 in July instead of 2.9 in
November and .03 in July.

4. Number of storms twice as great as now.

Such a change seems conservative. It is scarcely more than

402 CIVILIZATION AND CLIMATE

the normal variation from one year to another. Nevertheless,
when we calculate its effect upon health in the same way that
we have calculated the data for our map of climatic energy
based on the effect of the seasons in American cities (Figure 86),
the result is astonishing. From a climatic level only equal to that
of Augusta in Georgia and Vicksburg in Mississippi, Athens
rises to a level practically the same as that of New York and
Chicago, the best regions in North America aside from the coast
near Newport; a slight farther increase in storminess or a
farther lowering of the summer temperature two or three de
grees more would make Athens rival Paris and Berlin; and if
the storminess should be three times as much as at present, which
would be a small matter compared with the differences between
one year and another at places like San Francisco, the heali fa-
fulness and stimulating qualities of the climate of Athens would
rival those of southeastern England, which seems to be well-
nigh the most favored place in the whole world. When we recall
that according to Figure 41 the estimated death rate in Greece
is nearly twice that of England, the significance of such a
change is apparent. It would presumably get rid of malaria to
a large extent ; it would stimulate the Greeks to a degree of
persistent activity quite foreign to the country at present ; it
would raise the economic level and correspondingly improve the
diet of the people ; and it would give to the Greeks a spirit of
enterprise, a physical vigor, and a mental activity which would
probably soon enable them to take advantage of all sorts of
modern discoveries which they now use half-heartedly and in-
effectively, if at all. The slightness of the climatic change neces-
sary to produce important results seems to give to the conclu-
sions set forth in the present edition of this book a reliability
much greater than was possible in the first edition, before the
health of American and European cities had been studied in its
relation to the weather.

As we come to the end of this volume I am well aware that to

CLIMATIC HYPOTHESIS OF CIVILIZATION 403

those who accept the climatic hypothesis, it may seem depress-
ing. To the dweller in the less favored parts of the world it
may appear to sound the knell to his hopes for great progress
in the land that he loves. To his brother in the center of modern
activity a most disquieting vision of possible retrogression is
disclosed. If our reasoning is correct, man is far more limited
than he has realized. He has boasted that he is the lord of
creation. He has revelled in the thought that he alone among
created beings can dwell in the uttermost bounds of the earth.
One more of the bulwarks of this old belief is now assailed. Man
can apparently live in any region where he can obtain food, but
his physical and mental energy and his moral character reach
their highest development only in a few strictly limited areas.
The location of those areas appears to have varied greatly in
the past ; it may vary greatly in the future. In a thousand years,
for all that we can tell, so the prophet of evil will say, no
highly favorable region may exist upon the globe, and the
human race may be thrown back into the dull, lethargic state
of our present tropical races. Even without so dire a calamity,
the location of the regions of greatest climatic energy may in
a few hundred years change again to Egypt, Mesopotamia, and
Guatemala. The consequent rise of new powers and the decline
of those now dominant may throw the world into a chaos far
worse than that of the Dark Ages. Races of low mental caliber
may be stimulated to most pernicious activity, while those of
high capacity may not have energy to withstand their more
barbarous neighbors.

Even if such extreme disasters should not occur, the prospect
is depressing. Take such a favored country as the United
States. In the South we find less energy, less vitality, less edu-
cation, and fewer men who rise to eminence than in the North,
not because southerners are in any way innately inferior to
northerners, but apparently because of the adverse climate. In
the far West people seem to be stimulated to such a degree that

404 CIVILIZATION AND CLIMATE

nervous exhaustion threatens them. In the North we see still
another handicap. In spite of a wonderfully stimulating climate
most of the year, the people suffer sudden checks because of the
extremes of temperature. These conditions favor nervousness,
and worst of all they frequently stimulate harmful activities.
That, perhaps, is why American children are so rude and bois-
terous, or why so staid a city as Boston has six times as many
murders as London in proportion to the population. Our coun-
try takes immigrants of every mental caliber, and then stimu-
lates some to noble deeds and others to commit murder, break
down the respect for law, and give us city governments that
shame us in the eyes of the world. All these things would appar-
ently not happen to such an extent were our climate less bracing
and did not its extremes often weaken the power of self-control.
Other lands also have their drawbacks. Germany is much like
the eastern United States, although not so extreme. France on
the other hand is less stimulating. England suffers from too
great cloudiness, and in Ireland this becomes a factor of serious
import. If the best parts of the earth have such climatic dis-
advantages, what shall we say of Russia, weighted down with
benumbing cold and comparative monotony or with changes so
extreme that they are harmful? What of China under a much
heavier handicap of monotony ; or of tropical lands burdened
most heavily of all? If climatic conditions influence character
as we have inferred, does not our hypothesis weaken man's moral
responsibility? Will not people more than ever ascribe their
failings to nature, and so excuse themselves? In the favored
regions will not men become increasingly arrogant and over-
bearing, because they will be surer than ever that the rest of
the world cannot resist them? If all these sad results are possible,
is it well to know that climate so strongly influences us? We
cannot change the climate, so why ascribe to it such great effects
merely to destroy hope in some and moral responsibility in
others ?

CLIMATIC HYPOTHESIS OF CIVILIZATION 405

The answer to these questions may be put in the form of a
parable. Ages ago a band of naked, houseless, fireless savages
started from their warm home in the torrid zone, and pushed
steadily northward from the beginning of spring to the end of
summer. They never guessed that they had left the land of con-
stant warmth until in September they began to feel an uncom-
fortable chill at night. Day by day it grew worse. Not knowing
its cause they traveled this way or that to escape. Some went
southward, but only a handful finally returned to their former
home. There they resumed the old life, and their descendants
are untutored savages to this day. Of those who wandered in
other directions all perished except one small band. Finding
that they could not escape the nipping air, the members of this
band used the loftiest of human faculties, the power of conscious
invention. Some tried to find shelter by digging in the ground,
some gathered branches and leaves to make huts and warm beds,
and some wrapped themselves in the skins of the beasts that
they had slain. Soon these savages had taken some of the greatest
steps toward civilization. The naked were clothed ; the houseless
sheltered ; the improvident learned to dry meat and store it with
nuts for the winter, and at last the art of making fire was dis-
covered as a means of keeping warm. Thus they subsisted where
at first they thought that they were doomed. And in the process
of adjusting themselves to a hard environment they advanced
by enormous strides, leaving the tropical part of mankind far
in the rear.

Today mankind resembles these savages in certain respects.
We know that we are limited by climate. As the savages faced
the winter, so we are face to face with the fact that the human
race has tried to conquer the arctic zone, the deserts, and the
torrid zone, and has met with only the most limited success.
Even in the temperate zone he has made a partial failure, for
he is still hampered in hundreds of ways. Hitherto we have
attributed our failure to economic conditions, to isolation and

406 CIVILIZATION AND CLIMATE

remoteness, to racial incapacity, or to specific diseases. Now
we see that it is probably due in part to lack of energy or to
other unfavorable effects produced directly upon the human
system by climate. This is no reason for despair. We ought
rather to rejoice because, perhaps, we may correct some of the
evils which hitherto have baffled us.

Again and again in our discussion of factories and other
matters we have come upon ways in which a change in our
methods may do much to overcome the harmful effects of cli-
mate. I do not propose to enumerate them, for the specific
application of our results may well be deferred until we know
whether our main hypothesis is likely to stand. Yet one or two
general lines of progress may properly be pointed out. Take
the harmful winters of the northern United States. It is highly
probable that the loss of energy which occurs at that time may
be largely avoided, or at least greatly diminished. Much of it
arises from the fact that after the wonderfully stimulating
autumn weather, when we have been living under almost ideal
conditions of mean temperature, of humidity, and of variability
from day to day, we suddenly begin to heat our houses. We
create an indoor climate of great uniformity, of unduly high
mean temperature, and of the most extreme aridity. All these
conditions are harmful. If our houses were kept at lower tem-
peratures, if the temperature were varied from day to day, and
if the humidity were kept at the optimum, we should increase
our efficiency greatly. We should be comfortable, also, for with
proper humidity, and with changes from day to day, we should
not feel the need of the high temperatures which we now require
because the extreme dryness forces the body to give up much
more heat than would be demanded by air of greater humidity.
Moreover, the uniform dryness within doors does almost untold
harm in parching the mucous membranes and thus rendering us
peculiarly liable to colds, grippe, and similar ailments which
often lead to serious diseases such as pneumonia and tubercu-

CLIMATIC HYPOTHESIS OF CIVILIZATION 407

losls. Of course we could not entirely avoid colds by the method
here suggested, but we surely could diminish them. In the
autumn before our houses are heated, colds are comparatively
rare, and the same is true among people who live out of doors
in winter. If the conditions inside our houses could be like those
that prevail in the autumn, the general health of the community
would probably be much improved. In this one way there might
be a saving not only of millions of dollars' worth of valuable
time, but of an immense amount of nervous energy which is
wasted because persons who are irritated by colds do or say
things that they would scorn under normal conditions.

Along still other lines great improvements might be possible.
For instance, in many factories the same amount of work is
expected each month. Hence, at certain seasons many opera-
tives, especially girls, work harder than they ought, while at
others they do not work so hard as they easily could without
special effort. If factories were run in accordance with a well-
established seasonal curve of energy, we should find the ma-
chinery running slowly in winter, faster in the spring, and in
May perhaps 10 or 15 per cent faster than in January. Then
in the summer it would run more slowly than in May, but not
so slowly as in winter. Finally, in the autumn it would run at
greater speed than at any other time of year. The operatives
would scarcely be conscious of the difference, and they would
probably do more work and preserve their health better than
under the present system.

If our hypothesis is true, it is likely to prove helpful not only
to places where the climatic disadvantages are slight, but where
they are great. Consider regions which have a winter of great
severity, but an invigorating summer. Contrast them with
places where the summer is too hot, but the winter favorable.
Russia and Mesopotamia may serve as examples. Today we
already have a small number of people who move back and forth
each year between places of this sort, for instance northern

408 CIVILIZATION AND CLIMATE

Germany and the Riviera, New England and Florida. Unfor-
tunately, those who do this are usually not the workers, but the
idlers or those whose work is almost finished. In the future,
however, if the principles here laid down find acceptance, we
may expect that such interchanges will take place on a scale
to stagger the imagination. Not only the leisure classes, but
laborers and farmers may thus move back and forth. In winter
most of Russia's peasants have little to do, and their enforced
idleness is harmful. They might go to Mesopotamia where most
of the farm work is done between October and May. If people
could move thus from place to place, not only would there be
an enormous increase in production, but many other benefits.
The part of the population that moved would be stimulated, not
only by the change of climate, but by contact with other races
and new methods. They would be more tolerant, more pro-
gressive, and more eager for education. Both countries would
benefit by such an interchange of workers, and much of the
handicap of places like Russia might possibly be overcome.
Perhaps the day will come when only the poorest families will
stay in an unfavorable climate more than a few years at a time
without going at least for a season to some place where they
will receive new stimulus.

In tropical countries the chances for improvement are at a
maximum. Already most Europeans and a few natives appre-
ciate the necessity for spending part of the year among the
mountains or in a climate different from that where they usually
live. For the most part the lowlanders go to the highlands, but
in lofty plateaus like Mexico it is not uncommon for foreigners
to take a run down to the coast for a change of air. Of course
altitude has much to do with this, but even though Vera Cruz
has a bad climate for permanent residence, it is stimulating for
a short while when one comes from a wholly different environ-
ment. In the future we can scarcely doubt that this method of
overcoming the evil effects of a tropical climate will be resorted

CLIMATIC HYPOTHESIS OF CIVILIZATION 409

to on a vast scale, not only by foreigners, but by the more
intelligent portion of the natives.

In the warmer parts of the earth there in another side to the
question. Mankind needs not only seasonal changes, but varia-
tions from day to day, or week to week. Two methods of obtain-
ing these suggest themselves. One is by cooling the interiors of
houses. Today this is done on a small scale by shutting out the
sun and sprinkling water to cause evaporation. There is no
reason why the same result should not be produced on a large
scale. We already know how to cool houses as well as to heat
them. We do it in ice-plants. A thousand years ago men would
have laughed at the idea that hundreds of rooms would some
day be heated by a single fire, yet we see it in every office building
or hotel. In equatorial regions there is as much reason for
equipping the houses with coolers as there is in temperate regions
for equipping them with heaters. In both cases uniformity of
temperature is apparently a mistake, for moderate changes
from day to day appear to be favorable. Even though a man's
work may be out of doors, it seems probable that he would be
much stimulated and much better enabled to work hard in the
heat if he could sleep in a comparatively cool house. If he lives
where the climate is too damp, he would be benefited by having
the house relatively dry, just as the northerner in winter appar-
ently ought to have his house more moist than is now his habit.
Both need to enjoy the optimum conditions.

A second method of obtaining frequent changes may possibly
prove of much importance. Today the seacoast in many regions,
for example on the Atlantic shore of America from New York
to Boston, is bordered by an almost continuous line of houses.
At first people went to these only in the summer. Now many
go for week-ends at almost all seasons. Fifty years ago such a
thing was almost unknown. Fifty years hence it will probably
be many times more prevalent than now. In tropical countries
millions of people may not only move to other climates during

410 CIVILIZATION AND CLIMATE

part of each year, but many may move back and forth from the
lowlands to the highlands every few days. Their work may be
arranged so that almost every family can spend week-ends in
the highlands and the rest of the time in the lowlands. In Sar-
dinia, in order to escape the malaria of summer, many villages
are today completely duplicated in the lowlands and highlands,
churches, houses, shops, everything in duplicate. Moreover,
aside from farming, mining, lumbering, and the like, most of the
other kinds of work can be located where the climatic conditions
are best, a matter which is becoming increasingly easy as our
facilities for communication improve. Yet if these are to be
within the tropics, the people engaged in them must have an
opportunity to obtain the stimulus of changes. Perhaps they
will frequently go from their places of work in the highlands to
the neighboring lowlands or to the high mountains.

The expense of such a system of having two homes for a large
part of the population will doubtless be enormous, but that is
relatively unimportant. If the farmers of the tropics were as
efficient as those of the temperate zone, one man's labor would
produce two or three times as much as in Europe or the United
States. If white men can devise a means whereby they can live
in the torrid zone and retain approximately the energy which
they possess in their own countries, or if they can largely in-
crease the efficiency of the natives, they can afford to spend
t-normous sums in creating favorable conditions. How we shall
go to work in detail cannot yet be determined, but that will
easily be discovered. For the present it is enough to see that
the hypothesis of climate as a condition of civilization is far
from depressing. It holds out hope that the inhabitants of even
the most favored parts of the temperate zone may improve their
condition. It holds out still more hope that the people of the
less favored parts of that zone and of the subtropical zone may
be benefited. And it holds out far the greatest hope to those

CLIMATIC HYPOTHESIS OF CIVILIZATION 411

who dwell in the tropical regions which now are the most hope-
less.

If our hypothesis is true, man is more closely dependent upon
nature than he has realized. A realization of his limitations,
however, is the first step toward freedom. In suggesting pos-
sible ways of obtaining a new ascendancy over climatic handi-
caps we have dealt largely with material matters. Bound up
with these, and far more important, are great moral issues. We
are slowly realizing that character in the broad sense of all that
pertains to industry, honesty, purity, intelligence, and strength
of will is closely dependent upon the condition of the body. Each
influences the other. Neither can be at its best while its com-
panion is dragged down. The climate of many countries seems
to be one of the great reasons why idleness, dishonesty, immo-
rality, stupidity, and weakness of will prevail. If we can con-
quer climate, the whole world will become stronger and nobler.

APPENDIX

APPENDIX

(A) CONTRIBUTIONS TO THE MAP OF
CIVILIZATION

1. AMERICANS

J. Barrell, geologist, New Haven, Conn.

P. Bigelow, traveler and author, Maiden, N. Y.

I. Bowman, geographer, New York, N. Y.

W. M. Brown, geographer, Providence, R. I.

R. D. Calkins, geographer, Mt. Pleasant, Mich.

J. S. Chandler, missionary, Madura, India.

A. C. Coolidge, historian, Cambridge, Mass.

S. W. Cushing, geographer, Salem, Mass.

L. Farrand, anthropologist, Ithaca, N. Y.

C. W. Furlong, traveler and author, Boston, Mass.

H. Gannett, geographer, Washington, D. C.

E. W. Griffis, traveler and author, Ithaca, N. Y.

A. Hrdlicka, anthropologist, Washington, D. C.

E. II . Hume, physician and missionary, Changsha, China.

E. Huntington, geographer, New Haven, Conn.

M. Jefferson, geographer, Ypsilanti, Mich.

A. G. Keller, anthropologist, New Haven, Conn.

E, F. Merriam, editor, Boston, Mass.

J. H. Potts, missionary, Shanghai, China.

E. Sapir, anthropologist, Ottawa, Canada.

J. R. Smith, economic geographer, New York, N. Y.

E. V. Robinson, economic geographer, Minneapolis, Minn.

W. S. Tower, geographer, New York, N. Y.

R. H. Whitbeck, geographer, Madison, Wise.

S. W. Zwemer, missionary, Cairo, Egypt.

Anonymous, New York City.

416 APPENDIX

2. BBITISH

George Black, Sydney, Australia.

James Bryce, statesman, London.

Leonard Darwin, soldier, London.

T. H. Holdich, soldier, London.

H. H. Johnston, administrator, Arundel, England.

J. S. Keltic, geographer, London.

T. S. Longstaff, geographer, London.

D. Carruthers, explorer, Manningtree, England.

3. TEUTONS FROM CONTINENTAL EUROPE

S. de Geer, geographer, Stockholm, Sweden.
H. F. Helmholt, historian, MUnchen, Germany.

A. Kraemer, ethnographer, Stuttgart, Germany.

Mrs. M. Krug-Genthe, geographer, Chemnitz, Germany.

H. H. Reusch, geologist and geographer, Kristiania, Norway.

H. ten Kate, anthropologist, Holland.

F. von Luschan, anthropologist, Berlin, Germany.
K. F. Sapper, geographer, Strassburg, Germany.

4*. LATINS (AND OTHER EUROPEANS NOT ALREADY CLASSIFIED)

D. n Anoutchine, geographer, Moscow, Russia.

L. Gallois, geographer, Paris, France.

V. Giuffrida-Ruggeri, anthropologist, Naples, Italy.

G. Papillault, anthropologist, Paris, France.

B. y Rospide, geographer, Madrid, Spain.
G. Sergi, anthropologist, Rome, Italy.

S. Telles, geographer, Lisbon, Portugal.

5. ASIATICS

Katsuro Haro, Imperial University, Kyoto, Japan.
Inazo Nitobe, Imperial University, Tokyo, Japan.
Naomasa Yamasaki, Imperial University, Tokyo, Japan.
Jeme Tien-yow, Shanghai, China.
Wang Ching-chun, Tientsin, China.
Wu Ting-fang, Shanghai, China.

APPENDIX 417

SUMMARY OF CONTRIBUTORS

Letters sent Replies Contributors

Americans 64 57 25

British 43 88 8

Teutonic Europeans 42 28 8

Latin Europeans . . 27 8 6

Asiatics 21 8 6

Non-Teutonic and non-Latin Europeans .10 2 1

Latin Americans 6 2

Total 213 138 54

(B) THE RELATIVE CIVILIZATION OF THE
COUNTRIES OF THE WORLD

The following list contains the final results of the classifica-
tion made by the preceding contributors. In examining this the
reader should remember that the division into regions and the
grouping of the regions according to race make no claim to
perfection. Convenience in obtaining units small enough to
give a detailed map and yet to be known to people of many races
has been the primary object in dividing the different countries
of the world into smaller sections. The sections are small in
Europe and the United States because these regions are well
known, and large in Siberia because very few people can dis-
tinguish sharply between different parts of that country. In
a few cases I have added minor divisions such as southern
Alaska, with a special object not connected with the present
book. The grouping by races in Europe, Asia, and North
America has been guided also by motives of convenience. Such
places as Ireland, Asia Minor, Bulgaria, Baluchistan, and
others might properly be placed in other groups as well as in
those where I have put them. The reader can easily rearrange
for himself. My purpose has been merely to make a convenient
classification for our immediate purpose without respect to its
applicability elsewhere.

ORS

~ 2
< S S

05 OD GO GO

O5 O5 O5 O5

IR CIVILIZATIO]

H

O

EC,

V3
OH
t>

O

tf

fi

7 5
ITINS ASIATICS

nn IAA

\ I

i i

CO O >?
O5 O O5
rH

818

3 r>

05 05

8 8

OD Ct

O5 O5

SO
o

<O OD Q*
O5 O5 O5

O O
O O5 O

<3*

OJ

o

05

-* o

Ci Ol

o o

S

>NS OF THE WORLD ACCORDING TO THE

BY FIFTY CONTRIBUTORS

RANK ACCORDING TO VARIOUS <

25 7 6
AMERICANS BRITISH TEUTONS L,

inn inn inn i

5 O
>
4 rH

> 05
> 05

H

> CO
5 O5
4

C

05

O5 O5 fr

O5 O5 O5

OD OD I-
O5 O5 O5

O5 O5

05 <0
O5 O?

00 00
O5 O5

O5 O5

O5 O5

C! . .
03

pq : *

O CO G*
O5 O5 Oi

*? O G^
O5 O5 O5

CC

CT

ctf

CO O
O5 t-

ex o

rH
O5 OD

o

O5
b-

r^t

OS

c

d
ffi ' -

O)

J3

cfi

PQ . .
O

ill

53

'O I '

c

cd

C

fc
O

o

31 REGIO

PQ T-J

s 1

fH

=e : :
i-i

oj
ctf

JZ

WJ

CC

o
c

H

fe

tr

<L

1 2,

fi ^

ec

PQ : :
b : :

C

KJ

t

c

<J

<X
T
2

S 1 &

Northwestern Gei
Central Germany

O

<A

11

bfi 4^ S

S 8 8

XJ 03 Q

Holland
Switzerland ....

Southern Germai
Bingen)
Sweden

Belgium
Austria Proper
Norway

Southeastern Ger

Inno-itn^^ 11 TiM

Northeastern Ger
not including that
Ireland

Austrian Alps . .

w

in

ft
O
H

H

K

O
O

U-
C

O

o
O

o

2

o

H

Q

g

O

u

o

o

UJ

H

Oi o* o r^ oc "^

O5 O5 CO O CO CO

K

CQ

U

O 0> S CO CO CD

r-4

sss^s

s

ss

s

<

. . . . be

. . . . G .

S

: | g ^

o

X!

O
C

CO

of latitude

x v ^, rj 5j

C!

,-C

: 1 o : ; :

o -S ^ "3 <u

O

0*

*-

<]J '*H ^J C Q

"^

**

C

O

C si .* T3 ^

^ K**" C

PL, ^

,*

*+->:

^3

5

H

; o

i

o o c* "^

G

c3

S

^-^

"rt

22 W ^ * r5

rt

^4-

x

X'

c

o

t-

^o

<)

oj 03 ^ ."t! C
Q^ t|_ re s- -!

" ' cc S ^ -*->
03 - v ' O

J r^ M 'rt

W >-> fe PH 4^

S S g =" .

IIJS 2 ^

o | ? I '

i^ci^i^^y

. Central France (mounl

,C

"5

c

c/i

.S

'S

CH

CO

c3
^i

4J
C

<u
O

Guadalquiver River)
. Northern Spain (apprc

TOnnmania

.

: c/5

'. 'O

. c

. c3
'eS

* 4->

3 ^ H "
2 G
3 f->
<U J
3 O ,C
2 <V +J

5 <u S

30

. Southeastern Spain (A

. Portugal
. Corsica and Sardinia

. Albania and Monteneg

t GO Oi O I-H

Oi O -H
G^ CC CC

8 S

<5^ G^ O
^O t* ^O

es

&o
^

&

pQ

O
*0

<u

T3

^-s

4^

5

0?

^

.

4_(

03

t/J

O

<4-l

o

> * x

c

OJ

as
S

a
1

S B>

S 5 "

c * .5

113

s:*

i n

TJ

c

cj
^C

P>

(D

A
+->

ol

scarpathian Austri

5

eC o flj

pq o

llg

Be ^ H

CC CC CC GC

BUTORS

8

5 CO O

< 3

8rH
*0

2

3

H

O
O

s

2 co o

Mi S CO CO

O <O

CO Tp

o

<

0)

cu

1

O

00

5 o oi

r- H fc- co

3 S3

C/l

>J

D

?:

SS

g

, I

w
H

3

S

g loo

O " g *- t -

8

Et,

O

(5

I

8

i
^

.-a

bO r?

.S o

I 'S

.S to

T3 g

5 PH
'43 PH

50^

" "S
^ s

lei

OJ .

bD

a.

O

>>-> G o3

IH >-,.--;

O ' 3

n W

fc 00 O> O FH

CO CO CO "* r?

T3

1-4
OJ

o>

rH

O

CO t-

0> GO

CO *O
O> O5

0)

CO GO

=

j
1

^>
"

CO

i : :

T3 ; i

c . .

HH

CO

G

<u
PH

-sf
jS

ft

G . .
ed
bC . .

Ic I .

o

<u

5,

CO

-

c

c I
o

8,

.' V

o

e
2
"Si

<%>

Kl

S
"S

o

SO

3
s

CJ

*s

03
-t->

<
J5

rH
O

J25

?

SS
cu

^

cu

^

TS

C
OS

'bb
W

1

5?

^

?

^ ^ j

ail

CU <H

^ rO g

^ a .P8

M 03 O

Quebec
Washington and Or

S

5

S

0) O
-* *0

rH 0*

*o *o

1

b
O

00

OH

D

O

8B S

D
O

o

8

O

OJ CO t- I b^ 00 b-

CO O G*

CD Oi If

25
O

3

O

w
S5

ect

5 o :

.

o : :

.

S

*S

.

1

: M J 2 i

o

o

O aj
Sj X-N (H

S :

g

CO
CO

S/

> ' '

^

^ T3 52

co o

o c

.

3

to *

aS

-2

Q

CO

a

_id

s

00 S

&

CO

1

S

CU x-
*C c/

^ o;

8 1

^ ^ 1

rn Prairie States
i, Minnesota, Nebras
;rn Rocky Mountain
/Vominffl

Tie Provinces of Can

rioliimhia

o3
-C

.S

ctf <U

"S
T3

f

O O rr
O N p
P3 "S ex

11

a and Saskatchewan
rn Greenland and Ice
rn Alaska

O

a
^ |

c?

O

1 1 * i 1 1 * 1 1 $ ^ i

3 3

o o
en en

o

H

erf
H

O

O

O

oj

o

D
O

g

O

S

oj

8

U

55

2 <

o

fc
o

<

.$
to

e
8

'C

c

03

aJ

a

cu

PH

OH

tfj

G

oS

en

o

o

C

cj

2

CO

c^t

p

<v

i

.

^3

^

.s

3

V

'o

^

ecJ

*?

c

O

3

O ^

a

s:

'I s

Bs
Jfi

^

o

.y
*x

(U

s i

8

><

QJ r/

^i
c

; America
rn Mexico

CO |

r

"F*

55 ^

cC <U

\U ., c ^ +-

I S -S o 8

^ o s x ^

.

"tS O i-M G$

e - - -

OJ

Oj C

&
""a ct

O

3 C! cc -M

OH S3 g

a cf s i

^1 s ^

^3

G, fl o 13
c& fe ^ .S

Ca
th

cr

c

. ^ cc

03

^ S

s &

16

c ^ S S

<u c3 SJ

bo

nS

03

I

S G

S

t OO Oi O i-l

t- b- ir- OD OD

3

8 S3 3 S S33

8

00
**

O Q*
CO *

C/J

CL,

O
c4

O

en

5

O <Ot>G*ODOC*Q*COOD

H *O*a*O"**O'<*<-$<COCO

OS

8

O

Qi ^H t, rH CT>

t- rH CT> i (Cib-OOCiCO
OO^COThC^COCC^COCO

<c

s

OS

-2

g

/-N

,c

^r?

I

f !H

'3

O
CO

M

<u

G

'G

i r^ r^ t

REGIO>

oo

.1

|>

Pu

In

1

G

O T3 'O O
CM G G G^
"*"* ."t^

fc

^

c3

?

^ "cS "03^

"

O

KH

5 "~ H ^3

&3
S
<<
Z

I

1

oj
G

S

G
*-i

<L>

g

6

d>

-M

Manchuria

*o c6 'o "

^ 1 ~

g | |

O 03 ^ Cfi O
03 G G G

C

03

OJ

a

c3

2S

s 'C

<L> G

N |

u

O ^ S cd

7^

Q

|

-G SJ

S 5

_^
CO P<H

^

CO

fl c

s

03 CC

|

g i| i g|

&&&,

bf

o

r4 Q( 00

O) O) O)

$

9

S
a>

o
CO

c

1

T3 '

G

!Jj

4^ ' '

oj

G

.

-V

^

" s

G

03 * *

O

"So

2 : :

S-4

:

ce

-3

03 . .

H3

G

g

^

C/5

.S *

3
o

r^

OJ

*

CO

'C

SP

Cd

PQ
<ii

S :

03

: 1

^3 ' OQ

3

OS

1

1

.2

"C

J5

^

aj
H4
n

CO
CO

^ :

T3 "

^"^ . '"O

^ oj

J

di

1

OH

^

7}

+->

Tl

a s

- ' S

G

OJ

U)

1

"S

a.

^4-

S

1

1

P c3 ^

t>D O

g

PS O

03 o

.M * ^

e ^ 'ff
3tf

Armen

,o

'O

*n

Od

K

O)

O)

O) O)

O)

o

VARIOUS GROUPS OF CONTRIBUTOF

675 ALL
TONS LATINS ASIATICS CONTRIB

G*b-^F"AQCD*O*rtOOrHt-<O<OC0a*O G* G* O O O
CO*O*O^C?5COCOCOCOG*G*G*CS*G*G* G* Q* G* G* G*

O "*& O *f< CO G^ G< CO OD O O G^ O O G% *& OO OD O O
COCO*O-!}<COCOcOOOr-iG*G*G*G*G*G ; * CO rH rH G* G*

O5 G^ CO O OJ v ' t CO H?* CO ^ CO CO b- CO *O ^J* t-H CO "HP 1
w^OGOb-CTCOCOGO^OGOG^AOT^b- *fbOD*OCO

rH

rH

00
rH

CD

o

rH

rH

O

RANK ACCORDING TO T

25 7
AMERICAN BRITISH TEU

G^^^OSO^HCOOiCDrPCOt-b-b-OS OOOCOrHO
CC*O*O*O^-*^G<-*COCOCOG^G^G^ mG'tG^G^G*

G^^ODb-b-CO^OOSCOb-COOCOCO OtDO5OO5
**.*.*'* &

TM O

rH

00
G*

rH

CO
rH

::::::::: rg ::::: g :::: U
53 (
::::::::: ^ ::::: o ;::: I

::::::::: a : : : ^ : : g &

.* ^ eg < o 4:

LICA (Continued)

i - ; - s ^l * ' i * i

H | ' | < | ^ | -S : : g

'fi'S "^11 ^[Sri S ! t3^ fe

= .^^3w c wc/3 g ' S B

^* tf 8.^Sa tf ^rg.2gS ^ C85S)g

^ &'5 ^l^^^--l^s 1 S fl ^ a "3 5 |

*- t*Sn^ *^ -H >-* ^ *-<." * ^*"C S 1 ^ ^ ^^ fi fl T^ ^
e8 QO ^T o^'jU^JS o '^ csCfc^fu <u -^ O;L C o o ^

fcW^cnSH^Sfcffl^WOOwfcS^SriHW

a

1

CO

Kalahari Desert . .

bu

GiC2-^*OCO*bcdO5OrHGicO^*OCo"b- CDOsOrH

Q*

CO

CO
CO

C/l

otf
g

o
u

&4

o

o

a;

o

I

g

o

g

S

o
o
o

O

O
w

ft?

&,
o

Cd
<

00 CO *O CO "^f 1 CO CO CO CO ^

S5-*COGOG**i-tCOb-COO "H

H b* CO CO *O CO *O ^f 1 ^f *O ^f 1 ^f*

o o

CO 0*

K

t- h

S

CQ

St-cOaOCOCOOT-^Ob-O CO b- -Hf b- O O ^

0^40^^-^^CO^ COG^COCOCOCOCO

*O O5 GO CO Tf tj<

SOUTH AMERICA

o

co . .

<u

CJ

js

q-,
"I ^

.s ^

f-t n^ t-t

< O

Argentina (north of latitude 30)
Uruguay

Paraguay
Mountainous part of Peru

Mountainous part of Colombia
Ecuador

Mountainous part of Bolivia

Mountainous eastern region of Brazil (no
tude 20 ^

Mountainous part of Venezuela
Guiana (French, British, and Dutch) . . .

Northern desert portion of Chili
Southern Chili (including Tierra del Fuegc

Eastern lowlands of Colombia
Eastern lowlands of Bolivia

'3

PQ : :

o *

H
1

CJ
O

CO x S

O

o i

ij B

o

183. Eastern lowlands of Peru
184. Orinoco lowlands of Venezuela
185. Amazon Basin

CO CO CO

b GD

CO CO

Os

fH G* CO

S g

b- 00

Oi O

So

428 APPENDIX

NOTE. In the Andean countries of South America, except Chili, the
most advanced districts are found in the highlands. The eastern lowland,
which is heavily forested or else covered with savannah, contains only a
scanty population composed almost wholly of Indians. The way in which
each of these countries is divided into a "mountainous part," and an "east-
ern lowland" seems to have created confusion in the minds of many con-
tributors. They apparently thought of the lofty mountains, and not of the
plateau as was intended. Hence they rank the lowlands higher than the
mountains. As a matter of fact, the lowlands are in almost the same
condition as the Amazon Basin (No. 185). Therefore, after the figures for
Venezuela, Colombia, Peru, and Bolivia, I have added in parentheses a
series of numbers indicating the rank of these countries if in each case the
higher value in any given classification is taken as intended for the more
advanced portion, that is, the highlands. The numbers in parentheses are
probably nearer right than the others, and have been used in preparing the
map of civilization, Figure 45,

APPENDIX

429

CLASSIFICATION OF THE STATES AND PROVINCES OF THE UNITED
STATES AND CANADA

tfl CO JO

o o <y ^d O

*~ 1 " 4 ""' * O -+"j W_, C3"ri

4)3 o -< 3 C2 O CO oS ^

G-.O 2L-.0 :: fc. -5

UNITED STATES

A.

B.

C.

D.

1.

Alabama

2.1

1.9

0.2

2.0

2.

Arizona

1.2

1.7

0.5

1.6

3.

Arkansas

2.3

1.9

0.4

2.0

4.

California

5.3

4.8

0.5

5.1

5.

Colorado

4.1

4.1

0.0

4.2

6.

Connecticut and Rhode Island

5.9

5.8

0.1

5.8

7.

Delaware

4.0

3.9

0.1

4.0

8.

Florida

2.0

2.0

0.0

2.0

9.

Georgia

2.1

2.5

0.4

2.5

10.

Idaho

2.5

2.5

0.0

2.6

11.

Illinois

5.7

5.7

0.0

5.7

12.

Indiana

5.3

5.6

0.3

5.4

13.

Iowa

5.2

5.1

0.1

5.0

14.

Kansas

4.8

5.0

0.2

4.9

15.

Kentucky

3.2

3.3

0.1

3.3

16.

Louisiana

2.0

1.7

0.3

2.0

17.

Maine

4.9

5.0

0.1

4.8

18.

Maryland

4.1

4.0

0.1

4.1

19.

Massachusetts

6.0

6.0

0.0

6.0

20.

Michigan

5.4

5.6

0.2

5.4

21.

Minnesota

5.1

5.4

0.3

5.2

22.

Mississippi

1.7

1.4

0.3

1.7

23.

Missouri

3.7

3.5

0.2

3.7

24.

Montana

3.0

3.2

0.2

3.1

25.

Nebraska

4.4

4.3

0.1

4.4

26.

Nevada

1.3

1.9

0.6

1.7

27.

New Hampshire

4.8

4.8

0.0

4.6

28.

New Jersey

5.7

5.4

0.3

5.6

29.

New Mexico

1.6

1.5

0.1

1.6

30.

New York

5.9

5.7

0.2

5.8

31.

North Carolina

3.1

2.8

0.3

2.9

32.

North Dakota

3.8

3.8

0.0

3.8

33.

Ohio

5.8

5.9

0.1

5.8

430 APPENDIX

<n w w

O * j>

8 B Ss ga 3 g

a -d.-2 &" S r

ia us Il irl

S 3 <8 o^<J S&8

UNITED STATES (Continued) A. B. C. D.

34. Oklahoma 3.7 3.1 0.6 3.4

35. Oregon 4.6 5.0 0.4 4.9

36. Pennsylvania 5.6 5.5 0.1 5.6

37. South Carolina 1.6 2.2 0.6 1.9

38. South Dakota 3.8 3.7 0.1 3.7

39. Tennessee 3.0 2.6 0.4 $.8

40. Eastern Texas 3.1 2.8 0.3 3.1

41. Western Texas 1.8 1.6 0.2 1.9

42. Utah 2.4 2.3 0.1 2.5

43. Vermont 4.4 4.7 0.3 4.4

44. Virginia 3.8 3.8 0.0 3.9

45. Washington 4.8 5.0 0.2 5.0

46. West Virginia 2.5 2.9 0,4 2.8

47. Wisconsin 5.4 5.8 0.4 5.6

48. Wyoming 2.9 2.6 0.3 2.8

49. Southern Alaska 1.5 1.2 0.3 1.5

CANADA A. B. C. D.

50. Newfoundland 1.9 2.0 0.1 2.0

51. Prince Edward's Island . . 2.6 2.1 0.5 2.5

52. Nova Scotia 3.8 3.4 0.4 3.6

53. New Brunswick 3.3 2.9 0.4 3.2

*54. Quebec, east of longitude 72 30' 3.1 2.6 0.5 3.0

*55. Quebec, west of longitude 72 30' 4.1 2.6 1.5 3.4

56. Ontario, east of Lake Huron . 5.4 4.7 0.7 5.1

57. Ontario, north of Lakes Huron

and Superior 3.0 2.0 1.0 2.6

* In this case some contributors may have misunderstood what was

intended. At least, several of them placed the part of Quebec containing
Montreal lower than the sparsely populated portion from Quebec eastward.

If each contributor's higher figure be taken as meant for the part of the
province containing Montreal, the final numbers in column D become East

Quebec 2.5 and West Quebec 3.9. I am strongly inclined to think that
many of the Canadian figures are too low, because all but one of the
contributors were from the United States. For this reason Canada is not
included in Figure 43.

APPENDIX 431

a -0,0

CANADA {Continued) A. B. C. D.

58. Manitoba 3.5 4.4 0.9 4.0

59. Saskatchewan, southern half, i.e.,

south of latitude ,55 ... 3.0 3.0 0.0 3.1

60. Saskatchewan, northern half,

i.e., north of latitude 55 . 1.3 2.0 0.7 1.7

61. Alberta, southern half ... 2.9 2.9 0.0 3.1

62. Alberta, northern half ... 1.7 1.7 0.0 1.8

63. British Columbia, southern half 3.8 3.4 0.4 3.8

64. British Columbia, northern half 1.3 1.9 0.6 1.7

(C) EFFICIENCY AND MEAN TEMPERATURE

A TABLE BASED ON THE WORK OF 310 MEN AND 196 WOMEN
AT NEW HAVEN, NEW BRITAIN, AND BRIDGE-
PORT IN 1910-1913

TEMP. EFFICIENCY TEMP. EFFICIENCY

a c.

35 99.8% 21 93.5%

34 92.8% 20 93.6%

33 92.9% 19 93.6%

32 92.9% 18 93.7%

31 92.9% 17 93.8%

30 93.0% 16 93.9%

29 93.0% 15 94.0%

28 93.1% 14 94.1%

27 93.1% 13 94.2%

26 93.2% 12 94.3%

25 93.3% 11 94.4%

24 93.3% 10 94.5%

23 93.4% 9 94,6%

02 93.4% 8

432 APPENDIX

TEMP. EFFICIENCY TUMP. EFFICIEHCY

c. c.

7 94.9% 15 100.0%

6 95.1% 16 99.9%

5 95.3% 17 99.7%

4 95.5% 18 99.3%

3 95.7% 19* 98.9%

2 95.9% 20 98.4%

96.2% 21 97.9%

96.5% 22 97.4%

96.8% 23 96.9%

2 97.1% 24 96.4%

3 97.4% 25 95.9%

4 97.7% 26 95.3%

5 98.0% 37 94.6%

6 98.3% 29 93.9%

7 98.6% 29 93.2%

8 98.8% 30 92.5%

9 99.0% 31 91.7%

10 99.2% 33 90.5%

11 99.4% 33 89.0%

12 99.6% 34 86.8%

13 99.8% 36* 84.3%

14 99.9% 3 81.0%

INDEX

INDEX

Aboriginal America, 366 ff.

Abyssinia, Christianity in, 301.

Accad, 352.

Accidents, and climatic changes, 336.

Achaeans, 394; migrations of, 23 f.

Adults, relation to weather, 199 ff.

^Egean Sea, early culture of, 357.

^Esthetics, importance of, 251.

Afghanistan, 355.

Africa, civilization in, 261 ; sheep and
cattle in, 53 f.; status of, 298.

Africans, in World War, 274.

Age, relation to influenza, 188 ff.

Agra, climatic energy at, 231.

Agriculture, civilization and, 373;
effect on natural selection, 370; in
Bahamas, 47 f.; in Central Amer-
ica, 328 ff.; in Guatemala, 330; iron
tools and, 371; of whites vs. ne-
groes, 37 ff.

Alabama, Black Belt of, 312.

Alberta, climate of, 224.

Alcohol, in tropics, 71 f.

Alexandria, Greeks in, 27; rainfall
at, 200; rise of, 397.

Algae, optimum temperature of, 129.

Algonkian Indians, 369.

Alps, climate of, 220.

Altitude, physiological effect of, 77 f .

Amaru, 342 f.

Ambition, relation to national de-
cline, 27.

America, aboriginal, 366 ff. See also
North America and South Amer-
ica.

American Indians, standards of, 254.

Amorites, 342 f .

Amsterdam, death rate of, 62.

Andean highlands, status of, 298.

Anger, in tropics, 70 f .

Anglo-Saxons, in cotton mills, 97 f.

Animals, optimum temperature for,
130.

Annapolis, changes of mean tem-
perature at, 140 ff.; students' rec-
ords at, 14 f., 104 ff.

Anopheles mosquito, 392.

Antevs, E., cited, 320, 345 f.

Anthropologists, and climatic
changes, 18.

Appetite, atmospheric pollution and,
170.

Arabia, ruins in, 367.

Arabs, migration of, 388; religion
among, 390.

Archaeologists, and climatic changes,
17 f.

Argentina, climate of, 223 f.; index
of civilization, 252; status of, 298.

Aridity, effect on races, 394 f.; natu-
ral selection and, 394; relation to
Mohammed, 390.

Aristotle, 3, 76.

Arizona, climatic energy in, 231;
rank in progress, 288.

Arkansas, eminence in, 311.

Aryans, 355.

Asexual reproduction, 131.

Asheville, climate of, 161.

Asia, archaeological work in, 9; civi-

436

INDEX

lization in, 259 f.; climatic curve
of, 8; climate energy in, 236; des-
iccation of, 11; relation to ideal
climate, 223 f., 226.

Asia Minor, civilization in, 857;
Greek settlement of, 24; stonni-
ness in, 362.

Asiatics, in World War, 274.

Assyria, early greatness of, 356.

Athens, ancient climate of, 401 f .

Atlanta, weather and deaths in, 208,
213 ; workmen in, 50.

Atlantic City, climate of, 161.

Atmosphere, chemical pollution of,
169 f.; health and, 153 ff.

Atmosphere and Man, committee on,
187 ff., 194, 203.

Atmospheric moisture, ancient, 401.

Atmospheric pressure, effect of, 78,
137.

Attica, settlement of, 23 f .

Augusta, climate of, 402; effect of
seasons in, 93 ff., 97 f.

Aujeh, 333.

Australia, aborigines in, 378; civili-
zation in, 260, 263 ; climate of, 223 ;
death rate of, 67, 378; status of,
297, 351 ; tropical, 58, 376 ff .

Australians, nativity of, 882.

Austria, climate of, 224.

Austrians, in World War, 274.

Baber, 391.

Babylonia, 343; civilization in, 353;
early greatness of, 856.

Bacteria, humidity and, 190, 203; re-
lation to influenza, 189.

Bacterial infection, atmosphere and,
168.

Bahamas, climate of, 48 ff . ; white vs.
colored, 83 ff., 46 ff.

Balfour, A., cited, 44.

Balikh River, 843.

Baltic region, climate of, 224.

Baltic Sea, ice in, 326.

Baltimore, humidity and death rate
in, 178; storminess at, 217.

Baluchistan, 355.

Bantus, 43, 45.

Barbarians, migrations of, 11, 390.

Barrell, J., cited, 338.

Barrett Expedition, 7.

Bartholomew, J. G., cited, 223.

Basutos, Drakenberg Mountains, 43.

Baths, stimulus of, 146.

Beaches, at Owens Lake, 325.

Beasts of burden, Indian civilization
and, 371 ff.

Beersheba, 333; drought at, 387.

Belgium, in World War, 274.

Berg, cited, 337.

Berlin, climate of, 402.

Berliner, cited, 14.

Besson, L., cited, 174 f., 203.

Big business, 350.

Big trees, of California, 8 f., 820,
341, 346.

Binet tests, white vs. colored, 81.

Birth rate, 25; in Australia, 378, 388.

Bison, 373.

Black Belt, 312.

Blondes, in Greece, 23, 394.

Blood, flow of, 166 ff.

Blood pressure, 166 ff.

Body temperature, 165.

Bolivia, status of, 298.

Bombay, climatic energy at, 231;
death rate of, 62; deaths and
weather in, 181.

Bonuses, effect of, 89 f.

Books, and climate, 61.

Boston, climate of, 161; Health De-
partment, 158; hospitals in, 21;
humidity and death rate in, 178;
influenza in, 185 f., 189; murders
in, 404; operations in, 179 ff.; pneu-
monia in, 177; school children in,

INDEX

437

158, stonniness at, 214 f.; weather

and deaths, 205 f .
Boys, in school, 283 f.; proportion

in Australia, 380.
Brass fittings, making of, 82 f.
Breasted, J. H., cited, 342 f.
Breath, carbon dioxide in, 133.
Bridgeport, changes of temperature

in, 139 ff.; factory operatives in,

14 f., 80 ff.
B right's disease, influenza and, 188

ff.

Britain, climate and, 224.
British, in World War, 2T4.
British Columbia, climate and, 221,

224; effect of temperature, 162;

humidity and death rate in, 178.
Brooks, C. E. P., cited, 8, 17.
Brown, L., cited, 94.
Bruckner, E., cited, 25.
Brunhes, cited, 341.
Brunnow, cited, 132.
Bryce, J., cited, 288; letter of, 253.
Buckle, cited, 3.

Buddhism, 355; progress and, 301.
Buffalo, deaths and temperature in,

213.

Burans, 237.
Butler, H. C., cited, 344, 388 f.

Cairo, rainfall at, 400; deaths in, 181.

Calcium, in blood, 157.

Calcutta, climatic energy at, 231;
death rate of, 62.

California, big trees of, 8 f ., 320 ; cli-
mate of, 220, 225; climatic curve
of, 8; eminence in, 308; nervous
disorders in, 225; status of, 289,
297, 351 f.; stonniness in, 334;
temperature of, 164 f .

Cambodia, storms in, 367.

Cambridge, influenza in, 190.

Campani, A., cited, 163.

Campeche, malaria at, 829.

Canada, climate of, 223 f.; loyalists
in, 46; progress of, 396; provinces
of, 429 f.; status of, 267, 297, 396.
Canal Zone, 62, 64.
Cancer, and influenza, 188 ff.
Cape Horn, storminess at, 230.
Carbon dioxide, in breath, 133.
Carnegie Institution of Washington,

7 f., 194, 320 f.
Carpenters, seasonal work of, 93 ff.,

98 f .

Carr-Saunders, cited, 25.
Carthage, capacity of, 255; culture

of, 357.
Caspian Sea, climate of fourteenth

century, 326.

Cattell, J. McK., cited, 127.
Cattle, in Africa, 53 f .
Centers of civilization, 347 ff.
Central America, archaeological
work in, 9; disease in, 12; mis-
sionary in, 75; morality in, 75;
negroes in, 67; status of, 298,
storms in, 367.
Central Asia, climate of, 7 f ., 390 f . ;

migrations in, 25.
Ceylon, ruins in, 366.
Chaco Canyon, 338.
Chad, Lake, 327.
Changes of climate, see Climatic

changes.
Changes of temperature, 138 ff., 140

ff., 188 ff., 196 f., 222.
Character, of Europeans in tropics,

68; relation to climate, 404.
Character of Races, 7, 23, 29.
Chemical pollution, of atmosphere,

165.

Cherokees, 875.

Chicago, climate of, 402; death rate
in, 178; storminess at, 215; wind
at, 204.

Children, British, in India, 58; gain
in weight of, 158 f.; in Australia,

438

INDEX

377, 388; in Denver, 137; in
tropics, 60, 378 ff . ; malaria among,
393; mental tests of, 31, 103; rude-
ness in America, 404; selection of,
24; weather and, 199 ff.

Chile, climate of, 221 ; death rate of,
62.

China, capacity of, 255; civilization
in, 261, 357; climatic advantages,
404; climatic energy in, 238; early
greatness of, 354; famines in, 11;
influence of United States, 350;
journey to, 11; migrations in, 25;
past climate of, 15 f.; storminess
in, 362 f.

Chinese Turkestan, autumn in, 226;
climatic energy in, 231 ; expedition
to, 7; forests in, 317.

Chlorine, in lakes, 323 f .

Churchill, W., cited, 53.

Cigar making, 32; and day of week,
125; and seasons, 93 Iff., 99 f.; and
temperature, 162.

Cincinnati, deaths in, 208.

Circulation of blood, 165 ff.

Cities, deaths in, 178, 204 ff., 279;
influenza in, 189 ff.

Civilization, agriculture and, 373;
barometric changes and, 13; cen-
ters of, 347 ff.; climate and, 12,
313 f., 387 ff.; conditions of, 291
ff., 387 ff.; definition of, 242, 255;
distribution of, 240 ff., 352 ff.;
higher vs. lower, 67 ff.; in United
States, 288; map of, 15, 277, 294 f.,
415 ff.; race and, 258 f.; storminess
and, 12 f.; table of relative, 417.

Classification of countries, 248.

Clay, A. T., cited, 842 f., 389.

Cleveland, deaths in, 208, 213; in-
fluenza at, 187, 190.

Climate, of Bahamas, 47; of Central
Asia, 390 f.; character of ancient,
400 ff.; civilization and, 2, 12, 813

f.; in Denmark, 96; and deaths,
20 f.; energy and, 302, 312; and
farm efficiency, 41; of Florida, 50
f.; of ancient Greece, 22 ff.; and
health, 3; ideal, 220 ff.; vs. inherit-
ance, 30 ff., 312; of Japan, 95 f.;
magnitude of effects, 149 ff.; and
mental activity, 51; methods of
overcoming, 406 ff.; migrations
and, 25; of New York, 95 f.; op-
timum, 6, 15, 20, 161; at Panama,
67; physiological effects of, 77 ff.;
psychological effects of, 77 ff.;
and slavery, 41; of South Africa,
45; summary of effect of, 10, 11,
14, 52 f.; uniformity of, 18 f., 338
f.; vs. weather, 136.

Climatic changes, in Asia, 8 ; in Cali-
fornia, 8; diverse views of, 5 f.;
economic effects of, 387; evidence
of, 17 f., 316, 339; in Greece, 26 f.;
growth of idea, 335; and history,
18, 333; migrations and, 11; nature
of, 10, 326 ff., 333 f., 335, 340; and
political disturbances, 11.

Climatic energy, ancient distribution
of, 359 ff.; and influenza, 188 ff.,
191; distribution of, 276, 278, 291
ff., 294 f., 347 ff.; map of, 228 ff.

Climatic Factor, 8.

Climatic hypothesis of civilization,
364 f., 387 ff., 403 ff.

Climatic pulsations, 7 ff.; glacial
epochs and, 319 f.

Climatic zones, shifting of, 315 ff.,
327 f .

Climographs, 161, 182 ff.

Cloudiness, effect of, 79, 145, 149 f.

Coal, relation to manufacturing,
289 f .

Coca, in Peru, 368.

Cold, resistance to, 396.

Cold Spring Harbor, 194.

Coldward Course of Progress, 17.

INDEX

439

Colombia, status of, 298.

Colon, death rate of, 62.

Colonization, in tropics, 236.

Colorado, storminess in, 398.

Color line, 32 ff.

Columbia, effect of seasons in, 93 ff.,

97; mental tests at, 31.
Columbus, O., deaths in, 213; in-
fluenza in, 190.
Comfort, and atmospheric pollution,

169 ff.; and temperature, 167.
Conception, seasonal variations,

159 f.

Congo, climate of, 223, 231.
Connecticut, death rate of, 64, 66;
effect of changes of temperature
in, 140 ff., 144; effect of humidity
in, 112; effect of light in, 109 f.;
effect of seasons in, 84, 92 ff., 154
f.; effect of temperature in, 115 ff.,
162; efficiency in, 94 f.; eminence
in, 306, 308; piece-workers in, 80
ff.; rank in progress, 289.
Consumption, selective action of,

391.
Contagious diseases, and humidity,

178.
Contributors, to map of civilization,

256, 415 ff.

Cooling of houses, 409.
Copenhagen, children in, 96; test of

strength at, 128.
Correlation coefficients, 187 ff., 195

ff., 289.
Cotton factories, effect of seasons in,

93 ff., 97 f., 290.
Countries, relative civilization of,

415 ff.

"Crackers," 42, 52, 311.
Crampton index, 165 ff.
Criticisms, of climatic hypothesis,

366 ff.

Critics of pulsatory hypothesis,
355 ff.

Cross-breeding, 53 f.

Cuba, civilization in, 266; negroes
in, 32.

Cubans, as cigar-makers, 99; effect
of temperature on, 162; and holi-
days, 125; optimum temperature
of, 164.

Cultivation, as factor in civilization,
2.

Culture, and civilization, 1, 387; and
climate, 16 f., 395; test of, 281.

Cuneiform writing, 353.

Curve of climate, 319.

Cuzco, climate of, 368.

Cycles of climate, 320.

Cyclonic storms, civilization and, 12
f.; and climatic changes, 10, 334;
cycle of, 352; effect of, 136 f.;
location of ideal, 222 ff.

Cyprus, forests in, 341.

Daily wages, at New Britain, 122.

Dakotas, climate of, 222, 224; edu-
cation in, 285, eminence in, 311.

Dark Ages, 11; religion in, 389 f.;
temperature in, 397.

Davis, W. M., cited, 7.

Death rate, in Australia, 376 ff., 381
ff.; in cities, 279; and climate, 20
f.; comparative, 62 ff.; in Con-
necticut, 154; in Europe, 291 ff.;
and humidity, 177 ff.; and influ-
enza, 188 ff.; in Japan, 95, 159 f.;
of miners, 279; in New York, 94 f.;
after operations, 179 ff.; in Penn-
sylvania, 154; from pneumonia,
175 f.; and seasons, 92 ff., 159 f.;
selective, 62; and temperature,
162; in tropics, 378; and weather,
21.

Death Valley, climatic energy hi,

231.
Deforestation, hypothesis of, 316 f.

440

INDEX

De Geer, S., cited, 320, 339.

Delaware, eminence in, 806.

Denmark, climate of, 96; death rate,
376; seasonal effect on children,
92, 96.

Density of population, and influenza,
188 ff.

Denver, children in, 137; deaths in,
208.

"Desert bay," 342 f.

Desert Botanical Laboratory, 8.

Deserts, ancient, 400; drought in,
388; shifting of, 327 f.

Detroit, deaths in, 213.

Dexter, O. E., cited, 13, 137.

Diet, in Bahamas, 47.

Disease, and aridity, 391; relation
to atmosphere, 168; and civiliza-
tion, 60; among Mayas, 331; se-
lective action of, 391.

Dixon, R. F., cited, 369.

Dog, among Indians, 373.

Domestic animals, and civilization,
373 ff.

Dorians, migrations of, 23 f .

Douglass, A. E., cited, 320, 338.

Drought, in Palestine, 387 f. See also
Aridity.

Drunkenness, in Central America,
71 f.

Dryness, effect of, 113, 181 f.

Dublin, L. I., cited, 194.

Dynamometer, tests with, 78, 128.

Earnings, of whites vs. colored, 32,

34.
Economic effects of climatic changes,

387 ff.
Economists, and climatic changes,

18.

Ecuador, temperature of, 227.
Education, as test of progress, 282;

and vitality, 275 ff.

Efficiency, relation to deaths, 154;
seasonal, 85 ff.; and temperature,
431 f.

Egypt, Alexandrian period in, 397;
ancient climate of, 15, 16, 363 f.,
400; capacity of, 255; character
in, 30 ff.; civilization in, 297, 357;
decay of, 13; disease in, 12; effect
of climate, 51 f.; energy in, 13;
Greeks in, 27; rainfall in, 398;
storminess in, 362.

Egyptians, ancient culture of, 354,
356.

Einfluss von Klima, 14.

Electrical apparatus, 102.

Eminence, distribution in United
States, 304 ff.

Energy, and climate, 312; and emi-
nence, 305 ff.; and health, 154;
importance of, 302; map of, 234.

England, climate of fourteenth cen-
tury, 326; climatic advantages, 220
f., 224, 226, 404; coal in, 289; death
rate, 376; rank in civilization, 257.

English, birth rate in Australia, 382
ff. ; seasonal marks in, 104 ff.

Environment, social vs. physical,
57 f.

Equatorville, climatic energy at, 231.

Eskimos, character of, 235.

Eugenists, 310.

Euphrates region, ancient climate
of, 343 ff.

Euphrates River, 342 f .

Eurasia, effect on storms, 361 ff.

Europe, civilization in, 255, 257 ff.,
357; climate of, 221, 223 f.; coal
in, 289; death rate, 376; impor-
tance of, 347 ff.; maps of, 291 ff.;
nervous disorders in, 156.

Europeans, capacity of, 255; effect
of tropics on, 56 ff., 68.

Evolution of Climate, 8.

Exiles, in Siberia, 287.

INDEX

441

Expansion of Races, 25.
"Explorations in Turkestan, 7.

Factories, piece-workers in, 80 ff.;
seasonal variations in, 92 ff., 407.

Falkland Islands, climate of, 224.

Famines, in China, 11, 319.

Farmers, white vs. negro, 37 ff.

Fatigue, of nerves, 157.

Ferrell, J. A., cited, 47.

Fertile crescent, 342 f.

Fevers, in Central America, 328 ff.

Finland, progress of, 260, 397.

Floods, in France, 325.

Florida, changes of temperature in,
140 ff., 143 f.; climate of, 50 f.,
220, 223 f.; factory operatives in,
14 f.; manufacturing in, 290.

Fontein, climate of, 367.

Food, as factor in civilization, 2.

Foreign born, influenza and, 188 ff.

Forests, cutting of, 316 f.; in Cy-
prus, 341; in Mesopotamia, 844;
and primitive agriculture, 372 ff.;
and primitive civilization, 374 ff.

Fourteenth century, climatic stress
of, 325 f.

Fraas, O., cited, 13.

France, climatic advantages, 224,
404; climographs of, 182; death
rate in, 20 f., 178; floods in, 325;
index of civilization, 252; tem-
perature of, 130, 162.

French, in World War, 274.

Fresno, climate of, 221.

Gale, H. S., cited, 323 f.

Garden of Eden, 344.

Gamier, cited, 156.

Gautama, 255, 355.

Genius, geographical distribution of,

802.
Geographers, Association of, 18.

Geographers, opinions of, 17 f., 19,

337.

Geologists, and climatic changes,
17 f.

Oeopsychische Erscheinungen, 14.
Georgia, 253; Black Belt of, 812;
climate, 50; eminence in, 807; high
school graduates, 283; school at-
tendance in, 282; seasons in, 97
f.; storminess in, 398; teachers'
salaries, 283.

Germans, birth rate in Australia,
383 ff.; in World War, 274.

Germany, ancient climate of, 863 f.;
civilization in, 252, 267; climatic
advantages of, 224, 226, 404; cold
winters in, 326; migrations from,
25.

GilFillan, S. C., cited, 17, 395, 397.

Girls, in Bahamas, 50; and bonus,
89 f.; factory work of, 97; pro-
portion in Australia, 380; in
school, 283 f.; sensitiveness to
climate, 126; and weather, 92 ff.,
139 ff., 141 f.

Glacial epochs, number of, 338.

Gouldsbury, cited, 73.

Government, and progress, 301.

Grand Rapids, influenza in, 189 f.

Grasslands, and primitive agricul-
ture, 372 ff.

Great Britain, index of civilization,
252; influence of, 297.

Great Lakes, climate of, 223;
storminess on, 230.

Great Plains, education in, 285.

Great Salt Lake, 327.

Greece, ancient climate of, 16, 22 ff.,
363 f.; civilization in, 367; climate
of, 223, 402; culture of, 357; de-
cay of, 13; disease in, 12; malaria
in, 392 f.; natural selection in,
23 f., 26; in World War, 274.

442

INDEX

Greek Christianity, and progress,

301.
Greeks, capacity of, 255; in Egypt,

397; energy of, 13.
Greenburg, D., cited, 176, 203.
Gregory, H. E., cited, 78.
Gregory, J. W., cited, 9, 339 ff.
Grenfell, W. T., cited, 235.
Growth, and atmospheric pollution,

170; and influenza, 188 if.
Guatemala, drunkenness in, 71;

past climate of, 16; ruins in, 9,

328 ff.

Guinea pigs, and vitiated air, 170.
Gulf of Lions, storminess of, 362.

Hampton Institute, 35.

Hann, J., cited, 229, 368.

Harper, R. M., cited, 312.

Harris, J. A., cited, 194.

Health, and atmosphere, 153 ff.; in
Australia, 381; and climate, 10
ff.; and energy, 164; map of, 291
ff.; and national decline, 27; and
weather, 194 ff., 277.

Health resorts, climate of, 162.

Hearn, Lafcadio, cited, 54 f.

Heart, relation to influenza, 188 ff.

Heart diseases and influenza, 191.

Heart-rate, 165 ff.

Heat, effect of, 126 f., 171; effect
on death rate, 117; protection
against, 396.

Heating and Ventilating Engineers,
181.

Hebrides, climate of, 221.

Hebron, 333; hospital at, 388.

Hellpach, cited, 14, 73, 148.

Herbette, cited, 337.

Heredity, and eminence, 307; op-
portunity vs., 312.

Hermans, cited, 165.

Hess, A. F., cited, 167.

Hieroglyphs, 352.

Highlands, effect on storms, 860.

High School, graduates of, 282 f.

Hill, L., cited, 166.

Himalayas, storminess in, 362.

Hindu, capacity of, 255.

Hinges, work with, 82.

Historians, and climatic changes,
18, 241 ff.

History, basis of, 369 f.; and cli-
matic changes, 18, 333.

Hittites, 356.

Holidays, and Cubans, 125.

Holland, storms in, 325.

Honore, M. F. C., cited, 53.

Hookworm, 60; in Bahamas, 47.

Horse, metaphor of, 152, 161, 225,
368.

Horse-races, in United States, 78.

Hospitals, in Boston, 21; psychia-
tric, 156.

Houses, cooling of, 409; effect of,
110 f.; heating of, 406.

Huguenots, 809.

Human activity, and changes of
temperature, 140 ff.; and seasons,
92 ff.; and temperature, 124 ff.

Human energy, map of, 15, 234.

Humboldt, cited, 76.

Humidity, ancient, 401; and carbon
dioxide, 133 ff.; curves of, 212;
data for, 233; and death rates,
177 ff., 196 ff., 202 ff.; effect of,
14 f., 109 f., 167 f., 209 f., 214,
219; and health, 204 ff.; ideal,
222; and influenza, 188 ff.; mag-
nitude of effect, 149 f.; and op-
erations, 179 ff.; optimum of, 161;
optimum for negroes, 164; in
Paris, 174.

Hungary, in World War, 274.

Hunting, and natural selection, 870.

Iceland, civilization in, 270, 801;
progress in, 299, 897.

INDEX

443

Idaho, education in, 286; eminence
in, 311; school attendance in, 282.

Illinois, education in, 285; eminence
in, 308; teachers' salaries, 283.

Illiteracy, as test of progress, 282.

Immigration, racial character and,
309 f.

Immorality, in tropics, 73.

Incas, 329.

Index of civilization, 250.

India, ancient climate of, 363;
British in, 68; capacity of, 255;
Christianity in, 301; civilization
in, 357; deaths and weather in,
181; early greatness of, 354 f.;
European domination in, 296;
growth of population, 58; hill
tribes of, 300; humidity data for,
233; influenza in, 191; and Japan,
351; Moguls in, 391; past climate
of, 15 f.; storminess in, 362.

Indiana, eminence in, 308.

Indianapolis, deaths and tempera-
ture in, 213.

Indians, contact with whites, 56;
energy of, 369 ff.; inheritance of,
309, 370; in South America, 57.

Indo-China, disease in, 12; early
greatness of, 354; European domi-
nation in, 296; ruins in, 866;
storms in, 367.

Industry, tropical, 68 ff.

Infanticide, in Attica, 24.

Infants, blood of, 157; death rate
in Australia, 378.

Infection, and atmospheric pollu-
tion, 169 ff.

Influenza, and humidity, 178; and
weather, 21, 187 ff.

Infusoria, temperature relations of,
130.

Inheritance, as factor in civilization,
1 ; and climate, 3, 80 ff . ; and emi-
nence, 305 ff.; North vs. South,

42; vs. opportunity, 312; of Peru-
vians, 368.

Insanity, in California, 225; in Lon-
don, 156; in tropics, 227.

Insurance companies, data of, 278.

Intermarriage, among Greeks, 24.

International Institute of Statistics,
63 f.

Inventions, for resistance to cold,
396.

lonians, migrations of, 23 f.

Iowa, eminence in, 308; immigrants
to, 309.

Ireland, changes of climate in, 319;
climatic advantages, 221, 231, 404;
rank in civilization, 258.

Irish, birth rate in Australia, 383
ff.; migration to America, 25.

Iron, and Indians, 371 ff.; and
manufacturing, 289 f.

Iroquois Indians, 369.

Islam, rise of, 9.

Ispahan, rainfall at, 400.

Italians, in factories, 82 f.

Italy, civilization in, 252, 357; cli-
mate of, 223 f.; climographs of,
182; deaths in, 20 f.; humidity in,
178; malaria in, 892 f.; optimum
temperature in, 163; sexual ir-
regularities in, 74; storminess in,
361; temperature in, 162; in
World War, 274.

Jacksonville, carpenters in, 98 f.;
cigar-makers in, 99; hospitals in,
156; manufacturing in, 290; sea-
sons in, 93 ff.; whites vs. negroes,
32.

Jamaica, negroes in, 57.

James, W. B., cited, 145.

Japan, civilization in, 201, 357; cli-
mate of, 224; climatic energy in,
239; death rate in, 95; importance
of, 350 f.; index of civilization,

444

INDEX

252; manufacturing in, 290; races
of, 253; storminess in, 361 ff.

Java, disease in, 12; growth of
population, 58; ruins of, 366.

Jefferson, M., cited, 281.

Jenghis Khan, 391.

Jewish atrocities, 260.

Jews, capacity of, 255; early cul-
ture of, 356.

Johannesburg, deaths and weather
in, 181.

Johnson, D. W., cited, 345.

Jones, W. H. S., cited, 12, 393 f.

Kansas, climate of, 224; rank in

progress, 288.

Kentucky, eminence in, 311.
Khabur River, 343.
Khartum, climatic energy at, 231.
Kropotkin, cited, 7, 11.
Kullmer, C. J., cited, 13, 230, 398.

Labrador, influenza in, 190.

Lag, of health after weather, 161.

Lakes, chlorine in, 323 f.; saline,
323, 327; sodium in, 323 f.

Latin America, Christianity in, 301;
whites in, 58.

Latitude, and influenza, 188 ff.

Leadership, among whites vs. ne-
groes, 35.

Legrain, cited, 843.

Lehmann, cited, 13, 78, 96, 103, 128,
137.

Letters about map of civilization,
242 ff.; as test of progress, 281.

Life insurance, in South, 279; sta-
tistics, 276.

Light, effect of, 109 f.

Lincolnshire, storms in, 325.

Liverpool, climate, 221.

Liverpool School of Tropical Medi-
cine, 893.

Livingstone, D., cited, 73.

London, climate of, 220; efficiency
in, 229; insanity in, 155; mur-
ders in, 404.

Longitude, and influenza, 188 ff.

Lop Nor, climate of fourteenth cen-
tury, 326.

Lorain, cited, 32.

Los Angeles, climate of, 164, 221;
influenza in, 189; suicides in, 225;
water supply of, 323 f.

Louisiana, school attendance in, 282

Loyalists, in Bahamas vs. Canada,
46.

Lucknow, deaths and weather in,
181.

Lumber industry, in Bahamas, 34.

Lumber trade, of Mediterranean,
341.

MacDougal, D. T., cited, 8, 129.
Madras, deaths and weather in, 181.
Madrid, deaths and weather in, 181.
Magazines, and climate, 51.
Maine, eminence in, 306, 311; kigh

school graduates in, 283.
Malaria, 12.
Malaria, in Bahamas, 47; in Central

America, 328 ff.; changes in, 12;

climate and, 22; in Greece, 26;

selective action of, 391 ff.; tropi-
cal, 60.

Male death rate, in Australia, 878.
Manchester, tests of carbon dioxide

in, 133.
Manufacturing, and influenza, 188

ff.; and progress, 281, 289.
Map, of civilization, 240, 256 ff.,

415 ff.; of climate vs. eminence,

812 f.; of climatic energy, 228 if.,

232 ff.
Mari, 842 f .

Maritime provinces, climate of, 224.
Mark Twain, cited, 136.
Marriage, 810.

INDEX

445

Maryland, boys in school, 283 ff.;
climate of, 223 f,; school atten-
dance in, 282.

Massachusetts, advantages of, 313;
death rate in, 117; eminence in,
303, 306, 308; hospitals in, 156;
rank in progress, 288; railroads
in, 281; school attendance in, 283;
teachers' salaries, 283.

Massaua, efficiency in, 229.

Mathematics, seasonal marks in,
104 ff.

Maya history, tree growth and, 332.

Maya regions, past climate of, 16.

Maya ruins, 9, 329.

Mayas, architecture of, 329 ff.; cal-
endar, 329; capacity of, 255;
civilization of, 359; climate, 362;
sculpture, 329 ff.

Mayo, cited, 32.

McHattie, Dr., cited, 47.

Mecca railroad, 388.

Medes, 353.

Medical practice, and influenza,
188 ff.

Mediterranean region, early culture
of, 357; vegetation of, 9 f.

Melon, and wind, 138.

Memphis, storminess at, 217;
weather and deaths in, 208, 213.

Men, distribution in Australia, 380;
effect of weather on, 141 f.; op-
timum temperature for, 130; in
tropics, 378 ff.

Mental activity, and climate, 51;
and seasons, 103 ff . ; and tempera-
ture, 128 f.

Mental disorders, seasons and, 156 f.

Mental efficiency, and atmospheric
pollution, 169 ff.; and tempera-
ture, 167.

Mental tests, white vs. colored, 31.

Mental work, changes of tempera-
ture and, 142 f.; and climate, 103

ff.; optimum temperature for, 168.

Mesopotamia, ancient climate of,
343 f., 363 f., 400; capacity of,
255; civilization in, 352 f., 357;
early greatness of, 355 f.; migra-
tions in, 407 f.; ruins in, 889;
storminess in, 362.

Metabolism, and temperature, 134,
167.

Metropolitan Life Insurance Com-
pany, 194.

Mexicans, inheritance of, 309.

Mexico, civilization in, 266; climate
of, 223, 408; deaths and weather
in, 181; drunkenness in, 71; index
of civilization, 252; status of, 298.

Mexico City, deaths and weather in,
181.

Micro-organisms, see Bacteria.

Middle Atlantic states, agriculture
in, 40.

Migrations, aridity and, 390; from
Bahamas, 46 f.; and climate, 3,
25; and drought, 388; effect in
tropics, 59; into Greece, 23 f., 27;
of Indians, 370; and natural se-
lection, 52 f.; nature of, 26;
seasonal, 407 ff.; weekly, 409 f.

Milan, death rate and weather in,
182; weather in, 103.

Milwaukee, influenza in, 187.

Mind, effect of tropics on, 69.

Miners, death rate of, 279.

Ministers, in Bahamas, 49.

Minneapolis, weather and deaths in,
182, 208.

Minnesota, death rate of, 63.

Minoans, 23.

Missionaries, in Central America,
75; in China, 350; in tropics, 73;
in Turkey, 74, 350.

Mississippi, education in, 287;
school attendance in, 288;
teachers' salaries, 283.

446

INDEX

Moab, drought in, 387 f.

Moguls, in India, 391.

Mohammed, aridity and teachings
of, 390.

Mohammedanism, and progress, 301.

Moisture, effect on health, 174 ff.
See also Humidity.

Monotony, of tropical climate, 227.

Montana, boys in school, 285; edu-
cation in, 286; eminence in, 311.

Montesquieu, cited, 3, 76.

Moorhead, cited, 369.

Morality, in Central America, 75.

Mormonism, education and, 286.

Morse, cited, 31.

Mortality, in Europe, 291 ff.; and
weather, 174 ff.

Mosquito, anopheles, 392.

Mountain sheep, 374.

Mucous membranes, effect of tem-
perature on, 168.

Murders, in Boston, 404; in London,
404.

Mutations, and climate, 3.

Naivasha, stock at, 53.

Nansen, F., cited, 237.

Nantucket, climate of, 161.

Naples, deaths and weather in, 182.

Nasal mucosa, temperature and,
168.

Nashville, weather and deaths in,
208, 213.

National Research Council, 187, 194.

Natural selection, 52 f., 365 ff.;
aridity and, 394; climate and, 6
f.; death rate and, 379 ff.; in
Greece, 21 ff., 26; among Indians,
370; among Mayas, 331; in Meso-
potamia, 352; in Peru, 368; in
tropics, 385.

Nearing, S., cited, 302 ff.

Nebraska, eminence in, 306, 808.

Neglected Factor in Race Develop-
ment, 25.

Negroes, among cigar-makers, 99;
education and, 287; efficiency of,
57; humidity and death rate
among, 178; influenza and, 188 ff.;
mentality of, 31; optimum for,
21, 164; vs. Teutons, 30 ff.; in
United States, 395.

Nephritis, and influenza, 188 ff.

Nerves, fatigue of, 157; in Peru,
368; and wind, 137 f.

Nervous disorders, in California,
225; in Europe, 156.

Nervousness, origin of, 115.

Netherlands, index of civilization,
252.

Neurasthenia, selective action of,
391.

Nevada, education in, 286; eminence
in, 311; railroads in, 281; rank in
progress, 289.

Newark, pneumonia in, 177.

New Britain, changes of tempera-
ture in, 139 ff.; factory operatives
in, 14 f., 80 ff.; wages in, 122.

New England, climate, 224; educa-
tion, 285; eminence in, 303, 306,
308; inheritance in, 42; loss of
ability, 310; manufacturing in,
290; weather of, 146.

Newfoundland, civilization of, 267.

New Hampshire, high school gradu-
ates, 283; eminence in, 306, 311.

New Haven, changes of tempera-
ture in, 139 ff.; factory operatives
in, 14 f., 80 ff.; influenza in, 190;
weather and deaths in, 205 f., 208.

New Jersey, rank in progress, 289;
school attendance in, 283;
teachers' salaries, 283.

New Mexico, education in, 285 ff.;
eminence in, 308; rank in prog-
ress, 288.

INDEX

447

New Orleans, climate of, 161; in-
fluenza in, 189 f.

Newport, climate of, 402.

New South Wales, birth rate among
natives of, 383 ff.; death rate,
381 ff.

New York, climate of, 220, 402;
death rate of, 63 f., 248; deaths
from pneumonia, 175 f.; educa-
tion in, 285; eminence in, 306;
hospitals in, 156; humidity and
deaths in, 178, 214; influenza in,
185 f., 189; negroes in, 32; pneu-
monia in, 177, 208; rank in prog-
ress, 289; seasons in, 92 ff.;
storminess in, 215, 334; tempera-
ture variations, 183 ff. ; weather
and deaths in, 194 ff., 205 f., 208.

New York Life Insurance Company,
194.

New York State Ventilation Com-
mission, 100, 127, 165, 177, 181,
202, 214.

New Zealand, climate, 221, 224;
death rate of, 64, 376; importance
of, 351.

Norbury, F. P., cited, 156.

Norse, in Greece, 23.

North, changes of temperature in,
144; slavery in, 41; vs. South, 36
ff., 403 f .

North Africa, climate of, 223; in
Dark Ages, 397.

North America, civilization in, 263;
climate of, 223.

North Carolina, eminence in, 307,
311; seasons in, 97.

Norway, progress of, 396.

Nova Scotia, ability in, 267 f.

Ohio, eminence in, 308; rank in
progress, 289; teachers' salaries,
283.

Old Testament, 344.

Operations, weather and, 21, 179 ff.
Opportunity, eminence and, 305 ff.,

311; heredity vs., 312.
Optimum climate, 161.
Optimum temperature, 126 ff., 163.
Oregon, eminence in, 306, 308; high

school graduates, 382; rank in

progress, 289; teachers' salaries,

283.

Ostiaks, character of, 235.
Overheating, 171.
Owens Lake, 321; salt in, 323 ff.
Oxygen, absorption by blood, 78;

consumption by crayfish, 132.

Pacific coast, climate, 221, 223; edu-
cation on, 285; manufacturing on,
290.

Palestine, climatic changes in, 389;
capacity of, 255; drought in, 887;
palm and vine in, 339 f.; popula-
tion of, 389; storminess in, 362;
tree growth and, 332 f.; vegeta-
tion of, 9 f .

Palestine and Its Transformation,
8, 28.

Palm, climate and, 9 f., 339 f.

Panama, death rate of, 62 f.; health
in, 61 ff.; malaria in, 329, 392.

Parable, of savages, 405.

Paraguay, index of civilization, 252.

Paralysis, in Paris, 156.

Paramoecium, temperature and, 180.

Parasites, and civilization, 2.

Paris, climate of, 220, 402; paraly-
sis in, 156; humidity in, 174.

Parsis, progress among, 300.

Partial correlations, 187, 195 ff.

Patagonia, climate of, 221.

Patterson, Dr., cited, 888.

Pearl, R., cited, 187.

Pedersen, cited, 13, 78, 96, 103, 128,
137.

Peking, climate of, 220.

448

INDEX

Pellagra, 60.

Penck, A., cited, 10, 326.

Pennsylvania, education in, 285;
eminence in, 308; immigrants to,
309; piece-workers in, 102 ff.;
seasons in, 154 f.

Peonage, 59.

Persia, 355; ancient climate of, 400;
early greatness of, 354, 357;
palms in, 340; storminess in, 362;
wind in, 138 f.

Persians, in Mesopotamia, 353.

Peru, climate of, 368; civilization in,
357; ruins of, 367; status of, 298.

Peten, malaria in, 327.

Pfeffer, cited, 129.

Philadelphia, influenza in, 187;
weather and deaths in, 208.

Phoenicia, ancient climate of, 299.

Phoenicians, early culture of, 356.

Phosphates, in blood, 157.

Physical activity, vs. mental, 105 f .,
148.

Physical work, and atmosphere, 168,
170; and change of temperature,
142 f.

Physicians, busy season of, 153 f.

Physiological reactions, and atmos-
pheric pollution, 169 ff.

Physiology, aridity and, 391.

Piece-work, and temperature, 162.

Piece-workers, day of week and,
122; in New England, 80 ff.

Pilgrims, 309.

Pittsburgh, factory operatives in,
14 f., 102 ff.; seasons in, 84; tem-
perature in, 162; weather and
deaths in, 208, 213.

Place, vs. race, 30 ff.

Plants, temperature and, 128 f.;
variability and, 145.

Pneumonia, deaths from, 20 f., 175
f.; variability and, 183 f.; weather
and death rate from, 178.

Poles, in World War, 274.
Political corruption, and national

decline, 27.

Poor whites, 42, 52, 97, 811; in Ba-
hamas, 33.
Population, decline in, 389; early

density of, 337; of Japan, 160; in

tropics, 58.

Population Problem, 25.
Porter, W. T., cited, 158.
Portland, Ore., climate of, 161.
Port Said, rainfall at, 400.
Portuguese, in World War, 274.
Potatoes, in Ireland, 25.
Practice, effects of, 86.
"Premium" plan, 89.
Presbyterian Hospital, at Hebron,

388.

Probable error, 197 f.
Professional men, climate and, 385.
Progress, map of, 291 ff.; railroads

as test of, 281.
Progressive desiccation, hypothesis

of, 316 f.

Protestantism, and progress, 301.
Providence, pneumonia in, 177.
Provinces, rank of Canadian, 429 f.
Psychiatric hospitals, 156.
Psychological vs. physical effects of

weather, 148.

Pulsations of climate, 17, 338 f.
Pulsatory hypothesis, 7, 315 ff.,

335 ff.

Pulse of Asia, 7, 28.
Pumpelly, Raphael, 7.
Pumpelly Expedition, 7.
Puritans, 308 f.; slaves among, 41.
Putter, cited, 134.

Quakers, 309.

Queensland, climate of, 384; death

rate, 376, 378, 381 ff.
"Queensland walk," 885.
Queenslanders, birth rate among,

INDEX

449

383 ff.; in World War, 377.
Quintana Roo, malaria in, 329.
Quito, relative efficiency in, 229;

temperature at, 227.

Rabbits, and temperature, 168.

Race, and civilization, 258 f.; maps
of, 300 f.; vs. place, 30 ff.; vs.
territory, 254.

Races, and aridity, 394 f.; and in-
fluenza, 188 ff.; capacity of, 296.

Racial character, and immigration,
309 f.

Racial inheritance, and civilization,
27,387; in Greece, 21 ff.

Racial mixture, and climate, 3.

Racial selection, and climate, 6 f.

Railroads, as test of progress, 281.

Rain, ancient, 400 f.; and climatic
changes, 9 f.; in Egypt, 398; and
rate of work, 98 f.; relation to
storminess, 398 ff.; in Turkey, 79.

Range of temperature, ideal, 222.

Ratzel, cited, 76.

Reclus, cited, 7.

Rectal temperature, 165 ff.

Red corpuscles, altitude and, 78.

Relative humidity, see Humidity.

Religion, among Arabs, 390; in
Dark Ages, 390; and progress,
301.

Reproduction, in paramoecia, 131.

Respiration, rate of, 166 ff.

Respiratory diseases, ideal climate
for, 176; and moisture, 174 ff.

Rhode Island, eminence in, 308;
railroads in, 281.

Rhodes, early culture of, 367.

Rhodesia, backwardness in, 73; cli-
mate of, 367; ruins of, 367.

Richmond, children in, 63; influenza
in, 190.

Rickets, seasonal course of, 157.

Rirers, freezing of, 325 f.

Riviera, climate of, 220.
Rochester, deaths and temperature

in, 213.

Rockefeller Health Commission, 47.
Rocky Mountain states, education

in, 285.

Rodishev, cited, 238.
Roman Catholicism, and progress,

301.
Romance nations, civilization of,

258 f.

Romans, capacity of, 255.
Rome, decline of, 397; disease in,

12; malaria in, 392 f.
Ross, Sir Ronald, cited, 393 f.
Ruins, in Central America, 328 ff.
Russia, climatic advantages, 404;

humidity data for, 233; possible

migrations in, 407 f.
Russians, in World War, 274.

Sabseans, 354.

Sacramento, suicides in, 225.

Saint Louis, weather and deaths in,
178, 208.

Saint Paul, capacity of, 255; in-
fluenza in, 189 f.; weather and
deaths in, 182, 208.

Salines, in lakes, 323.

Sand dunes, 327.

San Diego, climate of, 221; suicides
in, 225.

San Francisco, climate of, 164, 221,
224 f.; efficiency in, 229; stormi-
ness in, 334; suicides in, 225;
weather and deaths in, 178, 182,
208.

Sanitation, and influenza, 188 ff.

Sanscrit, 355.

Saranac Lake, seasons at, 92 ff.;
tuberculosis at, 94.

Sayles, R. W., cited, 839.

Scandinavia, climate of, 224.

450

INDEX

Scandinavians, in United States,

395.

School children, strength of, 92, 96.
Schools, attendance at, 282 f.; effect

on growth, 158; as test of prog-
ress, 282.

Scotch, in Australia, 382 ff.
Scotland, climate of, 222; progress

of, 396.

Screws, work with, 82.
Scythians, 353.

Seasonal migrations, 407 f., 409 f.
Seasonal variations in energy and

health, 154.
Seasons, ancient, 400; and children's

weight, 158 f.; effect of, 76 ff.;

human activity and, 92 ff.; in

Japan, 159 f.; rhythms of, 96.
Seattle, climate of, 161, 221 ; stormi-

ness in, 215.

Seistan, 340; wind in, 137 f.
Selection, in Greece, 27.
Semites, in Mesopotamia, 352.
Semple, E. C., cited, 341.
Sequoia Washingtoniana, 8 f.
Sequoias, 321.

Seventh century, dryness of, 390.
Sex, and influenza, 188 ff.; and

weather, 139 ff.
Sexes, proportion in Australia, 380

ff.; relation of, in tropics, 72 ff.
Sexual reproduction, 131.
Sheane, cited, 73.
Sheep, in Africa, 53 f.
Shifting, of civilization, 347 ff.; of

climatic zones, 315 ff., 327 f.
Siam, status of, 296.
Siberia, energy in, 236 ff.; humidity

data for, 233; status of, 296.
Sirocco, in Italy, 74.
Slaves, 41; in Bahamas, 47.
Slavic nations, civilization of, 260.
Slowness, of tropical people, 70 f.
Slubbing, 97 f .

Smoothing, of curves, 86.

Snow, and palms, 340.

Social degeneracy, and national de-
cline, 27.

Social standards, in tropics, 73 f.

Sodium, in lakes, 323 f .

South, books in, 51 ; changes of tem-
perature in, 144; human capacity
in, 86 ff.; life insurance in, 279;
slavery in, 41; teachers' salaries,
283.

South Africa, climate of, 45; fu-
ture of, 53; status of, 297; whites
vs. negroes, 43 ff.

South America, civilization in, 262,
298; climate of, 221, 223 f.; horse-
races in, 78; Indians in, 57; rela-
tion to Europe, 348.

South Australians, birth rate
among, 383 ff.

South Carolina, advantages of, 313;
education in, 287; effect of sea-
sons in, 97 f.; eminence in, 306,
311; high school graduates in,
283; school attendance in, 283.

South Dakota, education in, 286;
teachers' salaries, 283.

Southern California, climate of, 224.

Spain, death rate of, 62; in World
War, 274.

Specialists, opinions as to climate,
339 ff., 345 f.

Speeding, 97 f.

Spleen, effect of malaria on, 393.

Spooling, 97 f.

Spring fever, 128.

Standard Oil Company, 350.

Standard population, 292, 376, 381.

Standards of civilization, 254.

States, relative rank of, 429 f.

Stefansson, V., cited, 395.

Stevens, cited, 44.

Stock, P. G., cited, 44.

Storminess, ancient, 401; changes

INDEX

451

in, 398 ff.; civilization and, 12;
and climatic changes, 9 f.; distri-
bution of, 359 ff.; effect of, 214
ff.; and health, 181 ff., 204 ff.,
219; map of, 858 ff.; optimum,
161; and rainfall, 398 ff.; shifting
of, 327 f.

Storms, and deaths, 214 ff.; correc-
tion of curves of, 212; and health,
210; ideal, 223 f.; stimulus of,
147 f.

Strands, at Owens Lake, 325.

Strong, cited, 31.

Students, marks of, 104 ff.

Subtropical storm belt, 861 f.

Suicide, in California, 226.

Sumer, 352.

Summer, effect on growth, 158 f.;
effect on work, 115 ff.

Sunspots, 361 f.

Surgical operations, weather and,
179 f., 186.

Sweden, cold winters in, 326; manu-
facturing in, 290; progress of,
396.

Swedish bath, 146.

Switzerland, index of civilization,
252; manufacturing in, 290.

Syria, ancient climate of, 344 f.,
388 f., 399 f.; civilization in, 357;
desiccation in, 317; population of,
389; storminess in, 362.

Syrians, early culture of, 356.

Takla-Makan, autumn in, 226.
Tampa, cigar-makers at, 99; climate

of, 161; manufacturing in, 290;

seasons in, 93 ff.; temperature in,

162; whites vs. colored in, 83;

work by day of week, 125.
Tasmanians, birth rate among,

383 ff.

Taylor, F. B., cited, 820.
Teachers, salary of, 283.

Teheran, rainfall at, 400.
Temperature, ancient, 400; carbon
dioxide and, 133 ff.; changing
adaptation to, 397 ff.; and cli-
matic changes, 10, 334; correction
of curves of, 211 f.; and culture,
17; and daily deaths, 196 ff.;
effect of, 14 f.; and efficiency, 109
ff., 115 ff., 431 ff.; effect of high,
165; of glacial period, 340; and
health of cities, 204 ff., 208 f., 219;
and human activity, 124 ff.; ideal,
220; and influenza, 188 ff.; mag-
nitude of effect of, 149 f.; and
mental activity, 107 f.; and opera-
tions, 179 ff.; optimum, 161;
range of, 137; and vital processes,
130 ff.

Ten Kate, H., cited, 252.

Tennessee, eminence in, 311.

Teutonic regions, 258.

Teutons vs. negroes, 30 ff.

Texas, rank in education, 287.

Thomson, cited, 133.

Thucydides, 24.

Tibet, 355.

Tierra del Fuego, climate of, 224.

Tobacco, 290; and weather, 101.

Tobacco factories, seasons in, 97.

Toledo, deaths and temperature in,
213; influenza in, 189 f.

Tools, stone vs. iron, 371 f.

Torrid zone, effect on Europeans,
56 ff.

Transportation, and optimum cli-
mate, 396.

Tree growth, as measure of climate,
8, 320, 333.

Tree rings, in ruins, 338.

Tropical Australia, disease in, 377 1

Tropical diseases, 60 ff.

Tropical inertia, 70.

Tropical labor, 57.

Tropical medicine, 67.

452

INDEX

Tropics, adjustment to climate of,
408; character of Europeans in,
68; colonization in, 70; death rate
in, 378; influenza in, 192; insur-
ance in, 279; natural selection in,
385; relation to ideal climate, 226;
white men in, 66 ff., 376 ff.

Tubercular patients, seasonal effect
on, 92 ff.

Tuberculosis, in Adirondacks, 94;
and influenza, 188 ff., 191.

Tumors, and influenza, 188 ff.

Turk, rule of, 9.

Turkestan, expedition to, 7.

Turkey, effect of rain in, 79; in-
fluenza of United States in, 350;
winds in, 74.

Typewriting, and temperature, 127.

Typhoid fever, and influenza, 188 ff.

Ultra violet light, effect on blood,
157.

Uniformity, of tropical climate, 227.

United States, civilization in, 266;
climate of, 221, 223 f.; climatic
advantages, 404; climographs of,
182; coal in, 289; death rate of,
62 f., 376; deaths vs. climate in,
20 f.; distribution of civilization
in, 288; early inhabitants of, 369
ff.; effect of temperature in, 162;
humidity data for, 233; humidity
and death rate in, 178; impor-
tance of, 348 ff.; index of civiliza-
tion, 252; influenza in, 21; map of
civilization in, 277; maps of, 291;
negroes in, 395; psychiatric hospi-
tals in, 156; rank of states, 429 f.;
Scandinavians in, 395; South vs.
North, 36 ff.; statistics in, 275;
status of, 297; storminess in, 361;
vitality and education in, 275 ff.

United States Commissioner of
Education, 282.

United States Military Academy,
students' records at, 14 f.

United States Naval Academy, stu-
dents' records at, 14 f.

United States Steel Company, 350.

Utah, boys in school, 283 ff.; edu-
cation in, 285 f.; eminence in, 311;
rank of, 289; school attendance
in, 282.

Vacations, effect of, 106 f.; effect
on growth, 158; effect on work,
114.

Van't Hoff's law, 131 f.

Variability, effect of, 173; and
health, 181 ff., 201 ff., 204 ff.;
ideal, 223; in Italy, 163; magni-
tude of effect, 149 f.; in New
York, 183 ff.; optimum, 161; value
of, 138 ff.

"Varves," 320.

Vasotone, 165 ff.

Vedas, 355.

Vegetation, and Indian civilization,
371 ff.; in Mediterranean region,
9 f.

Venezuela, status of, 298.

Ventilation, 170 ff.

Vera Cruz, climate of, 408.

Vermont, eminence in, 306, 311.

Vicksburg, climate of, 402.

Victoria, birth rate among natives
of, 383 ff.; death rate, 381 ff.

Vine, climate and, 9 f.

Virginia, books in, 51; factory op-
eratives in, 14 f.

Vital processes, and temperature,
130 ff.

Vitality, in United States, 276,
278 f .

Wages, of factory workers, 83 ff.;
per day of week, 121 ff.; varia-

INDEX

453

tions in, 103; of whites vs. colored,
32 ff.

War of Secession, 41.

Ward, R. DeC., cited, 339.

Washington, D. C., colored children
in, 31 f.

Washington (state), death rate of,
64; progress in, 289, storminess
in, 217; teachers' salaries, 283.

Weather, vs. climate, 136; and
deaths, 20 ff.; and health, 194 ff.,
277; and influenza, 188 ff., 191;
numerical expression of, 190; and
operations, 21; and work, 15,
136 ff.

Weather Influences, 13.

Weavers, work of, 97 f.

Week, effect of day of, 120 ff.

Weekly migrations, 409 f.

Weight, gain of, 94; gain among
children, 158.

West, teachers* salaries, 283.

West Indies, civilization in, 266.

West Point, students' records at,
14 f., 104 ff.; changes of tempera-
ture at, 140 ff.

West Virginia, education in, 287;
eminence in, 311.

Wheat, growth and temperature,
129.

"White Australia," 377, 384 f.

Whites, humidity and death rate
among, 178; mentality of, 31; in
tropics, 56 ff., 376 ff.

Who's Who in America, 302, 310.

Will, weakness of, in tropics, 68.

Wind, ancient, 401; correction of
curves of, 212; effects of, 137 f.;
and health, 204 ff., 210; and re-
spiratory diseases, 175.

Wind of One Hundred and Twenty

Days, 138.

Winslow, C-E. A., cited, 165.
Winston-Salem, seasons at, 92 ff. s 97.
Winters, cold, 326; and growth, 158

f.; and work, 118 ff.
Wisconsin, immigrants to, 309.
Wissler, C., cited, 369.
Woeikof, cited, 337.
Women, in Bahamas, 50; birth rate

in Australia, 378; climate and,

385; death rate in Australia, 378;

distribution in Australia, 380;

dress of, 73 f.; in Japan, 160;

selection of, 24, 380; in tropics,

73, 378 ff.

Woodruff, C. W., cited, 26, 109, 130.
Work, and weather, 136 ff.
World, maps of, 294 f .
World Power and Evolution, 20 f.,

28.
World War, capacity of races in,

271 ; Queenslanders in, 377.
Wyoming, boys in school, 285; edu-
cation in, 285; eminence in, 311.

Yale Expedition to Palestine, 8.
Yale University, death rate at, 64,

66 f.; Nova Scotian students at,

270.

Yangtse valley, storminess in, 363.
Yellow fever, changes in, 12.
Yemen, climate of, 367.
Yucatan, capacity of, 255; ruins in,

8, 328 ff., 332.

Zimbabwe, climate of, 367.
Zulus, 43; backwardness of, 73.