‘ ar Ou UFO TK

a onl

= = 3 (=e
Oceanicadese Llaflis

ee e
EprpgeT aS
ecard Z
= Ri ha ’
py .

ae

CORNELL UNIVERSITY LIBRARY
Gift of
HERBERT FIsK JOHNSON ’22

Cornell University Library

RC 116.L59

ntrol in Panama; the eradicati

olin

Cornell University

The original of this book is in
the Cornell University Library.

There are no known copyright restrictions in
the United States on the use of the text.

http://www.archive.org/details/cu31924012513176

Mosquito Control
in Panama

The Eradication of Malaria and Yellow Fever in
Cuba and Panama

_® BY
a .
Joseph A. Le Prince, c.z., a.m.
Chief Sanitary Inspector, Isthmian Canal Commission
1904-1914

and

A. J. Orenstein, M.p.

Assistant Chief Sanitary Inspector, Isthmian Canal Commission

With an Introduction by

L. O. Howard, LL.p.

Entomologist and Chief, Bureau of Entomology
United States Department of Agriculture

With 100 Illustrations

G. P. PUTNAM’S SONS

NEW YORK AND LONDON

The Knickerbocker Press
1916

eo
1iG |
$O4

CopyRIGHT, 1916
BY
JOSEPH A. Le PRINCE

A-SEL FRM

The tknickerbocker Press, Hew Work

INTRODUCTION

RESIDENT ROOSEVELT early appreciated
the fact that great as were the engineering
difficulties to be surmounted in the making of the
Panama Canal, the difficulties of sanitation would
be fully as great, if not greater, and, before the
first Canal Commission was appointed, he told a
committee of the American Medical Association,
the American Association for the Advancement of
Science, and the New York Academy of Medicine,
consisting of Doctor Welch and Doctor Osler of
Baltimore, Doctor Musser of Philadelphia, Doctor
Bryant of New York, and the writer, that it was his
intention to seek for the best man in the world for
the task, to pay him whatever would be necessary,
and to give him full power. Fortunately, the man
needed was found in the person of Colonel (now
General) Gorgas, fresh from his triumphant clean-
ing up of Havana with the consequent elimination
of yellow fever and malaria. General Gorgas took

with him from Cuba to Panama a man who
iti

iv Introduction

had been his right hand in the cleaning up of
Havana, J. A. Le Prince, the author of this
book.

The writer well remembers one morning when,
before starting for Panama, General Gorgas and
Mr. Le Prince called on him at his office in Washing-
ton to talk about mosquitoes, and the confidence
which they felt of accomplishing their gigantic task
was inspiring. He asked General Gorgas to have
specimensof allthe Panama mosquitoes sent to him
for naming, and the General said, ‘‘I will assign Mr.
Le Prince to see that it is done.’”” Upon which
Mr. Le Prince remarked, ‘‘I will have to do it
soon, Doctor, for in a year or so there will be no
mosquitoes there!”

The brilliant results of the work of General
Gorgas and his assistants are now known over the
whole civilized globe. It has been an object-
lesson for the sanitarians of the world and has
demonstrated the vitally important fact that it
is possible for the white race to live healthfully in
the tropics.

And now, the whole world wants to know
how they did it. Every detail becomes of the
greatest interest, and, in consequence, what
Mr. Le Prince has written here will be not only
of great practical importance as a guide to

Introduction Vv

future work of the same character, especially
in the tropics, but also of permanent historic
value.
L. O. Howarp.
BuREAU OF ENTOMOLOGY,
U. S. DEPARTMENT OF AGRICULTURE.

CONTENTS

PART I. ANTI-MALARIA CAMPAIGN

CHAPTER
I.—Tue Status oF KNOWLEDGE oF ANTI-

MatrariA WoRK IN 1904 AND THE
PREVIOUS CAMPAIGN IN HAVANA

II.—THE SITUATION ON THE ISTHMUS IN 1904,
BEFORE AMERICAN OCCUPATION

III.—METEOROLOGICAL AND TOPOGRAPHI-
CAL CONDITIONS. ‘TOPOGRAPHICAL
CHANGES AND THEIR BEARING ON
IsTHMIAN ANOPHELES

IV.—TwHeE SpPEcIES oF ANOPHELES ON THE
IsTHMUS

V.—ANOPHELES PROPAGATION AREAS .

VI.—HARBORING PLACES AND Foop OF
ANOPHELES

VIIL—FLIiGHT AND ATTRACTION oF MosQuI-
TOES . : : ? : ‘i

PAGE

16

27

43
55

83

94

Vili Contents

CHAPTER PAGE
VIII.—AtTTAcK ON PROPAGATION AREAS BY

FILLING. j ‘ : . II5

IX.—ATTACK ON PROPAGATION AREAS BY
DRAINAGE . ‘ és ‘ . 122

X.—ATTACK ON PROPAGATION AREAS BY
OILING : : : : . ISI

XI.—AtTrackK ON PROPAGATION AREAS BY
LARVACIDES : , : . 173

XII.—ATTACK ON PROPAGATION AREAS AND
ApuLt MosquiTroEs BY NATURAL
ENEMIES . : ‘ . . 180

XIII.—AtTTacKk ON PROPAGATION AREAS BY
CLEARING BopIEs OF WATER - 190

XIV.—ATTACK ON PROPAGATION AREAS BY
REMOVAL OF JUNGLE . . . 195

XV.—SCREENING AND PRACTICAL DESTRUC-
TION OF ADULT ANOPHELES IN
Houses. : : : - 198

XVI.—THE ReEsULTS ACCOMPLISHED BY THE
ANTI-MALARIA CAMPAIGN . . 218

PART II. THE YELLOW FEVER CAMPAIGN

I.—Tse Campaicn IN HAVANA . . . 231

II.— THE SITUATION ON THE ISTHMUS BEFORE
SANITARY WORK WAS STARTED . 261

Contents

CHAPTER
III.—GEoGRAPHY, METEOROLOGY, ETC., AND

THEIR BEARING ON THE PRESENCE
oF AEDES CALOPUS .

IV.—Tue First Sanitary Work Done IN
PanaAMA—ITS PURPOSE

V.—TuHE ANTI-YELLOW FEevER CAMPAIGN
AND ITS RESULTS . 5 ,

VI.—MEAsvurRES TAKEN TO KEEP THE IsTH-
MUS FREE FROM YELLOW FEVER .

VII.—THE VALUE OF YELLOW FEVER ERaDI-
CATION IN THE CONSTRUCTION OF
THE PANAMA CANAL

INDEX

ix

PAGE

268

271

293

306

314

325

NOTE—The word mosquito refers to the Anopheles
in Part I., and to Aédes calopus in Part II., unless

otherwise stated. /

/

Z ;

{ ‘ {
Steger, Ui/ Lat de ICM fr
i f

¥
/ { vy

ILLUSTRATIONS
CHart SHowInNG MarariA DEATH-RATE AT
Havana, 1871 TO 1914
AntT-Guarbs: OLp HospiTaL AT ANCON . ;

A DitcH at ANCON: WEEDS RETARDING
WatTER FLow . : : : 3 ;:

JUNGLE COVERING A STREAM .

Map oF THE CANAL ZONE
METEOROLOGICAL TABLES I TO 9
RAINFALL: CURVES FOR 1909, I912, 1913
A TopoGRAPHIC CHANGE: NEAR EMPIRE

CONSTANTLY CHANGING TOPOGRAPHY: NEAR
MIRAFLORES

DaILy CHANGE OF TOPOGRAPHY

BLOCKED WATER-CouRSE CHANGED INTO AN
ANOPHELES BREEDING PLACE . ; :

AQUATIC VEGETATION SHELTERING LARV

SWAMP NEAR BALBOA, CAUSED BY BUILDING OF
DIKES . F : : ; : :

PAGE

13
18

20
22
26
28

37
38

38
40

42
48

52

Xil Tilustrations

THE RIsING WATERS OF GATUN LAKE . :
CATTLE ON Low PAsTURE LANDS: PANAMA.
Tue EFFECT OF FLoops: EMPIRE . : A

ALG& UNITED WITH OIL IN MAT-LIKE MAssEs:
EMPIRE . ‘ ; : : ; ;

RAILROAD TIE LEFTIN A TREE BY A SUBSIDING
FLoop . . : , 7 5

OILING OF BREEDING PLACES ALONG A RIVER-
BANK 4 . A ;: F 2 i

VEGETATION PROTECTING Mosquito LARVA .

A SEEPAGE OUTCROP, CONTROLLED BY AN INTER-
CEPTING TILE DitcH: ANCON

VEGETATION REMOVED FROM THE EDGE OF A
PonpD AT BoHIO : ; F ; :

RESULT OF BLOCKING THE RIO CARDENAS

PROPAGATION AREA OF GATUN MosQuITO
INFLUX WHERE Mosquito FLIGHT WAS
DISCOVERED . . : : F ‘

DETAIL OF ANOPHELES BREEDING GROUND
WHERE FLIGHT OBSERVATIONS WERE MADE

GRASS AND PLANTS ALMOST COVERING THE
WATER SURFACE AND PROTECTING Mosquito
Larve . : 5: ‘ : ; j

DIAGRAMS SHOWING BEST LOCATIONS FOR
SCREENED Doors . j 3 5 :

58

62

64
66

68

79
72

76

78

78

87

Illustrations xiii

PAGE
A woPHELES COLLECTED IN THE GROUND CRACKS

UNDER THE HOUSEs . : ; : - 90
THE RELATIVE PosiTIoNns oF PANAMA, ANCON,
CoROZAL, AND MIRAFLORES : : » 95
Mosourro Trap ATTACHED TO A WINDOW
SCREEN . ‘ . F ‘ ‘ - 96
Mosouiro Trap (SEEN FROM INSIDE THE
House) . ‘ . : ; ‘ . 96
EXPERIMENT STATION: COROZAL : . 98
THE SHORE OF GATUN LAKE . : . 100

Map oF GATUN: SHOWING ANwoPHELES PROPA-
GATION AREA AND HOUSES WHERE STAINED
SPECIMENS WERE RECAPTURED . ; . IOI

FLOATING ISLANDS BLOWN TOWARD GATUN
LAKE , ; ’ ' , : . 102

Masses oF AQUATIC PLANTS AND FLOATING
IsLANDSINGATUN LAKE . ‘ ‘ . 102

THE AwNoPHELES BREEDING GROUND NEAR
GATUN . : ; ; : ‘ . 106

A Device FoR DETERMINING THE FLIGHT
DIRECTION OF A NOPHELES ; ; . 108

TABLE SHOWING NuMBER OF ANOPHELES
CAUGHT IN Houses AT GATUN . ‘ . I10

STAINING ANOPHELES WITH ANILINE DYE TO
DETERMINE LENGTH OF FLIGHT : . (12

xiv Illustrations

ANOPHELES IN A MosqutTo-BAR EXPOSED TO
AIR CURRENTS AT NIGHT . j ; a

CRACKS IN A HyDRAULIC FILL ‘ f F

LONGITUDINAL SECTION AND CROSS-SECTION OF
AN INTERCEPTING TILE DRAIN . : P

DETAIL OF OUTLET OF A TILE DRAIN : ‘

PLAN AND SIDE VIEW OF A JUNCTION OF TILE
LINES. ; ; : : . ,

CONCRETE LINING OF A DITCH OF SMALL CRrOSS-
SECTION i : ; : :

REINFORCED THIN CONCRETE LINING OF A
ROADSIDE DITCH, AND SUPPORT OF THE
SLIDING TOE OF A HILL WITH CONDEMNED
Cross-TIES_. : : P : :

PLAN AND CROSS-SECTION OF A KEY-WALL AND
POSITION OF WEEP-HOLES : :

CONCRETE DitcH LINING : i : .
CONCRETE LINED DitcH AT BALBOA ‘ .

KrEyY-WALLS TO PREVENT WATER LEAVING
LINED DITCHES F ; ‘ és ;

Brancu DITCHEs . , ‘i p i

JUNCTION OF CONCRETE DITCHES, SHOWING
SPLASH-WALL . ‘ . , ; :

CONCRETE LINING FOR THE BOTTOM OF A SMALL
DitcH . ‘ 7 , ‘ F ‘

PAGE

142

142

Illustrations

JUNCTION oF A Branco DitcH witH A LINED
DitcH

BuRNING GRASS FROM THE SIDE OF A DITCH;
CRUDE O1L UsEpD As FUEL

CONDITION OF THE DitcH Two MontTHS AFTER
BuRNING.

DitcH CLEARED BY HAND Lasor, SHOWING
ConDITION Two MONTHS AFTER REMOVING
GRass

A GaTE To IMPOUND WATER FOR FLUSHING A
DitcH: EMPIRE

REMOVAL AND DESTRUCTION OF LARVZ AND
MatTtep ALG& IN A DiTcH BY FLUSHING:
EMPIRE

A Srmmpce Form oF Or Drip
Tue FLat LAMPWICK DRIP FOR HEAvy OILS
Or Drip APPLIED TO A HILLSIDE STREAM

OILING OBSTRUCTED STREAMS ON CuT-OVER
LANDS

BRUSH IN VALLEY OF PEDRO MIGUEL RIVER .

APPLICATION OF LARVACIDE OR OIL BY KNAP-
SACK SPRAYER: MIRAFLORES 2

Spray NozzLes . ° : 4 . 7
FreLp SUPPLY TANK . . . ‘ 5

Om-CarT FOR APPLYING OIL TO ROADSIDE
DITCHES : s ‘ : ‘ ‘3

xV

PAGE

143

144

146

148

150

150
158
160

162

164
164.

166
167
168

168

XV1 Illustrations

Pack-MULES, FOR OIL TRANSPORTATION IN
JUNGLE TRAILS : : ; _ ‘i

THE LARVACIDE PLANT AT ANCON . i: ‘

REMOVING VEGETATION FROM OVERGROWN
STREAMS : . : : ‘ F

SCREENED VERANDAS: ANCON : ; :

A Goop DEVICE TO PREVENT THE SAGGING OF
SCREENED Doors. ; : é z

Types OF SCREENED HousEs: COLON HOSPITAL
GROUNDS

TYPES OF SCREENED Houszes: CULEBRA

SCREENED CARS IN RAILROAD CONSTRUCTION
CAMPS

TABLE OF MALARIA INCIDENCE AT CONSTRUC-
TION CAMPS

SLAPPER, CHLOROFORM TUBE, ACETYLENE
Lamp, AND PiILL-Box; USED IN CATCHING
ANOPHELES IN HOUSES f ; .

Mosouito TRAP, WITH DETAILS oF Con-
STRUCTION . : ; : 2

INTERIOR OF LABORERS’ BARRACKS; WHERE
ANOPHELES WERE CAUGHT DAILY F :

CHART SHOWING PERCENTAGE OF EMPLOYEES
SICK WITH MALARIA IN THE WET SEASONS
OF I9II, 1912, AND 1913. . . :

PAGE

170

174

192

192

201

204

206

206

209

212

214

216

221

Illustrations xvii

PAGE
DIAGRAM SHOWING THE YEARLY REDUCTION

OF MALARIA AMONG CANAL EMPLOVEES . 226
CHART SHOWING THE MonTHLY MALARIA
SICK-RATE FROM 1906 TO 1914 . : . 228
Mar oF HAvANA; SHOWING THE YELLOW
FEVER DISTRICT : : : : . 236
CHART SHOWING THE PREVALENCE OF YELLOW
FEVER IN HAVANA, CUBA . : F . 244
CHART SHOWING YELLOW FEVER CASES ON THE
ISTHMUS FROM 1881 TO 1889 : 4 . 264
PART OF THE PANAMA FUMIGATION BRIGADE . 284
SCREENED VESTIBULE Door : ‘ . 286
A PoorLy CONSTRUCTED EAVE-TROUGH . . 288
A CLoGGED EAVE-TROUGH . ; , . 288
DIAGRAM OF A FIRMLY SUPPORTED EAVE-
TROUGH : ' ; : : . 289
A FrrMLy SUPPORTED EAVE-TROUGH WITH A
Goop SLOPE . , ‘ F . . 290
MosoguiTo-PRooF WATER BARREL . : . 290

A SHort LENGTH OF EAVE-TROUGH USED OVER
ADooRWAY . : ‘ ‘ ; . 310

PART I
Anti-Malaria Campaign.

CHAPTER I

THE STATUS OF KNOWLEDGE OF ANTI-MALARIA
WORK IN 1904 AND THE CAMPAIGN AT HAVANA

HE anti-malaria work of importance accom-

plished previous to 1904 was that done by

Colonel Gorgas in Cuba and Sir Ronald Ross in the
British Colonies in the East.

The necessity for mosquito control has been
appreciated since it was first demonstrated that
some mosquitoes may transmit disease. Since
1900 great progress has been made in the study of
the life history and habits of mosquitoes. As a
result, we find remarkable differences in habits,
even among species of the same genus. The
determination of such facts is of great importance
in the economic control or eradication of the dis-
ease-conveying species, and also in that of other
genera that affect personal comfort or real estate
values.

It is realized that conditions pertaining to
Anopheles control on the Isthmus and the habits

3

4 The Control of Mosquitoes

of the species encountered may differ from those
which will be found elsewhere, and that methods of
procedure for malaria control must be modified
accordingly; but if in a large territory like Panama,
where conditions are most favorable for the
multiplication of mosquitoes every day in the
year, malaria has been controlled at a reasonable
cost, the methods which accomplished the change
are worth consideration.

Health officials from foreign countries have
visited the Isthmus to obtain practical infor-
mation in order to apply this later in malaria and
yellow fever eradication abroad.

Much interest was shown by the thousands of
Americans who visited the canal, and many of
them asked why the same class of work could not
be attempted in malarial districts in the United
States, and also sought information about the
work accomplished and methods of procedure.

The canal will be used by the commerce of all
nations, and the progress made in sanitation in
Cuba and Panama will have far-reaching effects.

The character of the work at Havana was quite
different from that required on the Isthmus. The
general situation resembled more closely the condi-
tions and drainage problems occurring in the
southern part of the United States, and subtropical

The Campaign at Havana 5

countries not having an excessive annual rainfall.
This work was started in 1901 when the yellow
fever campaign was well advanced. No anti-
malaria work had been attempted elsewhere pre-
vious to the Cuban anti-malaria campaign. Ross
started his practical field work in the East at the
same time that work was begunin Havana. There
was no previously acquired information available
on the subject nor any known practical methods
that could be followed, or that could assist those in
charge of operations. At that time, very little was
known regarding the habits and life history of
Anopheles. It was practically all pioneer work,
and improvements and methods of procedure were
devised as the work progressed. Years of subse-
quent experience have shown that for pioneer work
the scheme was excellent.

A survey was first made to determine the princi-
pal sources of Anopheles affecting Havana and its
outlying suburbs. The most important propaga-
tion areas received first attention. Treatment of
pools with oil having a paraffin base was tested.
As the wet places harboring Anopheles were widely
scattered over a large area, a covered wagon was
used for transporting laborers, tools, and oil to
control as much territory as possible with the
small force employed. The plan of oiling was

6 The Control of Mosquitoes

arranged so that areas were not treated at regular
intervals, but when Anopheles and other mosquito
larve were present.

The system of inspection was well carried out,
and inspections were independently checked at
proper intervals. The result of Anopheles eradi-
cation by use of oil was fairly good, and better
results were obtained than expected under the
climatic and topographical conditions on the
Isthmus. Very few topographical changes were
being made in the vicinity of Havana. The class of
labor available was very good, and fairly intelli-
gent. The wages paid to the men in the Anoph-
eles brigade were a little above average laborer’s
wages; this enabled the sanitary department to
obtain the best class of labor and toretain it. The
men used their heads as well as their hands, and
worked with interest. The unit cost of ditching
and of oil application was very low, and much was
saved by the proper selection and payment of
labor. Poor labor on anti-malaria work is most
unsatisfactory, and no saving has ever been made
by its employment.

It was soon found that there were disadvantages
in using oil, as well as much repetition of work that
could be eliminated. As rapidly as the limited
appropriation would allow, the pools costing most

The Campaign at Havana 4

to control were drained or filled in, and the area
under control thus increased. Each place drained
or filled became one less to inspect and treat.

As time went on, the anti-Anopheles work
consisted chiefly of draining and filling depressions
in fields where water collected. The use of what is
known as the ‘‘herring bone’’ system of ditching,
the training of streams to more direct courses, and
the reducing of the average wetted section of
ditches, rapidly decreased the necessity for regular
application of oil and for frequent inspection.

The geological formation and topographical
conditions were favorable. The watercourses and
areas to be treated were often distant from one
another, yet the streams were of fair grade, and
most of the areas needing treatment were located
so that they could be drained. In some cases the
impervious surface clay rested upon a coral for-
mation. The latter was of a porous nature, and
absorbed a large quantity of water in a relatively
short time. In some places where there were
ponds, and surface drainage was out of the ques-
tion, they were connected by one or more ditches,
and led to shallow wells in the coral formation,
where the water was absorbed. When one well
was not sufficient, two or more were made and
connected, to increase the absorbent surface.

8 The Control of Mosquitoes

It was found that the bottom of these wells became
clogged with clay and silt, but that the sides
supplied sufficient absorbent surface to give
satisfactory results. Fish were used to keep some
of the wells free from mosquito life; in others, that
frequently became dry, oil was used. Sometimes
it was necessary to scrape well walls several times
during the wet season to increase absorption.

The limited funds available did not permit the
permanent lining of ditches, although in many
cases it would have been more economical than
treating and maintaining them for an indefinite
length of time.

Anopheles production in wet areas caused by
seepage occurs in Cuba, and one case near Havana
was especially interesting. In 1900, a few houses
were located near the top of the hill known as
Jesus del Monte. Malaria prevailed among the
tenants, and several deaths from pernicious malaria
occurred there immediately before the undertaking
of anti-malaria work. The hill appeared to be
dry, and was well above the elevation of the sur-
rounding ground. Many inspections were made to
locate the source of Anopheles with negative results.
It was supposed that they came from a distance,
but an extended inspection of the surrounding
country disproved this conclusion.

The Campaign at Havana 9

Finally, it was decided that the breeding place
must be local, and the labor force was put in line,
men eight feet apart so that every square yard of
the hill would be seen, and no wet place, however
small, could be missed. They marched along for
some time before a piece of partially soft ground
was found. There were no surface indications,
such as a change in the vegetation or its color,
to mark the outline of the wet area. A thin film of
water was found to outcrop along a contour line,
to run along the surface for a distance of about
twenty feet, and then disappear into the ground.
The mosquito larve were very plentiful. A ditch
was made above the seepage outcrop line, but
failed to intercept the water. Two short and deep
parallel intercepting ditches were made about nine
feet apart, but failed to prevent the water coming
up between the two ditches. It became evident
that the local porous water-bearing strata were
inclined and pointed upwards, and that a distant
pressure was forcing the water uphill. Ditching
was out of the question. At a small cost the
ground affected by the surface flow was covered
with gravel and cinders to a depth of about six
inches, so that the mosquitoes could not reach the
water which continued in its original path.

Since that time the importance of hidden mos-

10 The Control of Mosquitoes

quito-producing areas, and their relation to the
malaria situation, has been fully appreciated by
those engaged in anti-malaria work. Beginners
usually learn facts by temporary failure. Soon
after the completion of this work the adult Anoph-
eles disappeared from the vicinity. No new cases
of malaria occurred. The houses that had been
so long vacant because of the prevalence of malaria
were again occupied. When it became known that
malaria no longer existed in that locality, it was
plotted into lots, and has become one of the popu-
lar residential sections of the town. Real estate
values have advanced accordingly.

On low flat lands where cattle were pastured, it
was found difficult to prevent Anopheles propaga-
tion in the water retained by hoof-prints. It be-
came essential to prevent certain portions of fields
from being used for grazing purposes during the
wet season.

It was also found that plants along the shore
line, as well as aquatic plants, reduced the surface
velocity of water in streams and rivers to such an
extent as to afford protection and harboring
places for mosquito larve. The vegetation at the
edges of ponds and streams was kept cut short.
Aquatic vegetation was pulled up by the roots and
removed. Small fish were very efficient larve

The Campaign at Havana II

exterminators after the clearing and training of
natural watercourses was accomplished. Condi-
tions affecting growth of vegetation and alge were
less favorable in Cuba, and, in consequence, fish
are of more use there as destroyers of mosquito
larve than on the Isthmus.

Beds of watercress in the Chinese truck gardens
were a prolific source of Anopheles in Havana.
The cress beds were made by erecting small dikes
on flat land, and bringing water to them by ditches
from nearby streams. When the enclosure was
filled, the water was left standing for long periods.
In order to control the situation, small wooden
sliding gates were installed at each end of the
watercress beds, which were flooded at frequent
intervals for short periods, and the water then
allowed to escape back to the stream. This
scheme, when properly carried out, practically
stopped Anopheles development. The gardeners
who refused to adopt the new method were not
allowed to grow watercress. The crops grown in
this manner were as vigorous as those raised in the
continuously flooded patches.

Between the heavy showers that occur near
Havana there are dry periods, accompanied by
strong winds. When the ditches and watercourses
are kept fairly free from vegetation, the heavy

12 The Control of Mosquitoes

showers are often beneficial, as they remove any
Anopheles larve clinging to banks of ditches or
cleared streams. The trade winds assist evapora-
tion to a considerable extent, and help to dry the
depressions in flat areas that hold water. The
soil is of such a character that the trained streams
and ditches can be kept to a proper section with
less work than is necessary to attain this result
at Panama. Washouts and caving of ditch banks
did not happen as frequently.

In many places apparently well fitted for the
support of Anopheles larve, they were absent;
yet lived and developed when placed therein as
an experiment. The reason why Anopheles eggs
are not laid in certain areas apparently in every
way similar to those in which larve are found is
yet unexplained. The season of Anopheles pro-
duction in Cuba is much shorter and less continu-
ous than on the Isthmus. It was largely due to
these causes that oiling at irregular intervals gave
satisfactory results. Conditions of soil, topo-
graphy, and climate there do not favor the rapid
and continuous production of mosquitoes as much
as in the Canal Zone.

The drainage connected with the Havana
campaign was completed within a year of the time
anti-malaria work began, and its maintenance

453_deathe
1

ell
iy FI
[$0 deaths per 100004 11 5
}| Bi
yf eS
i Ei
{ fg
ATF { a
JLT |
i]
EH
lor deste per 10 papa TNT TTT A
fl ra
if 3] stelaiel=
L § ke
A Fim ped ES ) cal Co) ad
Bidcaths por
EI \
he
G)
7 iS death] per [10000
i
priieatnn 8 3 8 228 28,1 Lidests
ae ot. AR aap ulation
2 Deaths eae by Malaria in Havana 1871-1914 Z

a
Anti Malaria Campaign was Started in 1901 -Note the Curve from 1901 to 1914

1871 to 1880 - 3416 deaths

1881 to 1890 = 2050

1891 to 1900 - 5633 &

1901 to 1910 - 444»

191 to1914- 25» (for 4 yeara)
Te71to1914 (12,607 In 44 years)

Chart showing malaria death-rate at Havana 1871 to 1914.
13

14 The Control of Mosquitoes

since then has been mostly a question of economy.
The sanitary department of the Republic of Cuba
has continued the work started by Colonel Gorgas
and deserves great praise for the excellent record
it has maintained.

To review the results of the Havana campaign,
the records show:

Population of Havana, 350,000.

From 1890 to 1900,—5643 deaths from malaria.

From 1900 to 1910,—444 deaths from malaria,
with a largely increased population.

It is interesting to note the rapidity with which
malaria control was accomplished, as shown by the
following table:

TABLE OF DEATHS FROM MALARIA IN HAVANA, CUBA,
I90I TO IgI2

No. of No. of No. of
Year) Deaths | Yer | Deaths | Year| Deaths

1901 I51 1905 32 1909 6
1902 77 1906 26 1910 15
1903 51 1907 23 19II 12
1904. 44 1908 19 I9I2 4

Malarial fever is almost eliminated from Havana
and its suburbs.
The cost of maintaining this condition is very

The Campaign at Havana 15

low, and might be further reduced by lining the
bottoms of the small streams and ditches with
reinforced concrete, in order to lower the cost of
cleaning and ditch maintenance.

CHAPTER II

THE SITUATION ON THE ISTHMUS IN 1904, BEFORE
AMERICAN OCCUPATION

RESIDENT ROOSEVELT appointed Colonel
W. C. Gorgas as Chief Sanitary Officer of the
Isthmian Canal Commission, and it was under his
leadership that the first sanitary survey was made.
He established temporary headquarters for his first
assistants at Ancon, where the French had terraced
the hillsides and had established a hospital of
about twenty wards. The situation selected by
the French was excellent, but to-day, from our
more advanced knowledge, we can readily under-
stand why their sick-rate was so high, and can
also realize that it would have been equally so
under an American administration working with
the limited knowledge pertaining to malaria pre-
vious to the year 1900.
At the time the American government assumed
control of the Canal Zone, there was a succession

of villages strung along the line of the canal; most
16

The Isthmus in 1904 17

of them close to the camps occupied by the laborers
during the days of the French Canal Company.
They were located apparently with reference to
proximity to work and many were near a conven-
ient source of water supply. At the time these
camps were established, nothing was known regard-
ing the transmission of malaria. Better locations
were available, but they were in nearly all cases
far distant from the canal. Some of the camps
had been abandoned altogether and could not be
reached without cutting pathways through the
jungle.

In some cases, where the rafters had decayed
and the roofs had collapsed, trees with a diameter
of ten inches were growing inside the buildings.
The natural watercourses were overgrown by
dense jungle, and nearly all of the villages were
practically in the jungle itself. The French
engineers had, for drainage purposes, placed
masonry lined ditches near some settlements, and
the work was excellent; it is in as good condition
to-day as at the date of completion, and will prob-
ably remain so for many years to come.

In June, 1904, about 1100 laborers were em-
ployed in canal work at Culebra; they lived in
the adjacent villages, and very few isolated houses
existed in the bush. Many of the old French

3

18 The Control of Mosquitoes

buildings were occupied by laborers who had re-
mained after the financial failure of the French
company.

Malarial fever was prevalent among these people,
and they had neither the knowledge nor the means
to combat it. At Paraiso, in September, 1904,
there was a population of 350 of which 42 were
sick in bed. At Bohio, a village fifteen miles
south of Colon, blood samples were taken from
natives, and eighty per cent. showed malaria
infection. But even this poor and ignorant class
of people took steps to do the best they knew
how; they prayed for deliverance from fever and
shunned certain localities in which the sick-rate
was abnormally high as compared with the pre-
vailing high fever rate.
> At Corozal, the nearest village to Panama, only
one house was occupied, and many colored peo-
ple told us that ‘‘to live there is to be sick and
die.”

The situation at Ancon Hospital was extremely
interesting. The grounds were magnificent; there
were palms, beautiful crotons, and other deco-
rative plants near the wards and on the terrace,
but the insects, and especially the ants, destroyed
them repeatedly. The gardeners partially over-
came this difficulty by placing hollow earthenware

Ant-guards: Old hospital at Ancon

The Isthmus in 1904 19

Tings of a semi-spherical section around each
plant. They were kept filled with water and of
course contained many larvee of both Aedes calopus
and Anopheles as well as alge, vegetable debris,
etc.

The ditches near the wards were lined with field
stone; the spaces between the stones were filled
with earth and grass which retarded the current of
water. These ditches discharged onto the grass-
covered, clayey soils, and pools remained there,
distributed over a large area.

The natural topography of the hill gave a fairly
steep slope with a more gently sloping bench below
it. The highest part of the hill is about six hun-
dred feet above sea level. The water absorbed by
the upper part of the hill came to the surface at a
lower level on a clay-like formation covered with
grass, making the hillside resemble a bog in many
places. The outcrop of water, which oozed
slowly from the ground, often followed the contour
of the hill for a considerable distance. Conse-
quently the-area immediately below that contour
was wet almost continually.

There was a cow pasture on the top of the hill,
and throughout the rainy season the hoof-prints
remained wet or full of water. Ditches of varying
widths ran down the slope, and the water spread

20 The Control of Mosquitoes

over the ground and sometimes did not return to
the lower part of the ditch. In addition, the foot
of the hill formed the shore line of an extensive
swamp. ~ These adverse conditions, and wet areas
covered with vegetation, encouraged Anopheles.
A more prolific source would be hard to imagine.

The number of adult Anopheles present in the
hospital wards and buildings corresponded with
the numerous favorable propagation areas on and
near the hillside. At night, comfort was impos-
sible and mosquito bars indispensable. Not a
single building was screened. The malaria and
yellow fever patients had been accessible to
Anopheles and Aedes calopus, and the other patients
were near them. The patients in wards were
located according to nationality instead of accord-
ing to the nature of their illness. Had it been
intended to spread yellow fever and malaria with
the greatest rapidity among the patients as soon
as they arrived, no better plan could have been
adopted. Similar conditions have existed in other
tropical countries; in fact, were customary in past
years. -

Before sanitary work was fairly under way, with
one or two exceptions, the small hospital sanitary
force was down with malaria. During the first
year of work, all the force suffered. At one time

A ditch at Ancon: weeds retarding water flow

The Isthmus in 1904 21

it became necessary for the sanitary inspectors
to be employed all night in clerical work. The
Anopheles were so numerous that night work had
to be done in relays; one set of men using fans to
protect those working. We were two thousand
miles from the source of supplies and had to wait
for screening, its installation, necessary supplies,
etc.

When the hospital was first opened and treat-
ment given without charge to employees, the
native laborers who had known of the suffering
and deaths in previous years at Ancon refused
to go to the hospital, stating that they ‘preferred
to die at home.’’ We could readily sympathize
with their ideas, knowing what had happened in
the past. No attempt was made to approximate
the number of Anopheles present. The prevailing
species was Anopheles albimanus. Inside one of
the buildings near the hospital, fifty-four adult
Anopheles were noted on the upper panel of a
screen door.

It was desirable to know the time of the day
or night when the Anopheles would take blood.
In July of 1904, men were dressed in white clothing
and dark coats and made to lie on cots in an un-
screened ward at Ancon. They were furnished
with pill-boxes and a clock. Each time a mos-

22 The Control of Mosquitoes

quito bit them, or tried to, it was captured and
placed in a pill-box and the date and hour written
on the box. It was found at that time that
Anopheles in the ward attacked men at rest at all
hours of the day, and at night they became too
numerous to make the work pleasant. It was
noticed that they did not bite as freely near the
doorway, where there was more light during the
daytime, and that they absolutely refused to fol-
low a man out into the bright rays of the sun.
Before anti-malaria work was begun, a prelim-
inary survey was made and Anopheles production
areas were found near every existing settlement,
as well as all abandoned camps that had been
used by the French. The greater part of these
areas was covered with dense jungle, and their
extent could not be at once determined. There
were numerous seepage outcrops and their out-
put varied according to the rainfall; and prob-
ably the contour line of seepage varied also.
The character of the soil was often such as to
retain the rainfall for relatively long periods, and
innumerable small depressions existed, which
might produce mature pup or not according
to weather conditions. The survey showed that
it would be necessary to clear off large areas of
jungle near settlements before any effective results

oOlog jv IaATy soiseyD oy} suriequs wiealjs & BuTIaA0O afsun[

The Isthmus in 1904 23

could be obtained, and it was decided to begin
work at the camps that were first to be occupied
by the laborers and American forces. It was
thought also that oiling of wet areas would be
attended with less successful results than in the
Cuban campaign, because the rainfall of the
Isthmus was often sufficiently continuous to re-
move all traces of oil before it had had much
effect on the mosquito larve. When pathways
were cut through the jungle new mosquito pro-
ducing areas were found.

The statistics of the days of the French Canal
Company are incomplete, and include only a
portion of the deaths due to malaria, as only cases
in hospitals, or patients able to pay for hospital
treatment, were recorded.

In the city of Panama, which was not so badly
infested with Anopheles as Colon, the malaria
sick-rate and death-rate were high. The number
of deaths recorded as due to malarial fever was:

Date Number of Deaths Population
From 1884 to 1893 20,000 to
inclusive 3504 22,000
1885 687 20,276

During 1906, the year of highest malaria sick-
rate of the American period of occupation,’ among

z

24 The Control of Mosquitoes

21,934 malarial fever patients entering the hospi-
tals, 195 deaths occurred, or less than one per cent.
of the cases. It can be readily understood from
these death-rates how prevalent infection was in
the city of Panama before 1904. The hospital
physicians who were at Colon and Panama previous
to 1904 state that over seventy-five per cent.
of all hospital patients had malaria. Of course,
in the villages out in the Canal Zone surrounded
by jungle, breeding places for Anopheles were
quite numerous and close to the houses. The
question of controlling malaria appeared at first
sight to be utterly hopeless. The United States
Public Health Reports describe the situation as
follows:

From Panama to Colon, a distance of forty-seven
miles, along the railroad there are many villages, in
fact almost a continuous settlement the entire dis-
tance, the total estimated population being 15,000
in January, 1904. The population is almost entirely
negro and Chinese. All the villages are filthy, with-
out regulations or restrictions, without sewers, and
having the usual water supply of the country, viz., rain-
water during the wet season, and water from streams
during the dry season. No attention is paid to the
wholesomeness of the source of the water supply.
Mosquitoes are prevalent in all these villages, breeding
in rainwater barrels, in the swamps, along the streams,
and in ponds. Malaria, elephantiasis, and beriberi

The Isthmus in 1904 25

are always to be found, and yellow fever and smallpox
will occur when favorable clinical material presents
itself, unless proper precautions are taken. -

In order to obtain certain data concerning
malaria transmission, an experimental station was
established at Ancon, and non-immunes were
isolated there for the experiments, but so many
Anopheles al imanus were present that it was
extremely difficult to prevent natural malaria
transmission.

Low lying flat areas occurred for a distance of
six miles inland from the Pacific terminal and to
a still greater distance from the Atlantic Coast.
Much of this area was affected by the highest
tides, and Anopheles larvee were present in streams
on steep hillsides as well as on the lowlands.

The preliminary survey showed that Anopheles
and other mosquitoes were present near all the
camps, but the impassable jungle made it impos-
sible to determine the extent of the propagation
areas and it was evident that considerable clearing
would have to be done to obtain the data desired.
Innumerable small puddles were everywhere, and
their possibilities as mosquito producers depended
upon the frequency of the successive showers. The
rapidity of growth of vegetation was remarkable,
and showed that its control would be an important

26 The Control of Mosquitoes

factor in the anti-malaria campaign. Alge that
protected the larve from their natural enemies
was often found and appeared to develop in a
few day’s time. It was evident that work would
have to be done on an extensive scale and that
the climatic conditions would make the campaign
a strenuous one. One apparently discouraging
feature was the frequent occurrence of showers that
removed oil applied to breeding areas before it
could accomplish its purpose.

| Area of Gatun Lake a
) Flex 85 Ft, 163.38 Sg Mi

!
}
!
Bs

! as
Noor ebay
ISTHMIAN CANAL COMMISSION

MAP SHOWING

|ISTHMUS WITH COMPLETED CANAL

| SCALE

| 1 o s — ° s

1 s s oe 70 =

House Doc. 426; 63d Cong., 24 Sess.

CHAPTER III

METEOROLOGICAL AND TOPOGRAPHICAL CONDITIONS.
TOPOGRAPHICAL CHANGES AND THEIR BEARING
ON ISTHMIAN ANOPHELES

HE Isthmian Canal Zone has a tropical climate,
high humidity, heavy rainfall, and a short
dry season of four months or less. The continental
divide occurs at Culebra about twelve miles from
the Pacific canal terminal. North of Culebra all
drainage goes to the Atlantic Ocean through the
Chagres Valley. The attached Table No.1 de-
scribing the air temperature for 1912 shows the sea-
sonal variation to be very slight, with no excessive
heat, and no marked difference of temperature at
different points in the Zone. It always remains
within limits that are favorable to the development
of mosquito life.

The yearly rainfall is not equal in all parts of
the Canal Zone. The extremes are noted at Colon
and Panama. On lands adjacent to the Pacific
coast, near the city of Panama, the precipitation

27

ge"er OLT vI* FLT or’ £9 Or roars obit 96°IT 26 ~"*sre0 Z Jo] adeIAY
02 ‘EL *" 98° 6° TOT fesesteeefoeee Pase|aeeioses sigale| ce vaieseeeislateelste re ae sian “sre0d Z 10} 16L
98 OT 98° 10° 06° st SLO LOT Xs 2 0'IL 96°6 i -sread Z 10§ ZI6T
: :8uT}yy Uene
16 ‘OT S's wB* cg° ¥8°T 92 ‘ST 08 “Cr OS "OT 02 ‘21 FP ‘OT ~ saved Of Joy adeI0Ay
elt 10°L 80° 39° 69°% aoe se reseie|osee te evaialcs ate Kawsleetscrewee! S180 Oe 10} ST6L
¥6°L LL* (1) a It 90° 9s°9 09"eL SLOT ¥9'9L sivot 08 1 Deurep ‘as
196 9o"e ee og oi 66 ‘OT $0 FT Tes 2L°6 ~“*srvok g Joy aBeIeAV
PLT 06° 2° 6L° 19°L -- eg Peers Ped ore weeesleweeee errr * 7 "sive g 10j eI6t
1e"9 19'S 10° ¥e* To° FL yr OT 8L"eE es OL ~* "sro 8 J0y tag
Set sS°S gs* | 73° 02° #99 62 "ET FIT TU €F Or L8°0T “-*sxvad Q JOJ oseIZAY
£0 ST 2% 00° #° or oe aaaies|Ue oa teeseclacicagivesise|sarcsciaees|steu ceegey [ses oe= cues “g1e0 9 10} SI6L
St°9 P'S 90° g9° 0° oF 90° Lg°It FI bT Lue 28°0T wrreeoosipad 9 Berek
OT ‘It 08 ‘e 89° og* SLT 89°L 08 “st 7m 44 €O IT 8S "OT 296 - seat 7% J0j e2BI0EAy
FL IL TE Ls°* gL° Lg‘ iaatiaiac atl (adele tic aa (aR aca (ae pciatans (cme ee _ --sreok 2% JOS £161
ST's Fle 20° 8° s0° 13°F 6s°L 1) § oLet 88 “SE greek ZZ 10j ebm
16 ‘OT ao" ee 89° Se'T £6°9 SUIT. | Or'er SI'It £0 7 *-sread g 10y edBIOEAY
ee-2L ZL'l 08" Fe ee |e woteas |b Seeec ees] snecCesees|srsceseeste| cress aigtes -="greak § 10} ST6T
PI'SL 9°9 £:) eA £0°S €0° ge° 20° 0h 102 9b 6 O1’st 61 °8T
188 Te ‘Ot 6S OT 9°E We 80°T 66° 88°L 66 OT S “aL 168 i 88 °S ky
sretteeeeel 0966, 8F EL ¥0'T 00° 10° ort tn uses [siete vaseen|eee seeeees|sinrscsigeeis|sce-cs bx.
TL"Sh 69°6 &6°9 wb 20° a 00° 16°€ ers SF 'OE 92 "Ot 1S"6
ONS O1peg —
16 €8 OT 26 8L°E 06° 99'T 10'S 9T'8 tar | 66 ‘ZT “*sreod F 10] @8812AV
we etteees| 65g ad él" 00° ez - jee [ttt asaac| areas soesslaneeels ae? aeaas *-srea } 10} ST6L
6F 88 1gs‘L 09°% 6L°F 00° ¥e" 00° $L°9 68 IT rst **sreod F IO} ZI61
rso10geqye
29 °OL S6% BOL 60°F ¥° is° tt 86°S 606 so°6 --sivek FT Oy OFBIIAY -
Sreseseeeel Gey, $0'S 90° 00° 6r° Sze sousesesss|sesee owas|s35S5 eere soe|---2 grea FL JO} CTT
68 TL 88°9 95°6 88° 00° 8t° 00° r6'F 419 SL°st “sre0d FI 10} Z16L rq
2B0q| -
Go'TL ST'8 86°8 91% 8L° 98° e0'T StF Sh OT 26 OT 7 ~"sread CT IOJ eSt1OAY a
teas eeeeel OTeg 18 €0° er 2° £9°0 see eepesss “sxe CT 10} ST6T
BL°TL 03 °S TL0T 89°% 100 80°0 “L reese ssread GT If ZT6T
smoony ~
senuny| -oune | Aye | cqady | -qoreyy [Aremaiqeg|-Arenoer| 20 “10d | -xeqo709 | __ 2°49. | asngny | “4mm “noes
: : -me00q | -meAONT -u1a}deg e

(seqouy uf sonqe, }

*sabpisan uoynjs puv ‘gi-gr6r ‘nupung fo snwyjsy ayy uo yyofurps hyyyuoyy—"T ‘ON TTAVL

*4Juo 1vad euQ

“SI6T 1884 04} 1OJ SPIO00I OPNIOUT 4OU Op SesBIZAL WOILIS—"'ALON

qeeeee eeee

T9°LFT

BR RER LSS BRS LSS

3 aia

wes

oat wid’

ow Ad

SS Ai 8
id

sel wi

RES

an
383

B'S

s9"¢

88°ST

02 6T

That

£671

Or-Ot
ooerEE,

a1'6

£9°EL

*""sBeh ¢ IO} OsRIOAY

+7 7sxBaA G 10} ST6T

vroreeesread JO} ZI6T
sO] 070g +

--svod Zp Oy BsBI0AY

2UOTOD
~*-sivek 9 Joy osBIDAY jj
c777 +" sread 9 10} SI6T
-sivok 9 10} ZI6E
syooIg sozergl

OL] 3400:
~-"g1eod ¢ Joy oateAy re
v7r to *sreed ¢ JO} ST6L
ee: sread ¢ 105 ZI6T
:peproLyy,

30 The Control of Mosquitoes

is from seventy to eighty inches per year. Toward
the divide the rainfall increases, and between
Culebra and Colon is very heavy, increasing as the
Atlantic coast is approached. The annual precip-
itation at Colon is between 120 and 180 inches.

The country along the line of the Canal is low
and near tide level for a distance of six miles from
the Pacific, and fifteen from the Atlantic ocean.
The remainder of the country is thickly studded
with hills and contains numerous springs, seepage
outcrops, swamps, and streams. The Canal ap-
proximately follows the line of the Chagres River as
far as the divide at Culebra Cut, and then down
the valley of the Rio Grande to the Pacific. There
is very little porous soil on the Isthmus, and a large
part of the surface soil is of a claylike character
and is so densely clothed with luxuriant vegetation
that travel is impossible except along the trails,
which if not constantly used soon become over-
grown and disappear. The lands immediately
adjacent to the Canal produce more mosquitoes
than places a mile or more east and west of it.
There are extensive areas toward the Canal Zone
boundary line and in the Republic of Panama
in which Anopheles do not exist.

The meteorological tables, on the pages pre-
viously referred to, show the large percentage of

TABLE No. 2.—Monthly rainfall by sections, Canal Zone— Year 1912 and averages.

(Values in inches.)
Pacific section. Central section. Ailantic section.
Month.”

1912 | Average.| 1912 | Averago 1912 | Average.
0.01 1.30 0.44 1.62 0.62 4.79
«23 +97 1.10 1.69 2.02 2.68
-O1 +66 . 1.21 +58 2.64
3.52 3.65 1.51 3.65 1.58 4.46
9.37 9.45 12.20 13.49 14.11
7.10 9.32 11.65 10.99 17.34 15.14
9.81 9.48 1.1L 11.22 16.51 16,31
9.43 8.70 12.4 11.89 12.60 15, 42
11.74 9. 12.16 11.37 11.05 11.86
11. 82 14.81 3.99 16.79 14.34
7.97 10.78 il. 15.64 22.17 24.64
‘ 4.58 6.33 3.88 7.90 10.61 15.55
stv eeeccereeeeeeeseteres 76.68 81.36 89. 68 103. 36 125.36 | 141.94

based on the records from 5 stations in the Pacific section, 11 in the Central section,

NortE.—Means are
and 4 in the Rtlantic section. All available records are used in computing averages.
Tasie No. 3.—Marimum rainfall in Canal Zone Oct. 1, 1905, to June $0, 1913.

Maximum rainfall,
Stations. 5 minutes. , Lhour. 24 hours.
Inches.} 4 Date. Inches. Date. Inches. Date.
Ancon (Oct. 1, 1905).......] 0.64] Aug. 7,1908 3.98 | Oct. 9,1911 7.23 | May 12-13, 1912.
Balboa (June 10, 1906 90} May 12)1912| 6.80] June 21: 7.87 | Nov. 16-17, 1906.
~60 | Nov. 11, 1908 3.30 | Aug. 27,1908 4.56 | Sept. 30-0ct, 1, T5009.
75 | July Py 1908 3.10 | Sept. 21,1912 6.00 | Dec. 2-3, 1906.
. May 2,1908 3.69 ict. 16, 5.55 | Dec, 3, 1906.2
«60 | July 25,1906 3.63] Oct. 1, 6.15 Do.
-59 | July 27,1008 3.32 May 1, 1911 6.56 | Dec. 2-3, 1906.
. July 20,1909 3.40 28; 1909 8.19 | Dec. 3, 1906.2
+62 | Aug. 3,1912 3.82 May 26, 1910 | 10.48 10.
Co ict. 1, 1905). 64 | Aug. 25,1909 4.90 | Oct. 8,1909 8.53 | Dec. 2-3, 1906.
Porto Bello (May 1, 1908).| #2:48 | Nov. 8, 1911} 4.53 | Nov. 29,1911 | 10.86 | Dec. 23-29, 1909.
fall in sn:
: No automatic recor: on Mines date; ‘otal for’ a pels ending at Doon,
record i jue to unusually Sah rate of fall.
Note. one in parentheses opposite iieelatiod names oe toi rainfall regi
Tasie No. 4.—Hourly distribution of rainfall in mae Suc Year 1912.
[Values in inches.]
_ Rainfall durin, Hourly rainfall.
working hours,
Total a.m. tod p.m. +
Btation. ——
; Accumu- Accumu-
Percent | Hour of Hour of
Amor i lated lat
tints | of total. amount, amount.
6.37 63 | 2-3 p. 0.32
44.3 62 | 2-3p. 2
57.95 77 | 2-3p. 16
63.13 71 | 2-3p. 2
43.45 2-3 P74
68.14 77) 34 3 27
65.70 78 | 23p. 134
. 28 43 | 3-4 p. 2.56
44.26 5-64. 1.91
72.58 49 | 23p. 2.65

31

32 The Control of Mosquitoes

rainy days that occur each year, and particularly
during certain months. At Panama from July to

‘Taste No. 5.—Monthly meteorological data—Ancon, Canal Zone, year 1912."

4 yp lS
Atmospheric
ressure Air (degrees ) a 5 ce
Uinches). | Fe ag
de eg
Month, |e Hl a F = m | 3g &§ & 3 3
e ¢ |[s.| § g.| 4 as 5-. 3 aa
e | E lzala| . |e glaze | 8 | Ee
3 2 {5 : a]. ga lacala |g. | 32
2 | ¢ |s8|3| 2/881 8 | 3 | ga lgck 3. | 3
a a |A ala sa a/al/a ff A |aA |a
|. 742 | 29.838 | 80.6 93 22 91 63 26 71 2 72 70 82
762 | 29.858 | 81.3 94 19 o1 66 1 7 26 72 70 82
. 749 | 29.844 | 83.0 96 7 69 12 2 24 B 7 7
770 | 29.865 | 82.1 97 7 92 69 2 72 26 4 72 8
732 | 29.828 | 82.0 96 7 90 7 20 vA 21 78 75 87
. 742 | 29.837 | 80.8 94 21 89 70 16 768 75
. 728 | 29.823 | 80.6 94 27 88 70 14 19 15 74 o1
). 728 | 29.624 | 80.2 4 31 88 69 7 72 21 vr) re
). 726 | 29.822 | 79.4 92 87 4 75 74 91
735 | 29.830 | 79.0 92 86 60 29 19 75 a 93
. 736 | 29.831 | 79.0 91 26 86 69 7 74 21 4 74 92
722 | 29.818 | 80.8 93 90 18 21 4 B 89
739 | 20.835 | 80.7 7 47 | 89.3 66 61) 72.3 26 | 74.2 | 73.1] 86.8
Precipitation (inches). Wind. Number of days.
°.
g 8 | Maximum veloc- q
: a ols y- 4
Month. 3 . | Bs 7 i i 3
3 & | #3 | wd | s F i) 2/8
>» | 2 | 3] e/g) 4. § ei.) 8/3
4 FI b “IS 3/3 ‘i -|/3/el1gie
2/6 (2/2 18 (24/fleia/a}213/2
a a = /e [ma ja A}Alo{[/a /o}/ale<
T. 1.02 0} 8, NW.| 24)/NW. 3 17 14 (i) 1] 3.8
0.08 +86 2/7,705|NW.| 29|/NW.| 23 3 17 9 0} 61
-O1 +78 1/8, NW.| 30) NW. 1 3 26 2 0} 5.6
2.68 2.76 5 | 6,300|}NW.) 25] N. 5 8 16 6 6] 5.1
10.71 8.98 14/ 4,999 |NW.| 23) NW. 2 )) 15 16 14| 7.7
5.80 8.15 13] 4,118]NW.| 34] 8S. 15 U) 18 12 20) 7.4
10.25 8.16 201 5,178|NW.| 24] BE, 4 1 10 20 18) 7.8
6.33 7.53 20 | 5, NW.| 241 _N. 18 1 14 16 18] 7.3
8.38 7.48 20 | 4,451) NW.| 31 | NE. 23 t) 14 16 7.6
17.89 | 10.92 23 | 4,538 | NW. 23) 8. 28 1 iL 19 4) 7.5
6.38 | 10.45 2 | 5,105 | NW. 25 | NW. 7 2 13 15 6.9
3.27 4.18 13 | 6,007 | N 20 | Nw. 16 6 16 9 9] 5.8
71.78 | 71.25 | 153 170,042 |NW.| 34] S. $15 42] 134] 140} 125| 6.6
1 Elevation of barometer 92 feet above sea level. 4 April.
3 Average for 14 years’ record. 6 February.
8 Tenths of sky. ¢ June,

November, 1912, inclusive, twenty or more rainy
days occurred each month. At Culebra, from nine-
teen to twenty-four, and at Colon from twenty

Conditions 33

to twenty-seven. These conditions favored rapid
development of vegetation and kept in existence

Taste No. 6.—Monthly meteorological data—Culebra, Canal Zone, year 1912. -

Atmospheria é 3
presure Air d Fahrenheit). E] i ca
inches). A | fs | 22
g.182| 28
Month. b ; 5 2 = Ian | 2 gl]a®ls
- | ¢ lsa/4 42/5 Balsa é [sz 28
q b 29 q | —g) Ee Jeg) s
3 3 a . g . Py 2
3 | 3 |38 e/e8) a) se /Ee does jd |g
a a |a a |a ala a ja ja
-| 29.438 | 29.851 | 79.0 90 8 87 68 27 70 21 7m 70 85
29.460 | 29.874 | 79.2 91 10 88 65 2 7 a 72 70 84
29.442 | 29.854 | 80.4 92 7 89 68 12 72 2B 72 70 nm
29. 464 | 29.874 | 81.4 96 17 1 67 2 72 27 B 70 nm
29.430 | 29.840 | 81.2 96 5 89 70 2% 74 22 1] 4 87
29.436 | 29.848 | 80.2 ol 21 88 70 16 3 2 15 “4 92
29.420 | 29.832 | 79.8 ol R 86 70 7 3 18 4 74 92
29.423 | 20.836 | 79.5) 92 8 87 69 12 2 19 4 92
29,422 | 29.834 | 78.7 90 1 86 67 a 72 21 74 23 93
29.430 | 20.844 | 78.6 90 5 86 68 2 72 17 B 93
29, 428 | 29.842 | 78.0 88 22 84 69 17 72 17 73 72 93
29.414 | 29,828 | 78.8] 88 1] 86) 67] 18] 72 91
29. 434 | 20.846 | 79.6 96; 45) 87.2 65] 82) 721 27 | 73.2] 72.1) 88.3
Precipitation (inches). Wind. Number of days.
g $ Maximum veloc- FA
g a ity. i d |
Month. 3 BS ; ‘
21. |B Fe Ee 2 | 4 3 : 3
ba pad bad : 9 i} g o
4 Be ~ 3 bef -{3]/ 39g
| E|ela le l28lelelalsleiale
a z2 /el/e |f& Ja jalaloj as oyalse
0.05 L75 2) 7,697|NW.| 2/NW.} 31 17 4 0 oO} 3.5
33 «56 4| 7,864)NW.| 30|/NW.| 24 9 2 ) Oo} 4.3
02 +68 1 | 8,893 | NW. 35] N. 4 9 22 ) oO) 4.1
3.14 3.80 4] 7,059|NW.| 30] N. Py 6 21 3 2] 4.9
8.15 | 11.16 15 | 5,538 | NW. 27| SW. 8 0 u 20 15] 7.4
7.87 8.86 21 | 3,715 | NW. 25 | SE. 15 0 9 21 21] 7.8
10.12 9.52 22| 4,962 |NW.} 36] NE. 30 i) 8 18 M4 7.7
12.88 | 10.58 19 | 4,592|NW.} 33] NE. 2 2 12 7 23) 7.3
14.12] 11.23 20 | 3,975 |NW. 32} NE. B 1 10 19 22) 7.5
10.15 | 11.54 25 | 3,505 | NW. 21) N. 7 0 9 22 22] 7.6
7.59 | 12.30 24 | 5,039 | NW. 27 |Nw. 17 0 16 15 | 7.4
4.87 7.68 15 | 6,567 | NW. a! N, 9 6 2 5 7} 5.2
78.99 | 89.66 | 172 68,406 |NW.| 36] NE.| $30 50} 176} 140] 137] 6.2
1 Elevation of barometer 404 feet above sea level. 4 May.
3 Average for 22 years’ record. 6 February,
4 Tonths of sky. ‘July.

innumerable small collections of water which could
not evaporate before being replenished. - The con-

tinuity of the rainy periods, rather than the total
3

34 The Control of Mosquitoes

volume of rainfall, is the important factor affect-
ing Anopheles propagation.

Taste No. 7.—Monthly meteorological data—Colon, Republic of Panama, year 191%.

Atmospheric a os 8 § Z
Ut ae 5 [ed |28
Month. z “e/a | 85
. g |2,| 4 .| g Bg2 | > | #5 | 38
€ ja q | =| B o| eT
B \es El. fae (ges/a |g? | ge
g |o8 8/38] a | § | 88 |so8] 3 g | 8"
a |a Ala Alala ja a ja |
29.872 | 82.2 88 22 | 86 76 8| 78 11] 74 72 77
29.892 | 80.8 88 27 | 8 7 10 | 77 15 | 75 73 82
29.882 | 82.4 88 29 | 87 76 11 | 78 11 | 75 3 79
29. 897 | 82.8 90 20 | 83 4 18 | 78 15 | 75 7 7
29.854 | 82.0 91 5 | 87 73 21/77 14/77 76 85
29. 860 | 80.2 89 9 | 84 73 16 | 76 15 | 77 (1) 88
29.852 | 80.6 89 27 | 84 7 7 14/77 76
29.846 | 80.2 89 9] 84 72 7) 76 14 | 76 75 87
29.840 | 79.4 89 17 | 84 71 5 15 | 76 75 87
29. 846 | 79.0 88 12] 84 72 29] 75 14] 75 74
29.850 | 78.6 88 5 | 82 73 14 | 75 14] 76 74 89
29. 845 | 80.4 87 26 | 84 B 17 | 76 12 | 76 7A
29,861 | 80.7 91 45 | 84.9 71) 610] 76.5 15 | 75.8 | 74.2 | 84.3,
Precipitation (inches). Wind. : Number of days.
C4 °
g 3 Maximum veloc-
a g a ity. , | 8
Month, 3 . | bs ; a g 3
3 x B | gS | of] s . i=) 3
Z| 4 S| e/a") "1 & | Seles
> | a7 |'s 5] 2 o u J Ls} 3
a| Bb /2/2 (E 2812/8) e]/el2 2/8
a a ele | la AlJA;Sia |oO;la]s
3.96 9 110,772) N. 32 | NE. 5 7 23 1 oO} 49
1.47 15 [30,066 | N. 321 N. 8 5 21 3 0} 5.5
1.67 11 12,550 | N. 33 | NE. 5 13 16 2 O| 4.4
4.08 6 |10,620 | N. 33 | NE. 24 16 12 2 0} 3.7
12.41 17] 7,614] N. 26 | NE. 18 2 17 12 12] 6.6
13. 35 241 5,165 | W. 26 | NE. W7 1 15 4 19] 7.3
16.38 26 | 7,574] N. 25 | NE. 28 3 12 16 TA
15.01 20 | 6,892] N. 29) W. 19 1 18 12 21) 6.8
12.53 22 | 5,473 | SE, 37| W. 2 3 10 17 16] 7.0
14.29 26 | 5,490 | SE. 33 | SW. 23 1 12 18 14] 7.5
2! 27 | 7,519 | W. 34/NW.) 17 0 10 20 10| 8.4
12.31 21 | 8,992 | NE. 29] NE. 13 9 Ww 5 1] 5.3
-| 117.59 | 129.32 | 224 |98,727 | N. 38 |SW.| 623 G1} 183] 122] 101) 6.2
1 Elevation of barometer 10 fect above sea level. 4 May.
2 Average for 42 years’ record. 6 February,
4 Tenths of sky. £ October.

The Isthmus, as may be seen by reference to the
map of the Canal Zone, is well supplied with
streams and natural watercourses. Some of them

Conditions 35

are of great length. At Gatun many flow from
the hills toward the settled area. During many
rainy periods, when all the soil is well water-
soaked, the hillside streams give least trouble, as
the larve are washed out of them. Unfortunately

Taste No. 8.—Comparative wind records—Ancon and Sosa, fiscal year 1918-18.

Ancon, Sosa.
5 c Expos
Aver- Maxtmum velooity. | Aver- Maximum velocity. |
Month. age | Per age | Per Ree,
Mand | Sw. | xan Mend | Sew | an Po
7 es] les
Direo- Diree- cent).
move- | wind. | per Day. | move- | wind. | per Day.
ment. nour.| Hon. ¥ | nent. Hour.| ton. 7
7.0 70| 24] SE. 4 10.5 90 44) SE 14 50
6.8 65 pzy N. 18 9.9 16 40| NE. 4B
6.2 48 31| NE. 2B 8.9 rit 39] SE 14 “4
6.1 37 23 &. 28 7.6 59 28) NW. 26
Tt a7 25| NW. 17 9.7 66 36 | NW. 17 37
6.7 58} 20) NW. 16 9.8 NW. 16 46
7.3 64 21] N. 29 10.7 89 30 | NW. 7 45
9.3 26 N. 10 13.3 36 | NW. 17 44
10.6 WW 26) NW. 23 15.6 96 36] NW. B 47
9.7 69 uw N. 2 13.5 36) NW. 5 40
6.7 63 20 N. 1 9.4 79 34| NW. 1 40
5.5 4a 32 8. 11 8.0 61 48 8. L 45
7.4 58 32 8. all 10.6 80 48 3 aii 43
1 Sosa records compared with Ancon. 3 June,

Nore.—Elevation of Ancon anuemometer 69 feet above ground and approximately 160 feet above mean
sealevel. Sosa anemometer 25 feat above ground and approximately 220 feet above mean sea level.

these hillside streams are often transporters of
larve and pupe, and become an important source
of mosquitoes when they reach a lower grade, or
enter larger streams or rivers near a settlement.
During dry periods streams become prolific
sources of mosquitoes; and even after the stream
becomes dry many depressions still contain water
in which larve rapidly develop. As the rain-
fall is heavy, there is a rapid run off, and erosion

36 The Control of Mosquitoes

and the formation of pot-holes frequently result,
often making it difficult to keep a stream properly
trained, and unless controlled, larvee will be found
at its edges where there is least current velocity.
Where branch streams occur, debris, sand, etc., are
deposited at their junction with the main stream,

TABLE No. 9.—Monthly evaporation—Canal Zone, years 1912, 1913, and averages.

[Value In inches.)
Ancon, Rio Grande. Gatun. Brazos Brook,
Month. rn x ik
ver- ver- ver= ver-
1912 | 1913 age. 1912 | 1913 age. 1912 | 1913 ago. 1912 | 1913 age.
January... B 6.363 7.208 6.066 | 6.387 | 5.842
February 6. 6.134 6.048 5.672 | 6.616 | 5.493
‘he... 795, 77.099 7.649 7.081 | 8.455 | 7.140
A \. 6.732 7.394 7.321 | 7.468 | 5.663
May. \. 5.350 5.335 5.707 | 4.167 | 4.474
June. > 2.708 | 3.836 3.263 3.729 3. 492
July. 3. 6 3.908 3.889 4,425 3.710
August. 3.648 3.021 | 3.983 4.316 4.611 3.906
| 3.395 3.120] 3.335 3.799 4. 487 140
October....] 3.458 3.314} 3.763 4.123 3.970 |. 063
-| 3.723 2.956 | 3.275 3.351 3.100 |. . 834
b 5.333 4.025 | 4.723 4. 809 |. 4. 860 |. . 790
Year.) 57.428 46.877 | 58.501 |.. 160. 929 |.
Daily moan] .157 |....... «128 160 +166

Note.—Averages st Rio Grandoe and Brazos Brook aro based on four years’ record; Ancon five years’
record; and Gatun two years’ record.

Evaporation measurements are from floating pans 4 {ect in diameter ord 10 Inches deep at Rilo Grando,
Gatun and Brazos Brook, and insulated tank 10 inchos in diameter at Ancon.

For monthly evaporation during past years, seo provious annual reports.

and afford collecting places for larve brought
down-stream.

Various accumulations of vegetable matter,
twigs, branches, etc., are found at the edges of
rivers and make suitable hiding places, often
inaccessible to the fish that prey upon mosquito
larve. At some points, seepage outcrops occur
on the stream banks and are constant sources of
larve throughout the year. It was not at all

Conditions 37

unusual for the water in a stream to disappear
into the substrata during the dry season and
reappear at a lower level. As the areas so affected
vary from year to year, it was necessary to inspect
the stream beds up to their sources. There are

180
170
186 [sos 1909 _) or
M40 cop Lo
150 | Years of |Record
= oS
2 no — otto} Ralpgay | 38S pare
& 00 Po Rae de
90 pa yh bane
a peor P= f ee
2 a 2 =a Anouat pe
e 10 fase
& oo Pe
50
40
30 7
20 ~
10
0
reere of 1G 16 G9 230 a1 8 2 a 4 9 18
=~ Bao a on a at
838 2822 2 818 3 2784 & 8
3 0 A wa 2 ‘2 842 a 3
SAS 2205 55 it
g4i4 da caa 6 G3 fg aesg 6 8
3
er hil 6.9. “Am L_40_1.23,/ 6.4 _,| a ee ee 14
suations {iles) wri TT [ i Tt {

RAINFALL CHART
many marshy areas of large extent within fifteen
miles from the Atlantic and six miles from the
Pacific, in addition to the inland marshes near the
occupied camps.

There is much marshy hillside caused by seepage
outcrops. The impervious character of the soil
enables small depressions to hold water for rela-
tively long periods, especially where rank vegeta-
tion proves a barrier to evaporation by sun and

38 The Control of Mosquitoes

wind. The rainfall tables show the number of days
a month on which rain fell. The depressions
referred to above take time to dry out, and mud
contains sufficient moisture to maintain life in
larvee of Anopheles and other mosquitoes. Admit-
ting that the Anopheles albimanus readily travels
half a mile or more, the numerous small depres-
sions hidden by vegetation within a circle of half
a mile radius are of importance, especially during
very wet periods.

The climatic conditions of the Isthmus, except
during the brief period of a few weeks toward the
end of the dry season, are ideal for the propagation
and multiplication of Anopheles. It is difficult to
conceive of a locality where topographical changes
have been made more frequently and rapidly.
The cost of anti-malaria work depends largely upon
topographical and climatic conditions.

Many miles of temporary track were frequently
changed and often each tie removed left a water-
holding depression. The excavation and trans-
portation of material excavated was continuous
for ten years. As the depth increased, more
water seeped into the new channels or benches
being cut by the steam shovels, and the former
ditch sites were excavated.

As the bottom of the Canal was lower than the

4
4,
%

4
%

A topographic change: near Empire

Constantly changing topography: near Miraflores

Conditions 39

surrounding country most of the sub-surface
water went toward it. The,dumps where the
excavated material was ‘‘wasted’’ often blocked
natural drainage channels and ditches. Water
collected constantly and proposed excavations and
fills interfered with or prevented permanent drain-
age schemes. As the Canal neared completion,
the bottom became too flat to drain economically
and numerous small pools were left. Hundreds
of acres of this territory had to be inspected
and oiled each week. The hilltop of to-day be-
comes a depression a month later, and permanent
drainage is often out of the question. The history
of tropical malaria is generally alike east or west..
It develops most rapidly when the soil is dis-
turbed by large and extensive excavations and
fills accompanied by the introduction of non-
immune labor housed near the site of their
work. Until now, no precautions had ever
been taken in similar construction work in
the tropics to prevent an increase in the num-
ber of places suitable for the development of
malaria-conveying mosquitoes. Pools and stag-
nant water were brought into existence by fills
that had settled. By the time the surface of
a dump was in condition to hold water it was
covered with rank vegetation, often more than

40 The Control of Mosquitoes

six or eight feet high, and the pools formed
were hidden.

If inadequate provision is made for taking care
of natural watercourses blocked by uninterested
engineers, acres upon acres of new swamp lands
may be formed, which are often ideal breed-
ing grounds for mosquitoes. They contain few
aquatic enemies of mosquito larve, plenty of
food supply, and the tangle of vegetable growth
protects the eggs and furnishes hiding-places for
larve and pups. Such places are ideal for mos-
quitoes as soon as made, and outrival the freshly
dug excavation, which is not frequently used by the
Anopheles for ovipositing until conditions, such as
protection for the young, food supply, and vege-
table matter, tempt their presence.

It can be readily understood that if the area at
the rear of the new dump or blocked portion of the
drainage channel is flat or gently sloping toward
the dam, a much larger area containing vegetation
will become flooded. If the final water surface
does not rise above the vegetation, the mosquitoes
are better pleased. In many cases, nature lends a
helping hand, and the vegetation under the water
grows rapidly until it reaches and extends beyond
the surface. Under these conditions, aquatic
plants develop and fish find the larve with diffi-

Daily change of topography

Conditions 41

culty and remove few of them. The water-fowl
appear and may be the cause of the introduction
of the kind of fish they come to catch. It is not
at all improbable that fish-eggs stick to their legs
and are thus transported from one place to another.

Extensive construction work, as a rule, is per-
formed by contract. Each of the contractor’s su-
perintendents is responsible for one particular
phase of the work and his entire time and attention
are devoted toit. The foreman on the dump must
unload his cars and return them within a minimum
of time, the tracks must be moved with equal
rapidity. If a culvert is to be elongated and no
pipe is available, some makeshift is used and prob-
ably no attention is given to grade when the pipe
is laid. The work is performed in record time, or
possibly a record is broken, and the cost of the
work for that day, week, or month is reduced an
appreciable amount. Probably such saving and
efficiency are properly rewarded, and the increased
salary or bonus acts as a spur to even more rapid
work. The makeshift culvert is soon covered up
and does not show. Ifit collapses some time after-
wards, the cause of the collapse is unknown. A
lagoon is formed at the back of the dam. Unless
the entire valley is filled, the artificial mosquito-
producing area created is permanent, and when

42 The Control of Mosquitoes

the country becomes settled those living within a
half-mile or so will be malarial, and people passing
through the district will be affected similarly.

It is not an economy to expedite construction in
sucha manner. If a contract takes along time to
execute and the laborers become infected by
Anopheles, created by the foreman’s energetic
though short-sighted action, what of the reduced
energy or vital force of the workmen? Labor
receives a daily wage and any decrease of output
increases the cost of supervision. How does
the loss of the laborers’ efficiency compare with the
saving made by neglecting proper care of the
natural watercourse? If all, or a large part, of
the force becomes infected what is the contractor’s
monetary loss?

History shows that important contracts have
failed in the tropics, and in some cases the entire
project has had to be abandoned. The anti-
malaria campaign on the Isthmus showed exactly
how and why these failures happen, and also pro-
vided opportunity for the study that may assist
in determining the best and most economical
means of overcoming future difficulties of similar
or related character.

Blocked watercourse changed into an Anopheles breeding-place

CHAPTER IV
THE SPECIES OF ANOPHELES ON THE ISTHMUS

T the beginning of the Isthmian anti-malaria
campaign available knowledge on this sub-
ject was very limited, and confined largely to the
experience gainedin Cuba. It was not known how
many different species of Anopheles existed, nor
was it definitely known which of them were the
important malaria carriers. The preliminary in-
vestigations demonstrated that they were not of
domestic habits similar to the Aedes calopus and
could not be considered house mosquitoes. Nu-
merous containers near dwellings were carefully
examined for Anopheles larve but none were dis-
covered. The propagation areas appeared to be
limited to seepage outcrops, pools, ponds, lagoons,
ditches, streams, and the edges of rivers.

It was not thought that the daily flight range
was extensive, nor was any information available
on this subject with respect to the different species.
It was easy to see that rank vegetation served as a

43

44 The Control of Mosquitoes

shelter or harboring place for Anopheles during the
daytime, as well as for other mosquitoes. The
climatic conditions were so different from those
previously encountered that it was evident much
investigation or pioneer work was required to ob-
tain data pertaining to the general life history
and habits of Isthmian Anopheles. We had no
means of determining how seasonal changes would
affect propagation, and the available data were
unreliable. It was generally believed at that time
that all mosquitoes traveled more or less with
gentle air currents, but there was no positive
knowledge of the habits of flight, and the length
of flight of Anopheles, or of any other Isthmian
mosquito. This was yet to be determined. It
was not known if or how topography affected the
distribution of species, whether Anopheles larve
thriving in small collections of water held by plants
were of economic importance, or whether certain
species were confined by fixed geographical limits.

Dr. O. L. Howard assisted in the work in many
ways and furnished valuable suggestions and
advice.

Important collections of available species of
Zone mosquitoes were made by Mr. A. Busck of
the Smithsonian Institute and by Mr. A. H.
Jennings, entomologist of the Isthmian Sanitary

Species of Anopheles 45

Department. The determination of species found
was made by Mr. F. Knab. Detailed studies
of the relation to malaria of Isthmian Anopheles
were made by Dr. S. T. Darling,' and portions of
his published work are quoted.

In order to have sufficient adults for the experi-
mental work, the sanitary inspectors along the line
of the Canal sent pupe and well developed larve
to the laboratory daily. Anopheles larve and
pupze are delicate organisms and must not be
subjected to rough treatment. They were trans-
ported in wide-mouthed jars, not too many in a
jar, containing about an inch of water and shaded
fromthe sun. If several inches of water are placed
in the temporary container and subjected to the
direct rays of the sun, or if the larve are too much
overcrowded and subjected to constant shaking,
many willdie. By having shallow water but slight
effort is needed for the larve to reach the surface.
In collecting larve in the field white enameled
saucers or dippers were found convenient, as the
larve are easily visible against a white back-
ground.

In connection with the experimental work con-
ducted at the laboratory, it became necessary to
collect live adult Anopheles from houses at the

1 See his Studies in Relation to Malaria.

46 The Control of Mosquitoes

various villages, ship them across the Isthmus, and
have them arrive in their normal healthy condi-
tion. It was found absolutely essential to keep
them shaded from sunlight, rain, and strong air
currents. They appear to die rapidly if placedina
drying wind, which apparently causes evaporation
that is fatal to them. The adults die if left
overnight in mosquito-bars out in the open. If
transported in cages few will die if the cage is
completely covered with paper.

’ In the biting experiments made by Dr. Darling
to determine which of the Isthmian Anopheles
were malaria carriers, seventy per cent. of the Ano-
pheles albimanus used became infected, about sixty
per cent. of the Anopheles tarsimaculata, and thir-
teen per cent. of the Anopheles pseudopunctipennis.
No Anopheles malefactor became infected although
several bit the same person who infected the Ano-
pheles albimanus. The conclusions drawn were:
that Anopheles albimanus was the all-important
transmitter of malaria fever in the Canal Zone at
the time the experiments were made. Anopheles
tarsimaculata was probably as susceptible to in-
fection as Anopheles albimanus. The former,
however, had a much more limited geographical
distribution at that time, and appeared to be
numerous only near Colon on the Atlantic coast.

Species of Anopheles 47

Anopheles pseudopunctipennis appeared to be
only slightly concerned in the transmission of
malarial fever, if at all, as only about twelve per
cent. of the specimens used became infected under
the most favorable artificial conditions.

Anopheles malefactor did not become infected.

It also appeared in some cases that even the sus-
ceptible mosquitoes possessed immunity toward
the malaria parasite and failed to become infected.

Unfortunately, Anopheles albimanus is the most
abundant of the Isthmian Anopheles, and it makes
stronger efforts to obtain entrance into inhabited
buildings than any other species. It seems to
prefer breeding places exposed well to the sun, and
the green alge in such exposed collections of water
provides food for its larve. The larve are found
in stagnant water but not when heavily charged
with sewage; and at the edges of streams and
rivers in places where there is practically no cur-
rent. Although they are found at the surface in
deep bodies of water when plenty of aquatic vege-
tation, debris, or other forms of protection are pre-
sent, yet they are frequently found in puddles
having a depth of only a fraction of an inch.
Even after the water disappears the larve often
continue to live for some time under a layer of
alge or inthe soft mud. The adult will fly a long

48 The Control of Mosquitoes

distance to reach occupied houses; but is seldom
found in vacant buildings.

The Anopheles pseudopunctipennis is found
throughout the Isthmus, is not as strong a trav-
eler as the other carriers, and is not so commonly
found in houses as Anopheles albimanus or Anoph-
eles tarsimaculata. Its season of abundance ap-
pears to be more limited. Running water that
is relatively clear seems to appeal to this species.
It does.not bite as persistently as the others, and
its relation to malaria on the Isthmus is not con-
sidered important.

Anopheles malefactor is apparently a non-carrier
but a vigorous biter. At times large numbers
of this species are found in houses. It recently
became the most numerous species at the huts
along the shore of Gatun Lake, when the water ele-
vation reached eighty feet above sea level. Several
months later it became relatively scarce.

The other species of Anopheles are not commonly
found in buildings, and so far as malarial fever
transmission is concerned are not of economic
importance. Regarding the identification of lar-
vee, Dr. Darling has supplied the following data:

I have made no attempt to determine in detail all
the anatomical characteristics of anopheline larve
of this region; that has been done for some species by

Aquatic vegetation sheltering larve

Species of Anopheles 49

Knab. The chief differentiating larval characters of
the common anophelines of this region are these:

A. albimanus or white-footed group:

Palmate hairs on all abdominal segments and some-
times on postero-external angle of the thorax.

Antenne without a tuft of hairs.

A. pseudopunctipennis group:

Palmate hairs on third, fourth, fifth, sixth, and
seventh abdominal segments, but none on the
first andsecond. On the latter two, however,
there is a rudimentary stalked tuft replacing
the palmate hairs.

Antenne without a tuft of hairs.

A. malefactor or spotted-legged group:

No palmate hairs on first and second abdominal
segments, but palmate hairs on all remaining
segments.

Antenne with a tuft of hairs.

These characters are very striking and sharply sep-
arate the malaria transmitting A. albimanus group
from other varieties. With care it is frequently
possible even in muddy water, from an examination of
the indentations of the surface film caused by the
palmate hairs, to at once determine the presence or
absence of members of the A. albimanus group. In
this group there is no break in the indented film, but
in the two other groups there is a well-defined non-
indented break in the film, due to the lack of palmate
hairs on the first and second abdominal segments.
The size of the palmate hairs on the postero-lateral
angle of the thorax, and the presence of these hairs on
the thorax, first and second abdominal segments, is

4

50 The Control of Mosquitoes

subject to some variation. It would seem that the
white-footed group is undergoing some variation with
regard to the size and location of these hairs, and
apparently they are becoming rudimentary on the
thorax and first abdominal segment.

The larve are easily frightened and remain
hidden under water during long intervals, and for
that reason are sometimes difficult to detect.
Experiments were made by immersing larvae in
cages in the body of water from which they were
taken, and it was found that some of them re-
mained alive under water for a period of two hours,
even though the surface of the water was covered
with oil after the cage was immersed. It is not
widely known that Anopheles rest and remain upon
dark clothing for relatively long periods although
the wearer may be walking rapidly. It has been
ascertained by observation that in this manner
people convey them into houses that have been
properly screened. Large numbers are carried
long distances by trains at night. One obser-
vation was made in a case where a mosquito
(genus unknown) resting on the shoulders of a
man wearing a dark blue serge coat was carried
from Colon to Porto Bello on board a steamer.
The time of the trip was nearly three hours, and
the distance about twenty miles. Observers were

Species of Anopheles 51

afraid of disturbing the specimen by close approach
but it was not an Anopheles.

There is no doubt at the present time that the
Anopheles albimanus is responsible for practically
all of the malaria transmission in the Canal Zone,
except at Gatun and Colon, where Anopheles
tarsimaculata sometimes exist in numbers large
enough to be of equal economic importance. The
decrease of malarial fever in Panama will be in
proportion to the eradication of these two species.

It is now known that both species travel in
enormous numbers over long distances (if other
suitable propagation areas are not available)
between their place of origin and the nearby settle-
ments. On the Isthmus, in two instances where
breeding was prolific, they have been known
to travel from half a mile to a mile to reach houses.
They do not seem able to remain long in a very
strong light, and as a rule prefer darker places.
No other Isthmian mosquitoes are so persistent in
attempting to enter screened houses, and yet once
inside they seem unable to find a way out, and
collect on the screening soon after daylight, and
again at dusk. When they find they cannot get
out they return to the interior of the building, but
make another effort to pass the screened openings
at the same hours next day. They are seldom

52 The Control of Mosquitoes

found on the screens between 9 A.M. and dusk.
When Anopheles in buildings are trying to find
openings in the screens of windows or balconies
they are not easily frightened and are readily
captured. They are so persistent in seeking an
opening that it is necessary to alarm them several
times before they will leave the screening.

When such extensive topographical changes
were made that it was temporarily impossible to
control the increased influx of mosquitoes at a
camp, the laborers of the oiling brigade destroyed
mosquitoes found in barracks in the early
mornings.

A description of Isthmian Anopheles as given in
Darling’s Studies in Relation to Malaria is quoted
below:

The following is a list of anophelines of the Canal
Zone:

Anopheles argyritarsis, R. D.

Anopheles tarsimaculata, Goeldi.

Anopheles gorgasi, D. K.

Anopheles albimanus, Wied.

Anopheles cruzii, D. K.

Anopheles apicmacula, D. K.

Anopheles punctimacula, D. K.

Anopheles malefactor, D. K.

Anopheles eiseni, Coquill.

Anopheles franciscanus, McCrack.

Anopheles pseudopunctipennis, Theob.

SoyIp
Ip jo Bur
Ipymq A
q pasneo ‘ev
oye
q iveu
dwe
AA
S

=
Sa S

Species of Anopheles 53

The above eleven species of anophelines have been
collected in the Canal Zone during the past eight
years. They are not taken nor do they exist in their
breeding places in anything like equal numbers.
For example: Only one specimen of A. gorgasi has been
found. Of the eleven species, the commonest ones
are A. albimanus, A. pseudopunctipennis, and A.
malefactor, but this again must be qualified by stat-
ing that the predominance of a species varies from
season to season and from place to place. In certain
villages, upon going through the barracks only A.
albimanus will be found, while in other villages from
five to ten per cent. of the mosquitoes will be A.
pseudopunctipennis, and at Ancon during October,
1908, twenty-seven per cent. were A. mulefactor and
seventy-two per cent. A. albimanus. A. tarsimaculata
appears to be distributed solely near the Atlantic.
Mr. A. Busck, of the Bureau of Entomology, United
States Department of Agriculture, who collected and
made observations on Zone mosquitoes during 1907,
gave it as his opinion that A. pseudopunctipennis was
the commonest anopheline during the period of his
stay.

The necessities of the Canal operations in excavat-
ing and filling change the topography of districts and
localities so as sometimes to convert salt marshes into
fresh-water ponds, or to make tracts of land contain-
ing few anophelines into a vast swamp in which they
luxuriate. On the other hand, swamps and breeding
places may be drained or filled in the work of excava-
tion. These factors, among others, influence the num-
ber and variety of species in a locality.

The commoner anophelines of the Canal Zone may
be divided into three groups:

54 The Control of Mosquitoes

(A) The white hind-footed group, comprising: A.
argyritarsis, A. albimanus, A. tarsimaculata.
(B) The leg-uniformly-colored group, comprising:
A. pseudopunctipennis, A. franciscanus.
(C) The spotted-leg group, comprising: A. male-
factor, A. apicmacula.
These groups present well-marked differences in the
markings of adults, in the breeding habits and ana-
tomical characters of the larve, and, as will be shown,
they possess varying susceptibilities to malaria.

The rising waters of Gatun Lake

CHAPTER V
ANOPHELES PROPAGATION AREAS
WATER CONTAINERS

HE larvee of Aédes calopus and Anopheles are
seldom found in the same water. On the
Isthmus the larve of the latter are seldom found in
containers, while Aédes calopus are found only in
containers near inhabited buildings. If containers
are upset or their contents enter into a pond or
ditch the larvee of Aédes calopus will develop to ma-
turity; again, if broad flat containers are so placed
that vegetation growsin them, or if they are covered
up in the grass and gather vegetable matter, dead
leaves, twigs, etc., resembling natural breeding
places, Anopheles will at times be found in them,
but these conditions rarely occur and are not
of practical importance. Isthmian Anopheles are
not container breeders, and the sum total produced
by water containers is very small. They are
found occasionally in large wells or cisterns where
55

56 The Control of Mosquitoes

the water is near the ground surface, and into
which dead grass, sticks, and leaves have fallen.
The number of these collections of water is limited.

HOOF-PRINTS

The hoof-prints of cows and horses become
important Anopheles producers. Riders often
take horses over soft ground and in the wet season
each hoof-print may become a source of larve.
Cows seem to be particularly fond of walking in
soft-bottomed ditches, and remaining at the edges
of ponds. In this climate they are not stabled.
Where the soil of the pasture is of a soft or clayey
character, the foot-prints may be one to six inches
or more in depth. These depressions often hold
water throughout the wet season, and retain
it from showers during the dry season. It is a
difficult and expensive task to locate and fill, or
treat, all hoof-marks holding water in a large
flat field. The conditions are not improved if
such a pasture is subjected to occasional flooding.
It has been necessary in some instances to place
fences around ponds and portions of pasture lands
that become prolific sources of mosquitoes. Ina
few cases, where wet pasture lands were close to
thickly settled suburbs, or villages, it became
necessary to have the cattle removed elsewhere.

Cattle on low pasture lands: Panama

Anopheles Propagation Areas 57

In the dry season pasturage is not good on dry
lands, and the owners of animals are then inclined
to turn them loose, or fasten them in low spots,
and near ditches. An active animal can spoil a
trained ditch in one night, and leave it in a condi-
tion to become the most prolific source of mos-
quitoes in the vicinity. This matter was found
to be so important that legislation became neces-
sary to keep ditches in proper shape. It is unlaw-
ful now for horses or cattle to be at large, or tied
in places where their presence interferes with
anti-Anopheles operations and becomes a sanitary
nuisance. Even so, it is necessary to put up sign
boards along ditches and in places where animals
are not desired. When sign boards were first
placed on soft ground to notify people not to tie
animals in that vicinity, some poor ignorant
natives who could not read thought the sign posts
were excellent objects to which to tie their animals.
Night policing is still essential to enforce the
ordinance.

WHEEL TRACKS OF WAGONS

On bad roads, not frequently used, the rut
marks produce Anopheles and Culex, but where
travel is frequent there are no larve. Where
wagons pass through pasture land in the wet

58 The Control of Mosquitoes

season and haul material over fields for construc-
tion purposes, areas of considerable extent are cut
up and the drivers seldom use the same route
twice; they find hauling easier over untrodden
ground, and temporary roadways may thus be
widened as much as a hundred feet. Near settle-
ments the matter is of sufficient importance to
make the filling of these new ruts necessary. It is
more economical to fill in than to treat with oil
weekly for an indefinite period.

EFFECT OF LONG RAINY PERIODS AND FLOODS

Floods occur during the rainy season and are
generally preceded and followed by showers; and
the rapid currents thus caused removed many if
not most of the larve from the banks of rivers,
streams, ditches, and the pools or depressions near
them. Unfortunately the overflow reached depres-
sions on adjacent lands and caused new temporary
breeding places. When the flood was of short
duration and was followed by dry weather and
strong winds it was even advantageous. The
latter condition rarely happened on the Isthmus.

Continuous rainy periods with daily showers
for twelve days or more, whether accompanied
by floods or not, are the cause of prolific mosquito
production on flat lands. Practically all Canal

Algee united with oil in mat-like masses: Empire

Anopheles Propagation Areas 59

Zone land is covered with luxuriant vegetation
whichretards evaporation. On flatlandsmosquito
eggs were laid in shallow water pools or films of
water and accumulated in the relatively low spots.
These places were grass covered and before they
dried sufficiently to kill off the developing larve,
another shower fell. If the wet mud was col-
lected and dissolved in water, mosquito larve in
all stages of development were often found. In
excessively long rainy periods it is not easy to
locate, reach, and treat small places dispersed
over large areas, but if they are neglected, enor-
mous numbers of mosquitoes ensue. Evapo-
ration and ground absorption may or may not
dry ground sufficiently to prevent the develop-
ment of the greater part of the mosquito crop.
Just what will occur cannot be foreseen, nor can
it be estimated in advance during which month
of the rainy season such ideal mosquito conditions
may prevail.

A variation of rainfall at villages three miles
apart often resulted in a large influx of adults
in one village and not in the other. At times the
areas apparently needing attention were so exten-
sive that the situation appeared hopeless, but
experience gained in each locality by the inspector
in charge was of great value. A new man requires

60 The Control of Mosquitoes

time to become thoroughly acquainted with the
conditions peculiar to his district, that each day’s
work may produce the maximum of effective
results. In time he becomes expert in omitting
much that is but apparently necessary.

STREAMS

Large and small streams and natural water-
courses, although dry for several months of the
year, often become important propagation areas.
Their relative importance or output of adults
varies with the season. Asa rule, during the wet
season larve are found in quiet places along the
banks where vegetation occurs or overhangs the
banks or where debris collects. It seems to
be an instinct of the Anopheles to lay her eggs
where there will be a maximum of protection, or
hiding places, and of food supply for her brood.
Small detached bodies of water adjacent to the
stream are favorable; and where the banks are
muddy, the footprints of those looking for larvee
may become filled with water by seepage from the
stream, and later contain larve. Where the water
is quiet or shallow, green alge develop rapidly
and by retarding the current assist the Anopheles.
These places offer favorable harboring places for
larve.

Anopheles Propagation Areas 61

“‘Pot-holes,’’ or circular depressions, often form
in the stream bed. The current may be swift only
at the center and larve may be present at the
edges where the debris collects. A sharp down-
pour will often collect most of the debris, alge, eggs,
larve, and pupe and carry them on downstream.
They may, and often do, collect in quieter places
in the larger stream or lagoon below. Where
streams flow toward a settlement and empty into
quieter waters near it, the adult mosquitoes in
houses often increase suddenly, due to the new
supply of pupe and larve thus transported toward
iis

It must not be assumed that alge in ditches
when treated with oil or larvacide will cease to
exist. A week or less may start and complete
a new growth. Often crude oil and alge unite
and form a mat that floats on the surface and in
a few days a part or all will settle to the bottom
of the stream; or may collect in irregular depres-
sions along the bank. Such conditions offer
hiding places for larve and make inspection
both slower and more difficult. In shallow streams
having coarse gravel or stones on the bottom and
along the wetted perimeter, when debris, grass, or
well-developed alge are absent, larve hide under
the small stones, and though none may be detected

62 The Control of Mosquitoes

at the water surface it is always well to remove
sufficient stones to be sure that they are not hiding
underneath. Frequently they remain hidden for
long intervals. When the stone under which the
larve are hiding is moved, they will rapidly travel
to the next nearest stone that affords a suitable
hiding-place.

As the dry season advances, the source of hill-
side streams moves farther downhill, the upper
part remaining dry. In case the stream bed is of
irregular grade and contains deep depressions,
as the bed becomes dry pools remain and become
breeding places. Should the stream be of great
length it is so easy to assume that its entire upper
length is dry that such places are apt to be over-
looked. Eventually the pools become dry, but
generally produce one or more crops of mosquitoes
before this happens. During the dry season, if a
shower falls, the stream bed will probably be dry
within a day or two, but it is never certain that
these depressions have not collected water which
they may retain for a week or more.

In the upper portion of the stream bed and
higher than its dry season source, the sub-surface
flow may outcrop at one or more points, and, due
to the soil formation, grade, etc., flow along the
stream for a short distance and again disappear

AIC A SuBVYD OY} UL pooy sulpisqns v Aq do} V UT JOT 91} pvorpIey,

Anopheles Propagation Areas 63

into the ground. The locations of the portions of
the stream that act in this manner may change
from year to year, and the observer is never sure
that a dry stream is not producing mosquito
larve until he has followed it throughout its
entire length.

A minimum of mosquito larve will be present in
a stream when it has been properly trained, that
is, reduced to the minimum width of uniform cross
section and freed from stones, grass, and debris
that would interfere with the flow or velocity and
furnish hiding-places for larve.

RIVERS

The larger rivers, where the banks are steep,
are not sources of larve. As a rule they are less
troublesome in the wet season than in the short
dry one. If the banks are not much above the
average water level, where flat lands are adjacent
and floods occur, we have conditions similar to
those in lagoons. At times of great floods, twigs,
branches, and even whole trees are carried down-
stream and when they come to rest may act as
collectors and hold grass, vines, and finely divided
debris. Such places become sources of mosquitoes
that larve-eating fish or top minnows often cannot
catch. Obstructions of this nature at times change

64 The Control of Mosquitoes

the direction of the river bottom and leave new
isolated pools as the water recedes. The high
water causes much silt containing plant food to
collect on the banks. This in turn produces
luxuriant vegetation at the water’s edge. Fre-
quently the vegetation extends into the water
and retards the current near the shore sufficiently
to create suitable areas along shore in which
Anopheles larve accumulate and remain. The
Rio Grande on the Pacific slope of the Isthmus
has been expensive to control because of these
conditions, and it has been necessary to have
workmen in boats continually on the river for
mosquito control.

Rivers may or may not assist mosquito develop-
ment in the wet season as above described. They
are generally a most prolific source in the dry
season. As the pools, ponds, streams, and marsh
lands dry, the rivers become more sluggish and
are the only available water sources remaining.
Naturally the mosquitoes use them. Where
the banks are kept clean, free from vegetation,
debris, and stones in shallow water, the larve
are seldom found. Mosquitoes instinctively avoid
places that fish and other enemies reach easily, and
lay their eggs in more favorable places. They
select shallow pools with stony bottoms in the river

Oiling of breeding-places along a river bank

Anopheles Propagation Areas 65

bed. Then vegetation grows around the edges
in shallow water giving the desired protection.
As the water is shallow, it becomes warmer and
this assists the rapid development of alge.

The larve can penetrate the alge but small
fish cannot. Alge containing larve may be
detached and floated downstream. Fish swim
-to and fro around it, and follow it, trying to seize
the larve, but they seldom appear to succeed.
Where the slope of the river bank is gradual, and
the water at the edges shallow, with gravel or
stone at the bottom, Anopheles larve are often
found, providing other nearby and better pro-
tected breeding places do not exist. With the
first heavy floods of the rainy season most of these
river production areas are promptly eradicated.

SEEPAGE

Seepage water is water that flows underground
along an impervious stratum and reaches the
ground surface when that stratum becomes exposed
at the surface, or approaches it. Or it may come
through a stratum that is more or less porous and
water bearing. The places where the water
appears at the surface are called seepage out-
crops. They are frequently found on benches, hill-

sides, and in small valleys. In areas where they
5

66 The Control of Mosquitoes

are present, sharp changes of grade, such as steep
hillsides terminating on flat ground, or on more
gentle slopes, should be carefully examined.
The seepage on a hillside outcrop may approxi-
mately follow a contour line. The escaping
water then flows downhill, and may again be
absorbed by the soil, or continue to flow until it
reaches a natural drainage course, or come to rest
in a depression and form a pool or pond. On hill-
side seepage areas only a thin film of water is
present, but small depressions in the ground give
a depth of water sufficient for mosquito develop-
ment.

Naturally the temperature of the water is
relatively high and frequently seepage areas con-
tain much alge and few enemies of larve. Cattle
complicate the situation by leaving deep hoof-
prints. The outcropping water may be present
throughout the year or only during short periods.
Each outcrop is a law unto itself. Strange as it
may seem, some places produce more water
immediately after the rainy season, than during it.
One such area was discovered at Balboa at the
foot of a hill composed largely of trap rock.

In several localities seepage outcrops had been
present only twice in ten years. Of course if
such places are neglected it will be extremely

espaip YOUsSIY plo ue siwou wArvy oymbsour Buryoojo1d uoryeyosa,,

Anopheles Propagation Areas 67

difficult to account for the presence of mosquitoes,
and more distant production areas, winds, etc.,
may wrongly be held responsible for a temporary
influx. Seepage may be found at most unexpected
places during or after long rainy periods; it almost
invariably produces mosquitoes, and in many
cases may not be noticed until some adults have
been produced. Nearly all seepage outcrops as
well as areas flooded by seepage water were covered
by the same class of vegetation that grew above
the outcrop, and there was nothing to denote their
presence. One cannot appreciate the number of
mosquitoes produced by even small seepage out-
crops without actually inspecting them.

LAGOONS, LAKES, PONDS, ETC.

Many parts of lagoons in the tropics are impass-
able and trochas or paths have to be cut through
the heavy growth to penetrate some parts of them.
Examination, or mosquito inspection, when the
bottoms are soft deep mud is tedious but interest-
_ ing, and the splashing of an occasional alligator
removes any sense of monotony. It is surprising
how many of these areas do not produce mos-
quitoes although conditions as to food, protection,
etc., seem to be ideal. After examining many of
this character the observer is apt to take it for

68 The Control of Mosquitoes

granted that Anopheles are absent. There are
some prolific breeding areas whose importance
cannot be determined without a thorough and
painstaking examination.

Many lagoons, ponds, etc., have been seen
where larve were not present near the shore line,
but were very numerous in protected portions of
deeper water. Water more than a foot deep, free
from vegetation, debris, alge, etc., is generally
non-productive. It often happens that numerous
larve are present only in particular parts of a large
and apparently suitable breeding place, and it
should be thoroughly investigated before any
conclusions are drawn regarding its possibilities.

Again, some areas may produce a species of
Anopheles larve only at certain times, and it is
wrong to conclude that because results are nega-
tive to-day the conditions will not change sooner
or later. Seasonal changes have an important
bearing upon mosquito propagation.

In clear water such as reservoirs with clean
edges and abrupt banks, more or less finely divided
debris collects and is often removed from one side
of the lake to the other by winds. Larve are
frequently present in this mass, and apparently
the small fish cannot reach them. Marked wave
action prevents mosquito production. If the

uoouy itp apy Buydeosezur ue Aq poyjoryuco ‘dozoyno a8vdeas y

Anopheles Propagation Areas 69

wind direction is uniform and continuous, larve
will be absent on the shore affected by wave action.
However, if there are small inlets along the shore
line the wave action may not affect the larve
in them, and they will need attention.

NON-PRODUCTIVE ANOPHELES PROPAGATION AREAS

In the Cuban anti-malaria campaign as well as
on the Isthmus, there were so many apparently
suitable places for mosquito development in
which the larve were absent, that attempts were
made to determine why Anopheles did not oviposit
in them. The attempts were not successful.
Larve were placed in many of these places and be-
came fully developed. Mosquitoes were some-
times induced to lay their eggs in non-productive
places by putting dead grass there. Investigation
and research along this line is essential. Probably
new facts will be discovered that may enable us to
make existing production areas unattractive as
breeding places and render them sterile. A dark
brown or reddish substance resembling iron rust
(name unknown), which is often accompanied by a
small amount of natural oil, is found occasionally
on the Isthmus at the edges of ditches or ponds in
the form of a limited seepage outcrop and appears
to be repulsive to Anopheles. Larve are never

70 The Control of Mosquitoes

found in that part of a body of water affected by
it, although they are found in other parts of the
same body of water to which this substance has
not extended.

PROPAGATION AREAS CREATED BY MAN

It is surprising how many mosquitoes are
unnecessarily brought into existence by people
who are aware that the construction work they are
doing will cause Anopheles to multiply. They do
this even when it would cost no more to arrange
for proper drainage. Borrow pits or excavations
are made where they tap seepage water planes,
and are sometimes left to collect water, in places
where they cannot be drained.

River channels are blocked up near settlements,
flooding hundreds of acres with shallow water.
Culverts are situated in a way that causes ponds
to be formed, and streams and ditches to fill with
silt. They are often entirely closed up, and the
water allowed to spread where it will. Such condi-
tions have been created by engineers who knew
the ensuing results, but were apparently in no way
interested providing they did not suffer personally
or financially.

The additional cost of controlling malaria in
these localities has been enormous and made the

Vegetation removed from the edge of a pond at Bohio

Anopheles Propagation Areas 71

cost of the anti-malaria campaign very much
higher than it would have been had sufficient
authority been given to the sanitary officials to
prevent the unnecessary formation of hundreds of
acres of artificially created mosquito-producing
areas, with the corresponding attendant costs for

their control, and for the care of sick laborers
affected.

DITCHES

Even where anti-malaria work is being per-
formed, parts of ditches or pools not kept in proper
condition will harbor larve. Any obstruction or
change of conditions that interferes with the flow of
water in a ditch, and any condition that brings
about the breaking of the continuity of the oil film
may attract mosquitoes. Production is increased if
the amount of water flowing in the ditch decreases.

Where the matted alge breaks up and some
portions of it sink, the direction of the small
current in the ditch bed changes frequently, and
increases the number of places and the extent of
area where larve may be found. Conditions are
aggravated if cattle have access to the ditch. If
treatment is neglected or faulty inspection is
made, the ditch becomes a menace and produces
large numbers of mosquitoes daily. The flatter the

72 The Control of Mosquitoes

grade the more likely are such conditions to occur.
The higher the temperature of the water the more
favorable it becomes for rapid development.
Anopheles larve have been taken in ditch water
having a temperature of 102 degrees Fahrenheit.

CRAB HOLES AND PLANTS AS SOURCES OF ANOPHELES

Anopheles larvee have been taken from crab
holes both in depressions on flat land and on river
banks which received larvee from receding waters.
It is known that Anopheles seldom if ever lay eggs
in crab holes where water is below the ground
surface. One species of mosquito, which does
not bite man, habitually lays its eggs in sub-
surface water incrab holes. Itisnot an Anopheles.

While larve of some species are found in water
held by stems of plants, and in tree holes, the
species that enter dwellings and are known to be
malaria carriers seldom oviposit in such places.
Propagation in water held by plants and trees was
not considered of economic importance on the
Isthmus.

CHANGE OF SPECIES CAUSED BY CHANGES OF SEASON
OR OF LOCAL CONDITIONS AT BREEDING PLACES

It would be natural to expect a prolific source
of Anopheles albimanus, or Anopheles malefactor, to

YY.
Wy

Result of blocking the Rio Cardenas. A change of species of Anopheles occurred here

Anopheles Propagation Areas 73

remain favorable to that species. We do not
think this to be true. Certain changes take place
the nature of which is not yet well understood.
The fact remains that within a short period one
species will sometimes disappear and be replaced
soon afterwards by a large number of larve of
another species. There are changes in the food-
stuffs present, formation of alge, plant growth,
etc., and also other factors and changes that are
at the present time unknown. Changes of species
have been observed in relatively large bodies
of water, such as lagoons, wide parts of rivers
containing much aquatic growth, and in large
ponds. In 1913 when the lake at Gatun was
formed, conditions appeared to be quite favorable
to the development of Anopheles malefactor and it
became the prevailing species at Gamboa from
November, 1913, toFebruary,1914, while previously
the Anopheles albimanus was more common in
houses. Since the last mentioned date the Anoph-
eles albimanus has replaced it and later constituted
about ninety per cent. of all Anopheles at Gamboa.

A similar change is taking place at Cafio Camp,
southwest of Gatun, on the shore of the lake. At
that point the jungle is flooded with water and
most of the trees and brush are now dead. Much
of this growth was alive when the camp was first

74 The Control of Mosquitoes

established, about December 1, 1913. The month-
ly Anopheles catch at this camp was, December,
1913, 320; January, 1914, 1969; February, 1914,
2834. Recently the Anopheles albimanus is be-
coming the predominating species.

Strange to say until January, 1914, there has
been no influx of Anopheles from the lake at Gatun
and practically no breeding along the shore of the
lake near that town, although within a mile
conditions along the lake shore appear quite
similar to those near Cafio and Gamboa. Condi-
tions may change and become favorable at a
later date.

OCCURRENCE OF ANOPHELES IN BRACKISH WATERS

At Cristobal, Beach Island, in the Rio Grande
valley, and at Gatun, Anopheles larve have been
found in brackish and salt water. In the first
three locations the propagation areas were affected
directly by tide water. At Cristobal, in that
part of the tidal flats covered by high tides and by
excessive tides, larvee were found to be numerous
wherever clearings were made and leaves remained
in the water. Clumps of plant stems afforded
hiding-places to the larve of Anopheles albimanus
and Anopheles tarsimaculata, even when small
fish were present. Invariably larvee were most

Anopheles Propagation Areas 75

numerous where the fallen leaves were most
plentiful. These leaves were about eighteen
inches wide and four feet or more in length. At
Beach Island, the wet area, a mangrove swamp,
was less brackish and well shaded by trees. It
had a depth of from two to twelve inches and was
filled with stumps of young trees relatively close
together. No alge and very little debris were
present. The young larve were more numerous
than those further developed. The older larve
kept closer to the stumps. Probably the fish
preferred them; this would account for the relative
absence of large larvee. In the swampy area in the
Rio Grande valley the percentage of salt water
varied with the tide and rainfall. Anopheles
albimanus was the prevailing species. The deep
water contained many mangrove trees and drift
from upstream, while the more shallow water
was well covered with grass, dead leaves, and
plants that thrive in brackish water in the tropics.
Larve could always be found in untreated portions
of this area where there were sufficient hiding-
places. The area was about a mile in length.

GATUN PROPAGATION AREA IN IQI3

The largest influx of Anopheles that occurred in
the Canal Zone during the canal construction

76 The Control of Mosquitoes

period came from a flat depression north of Gatun
Dam. In making the channel near the entrance
to the lower lock the material was excavated by
dredges and passed through pipes onto the area
referred to.* | The canal channel at points where
the dredges operated was several hundred feet
wide and subject to the usual rise and fall of
the tide. Previous to the introduction of salt
water into this area it produced some Culex but
not many Anopheles. It was covered with high
grass and in many places with brush. The
brush and trees began to die soon after the intro-
duction of salt water. As the work of filling
progressed, the increasing body of water spread
over many acres and this sheet of water, about six
inches deep, was hidden by tall grass. The water
was deeper farther off shore. The area was more
than a mile long and about half a mile wide.
Flight observations indicated and fairly proved the
direction the influx was taking before it was known
by actual inspection of the swamp that this water
was the source of thousands of mosquitoes that
affected the settlement from half a mile to a mile
distant. The water in the grass around the edges
of the newly formed lagoon remained nearly fresh
and no salt could be tasted along shore. In some
See Map of Gatun No. 36

QOAOASIP SVM YS oyNbsour srayM xnpur oynbsour unyey jo vaie uoryesedor
Pp Ip 19! : poy w) F : d

Anopheles Propagation Areas 77

places where tall grass grew salt was not percepti-
ble to taste six hundred feet from the shore line.
In wading out from the shore the water was tasted
every few yards, and it was noted that young
Culex and Anopheles larve appeared with the first
indication of brackishness. In going farther from
the shore as the water became more brackish the
Anopheles larve found were more numerous and
more mature. When the water became salty
enough to be decidedly disagreeable to taste,
Anopheles larve were most numerous. They
were more numerous per unit of area than had
been noted anywhere on the Isthmus during the
previous nine years of anti-mosquito work. The
absence of Anopheles and the scarcity of Culex
larve in the wet zone not affected by salt water was
unique. Tests made along shore at many points
established the fact. The condition was so
uniform that by wading slowly from shore to shore
with eyes closed, and testing by taste alone, we
were able to reach the infested zone and secure
larve in collecting cups. The water surface
contained large quantities of leaves that had
fallen from the dying trees. Small larvee-destroy-
ing fish were quite numerous, but larve of Anoph-
eles and Culex were so plentiful in the salty
water that it was impossible for the fish to make

78 The Control of Mosquitoes

any reduction. As time went on the density of
larve increased. The observers studying the situ-
ation were bitten continuously by Anopheles of
both species while standing in the water in full
sunlight. The adult species present was chiefly
Anopheles tarsimaculata although Anopheles albi-
manus and Culex were very plentiful. This
was the only time or place on the Isthmus in which
Anopheles were known to come out into the open
and bite freely in full sunlight.

This production area continued in existence for
several months and frequent analyses of the water
for sea water content were made. In places where
the larvee were very numerous the water contained
sixty per cent. or more of sea water and at times
above eighty per cent.

Although the Anopheles tarsimaculata appears
to thrive best in salty water, yet the larve have
been found in fresh water in small quantities,
and a few adults have occurred ten miles away
from water affected by tide water.

METHODS OF DETERMINING THE PRESENCE OR
ABSENCE OF ANOPHELES LARVA

In general we may expect to find larve in
relatively quiet water where aquatic growth,
vegetation, dead leaves, debris, and twigs have

Grass and plants almost covering the water surface and protecting mosquito larve

Anopheles Propagation Areas 79

accumulated. If not disturbed or frightened
they can be seen at rest on the water surface but
often they dive or hide before the observer is
sufficiently near to notice them. It is not usual
to find them where such natural protection or
source of food are absent. Grass and plants
growing in the water are favorable, but dead
vegetation appears to be more satisfactory. After
a little practice an observer becomes expert in
judging whether parts of a body of water contain
larve. It is found convenient to use a small white
enamel-ware saucer to dip for larve; this can be
carried in a coat pocket when not needed. For
continuous inspection work a white enamel-ware
dipper is used to advantage. Where vegetation is
present and larve are hiding in it, the saucer or
dipper may be pushed up to the plant or grass
stalks rapidly, and the larve drawn out of theif
hiding-places by suction as the water enters the
saucer, or dipper. This method will obtain larvee
that are not detected in the water by the eye alone.
When either of these methods fail, the larvee may
be detected by the application of larvacide. This
substance spreads quickly under the water surface,
and the larve rise rapidly in their efforts to escape.
It is much used in the Canal Zone and many places
can be rapidly examined by using a small quantity.

80 The Control of Mosquitoes

The commercial kresol preparations on the market
have similar qualities but are more expensive than
the product we use. If these aids are not at hand,
fairly good results may be obtained in water
of less than a foot deep by thoroughly stirring
up the mud below the water. This generally
makes the larve come to the surface in a short
time. In shallow water where a large number of
places must be examined in a short time, inspection
is made by walking in the water and stirring with
the foot. If grass or debris is present it must be
pushed or held back to leave a clear area for
observation. Where the water has almost dis-
appeared from depressions, or only soft mud
remains, samples may be washed in clear water to
determine the presence of larve. The observer
should walk along the beds of streams, or the edges
of ditches or ponds, and examine all places that
appear favorable, including floating debris on the
surface of deep water. In ponds and lagoons in
which tall grass grows examination should be made
by boat, or if none is available the inspector must
wade out until he knows the condition of all places
that may harbor larve. It must always be re-
membered that the productive portions of a rela-
tively large area may be limited to certain small
ones, either in fresh or brackish water marshes.

Anopheles Propagation Areas 81

Seasonal changes must be also borne in mind.
The unexpected happens very frequently in this
class of work.

Only a few of the negro laborers wore hip boots
for work in deep water. In the wet season the
sanitary inspectors dressed in khaki, heavy leather

~shoes, and stiff pig-skin leggins. When heavy
showers fell unexpectedly it was impossible to
keep dry, and when bodies of water had to be
examined they waded into them, and let the hot
sun dry their clothing asit could. It was fortunate
that the inspectors’ enthusiasm lasted longer than
their clothing! They found being drenched to the
skin several times a day less fatiguing than wear-
ing surveyor’s boots for work in lagoons having
soft mud bottoms. Boots would have been too
heavy for the amount of walking and work of a day
in such a hot and humid climate. The leggins were
essential for protection from thorns, spines, and the
red bugs that make life a misery to some.

It is often stated that sewage and polluted
waters are favorable to the development of
Culex, and that Anopheles are not found in it.
This assertion did not hold true under the condi-
tions in Cuba and Panama. The farther from
the sewer outlet, the less will be the amount of

organic substance carried by the water, and a
6

82 The Control of Mosquitoes

point is reached where the quality of the water
becomes satisfactory to Anopheles.

In cases where sewers have broken on grass-
covered hillsides mosquito larve have been found
numerous within one to two hundred feet of the
point of discharge.

CHAPTER VI
HARBORING PLACES AND FOOD OF ANOPHELES

N uninhabited regions this species feeds on
the blood of birds and animals, juices of
fruits and of plants, and probably on the pollen
of certain plants. Some observers think that
blood is the normal food of female Anopheles
and that other food is taken only when blood
cannot be obtained. Anopheles apparently try.
to obtain more of it than other mosquitoes. We
have observed that they do not bite in full sunlight.
Constant observations on the Isthmus for a period
of eight years, 1904 to 1912, failed to establish a
single case. During these years, when numerous,
they bit freely in the shade of woods. In 1912, at
the breeding area of Gatun, direct sunlight on
brightest days did not prevent six or eight Anoph-
eles biting the face at one time. The species were
Anopheles tarsimaculata and Anopheles albimanus.
They attacked us while up to the knees in water
as well as on shore. It is interesting to note that
83

84 The Control of Mosquitoes

both species were to some extent afraid of the
light of a lantern at night in that particular lo-
cation. If the lantern were behind the neck the
face was bitten, and on bringing the light around
toward the face, they immediately ceased oper-
ations on it and collected on the back of the neck,
and continued biting vigorously. The same was
true of the biting of the hands. The hand farthest
from the light was the one bitten. They invari-
ably left the lighted side of the hand, if it were
slowly moved to about eighteen inches from the
lantern. The same rule held true with regard
to their settling on the observers’ clothing, which
continually occurred. This apparent fear of arti-
ficial lights had been noted in previous years at
Corozal, three miles from the Pacific Ocean, but in
that locality Anopheles tarsimaculata were then
absent and Anopheles albimanus was the prevail-
ing species.

When bright acetylene lights with reflectors
(automobile lamps) were used at Corozal, an
observer standing in the direct rays and holding
his bare arms in them ten feet away from the lamp
was never bitten. Yet when an obstruction was
placed in the column of light and one inch of finger
put behind the obstruction, in its shadow, several
Anopheles albimanus settled on that finger in less

Habits and Food 85

than one minute. Four observers were present
when these tests were carried out. One of the
men was known to be more attractive to mos-
quitoes than the others, but when, having started
with his arms literally covered with Anopheles, he
reached the zone of a certain intensity of light not
one remained. They fled with a rapidity similar
to the movement of people or animals whose eyes
had been suddenly hurt. In all cases the light
was sufficiently distant to cause no increase in
temperature. It is difficult to account for the
instance of biting in full sunlight at Gatun (near
the breeding place only) as compared with all other
cases of absolute repulsion caused by artificial light.
Of course it is possible that the Anopheles at
Gatun bit us while in the sunlight because other-
wise they could never have obtained any blood.
Their natural instincts are difficult to determine.
In general, they are most ravenous at dusk and
soon after daylight. When numerous they bite
freely in shady places near the production area
and at places distant from it where they rest during
the daytime. Many were noted less than a foot
above the ground on the leeward side of trees near
the breeding area. On one tree forty-three were
counted.

In a similar manner they are found on wire

86 The Control of Mosquitoes

screens on the leeward side of houses in large num-
bers while absent from screens on the windward
side. At Gatun, when the mosquito influx was at
its maximum, they were found at rest in brush and
under dead grass on the wind-swept hillside, in
large numbers under houses, and in any place that
afforded protection from sun and wind. This
condition existed for several months during the
dry season. No natural flight in the sunlight has
yet been observed. While no suitable hiding
places except vacant houses were without mos-
quitoes in the daytime, yet beyond the settled
area none were found. They only rested in places
between the brackish body of water and the vil-
lage limits; and when grass in which they were
resting was disturbed in the full sunlight they flew
about ten feet to the next nearest shaded place.

Large numbers were collected under houses
where the breeze was sufficiently strong to make
the lighting of a match difficult. These inhabited
houses were on posts from two to ten feet above
ground. The dry weather ground-cracks under
the houses were several inches deep and the mos-
quitoes collected in them, but when small bunches
of grass or dry hay were placed under the houses
they hid in these in preference to the cracks.
None rested on the floor beams under the houses,

Influence of Prevailing Wind Direction
and its relation to Anopheles entering houses.

Front

>
>,
Direction of
prevailing wind

tt? PN PR

A.Attime of ‘wind Anopheles collect at rear of house.
(leeward side).
Door ( b) is the danger point.
Door (a) may be held open by wind,
Arrows denote resting or aollecting places of Anopheles,

POAC Sete Boe

B.- Screened porch advantageously located,

Diagrams showing best locations for screen doors

87

Influence of Prevailing Wind Direction
and its relation to Anopheles entering houses.

os
3/4
BP
¢ Ae v
ald
a As
ioe
y
tot
C) A Door in a re-entrant
angle facilitates entrance
Pe PsA tt of mosquitoes.
They enter when the dvor
C-Anopheles collect on screen is opened,
and enter if door is left
open long enough,

t T ¢
The.door of screened porch (E.) is better located than thatat(F),
4¢tftAnopheles

-

Diagrams showing best locations for screen doors

88

Habits and Food 89

where there was enough light to read by with
comfort.

In order to determine to what extent the mos-
quitoes rested in the shade under houses, mos-
quito bars were hung up to clear the ground by a
few inches. A laborer was placed under each net.
The mosquitoes did not hesitate to leave their day-
time resting places and soon reached the nets,
where they were collected. Account was kept
of the number taken between six-thirty a.m. and
five P.M. During a period of twenty-two days it
was observed that thirty-three per cent. of their
total number entered the nets between six-thirty
and eight a.m. The total catch by this method
under five houses in the same period was 14,322
adult Anopheles, mostly Anopheles tarsimaculata.
These came from ground-cracks under the house
and from the small quantity of dead grass that had
blown there. In another five-day test, with a
man under a mosquito bar beneath each of the
five houses, 4389 female and 132 male Anopheles
tarsimaculata were captured inside the nets. All
these mosquitoes had been resting on the ground or
in the ground-cracks under the houses. By poking
the cracks with a stick they were disturbed, but
flew away only a few feet to another hiding-place;
none went out into the open where the sun was

90 The Control of Mosquitoes

shining. Although so many were under the houses
none came out to bite anyone standing in the sun.

The Gatun salt-water swamp was the source of
these mosquitoes.t The houses affected were
numbers 3, 184, 185, 195, and 220, located near the
railroad depot.? There are many other houses
at Gatun, some less and some equally distant from
the propagation area. Under most of them no
mosquitoes were present in the daytime.

Although it is customary for mosquitoes to bite
soon after sunset, cases have been known where
observers remaining in one place from six P.M. were
not bitten until eleven p.m., when Anopheles
albimanus became suddenly numerous and fed
freely. When the jungle surrounding houses was
cleared, for several days mosquitoes entered them
in much larger numbers than before the clearing
was made. It is said that a fairly brisk wind
prevents the flight and biting of mosquitoes, but
on the Atlantic slope where the trade winds blow
nearly all the year round, they often bite at dusk
whether the wind blows or not. In infested settle-
ments mosquitoes are seldom found in vacant
houses.

When houses are imperfectly screened Anopheles
are more apt to find defects than other mos-

1See map p. IOI. 2 See map p. IOI.

unqery jo deur oy} uo poyseut ore oS8parp pue sasnoy syy, ‘espoip jueysIp oy} Ivou sem vole Sulpoalq
SUL “WY ydeas-pura oy} Uo punosse10j oy} Ul sasnoy ay} JapuN syowIN punoss oy ut poyarjoo sapaydou yt”

Habits and Food gI

quitoes, but they have difficulty in finding their
way out of buildings. In the daytime, they are
more commonly found in the bedrooms than in any
other part of the house.

Some people are more attractive to Anopheles
than others. One of the sanitary inspectors was
known to possess this quality, and the arms of
three men were placed around his bare arm, so
that any Anopheles desiring to bite him had to
find their way to him through an inch wide opening
between the bared arms of two other men. Several
Anopheles albimanus did this. Some time later
the same person and two observers trying to locate
the principal source of Anopheles causing the
influx at Cristobal were not bitten once from dusk
to nine o’clock. A brown horse nearby was
examined and found to be well covered with
Anopheles albimanus. The observers sat down
near the animal, but failed to attract a single
mosquito in thirty minutes. They then began
to catch the Anopheles that were biting the horse,
using chloroform tubes: the mosquitoes attacking
the horse arrived more quickly than they could be
captured and kept the party busy. The catch-
ing was continued for an hour, and during that
period none of the men were bitten. They had
lamps and noticed that no mosquitoes settled on

92 The Control of Mosquitoes

their clothing, although they frequently examined
the parts which were not exposed to the direct
rays of light.

Mosquitoes not only enter trains to bite people,
but remain in passenger cars that are in motion
for long periods during the night time; they usually
leave the car soon after daylight. Some are
accidentally trapped in train closets and freight
cars and are then transported long distances.

The importance of Anopheles settling on people’s
clothing, and being transported long distances and
into screened houses has not received the attention
it deserves. Cases have frequently been noted
in which they remained on a man’s coat during a
stiff wind while he walked a hundred yards; and
even longer transportation by this means, or on the
backs of animals, is quite possible.

While flight observations were being made at
Gatun, a chicken coop was examined at eight
o'clock on two consecutive nights. The observers
were bitten by Anopheles while watching the
chickens, but failed to note any mosquitoes trying
to bite the fowls. The Culex resting in large
numbers on the chicken roost were not gorged and
did not bite the fowls, but their position gave
them shelter from the breeze. Those outside this
shelter bit us freely.

Habits and Food 93

In one instance three mosquito bars were placed
in the woods for one night with a man under one
bar, a dog under the second, and a coop containing
fowls under the third. The net with the man
collected 274 Anopheles and fifteen Culex, the
dog’s net had five Anopheles and nine Culex, and
the fowls’ net contained twelve Anopheles and six
Culex.

CHAPTER VII
FLIGHT AND ATTRACTION OF MOSQUITOES

HE various species of mosquitoes have not
the same habits. The differences of time
of biting, of selection of propagation areas, quality
of water in which larve are found, etc., are well
known. Every night at certain periods of the year,
hundreds of Culices are found dead in the globes of
electric arc lights. Anopheles on the contrary are
practically never found in the globes and will not
fly close to a bright light. Apparently only a few
species of one genus of mosquito are attracted by
light sufficiently to be destroyed by the flame,
while the Isthmian Anopheles and possibly others
find a strong light repulsive. It is possible that
some Anopheles associate lights with man.

The general direction of the wind at Panama is
from north to south and it blows from south to
north on but few days in the year. Many years of
careful observation on the Pacific slope of the
Isthmus gave fairly conclusive evidence of a marked

94

SOIOPCPY puw ‘yezo1og ‘uoouy ‘eueuRg jo suorjsod oalqepor oy

@ 33 3 3

8 tH 3 e

"Yeeg wi 9ayvI9S

twmouY pus WOrUY jexor0, ‘arsayfoatty

{? wexyro/ aay eyed Grmoys WPS

F340)

fon,
©

Flight and Attraction of Mosquitoes 95

flight of Anopheles in a northerly direction, or
against the wind, and this was reported by Le
Prince at the International Congress of Hygiene,
1912.

Anopheles propagation areas occur near Ancon
and the city of Panama, and the wind blows from
those places toward the houses: if the flight of
Anopheles was assisted by gentle winds the houses
should contain many.? As a matter of fact,
there are few at Ancon as compared with other
settlements, and the production area there has been
under observation for nine years.

Panama touches the Pacific Ocean. Beginning
at the base of the hill on which Ancon, its suburb,
is located, a swampy area extends for about two
miles northwards, and the village of Corozal is
three miles north of Panama. Three miles north
of Corozal is the village of Miraflores, and a
prolific Anopheles-breeding area lies between the
two villages, extending over the larger part of
the intervening distance. The southern end of this
area is nearer Corozal than any part of the Ancon-
Corozal swamp.

In 1911, work was concentrated on the southern
portion of the Corozal-Miraflores swamp, in order
to reduce the Anopheles arriving at Corozal. The

«See map opposite p. 95.

96 The Control of Mosquitoes

work on the windward side was ineffective, and the
influx of Anopheles albimanus increased. Work in
this swamp was then stopped and the force con-
centrated in a larve-infested portion of the Ancon-
Corozal swamp about a mile south of Corozal.
Since the latter area has been controlled no large
influx of Anopheles has taken place at Corozal,
except during short periods when the wind changed
and blew from south to north.

The large swamp between Corozal and Mira-
flores was difficult to control in a satisfactory
manner. It was neglected for a certain time to
ascertain if, when producing Anopheles at its
maximum capacity, many of them would reach
Corozal, traveling with the wind. Apparently
this did not happen, as the number of adults at
Corozal did not increase while the other area a mile
to the south of Corozal was kept under proper
control.

Soon after the Corozal-Miraflores area was
purposely neglected the catch at the laborers’
barracks at Miraflores increased and the influx
continued. The maximum catch in a single trap
on one night was more than a thousand. This
trap was attached to an 8” x 24’ opening in the
screening and the total area of small openings in
the trap screen through which mosquitoes passed

(osnorl 94} SPIsUr WoO1j Usas) Dery oymbsopy uaelIS MOPUIM & 0} poyoezye dvsz oymbsoyy

Flight and Attraction of Mosquitoes 97

was less than eight square inches. During this
high influx at Miraflores the low catch at Corozal
continued. The most prolific part of the interven-
ing production area was much nearer to Corozal
than it was to Miraflores.

Is it the usual custom for this Anopheles to fly
against light breezes? Is it possible that they are
attracted by scent? Or when the number of
Anopheles being produced in a limited area amounts
to millions, do the habits of flight to obtain blood
then change? These questions are exceedingly
difficult to answer, but as they are of the utmost
importance in successful and economical control
of Anopheles they should be investigated.

The increase or decrease of electric lights in
buildings apparently does not make much difference
in the number of Anopheles entering. In several
instances, by increasing the amount of light in a
small experimental house at Corozal, the number
of Culex (species not determined) increased even
when vacant. When the building was empty or
without lights, only an occasional Anopheles
entered, but when one or two men were sleeping
in the building Anopheles crowded through small
openings and bit them whether the building was
lighted or not.

Where buildings are well screened and Anopheles
7

98 The Control of Mosquitoes

outside are numerous, they will try to force their
way through incredibly small cracks or defects in
the screen. Many cases have been noted where
they have become tightly wedged in their efforts
to pass through openings that were too small, and
then were unable to free themselves. While this
experimental work was being done it was found
that although the breeding of Anopheles pseudo-
punctipennis and Anopheles malefactor was taking
place much nearer to the buildings at Corozal than
Anopheles albimanus, mosquitoes of the former
species were not found at Corozal camp. It is
evident, from the long-distance flights described
above, that in some instances it was essential to
locate and control the principal source or sources
of Anopheles, which although more distant than
other propagation areas located relatively near,
may be, and in the two instances above mentioned
actually were, the source of malaria-carrying
Anopheles that reached the houses.

Gatun, about seven miles south of Colon, is one
of the largest settlements in the Canal Zone.
Above five thousand canal employees live there.
Between January and March, 1913, more mos-
quitoes were found there than had been found in
any settlement since work began on the canal.
The weekly catch of the Anopheles that gained

Experiment Station, Corozal, where studies
were made of mosquito attraction by light

Plight and Attraction of Mosquitoes 99

access to the interior of dwellings varied from 7000
to 22,000. It was necessary for office men to burn
smudges and wear leggins. The seats of cane-
seated chairs were covered with paper, and many
devices invented to make life more pleasant.
During the worst part of the period of influx any-
one out of doors at dusk was bitten many times
per minute.

It was evident that some new source of Anopheles
had been brought into existence. Wherever anti-
malaria work had been done before, all known
production areas were thoroughly examined, but
gave no clue to the new situation, as they were
in normal condition. The area under inspection
for propagation places was extended. Mosquitoes
were present in the daytime in all places sheltered
from the sun and wind, except at the east of the
settlement. To the south, was the shore of Gatun
Lake along which much plant growth, debris, and
floating islands collected. Certain aquatic plants
were growing in solid masses from the shore line out
into the water, and filled the spaces between the
floating islands that had collected near the shore.
Vines were growing on the debris, and the floating
mass was more or less continuous along the shore.
All of this new possible production area was care-
fully examined and very few larvae were found.

too The Control of Mosquitoes

The daily inspection was supplemented by night
observations; the shore and the lake just off shore
being observed. No mosquitoes were present on
the lake at night and people in small boats close
to the lake shore not far from the settlement were
not bitten. It was evident after making many
observations that the lake was not the source.
Careful examination of all the territory to the
north of the settlement failed to show any breeding
places of sufficient extent or importance to account
for the influx. The area to the west was examined
last, because of the extensive cleared area west of
the settlement which was known to be free from
possible breeding places.

To the west of the old French canal there was
some flat land into which the sea water and mud
from the American canal was being pumped. This
place had never before produced mosquitoes that
affected Gatun, and was so located that to reach
the settlement the adults would have to fly from
half a mile to a mile straight across or at right
angles to the stiff breezes which prevail at Gatun in
the dry season, over ground containing very little
protection from wind by high grass or bushes.
Flight of this length under such conditions was
not thought to be possible. Between the wet area

t See map.

dNVT unjzer) Jo a10ys ay

Indicates houses where stained
Anopheles were retaken

Agua Clara

Reservoir
L.L.POATES CO. N.Y.

Map of Gatun: Showing Anopheles propagation area and houses where stained specimens were recaptured

102 The Control of Mosquitoes

find blood sufficient to satisfy them. It appeared
within the limits of possibility that they might fly
high at night during the lulls in the wind, and not
be noted by people in boats on the French canal.
The assumption was also made that the period of
long flight might be of limited duration. Other
factors bearing on the problem and all previous
information obtained relating to Isthmian mos-
quitoes were given due consideration. We contin-
ued the investigation, beginning by making several
careful observations on the French canal lasting
more than twenty-four hours each before reaching
any partial conclusions.

On January 20, 1913, at 4.30 P.M., two ob-
servers were posted on the opposite bank of the
French canal from the propagation area, and, facing
the latter, watched carefully for any indication of
flights. They were prepared for a twenty-four
hour watch, although extra night work had robbed
them of much sleep in the past week. For two
hours nothing happened. At 6.20 p.m. birds ap-
peared in the air, apparently catching insects.
They were at an elevation of thirty feet or more
above the water surface. A little later it was
noted that these birds fed at a lower elevation.
At 6.30 interest deepened. The birds kept on
feeding and then flew rapidly back and forth at six

Floating islands blown toward Gatun

ae

ee 2

Masses of aquatic plants and floating islands in Gatun Lake

Flight and Attraction of Mosquitoes 103

feet or less from the water surface. It was then
that Le Prince, looking over the side of a flat
bottomed boat toward the clear sky line, dis-
covered the first appearance of the flight of Anoph-
eles accompanied by Culex. The flight was from
west to east and quite marked. He then selected
a place on the opposite bank of the canal from the
propagation area and faced it. As it became
darker, the quantity of flying Anopheles increased,
and, by bending low and looking past a dark
object at the clear sky line, hundreds of Anopheles
could be seen passing by in one definite direction.
They not only traveled in a fixed direction but
many appeared to hurry about it. After dark the
flight was reduced to practically nothing. During
the period of flight, the observers were bitten
continuously. Soon after the flight ceased one
could remain on the east bank (in the path of the
recent flight) and be attacked only once or twice
in an hour’s time. All night long on the west
bank, near the propagation area, and between it
and the French canal, hundreds of mosquitoes
surrounded and bit an observer. There was no
secondary flight period, although observers re-
mained to note if this occurred.

During subsequent evenings, the flight was
recorded; it started about the same time each day.

104 The Control of Mosquitoes

On the second evening, directly after the birds
arrived and began feeding, several were shot and
examined. It was found that they were catching
the Anopheles, some of which were found in their
throats. These birds are called ‘‘night-jars”’; they
were feeding between thirty and forty feet above
the canal and had not fed at a lower elevation.
After the flight direction near the canal was
ascertained points of observation were selected
between it and the settlement, and the direction of
flight was noted to be relatively constant. When
the winds came in short strong puffs, the Anopheles
headed directly into it, but ‘‘skidded”’ sideways
and were able to fly for short intervals at right
angles to the direction they were facing. Some
remained on the wing in a fixed location and as
soon as they succeeded in controlling themselves,
and conditions were right, dashed off eastward, at
right angles to the direction of the wind. This
forward flight of Anopheles tarsimaculata and
Anopheles albimanus was so decidedly marked
that after its discovery by the senior author it was
easily noted by many, including those who at
first scarcely believed it could be true. No one
in the entire area so thickly infested noticed the
flight direction until instructed how to observe it.
It was thought that with thousands of mos-

Flight and Attraction of Mosquitoes 105

quitoes traveling from the swamp to the settle-
ment each night, an appreciable number might fly
beyond the settlement.

Large numbers passed occupied houses and
appeared at more distant ones; but apparently
none passed the houses most distant from the
breeding area. We remained in the shade in the
day and thrashed the bush for a hundred yards
beyond the settlement and stayed there at night
without securing any specimens. Yet during
this period they were present in practically all
occupied houses, and particularly numerous in
screened houses where the doors were frequently
used, as at the hotel, Y. M. C. A. building, bache-
lors’ and laborers’ barracks, etc.*

The results of the numerous observations showed
that the Anopheles knew where they desired to go;
that they traveled in a direct route at practically
right angles to a strong breeze, and that large
numbers went forward between 6.30 and 7.00 P.M.
daily. It was not ascertained whether any of them
made more than one forward trip to the feeding
ground. It may be that several trips were taken,
but had none returned the thousands passing out
daily were sufficient to continue the influx at the
settlement.

tSee map opposite p. 101.

106 The Control of Mosquitoes

As it was definitely proved that the forward long-
distance flight was of limited duration and occurred
just at dusk, it was assumed that there should be a
visible return flight at or before dawn. The first
morning this was carefully waited for from mid-
night to 6 A.M. but did not occur, and the observers
noticed very few mosquitoes, although thousands
had passed them on the forward flight. It was
thought that there must be a return flight to the
propagation area of a nature not yet understood.
It had been noted that at the start of the forward
flight (z.e. over the canal) the Anopheles were forty
feet up in the air, and it was possible that the end
of the return flight was also above the range of
vision. Additional investigation, however, showed
that there was a marked return flight and though
this happened on several successive mornings, other
mornings showed practically no return flight from
the village, or houses, to the breeding place, so far
as could be noticed. This return flight did not
begin until 6.00 A.M. although there was sufficient
light to read by ten minutes before that time.
The return flight, once fairly started, was of much
shorter duration and more rapid than the forward
flight. The ‘‘night-jars’’ accompanied the return
flight, but were absent on those mornings when no
return flight occurred. As the daylight became

ABMV SOIT XIS SI DOULISIP 9} UT UBIO OTURHY oY “spunois Burpasiq sayaydou pr ayy JO o10ys
IVOU JY} ST [JVM Oo] oy} puodsq ouy eyed oy, “Surpying “y ‘O “JL “A OY} St punosSai0} OY} ur ‘WYsW 9yy OL

Flight and Attraction of Mosquitoes 107

stronger the speed of the returning Anopheles
increased. The termination of both forward and
return flight was remarkably abrupt, or as one ob-
server expressed it, ‘‘the flight stops with almost
mechanical precision when there is too much
daylight or too much darkness.”

As already mentioned, the Anopheles travel at
incredible speed toward the end of the return
flight. The only change taking place was the
increasing intensity of light. This, together with
the fact that when sheltering from the wind or sun-
light they will remain hungry rather than fly
three feet out into the sunlight to bite a person
standing in the sun, but will immediately attack
him if he steps into the shade, is at least sug-
gestive. Are both heat and a certain intensity
of light repulsive to Anopheles, or is it the light
only? The latter would seem to be the case under
natural conditions, as they do not come out into
the sun to bite after 7.00 A.M. One of the obser-
vers, Mr. Zetek, noted some males in the return
flight, and also found blood in the mosquitoes
returning.

During the flight observations Mr. E. F. Quimby
conceived the idea of using an apparatus for
registering the direction of the flight of Anopheles
with a view to determining the direction of, and

108 The Control of Mosquitoes

the area covered by, heavy flights. The appara-
tus is here shown. It consists of four glass plates
set in a metal frame; the latter mounted on a
tripod. The
plates are set in

S two vertical
planes at right
angles to each
other. The in-
struments can be

set up so that

the plates point
north, south,

east, and west.

The glass was
painted with a
mixture of resin

and castor oil
giving a practi-

- cally transparent
A device for showing flight direction of coating capable
unophelas of holding any

mosquitoes that came into contact with it. The
solution was made by adding small quantities of
pulverized resin to heated castor oil; constant
stirring was necessary. The proportions used
were one quarter pound of resin to a pint of castor

Flight and Attraction of Mosquitoes 109

oil. The instrument was used and the results
obtained checked with the results of personal ob-
servations on both forward and return flight, with
the exception of one case, where a male Anoph-
eles was apparently traveling at least temporarily
in an opposite direction from the main flight.

It is possible that where anti-malaria work is to
be taken up in badly infested regions observations
of flight made in connection with this apparatus
will indicate which, of several possible production
areas, is the principal source of the particular
species of mosquito that it is desirable to eradicate.

One interesting feature in connection with this
long-distance flight was that the malaria sick-rate
did not increase much although the number of
malaria-carrying species of Anopheles present
in houses increased enormously. Eight men were
employed daily catching mosquitoes in houses at
Gatun. The indoor catch of each week and
corresponding cases of malaria are given in the
table on the following page.

Although, as already stated, long flights of
mosquitoes on the Isthmus were known to some
of us, we were unfortunately unable to trace any
individual mosquito and find out by actual obser-
vation how far it went, and in what direction.
The necessity for this information was apparent.

110 The Control of Mosquitoes

The chief sanitary inspector assigned to Mr. J.
Zetek the task of marking mosquitves in such
a manner that they could be recognized. We had
previously tried to induce them to mark themselves

Number | Malaria | Percentage
of adult cases per | of employ-
Week ending— Anopheles} week per | ees sick
destroyed | 1,000em- | with ma-
in houses. {| ployees. laria.
Per cent.
207 2.7 0.27
149 3.5 235
199 4.5 245
404 2.5 «25,
666 6.8 - 68
779 6.8 68
3,397 6.1 61
3,150 10.1 1.01
3,296 6.1 61
5, 430" 5.1}, 51
9,415 3.9 239
11,698 5.0 -50
22,074 4.1 41
22,988 5.5 55
19, 873 6.2 62
15, 746 7.6 75
5, 580 8.2. 282
15,676 9.3 93
11, 441 6.8 68
11, 234 5.4 | 54

Table showing number of Anopheles caught in houses at Gatun
each week during the heavy influx

with red ink while escaping from cages, but the
experiment failed. Fortunately Mr. Zetek solved
the problem before the influx at Gatun took place.
The method devised for following up the individual
mosquito was as follows:

Larve and preferably pupe of Anopheles were
collected and developed into adults. These were
placed in cages, protected from sun or wind, and

Flight and Attraction of Mosquitoes 111

stained with an aqueous solution of an aniline dye.
They were given a period of rest and liberated
toward evening. Aqueous solutions of eosin, me-
thylene-blue, etc., were used, one gram of dry stain
to fifty cubic centimeters of water. An atomizer
was used to direct a fine spray upon the mosquitoes,
When spraying care was taken to avoid covering
them entirely with the stain; only minute particles
were allowed to touch them. It was found that too
much stain rendered the mosquitoes useless
for purposes of experimental flight. The stained
mosquitoes were liberated at known distances
from the laborers’ barracks at Corozal and to the
south of it. All mosquitoes noted in the building
at that camp were carefully collected; those from
each house placed in a separate box, dated and
labeled. Later they were placed on a glass plate
that rested on white paper and spread out. Each
one was treated with a small amount of a solution
that dissolved the dye on any stained specimen.
When stained specimens were encountered, the
color was seen to be present as soon as the testing
solution came in contact with it. The solvent
for testing consisted of three parts alcohol, three
parts glycerine, and one part chloroform. As a
result of this method of staining and consequent
ability to follow the movement of mosquitoes

112. +The Control of Mosquitoes

near Corozal, the previous northward flight al-
ready referred to (¢.e. against the wind) was
established beyond a doubt. One of the many dif-
ficulties in this work at first was due to seasonal
changes.

From time to time the supply of mosquitoes
gave out and we had to transport larve and pupe
across the Isthmus; many died of the rough han-
dling and shaking unavoidable with inexperienced
collectors, and when many larve must be collected
quickly.

At Gatun, where staining methods were again
used to check the flight already observed, we
collected mosquitoes in tents near the breeding
places, and also tried small paper houses. The
work was tedious, and it was difficult to move
about in such small places, or to keep quiet while
being bitten by hundreds of mosquitoes and still
more sand flies. This method of collection was
replaced by using mosquito bars. They were hung
up in the woods and the lower edges pinned up and
kept a few inches from the ground. It was soon
found that by occasionally brushing the mos-
quitoes from the hands, face, and clothing, they
would fly upwards and eventually come to rest
in the upper part of the net. Sometimes they
were but half an inch apart. This method was so

Staining Anopheles with aniline dye to determine
length of flight

Flight and Attraction of Mosquitoes 113

successful that other methods of collecting were
abandoned.

When the first net was installed the biting was so
continuous that after it became well filled with
mosquitoes the observers crawled out from under
and tied up the bottom of the net. Conditions
were becoming unbearable even to those persons
who were accustomed to be bitten frequently.
Next morning soon after daylight nearly all the
mosquitoes within the bar were dead, apparently
killed by the drying action of the wind during the
night. After this experience, as soon as a sufficient
quantity of mosquitoes collected in the upper parts
of the mosquito bars, they were sprayed very
lightly with aqueous solution of aniline dye and
liberated by turning the mosquito bar inside out.
This operation was repeated as long as the obser-
vers could keep their tempers, and then a general
retreat was made to the other side of the canal
where there were no mosquitoes.

Very patient negroes were necessary to act as
bait under the mosquito bars! Had the boat
used for crossing the canal been conveniently near
the bait might have escaped. Many of the stained
mosquitoes were recovered, and even two weeks
after the last catch was stained, some of them were

captured near the place where the nets had been.
8

114 The Control of Mosquitoes

Considering the small number that were stained
compared with the enormous number that took
part in the Gatun flights, it was not to be expected
that many stained specimens would be recaptured.
Forty stained specimens were recovered at dis-
tances varying from 1200 to 6250 feet from the
liberation station:

2 between 1000 and 2000 feet distant

ce oe ce

7 2000 ‘‘ 3000

Oo ce 3000 oe 4000 ia “
24 “ce 4000 ce 5000 ae ce

2 “ 5000 ce 6000 “ec “ce

5 at 6250 “ ‘i

At the time of the staining the Anopheles
tarsimaculata was the most numerous species,
although Anopheles albimanus and Culex were
present both at the breeding place and in the
flight. Among the specimens recovered, thirty-
three were Anopheles tarsimaculata; five Anopheles
albimanus; one Culex, and one undetermined.
Some of them were captured in mosquito bars
under houses. Of the five that made the longest
flight, four were Anopheles tarsimaculata and
one Anopheles albimanus. They were caught by
an observer under a net placed in the shade of a
building 6250 feet from the liberation station.

Anopheles in a mosquito-bar exposed to air currents at night

CHAPTER VIII
ATTACK ON PROPAGATION AREAS BY FILLING

N beginning anti-malarial work near a village
where this work has not been done before,
the filling of depressions, etc., that hold water
should be accomplished. These places may be
controlled by the application of oil or larvacides,
but this method is not as good, because the work
needs to be repeated at frequent intervals, and also
because the personal equation is involved and
some propagation areas may be left untreated suf-
ficiently long to produce one or more broods. All
small depressions that hold water, including cow-
hoof marks, badly cut-up land, wheel-track marks,
small ponds that cannot be drained satisfactorily,
and flat lands that do not dry with sufficient
frequency to prevent the development of pupe
during the wettest periods, should receive atten-
tion. There is nothing to be gained by filling
depressions having such absorbent qualities that
they become dry before a brood of mosquito
TI5

116 The Control of Mosquitoes

larve can mature. Lowlands that cannot be
drained and those parts of extensive swampy
land that produce Anopheles can often be filled
to advantage.

It is also important to enact laws preventing
excavation or filling from becoming the source of
new broods of mosquitoes. In borrowing material
for fills, relatively porous material should be se-
lected, if available, because depressions in a fill
made of clay will retain rain-water. If the ma-
terial is taken from a hillside, the bottom of the
borrow pit should be left properly graded.

If seepage planes are suspected, the proposed
site of the borrow pit should be examined with
an earth augur to determine the depth of any
seepage plane that may be present, and the
borrow-pit floor must be kept well above the damp
soil overlying the plane of seepage. If this is not
done it may become necessary, later, to install
seepage intercepting ditches to take care of the
borrow pit. The finished surface of fills should be
properly graded and allowance made for subse-
quent settlement, which always takes place after
filling.

Attempts to cover wet places where subsurface
water under pressure is the difficulty will often
fail when a fill is shallow, and the entire new fill

‘Attack by Filling 117

may become saturated. If there are but a few
springs or sources of water, the fill can be kept
dry by making small ditches care for the water.
It is more economical and satisfactory to treat
flat ditches than to attempt to oil wet grass-
covered areas. In the latter case, as the work is
performed by laborers who are fatigued by carry-
ing loads of oil in mud and water all day long, we
cannot count on their vigilance to ensure the entire
surface of every small body of water covered by
grass being completely covered by oil. Consequent-
ly it is essential to use large quantities of oil and
thoroughly saturate all of the wet land. The area
of the ditch as compared with the area of the wet
land to be filled will indicate the relative cost
of oiling each. Small areas are filled by pick,
shovel, and wheelbarrows. For larger areas
where suitable material is near, drag scrapers
or wheel scrapers drawn by horses are used.

On the Isthmus much excavated material was
available and many acres were filled by dump
cars and Lidgerwood cars from which the mate-
rial was removed by a plow drawn by a steel
rope. On some of the heavy or deep fills made
by dump cars the soft land at the toe of the new
fills was sometimes thrust up by the weight of the
fill or rolled up in front of it. In making fills of

118 The Control of Mosquitoes

this nature when the heavy rocks roll down the
slope faster and go farther than the accompany-
ing earth, it generally happens that a layer of
rocks and boulders lies under the heavy fills thus
made.

Much of the rock used in filling taken from Cule-
bra Cut decomposes rapidly when exposed to the
air, and turns into soilinafew months time. The
freshly made dumps are at first sufficiently porous
to absorb the rain-water that collects in the low
parts of the uneven surface of the dump. However,
in a year’s time settlement takes place and also the
impervious particles of the decomposed rock as
well as clay are washed by storm water into the
depressions. When the depressions cannot be
economically drained it is frequently possible to
get rid of the water by leading it down to the
spaces in the layer of loose stones at the bottom
of the dump. This is done by drilling a hole
near the pond and using a little dynamite.

Within a year’s time the new dumps on the
Isthmus were covered with dense vegetation and
had to be frequently inspected as the older they
became the more numerous were the small surface
pools.

The vegetation may make travel difficult.
Sometimes it is burned in the dry season, to facili-

Attack by Filling 119

tate the finding of newly formed breeding places
at the early part of the wet season when much
Anopheles production takes place.

After the deep fills are made there is sometimes
a considerable flow of water under them, which
outcrops near the toe of the fill and has to be cared
for, either by an intercepting ditch, or by oiling.

HYDRAULIC FILLS

In digging the canal channel near the Pacific
and Atlantic terminals, pipe line dredges were
used. These dredges cut the material near the
suction end of the pumps and draw it in with
the water. This muddy liquid, containing from
ten to twenty per cent. of solid material, is then
forced through pipe lines for long distances. The
lowlands are surrounded by dykes and the liquid
mass fills the enclosure. When the surface water
is at first drained off, the mass of silt and mud
dries very slowly. Its depth may be anywhere
from one to ten feet or more.

In many places pools and depressions produce
mosquitoes before the fill has time to settle and
dry. As the mud dries numerous cracks are
formed which catch and hold rain-water. As the
drying proceeds the cracks widen from a fraction of
an inch to three inches and sometimes are more

120 The Control of Mosquitoes

than afoot deep. The dried surface soil is at first
merely resting on liquid mud and if a person stands
on one of the solid portions it frequently sinks
beneath him. When the crust becomes thicker, the
cracks are more numerous, and so are the numbers
of Anopheles developing in the cracks after showers.
To control similar conditions we installed surface
drainage to assist the drying. This had to be
accomplished without drowning the laborers.
Large sheets of waste galvanized iron roofing were
thrown forward and heavy planks used to bridge
the space between these sheets and put only
light pressure on the thin crust. Shallow ditches
were then dug by laborers standing on the planks.
They do not prefer work of this nature! This
preliminary ditch rapidly dries the soil at its sides
so that it will bear a man’s weight if he does not
stay too long in one place, and it can then be made
deeper. The process is repeated until the desired
depth and grade are obtained. When the cracks
do not become filled by the action of rains, as
soon as the crust is sufficiently strong to support
a mule, an iron rail is dragged over the surface, to
level and fill the cracks. When the material
contains much sand or gravel, the cracks do not
form.

About one half of the area on which Colon is

Cracks in a hydraulic fill

Attack by Filling 121

built was filled by this hydraulic method at the
request of the sanitary department and is now
producing a revenue. The cost of this work at
Colon was charged against the sanitary depart-
ment, but the large ground rents accruing have
not been placed to its credit. The extensive area
on which the new town of Balboa is being built
was formerly a tidal swamp. It was filled by
dumping soil from Culebra Cut upon it. It
was more expensive to dump waste in that
locality than at other established dumps, where it
would have served no purpose and had no future
value. But the difference of cost of dumping
was also charged against sanitation, although the
sanitary department has not been credited with
the rise in value of property formerly worthless.

The work of filling the large swampy area north
of Ancon will be similarly charged. Of course the
paper cost of Isthmian sanitation is increased by
these charges. After the hydraulic fills near
settlements become thoroughly dry and receive
the necessary surface drainage, the value of
formerly useless property increases.

CHAPTER IX
ATTACK ON PROPAGATION AREAS BY DRAINAGE
GENERAL CONSIDERATIONS

ITHOUT doubt, proper drainage is the all-

important and most effective method of
eliminating malaria. It should be more generally
known that so far as mosquito eradication is con-
cerned, the drainage scheme should be planned with
a view to destroying them. The sanitarian and the
engineer often look at the problem from different
points of view. The sanitarian wishes to elimi-
nate all Anopheles-propagation areas permanently.
If that end cannot be obtained at a reasonable
cost, to ensure success he plans a scheme of drain-
age which would eliminate or reduce the possible
breeding areas as much as possible, keep the annual
maintenance expenses at the lowest figure, and
eliminate to the utmost conditions favorable to
mosquito development and the necessity for

continuous future inspection of ditches. Any
122

Attack by Drainage 123

scheme that prevents a possible failure due to
improper or neglected inspection, or which
abolishes the personal equation, is worthy of
consideration.

The engineer often has another point of view.
He plans the ditches for rapid removal of storm
water, but gives no thought to the condition of the
ditch between showers, and seldom has any interest
in or responsibility for the future success of the
mosquito drainage work. The sanitary officer in
charge may be replaced by a man inexperienced
in Anopheles control. The engineer who planned
the work may be in another part of the world and
probably has not designed his drainage scheme to
meet these conditions. In general, he tries to be
rid of a body of standing water at the lowest first
cost. To him the problem is very simple, and
apparently not worth much consideration. He
may not be interested in the habits of mosquitoes
and is generally ignorant of the fact that small
puddles or even a fraction of an inch of water left
standing will defeat the sanitarian’s object. He
may not care if the water in the ditches runs
rapidly enough to remove Anopheles larve, but in
the tropics the width of the bottom of a ditch often
determines whether it is to be a means of reducing
Anopheles or of producing an additional supply.

124. The Control of Mosquitoes

If a competent civil engineer devotes his entire
attention to these matters, he not only makes a
study of details affecting the problem, but will
probably invent new methods of procedure and
new short cuts. It is not unusual for an engineer
to leave minor ditching work to the judgment of a
gang foreman, who makes the ditches too wide,
spoils the grades, and frequently the work is not
only unsuccessful, but expensive to maintain, and
ultimately becomes a dangerous source of Anoph-
eles. It is possible for an improperly drained
area to produce more mosquitoes after the ill-
planned work is concluded, than before the drain-
age was begun.

- An ideal scheme of drainage would be a plan to

remove all standing water from the Axopheles-
producing area, and take care of all storm water
in such a manner that within a short period after
a storm the ground surface and ditches would
become dry. These conditions would eliminate
Anopheles and many other species of troublesome
mosquitoes.

In many places in the tropics this ideal cannot
be attained, because of such factors as geological
formation, texture of surface soil and subsoil,
topography, vegetation, the extent and distribu-
tion of rainfall, and air movement. The mere fact

Attack by Drainage 125

that open ditches are not always followed by
perfect ‘‘ Anopheles drainage” need not in any way
discourage the sanitarian. He must overcome
many apparently difficult problems, or his work
will not be successful. Open ditches should be
made as straight as possible, and have narrow
bottoms. The side slopes should be clean cut.

Drains with flat grades may often deteriorate
to conditions that actually produce Anopheles.
Sometimes it is difficult to prevent the formation of
“‘pot-holes”’ in ditches on heavy grades, and as
each foot of open ditch means an item of expense
for maintenance, the drainage scheme should be
planned to use the least total length of ditches.
To attain the best results, we must be sure the
drain is correctly located. There is often a choice
of locations for a drainage ditch. The character
of the surface soil may be such that the necessity
for certain branch ditches is doubtful. When
working in wet areas it is often best to locate and
install the main ditches before the laterals are
definitely located. This is especially true of
extensive areas covered with water and jungle
where the low places are not yet known, and their
drainage can be planned better after the deep
water has been removed.

Additional advantages ensue in cases where

126 The Control of Mosquitoes

seepage water and springs complicate the problem.
Where these conditions are found the use of the
herring-bone type of ditch is often unsatisfactory,
and the position of the branch ditches may depend
to a large extent on the location of the seepage
water and the points at which it comes to the
surface, ‘‘seepage outcrops.”

A man who has had extensive practice in drain-
age for anti-malaria purposes can do much of the
minor detail work without supervision, but where
grades are light and work extensive, it is essential
to have levels taken, to profit by all the existing
grade. If ground and relative elevations are
judged by the eye alone, wrong conclusions may
easily be drawn as to the possibilities of drainage,
and it is too late, or at least more expensive, to cor-
rect errors after the actual work is well advanced.

It must be kept in mind that some swampy
areas, for reasons not yet thoroughly understood,
are not sources of Anopheles. In other cases they
may be the source of non-malaria conveying
species. Where possible, without increasing the
cost of the drainage work, it is often advisable to
determine the most prolific sources of the malaria-
carrying Anopheles, and to give the drainage of
these places preference over other work.

Seasonal changes may affect the production of

Attack by Drainage 127

mosquitoes, and certain breeding grounds may be
harmful only at one short period of the year, or
only during the wettest part of an unusually rainy
year. The latter condition calls for attention
when Anopheles originate in water coming from
seepage outcrops.

Many problems on the Isthmus differ from those
encountered in other parts of the tropics where
mosquito eradication or control measures will be
undertaken in the near future, and it may become
necessary to modify Isthmian sanitary practice to
attain results rapidly and economically in other
localities. However, it is thought that an outline
of difficulties encountered, and a brief account of
the methods used to overcome them, may be of
assistance to other communities suffering from
malarial fever.

OPEN DRAINS OR DITCHES

This type of drain may be divided into two
classes: First, those that are intended to carry off
storm water during rainy periods, and becoming
dry a day or two after the rain ceases, are known as
storm-water drains. Secondly, those that carry
off water for a period of more than a week, or
which hold more or less water continuously.

Some storm-water ditches may fall under the

128 The Control of Mosquitoes

second classification for varying periods of time
during the wet season, depending upon the rain
distribution in respect to time or continuity.
Storm-water drains may fail to give satisfactory
results because of a tendency of the bottom or
sides to scour at times of heavy flow. The charac-
ter of soil at the bottom and sides of the ditch de-
termines its ability to remain standing with fixed
cross-section. It should be remembered that at
times of maximum rainfall the ditch water may
carry large quantities of gravel and stones, which
assist in causing more erosion than usual. Other
conditions being equal, the steeper the grade of
the ditch the greater will be the tendency to the
scouring of the sides and bottom. Soft spots or
places are often found along the line of the ditch
where the texture of soil lends itself to erosion;
and washouts may be expected at these points.
Again, where the ditch becomes temporarily
obstructed by stone or otherwise, excessive local
scouring action may ensue, removing soil from
below the grade line of the ditch bottom, and
causing a hole in it. During subsequent storms
one or more stones may collect in this depression,
and travel with a circular grinding motion, that
enlarges the hole. These cavities are known as
pot-holes, and retain water long after the storm

Attack by Drainage 129

ditch is dry. In many cases these depressions may
be filled with stone, well rammed into place and
chinked, and the original grade of the ditch bottom
reéstablished before much harm is done. There is
a tendency to neglect these holes, and to assume
that the ditch is dry and does not need inspection.
This assumption would result in Anopheles pro-
duction in the depressions below the ditch grade
line.

- When ditches contain water continuously, or
for periods long enough to bring Anopheles larve
to full development, they must receive regular
weekly inspection. During the part of the year
when large volumes of water pass off in short
periods, the ditches are generally swept free of
mosquito larve. When the average flow is fairly
rapid, the same effect is produced.

It is when the velocity of the water becomes
retarded that conditions become most favorable
to mosquito development. In open ditches having
a rather flat grade such conditions often prevail,
and vegetation within them may develop rapidly.
Both grass and aquatic vegetation assist in re-
tarding the stream flow, besides furnishing food
and hiding-places for mosquito larve. The retard-
ing of the current causes silt to deposit and this

affords sufficient plant food to make vegetation
9

130 The Control of Mosquitoes

grow rapidly, and also raises the grade of the ditch
bottom.

Ditches in soft soil and having a low grade are
the most expensive to maintain, and may become
a prolific source of Anopheles. Vegetation on the
sloping sides of open ditches prevents the banks
from scouring, but when the grass grows long it
falls into the water, retards the current, and often
assists in making conditions favorable to mosquito
propagation. Streams and ditches are treated
similarly; by straightening the channel and re-
grading parts of the bottom, we confine the normal
water flow and increase its velocity, which removes
the hiding-places of larve.

TILE DRAINS

When rock, hard-pan, or other impervious
material underlies the surface soil, the ground
water not being able to penetrate it follows down-
hill on its upper surface, and where the impervious
stratum comes to the surface of the ground or
close to it, there may be a source of water. When
small quantities of water appear in this way on the
ground surface, they are called ‘‘seepage outcrops.”’
The line where water seeps out of the ground may
be short, or may extend along the entire side of a
hill, and become the source of a hillside swampy

Attack by Drainage 131

area. The extent of the seepage outcrop line
may vary with the duration of a rainy season.
Some seepage outcrops may be permanent, others
intermittent, and yet others may be active only
for a short time, during periods of excessive rain,
and may not become active every year.

The character of the surface soil below the

Longitudinal Section
ae ne Tile Drain

(Ame ca

Oross Section
Intercepting
Tile Drain

Longitudinal section and cross section of intercepting tile
drain.

seepage area may vary. If more or less imper-
vious, water remains on the ground surface, and
Anopheles are produced. In places where seepage
water is commonly found, it is well to examine all
sharp changes of grade in surface topography, and
the ground near the toes of abrupt slopes. As the
rainy season advances, the line of seepage outcrop
often moves up the slope, and in controlling this
water the highest points of the outcrop must be
determined. Under the conditions existing at
Panama it was found best, as a general rule, to

132 The Control of Mosquitoes

control seepage water outcrops by tile drainage,
with a minimum grade limit of one foot in two
hundred. The plan adopted is to intercept the
seepage water by tile drains placed approximately
at right angles to the line of flow of seepage water.
The line of seepage outcrop is determined
during the wet season, when the seepage is most
pronounced, and levels taken with an instrument.
The proposed ditch to take the tile drain is staked
in the field, and a profile map made. The profile,
compared to surface conditions, will show any
changes that can be made with advantage, and at
times it is essential to try out several lines in order
to obtain the best and most economical location
for the tile drain. When the uphill portion of the
tile line had a long steep grade, no harm was done
by using a flatter grade for a short length toward
the outlet, as the excessive rains in Panama
caused the pipes to run full section or nearly so.
The tile ditches are made as narrow as conven-
ient, and their bottoms are kept at the established
grade. The openings left between the successive
tile lengths are from one eighth to one quarter of
an inch, and the joints are not wrapped with
muslin or covered in any way. If soft spots are
found in the bottom of the trench, stones are
rammed into place until a solid foundation is

Attack by Drainage 133

obtained. After the tile is wedged securely in
place the trench is filled with stone to the height
of a few inches above the original ground surface.
Earth excavated is placed only on the downhill side
of the excavation. Small stone is preferable for
the top layer of cover stone at the ground surface.
This scheme of drainage has given excellent
results.

There are conditions under which intercepting
tile drainage should not be used. If unforeseen
future changes of topography produce these con-
ditions, the tile line will probably operate in an
unsatisfactory manner. The soil over the greater
part of the Isthmus contains a large percentage of
clay. If the lands above the tile line are kept
covered with vegetation or left in their natural
condition, the surface water running down the
hillsides is not heavily charged with clay and silt,
and passes through the cover stone above the tile
line and then off through the tile. Should the soil
be bared or excavated above the tile line, or a hillside
road located across the tile line, large quantities of
impervious material will be washed onto the cover
stone, and fill up all the spaces between the stones,
and no water can reach the tile. Tile drains are
not planned to meet these conditions, and cannot
operate under them without costly maintenance.

134 The Control of Mosquitoes

Under normal conditions the subsurface drainage
systems cost practically nothing to maintain.
Many of the drains installed seven years ago have
received no attention whatever since they were laid,
and are doing their work now as well as when new.
Anopheles are not produced in subsurface drains.

A few essential points to be kept in mind for
intercepting tile drainage for Anopheles eradication
are as follows: No water should be allowed to
enter the upper end of the tile drain or of its
branches. The grade of the trench bottom should
be true; tiles must not be located on soft mud,
where they may sink. Where the tile line comes
near the surface, due to topographical variations,
proper bridge crossings must be made, so that
wagons will not pass over and crush the tile.
Greasy water and house waste must not be
allowed to discharge into any part of the tile line
system. If drainage from roadside ditches or
excavated areas is turned onto the cover stone, a
tile line will probably become useless.

The profile of a proposed line will indicate the
amount of material to be excavated, and the depth
of the trench. It may often be advisable to use
one or more branch lines to include all the seepage
water, as the cost of excavation and cover stone
may be less than that of a single deep ditch.

Attack by Drainage 135

The outlet or point of discharge of a tile drain
must be well above the ground surface or above
the body of water onto which it is to discharge its
contents. Much solid matter is transported
through the pipe, and may in time block the outlet
at the point of discharge. This condition might

J ae ee Cross Section of Outlet

‘le Drai
Wall at Outlet of Tile of Tile Drain
Drain

Detail of outlet of a tile drain.

result in the clogging of a part, or all of the line.
It is well to arrange for periodical inspection of all
outlets, especially where the lower end of the tile
line has a relatively low grade.

Branch lines should be connected to the main
line by means of Y joints and approach it at an
acute angle or on a curve. Water-bearing strata
may be deep enough to necessitate the installation
of several parallel branch lines placed to prevent
seepage water outcrop between the lines of tile.

136 The Control of Mosquitoes

Where seepage outcrops on bare hillsides, or at
the foot of slopes where there is insufficient grade
for the use of tile drains, open ditches are used.
In these cases the water is intercepted as in the

Side View

\ <— Junction of Two
O Tile Lines showing
A “Y Joint" and Grades

Y Joint

Plan of Junction
of Tile Lines,

Plan and side view of junction of tile lines.

case of a tile ditch, but the side of the ditch nearest
the hill must be given flatter slope than the other
side, because it is wet and more apt to break off and
fall into the ditch. If such a ditch is expensive to
maintain, it should be lined with concrete, and
plenty of weep holes left to care for seepage
water.

Attack by Drainage 137

PERMANENT LINING OF DITCHES

The method that is most economical and durable
for any special locality should be used. It is often
more economical in the long run to line ditches
with stone or concrete in permanent villages and
suburbs of towns. A ditch may be properly
shaped and roughly lined with field stones, and the
bottom finished with cement mortar. If a flat V-
shaped section is used, it is found best to round off
the bottom, so that obstructions will not remain in
the ditch and collect other debris. Elaborate
or very smooth finish for ditches is not essential.

It may be thought more economical to maintain
open ditches than to line them, and in this case the
actual cost of maintenance, including cleaning,
regrading, and oiling treatment, should be accu-
rately kept. If one or more ditches are lined, a
true comparison of cost of the open ditch versus
the lined ditch can be made.

Experience on the Isthmus has indicated that in
most cases it is better to line permanent small
ditches which would otherwise have to receive
treatment throughout the year. Before lining a
ditch an estimate of cost of lining was obtained,
and compared with the estimated cost of mainten-
ance and treatment necessary for the unlined ditch.

138 The Control of Mosquitoes

In some cases lime mortar and rough stone have
been used to advantage, with a small amount of
cement mortar for a finishing surface, or the bottom
was lined with flat stone, the interspaces chinked
with small stone and sealed with cement mortar.
Where it was necessary to line the bottom of the
ditch only, the sides were given a slope beyond
the angle of repose, and allowed to become
covered with grass. As stone is sometimes costly,
the quantity used should be reduced to a minimum.

REINFORCED CONCRETE LINING FOR DITCHES

Where broken stone ‘‘screenings’’ or gravel can
be obtained at a reasonable rate, a reinforced
concrete ditch of light section may be the best
method of ditch lining. In the Canal Zone, a
thickness of about two inches is used for small
ditches with two-inch mesh hexagonal poultry
netting for reinforcing. In one case where a steep
embankment slid, a section of seven feet of this
lining was lifted out in one piece. It was only
necessary to regrade the ditch and replace the
section where it belonged. With most ditches it is
sufficient to line the bottoms and a few inches up
each side, as we are interested in the flow line of
the stream only after the rain has ceased.

Work of this nature cost about twelve cents per

UY} rey JO 90} Surpys jo ys0ddns
pue ‘YO}P eprspeo1 B Jo SUIUT]T oJoINUOD UIYY PadsoOjulay UO]IS-SSO1D JVUIS JO YO}IP B JO Buruly ojarou0|

Attack by Drainage 139

linear foot, in place, for ditches a foot wide, with
unsatisfactory labor costing ten cents per hour.
Reinforced concrete lining should be selected for

Cross Section
of Key Wall

Plan of Key Wall

[Concrete Lining of Ditch

‘Weep Hole

["—~Key Wall

Le eet
Plan and cross section of key wall and position of weep
holes.

ditches near permanent settlements that are
possible mosquito producers throughout the
greater part of the year. The work should be
accomplished during the dry season, and the ditch

140 The Control of Mosquitoes

water temporarily diverted until the concrete has
set. A thin layer of concrete about one inch thick
should be laid on the prepared ditch bottom, the

Retaining Wall
used at Sharp
Curve of Concrete
Lined Ditch

Key Walls used at
Outside Edge of Sharp
Curve of Concrete
Lined Ditch
To Prevent Storm Water

Leaving Ditch

Key Wall

Retaining Wall

To prevent water leaving lined ditches at bends or sharp
curves, key walls are used to break the current, or
the outside wall is made higher.

wire mesh pinned down on it, and the second layer
of concrete installed, leaving the wire mesh
embedded in the concrete. River gravel is as good
as broken stone or screenings and sand for this
purpose. In larger ditches the sides are sloped,
and lined for about a foot above the floor line.

yO,

Concrete-lined ditch at Balboa

Attack by Drainage 141

When natural watercourses are treated in this
way advantageous realignment is first made; and
grades are used that fit the topography; in many
cases causing the actual length of the stream to be
considerably shortened, and its velocity increased.

Whenever ditches are lined, sufficient weep holes

a

CA Junction of Branch Ditch
and Main Ditch

Branch ditches should join main ditches at an
acute angle or on a curve.

must be installed to enable the water that follows
the sides of the concrete lining to enter the ditch.
When weep holes are omitted water is apt to stay
in puddles near the stone lining. Water has a
tendency to flow along the side walls and under
lined ditches. In many instances this is sufficient
to tear away the earth supporting the ditch sides
and at times even to undermine the lining. This
action is prevented by installing key walls at right

Junction of concrete ditches, showing splash wall to confine
water within the ditch

Attack by Drainage 143

Where sharp bends or curves occur in lined
ditches, and especially on heavy grades, provision
must be made to prevent storm water climbing
out of the ditch, and undermining a considerable
section of the lining. This is easily arranged by
introducing one or more key walls on the outer
side of the curve, widening the ditch near the
point of curvature, or by raising the outer wall in

ff

Splash Wall
Oross Section

Bottom Branch Ditch Bank-of Main Ditch Showing
222 Junction of
Concrete
5 Lined Ditch
Concrete Lining ; 7h. errs 7 land A Branch
Reinforcing Wire “a,-===—" | Ditch
Reinforced Concrete Lining} os

Junction of branch ditch with a lined ditch.

the vicinity of the point of curvature. In small
ditches the same result is obtained by decreasing
the slope of the outer wall, or making it vertical
near the point of curvature.

It is not advisable to have a branch ditch meet a
larger ditch and be at right angles to it. It is
better to curve the small ditch near the junction
of the two so that they meet ina Y joint. If this
is not done, much vegetable debris, mud, and stone
may be deposited in the smaller ditch. It is well
to make the main ditch somewhat wider at junc-
tion points, sharp bends, and points of curvature.

Attack by Drainage 143

Where sharp bends or curves occur in lined
ditches, and especially on heavy grades, provision
must be made to prevent storm water climbing
out of the ditch, and undermining a considerable
section of the lining. This is easily arranged by
introducing one or more key walls on the outer
side of the curve, widening the ditch near the
point of curvature, or by raising the outer wall in

Splash Wall
\Oross Section
Showing:
Junction of
Concrete
= Lined Ditch

777 Serene of land A Branch
Reinforcing Wire 9“ ~===—"_ Ditch
Reinforced Concrete Lining 1

Junction of branch ditch with a lined ditch.

B
Bottom of Branch Ditch

Concrete Lining

the vicinity of the point of curvature. In small
ditches the same result is obtained by decreasing
the slope of the outer wall, or making it vertical
near the point of curvature.

It is not advisable to have a branch ditch meet a
larger ditch and be at right angles to it. It is
better to curve the small ditch near the junction
of the two so that they meet ina Y joint. If this
is not done, much vegetable debris, mud, and stone
may be deposited in the smaller ditch. It is well
to make the main ditch somewhat wider at junc-
tion points, sharp bends, and points of curvature.

144 The Control of Mosquitoes

In wide ditches the stone or concrete floor lining
should slope toward the center.

MAINTENANCE OF DITCHES

Maintenance of a ditch implies keeping it in
a condition to carry off water and not produce
mosquitoes. The work in Panama consisted of
keeping the bottom to proper grade, the cross
section of uniform width, the removal of all ob-
structions that affected the velocity of the water,
and also vegetable growth and alge that furnished
food and protection for the larve. It included
treating the ditch with oil and other forms of
larvacide when necessary to destroy larve, and
periodical inspection to make sure that the ditch
was in satisfactory condition and free from mos-
quito larve.

In comparing an open earth ditch with a con-
crete lined one, we found the following advantages
in favor of the lined ditch: The velocity of water is
increased to such an extent that mosquito larve
cannot live in it but are washed away or destroyed.
The increased velocity prevents the deposit of silt,
etc. Each shower cleans the ditch out, and gen-
erally removes any debris that has collected. The
cross section of the ditch remains uniform. No
grass or other vegetation clogs the ditch. Alge

Burning grass from side of ditch; crude oil used as fuel

Attack by Drainage 145

only occurs in lined ditches having practically no
grade, and then only at periods of minimum flow.
It is rapidly removed by the application of a small
quantity of copper sulphate used at the head of
the ditch. Food and protection for Anopheles
larvee are absent, and they appear to avoid in-
stinctively ovipositing in places where these
conditions prevail.

On fair grades there is no necessity for oiling or
treatment of a lined ditch. Inspection may be
made less often and as rapidly as one can walk.
Defects may be noted at a glance. If the lining
has been properly planned and installed, there is
but little maintenance cost, and in most cases
none at all.

Where ditches are practically without grade,
and the soil in the higher lands above the ditch is
heavily eroded by storm water, as the velocity
decreases, the depositing of matter in suspension
increases, but generally only a fraction of the
amount is deposited in a lined ditch that remains
in an open earth ditch. In fact, concrete lined
ditches often take care of themselves, and they
need very little attention.

The disadvantage of concrete lined ditches is
the higher first cost, and in case of construction
work, where the topography is being constantly

ro

146 The Control of Mosquitoes

changed, it is not advisable to install permanent
work that must soon be destroyed or covered up.

Under the conditions existing on the Isthmus,
the proper maintenance of open ditches is often a
difficult task. It is necessary at all times to keep
them free from obstructions. Even a small twig
caught by an irregularity on the bank will cause
other matter to collect and form a temporary
dam. When this happens in a ditch of low grade,
it may result in the deposit of silt, sand, and clay
for one or several hundred yards above the
obstruction.

In a deep ditch regrading and cleaning are
expensive. The current velocity may be increased
on the side of the ditch opposite an obstruction,
undermining that bank completely, and deposit-
ing large amounts of material at points lower down.

Where a ditch passes through soft material, the
channel frequently becomes out of alignment, and
constant cleaning and regrading tends to widen it.
During dry periods it may develop into a semi-
stagnant pool, in which alge forms rapidly, making
a new small temporary ditch necessary within the
larger one. The small one has a tendency to close
up, and is often destroyed by the first shower. If
cattle walk in the ditch, the water in each hoof-
print must be separately oiled. In general, as the

Condition of ditch two months after burning

Attack by Drainage 147

dry season advances, the cross section of the
wetted portion of a ditch grows smaller, and little
puddles capable of mosquito production are left
in the stream bed detached from the oiled water
in motion. All these depressions must be sepa-
rately oiled, and only an intelligent and inter-
ested laborer can be trusted to do this. As the
quantity of ditch water decreases, the rate of
growth of vegetation increases, and it is then nec-
essary to remove it, or smear it thickly with oil to
leave the film free from air holes. Then the work
must be thoroughly inspected to see that it
has been properly accomplished. A sudden show-
er may remove all the oil and the work have to
be repeated.

When the grade becomes flat by removing the
vegetation, the ditch may gradually change into
a long stagnant pool, and the advantage gained by
velocity is lost. It can readily be seen how im-
portant it is to have for foreman in charge of the
ditching maintenance gang a man who knows the
necessity of working correctly, and who can be
held responsible for results.

Spoiling the grade of ditches causes more costly
future oiling and ditch maintenance. Ditches
and streams should have a uniform grade, and as
straight a course as local conditions permit. Least

148 The Control of Mosquitoes

larve will be found where steep banks stretch
above and below the normal flow line. The width
of ditches should be no more than is absolutely
essential, and the water in them should be kept in
motion. Prompt removal of obstructions and
attention to minor detail save much future ex-
penditure and reduce propagation. Laborers at
work cleaning or regrading a ditch frequently pile
the excavated mud in such a manner that sooner
or later it returns to its original position. This
practice is common along railroads, and laborers
should be closely watched to prevent it.

Ditch maintenance gangs often make ditches
wider and deeper. On hillsides the banks of ditches
are apt to collapse from constant deepening of flood
water. This can be partially controlled by giving
a proper slope to the ditch sides, allowing grass to
grow on them, and preventing holes forming in the
bottom of the ditch. As soon as the soft spots are
formed, they must be filled with stone, thoroughly
tamped into place. When it is definitely known
that the erosion cannot be stopped, and that the
material carried away will be deposited in the
ditch at a point lower downstream, to be removed
at regular intervals, an estimate should be made
of the cost of constant removal compared with the
expense of lining the parts of the ditch being

ssvis BULAOUIO Io]ye SYJUOW OM} UOIIpPUOS SuUTAOYs ‘IOqe, puey Aq perea[d yo

Attack by Drainage 149

washed away. ‘The grades of pipe lines conveying
ditch water under roads or buildings should be
slightly more than that of the ditch leading to
them. Their entrances should be screened to
prevent debris entering the pipe without causing
water to be impounded above the pipe at time of
floods. A stone or board floor, or apron, at the pipe
outlet will prevent erosion.

The stability of open ditches and the cost of
keeping them to true grade and uniform cross
section with freedom from larve depends upon the
following factors:

Character of the soil.

Frequency of heavy rainfall.

Grade of the ditch bottom.

Change of grade with corresponding deposit of
material carried in flood time.

Presence of seepage water in banks of ditch.

Natural angle of repose of ditch banks when wet.

Presence or absence of vegetable matter washed
downstream.

Absence or presence of grass on the banks.

Natural watercourses are classified as ditches,
and have to be trained to keep the water at maxi-
mum velocity during dry periods.

The question of uniform cross section of ditches
has already been referred to, and variations of
width in a body of flowing water may not only

150 The Control of Mosquitoes

cause silt, etc., to be deposited, but decreases the
surface velocity near the banks in the wider section,
and in such places alge will grow and mosquito
larve thrive. Near Empire the topography was
suitable for the temporary impounding of water,
and by placing a gate at the entrance to a culvert,
sufficient water was obtained to thoroughly flush
the ditch below it and remove all mosquito larve
whenever desired.

sitdurg :3urysny Aq yoqrp eB ut aidurg :Yyoqrp
ees[8 pojjVUl PUL VAIL] Jo UOTjONIYsep puv [eAouIay e Zulysny 10} 1078. punodun 0} 9783 y

CHAPTER X
ATTACK ON PROPAGATION AREAS BY OILING

NDER unfavorable conditions larve will prob-
ably be found in various bodies of water
notwithstanding all that may be done by filling and
drainage to reduce to a minimum the areas favor-
able for breeding, and if they are to be prevented
from maturing into adults some other method of
destruction must be adopted. Oil is the most
commonly used larvacide, and being generally
applicable is the most useful. Kerosene, crude
oils of paraffin and of asphaltum base, and the
various distillates have been used, and also crude
creosote, eucalyptus, and juniper oil.*

Just how oil kills mosquito larve and pupe, the
writers are not prepared to assert. The generally
accepted theory is that oil clogs the breathing
tubes of the larve and pupe. Another theory is
that by reducing the surface tension, the oil film
makes it difficult, if not impossible, for the larve

tSee Ross on the Prevention of Malaria.
151

152. The Control of Mosquitoes

to hold themselves at the surface. Possibly a
combination of three causes kills the larve:

I. Specific toxicity of the oil to the larve and
pupe.

2. Minute particles of oil clogging the breath-
ing tube.

3. Reduction of surface tension, making it
difficult for the larvee to remain long enough at the
surface to make a rupture in the oil film and thus
obtain air.

Oil appears to be toxic to mosquito larve. We
have observed that a number of larve die shortly
after coming in contact with it. Mere deprivation
of air does not cause death very rapidly in mos-
quito larve. In submerged cages the larve and
pupe of Anopheles often survive more than an
hour.

Probably the minute particles of oil find their
way into the breathing tubes, and cause death by
clogging. We have seen larve take their breathing
tubes into their mouths after the contact with
the oil and apparently make frantic endeavors to
remove some offending substance.

The decrease in surface tension is also a factor,
although its importance varies with the various
genera of mosquitoes. The Aédes calopus can
remain at the bottom of a water vessel for a long

Attack by Oiling 153

time; the Anopheles larve remain on the surface of
the water unless frightened, and are more rapidly
killed by the oil film than others.

There is little, if any, ovipositing by mosquitoes
on water heavily treated with oil.

Kerosene has its merits. Its especial desir-
ability is its property of rapidly forming a thin
film. It was tried in Panama, but rejected in
favor of crude oil. The objections to its use are:

1. The film is so thin that very slight disturb-
ances of the water surface, by flotsam, vegetation
projecting through the surface of the water, ripples
caused by wind or current, etc., break the con-
tinuity of the film.

2. Kerosene is expensive.

3. It is transparent, and is wasted by oilers
because it is difficult to see where the film is
satisfactory.

4. Liability to fire, where sparks may drop
into it; for instance, near a railroad.

Crude oil of asphaltum base is used extensively
in Panama. Its great advantage is its low cost,
and because of that, its poor spreading qualities and
high specific gravity may be overlooked in a warm
country. We doubt if the quality of oil used in
Panama would serve in a cooler country during
spring and fall months. The oil delivered in

154 The Control of Mosquitoes

Panama is imported from California. It averages
20° Baumé. Oil from the same source, testing
30° Baumé, was much more satisfactory, but only
a limited amount could be obtained. The crude
oil was mixed with kerosene in varying propor-
tions to increase its spreading qualities, but the
resulting mixture proved more expensive and less
effective for practical field work than treating this
particular grade of oil with larvacide which con-
tained phenol compounds. It is possible that some
other grades of relatively heavy oils may be more
advantageously used after mixing them with
kerosene, when its cost is not excessive.

METHODS OF APPLICATION

The methods of applying oil for larvacidal
purposes may be classified as continuous or inter-
mittent.

The most serviceable for continuous effect is the
‘drip method”’ in which drops of oil fall upon the
surface of the water from a specially designed
container with sufficient frequency and in such a
manner as to form a continuous thin film of oil
over a certain area of water. This layer of oil is
usually known as the ‘‘oil film.”’

“‘Drips”’ are used advantageously where there
is a moderate surface current, where the water

Attack by Oiling 155

surface flows smoothly, where the channel is fairly
free from obstructions and flotsam, and where
there is very little vegetation and alge. In a word
“‘drips”’ are most useful where there are few, if any,
impediments to the formation of a good unbroken
film. An important advantage gained by using
oil drips is that oil is transported by the stream,
and left in the form of an oil film on the quiet
water along the bank where there is little or no
current, and where mosquito larve are most apt to
be found.

Circumstances occasionally arise which call for
the installation of a drip on a stream that does not
offer the requirements mentioned. For instance, a
rapidly moving, tortuous stream may empty into
a quiet, wide pond, in which conditions are favor-
able to mosquito breeding. This stream may flow
close to a road along which oil may be easily
transported. The pond may be at an inconvenient
distance from the road, and difficult of access for
wagon or cart. Under these circumstances, it may
be advisable to install a drip on the stream at some
point near the road, and by this means carry the
oil to the pond, where it is needed. This example
shows the possibilities of utilizing a stream as an
oil carrier. Our construction camps were fre-
quently situated near sluggish streams or branches

156 The Control of Mosquitoes

of rivers. Hillside streams from distant hills
joined the quieter water near the camp. Larve
and pupz were frequently carried by storm water
toward the settlement and remained to develop
in the quieter water in its vicinity. A rainstorm
thus produced a sudden influx of mosquitoes at
the settlement. This was prevented by installing
oil drips on all streams flowing toward settlements.

The drip may be used on temporary lagoons, as
an auxiliary to intermittent treatment. In the
rainy season in Panama lagoons are formed in
certain depressions from which drainage is difficult.
These lagoons rise and fall several feet in a few
days. Each fall of the water surface causes the
deposit of a large quantity of oil upon the banks,
where it is soon absorbed by the soil. After every
fall the oil film on the lagoon may have to be
restored by reapplication. This is, of course, a
great waste of labor and material. In many
instances labor and oil can be saved in such places
by introducing a few crude rafts carrying drips
and moored at suitable intervals in such a way
that they will readily rise and fall with the move-
ment of the water, and yet remain in about the
same position on the lagoon. The drips auto-
matically maintain the desired oil film, and are
replenished from time to time from a punt or boat.

Attack by Oiling 157

Breeding places for mosquitoes have been found
in catch basins of sewers, particularly during long
intervals between rains. This condition has been
remedied by small drips, but it would be better
to use a device making it impossible for mosquitoes
to enter catch basins except at the time water
flows into the basin.

The measure of success in the operation of a drip
is that it shall work as nearly automatically as
possible. The ideal drip is one that, once adjusted,
requires no further attention except refilling. With
heavy oil, the ideal has not been attained in
Panama. We were compelled to use drips that
worked in a fairly satisfactory manner. Although
far from perfection, the drips were extremely
useful.

The difficulties encountered in devising a satis-
factory drip are:

That crude oil is too thick to permit of making
use of capillary attraction, as may be done with
kerosene and light oils.

It contains a large quantity of suspended solids,
which in time block a small hole or wick, and the
flow either stops altogether, or is greatly impeded.

The oil becomes more viscid in the cooler
temperature of the night, and may stop flowing
until it is warmed later in the morning.

158 The Control of Mosquitoes

These difficulties have been partially overcome
by using the following drips:

The simplest form of drip is a vessel with a small
hole punched in the bottom. This form is still
much used in places where the installation is
temporary, as on temporary ditches that are wet

. ( #) Nall
Ps Cot
: aS ow estes of Of] Can

1) -

_———— Oil Drip made by
Kea Inserting Nail in _

Bottom of Container

A simple form of oil drip.

for relatively short periods, small pools shortly to
be drained, etc. A five-gallon kerosene can is
commonly used, having a hole punched in it by a
nail. A small quantity of cotton waste is wound
around the nail just below its head. The nail is
then pushed through the small hole from the inside
of the can. The quantity of oil allowed to drip
through the opening is regulated by pushing the
nail point upward or pulling the nail downward.

Attack by Oiling 159

Barrels fitted with various adjustable spigots
are also used, and work fairly well. The ordinary
wooden spigot serves for barrels located on the
larger streams.

On the whole, perhaps the most satisfactory
drip is one made of a standard garbage can of
thirty gallons capacity. A slot 3% inch by 1%
inch is cut in the side of the can about five inches
from the bottom. Into this slot a flat spout about
three inches long is soldered, and an ordinary lamp
wick inserted and made to project inside and out-
side the spout.

Water is poured into the bottom of the can
until it reaches within an inch of the flat spout.
Oil is then slowly poured in after previous mixing
with about five per cent. by volume of larvacide,
the latter thinning the oil. The amount of oil
flowing from the can is regulated by compressing
or prying open the spout until the drip gives the
desired number of drops of oil per minute.

In order to spread, the drops of heavy oil must
strike the water surface with considerable force,
hence it is best to elevate the drip so that the oil
drops at least three feet before striking the surface
of the water. On streams having an average width
of one to two feet, from ten to twenty drops of
oil per minute are applied. The quantity of oil

160 The Control of Mosquitoes

Oil Can used to
Drip Light Oils,

Wick Holder

is Y jee a
Flat Lamp Wick //

Detail of Drip
cased for Heavy Oils

* Oil Drip for Heavy Oils

Oil Drip Can Support

The flat lamp wick drip for heavy oils.

Attack by Oiling 161

required depends upon the spread of the oil, the
alignment of the stream, roughness of banks, grade,
alge present, obstructions, etc. For economic
control a trial should be made at each ditch or
stream where a drip is used to determine the re-
quisite rate of flow. In many cases the drip need
only be operated continuously for one or two days
a week. On long streams or ditches it was some-
times necessary to use several drip cans, so placed
that where the effect of the drip at the source of
water ceased, the next drip was installed. Settled
dry weather may permit the discontinuance of
some drips and allow their location to be changed.
The isolated pools remaining in the drying stream
bed were treated by using knapsack sprayers.

In practice it was found that drips required
periodic attention. Each drip should be visited
and adjusted at least twice a week; when new,
three or four times a week. Even the best designed
drip has required this periodic attention.

Occasionally it may be necessary to install drips
on streams subject to freshets. On such streams
they should be secured to prevent their being
carried off by a flood.

Another way to obtain a continuous application
of oil is to use cotton waste impregnated with it.
This method has a limited field of application, but

It

162 The Control of Mosquitoes

may occasionally be used with advantage. Dis-
carded oil-soaked cotton waste is tied into small
bundles, immersed in crude oil, and then placed in
small seepage streams and outcrops. If there is
danger of its being washed away, it can be tied toa
stone. These oil-soaked bundles of waste give a
thin film of oil to the water passing by or under-
neath them during seven to ten days. They may
be resoaked in oil and used many times. This
method of oiling is used where the volume of water
is insufficient to warrant the use of a drip can.
Other uses may suggest themselves under varying
local conditions.

As the conditions under which continuous oiling
can be successfully carried out are rather limited,
it follows that the larger part of oil application to
mosquito breeding areas must be done by the
periodical application of oil or ‘‘intermittent
method.”

In the intermittent method the aim is to produce
a continuous film, and to retain it in place suffi-
ciently long to kill all larve in the water covered
by the oil film. Given a perfect film, it must
remain unbroken at least several hours to insure
a marked reduction in the larve under it. In-
tentionally we do not say ‘‘destruction of all the
larve under the film,” because in the tropics it is

Oil drip applied to hillside stream

Attack by Oiling 163

not always possible, in the field, to produce and
Maintain an oil film on a large body of water at
reasonable cost that will insure the death of all the
larve under the film. We have found Culex
larve under an apparently good oil film that had
been in place several days in succession. The
explanation may be found in the fact that the oil
film, in the field, is frequently not continuous, and
that defective places may be found in it by which
the larve and pupe obtain air. However, a very
great reduction in the larvae may be achieved by
proper oiling, and Anopheles larvee succumb much
more rapidly than the Culex.

The question of how frequently the oil must be
applied, can be answered only after a knowledge
of local conditions has been acquired. The rule is
to oil at least once within the minimum of time
required to mature an egg into an adult. In the
Canal Zone for the mosquitoes most frequently
encountered, this period is about eight days, and
oil was therefore applied weekly.

In the intermittent method enough oil must be
applied to produce a continuous film, heavy enough
to withstand the tearing action of small ripples,
and the light flotsam and vegetation projecting
through it.

It is difficult, if not impossible, to say just how

164 The Control of Mosquitoes

much oil must be applied to a given area of water
surface. Such an estimate may be made for
laboratory work, but it cannot be made for success-
ful field application. Many factors operate to
make such an estimate difficult: Wind and wave
action; presence of vegetation, grass and brush
projecting through the water surface or floating on
it, and alge; the varying density of the heavier oils,
the diverse quality of various lots, and the varia-
tions in a given lot under differences in tempera-
ture. In Panama just enough is applied to attain
the desired result.

Where large quantities of oil are to be used, the
first consideration, from a standpoint of economy,
is the distribution of the oil from its source to the
places where it is to be used. The following meth-
ods of distribution were used in Panama.

About 600,000 gallons of crude oil have been
used annually for the sanitary work in the Canal
Zone. The oil was brought from California in
tank steamers, principally for use as fuel, and the
supply for sanitary purposes was obtained from the
oil company. The oil steamers discharged their
cargo into storage tanks situated at the Pacific
Ocean canal terminus. From these tanks it was
pumped across the Isthmus of Panama through
pipes owned by the oil company. At various

[LO OY} JO ONIVA [VIPUr ay} SOUT] [e1IDAOS 0} JUOWI}VOIY PUL
uor}eyIodsuv1} JO 4SOO dy} posvarour suor}puoo Teorydess
-odo} URIUIYIST BUIOG “JAY Jans O1pog jo Aeyyea Ur Usnig SpUuv] JOAO-jNd UO sUIvAI}S PopoNIjsqo BUITIO

Attack by Oiling 165

points on the Isthmus this oil main was tapped,
and smaller storage tanks connected with it.
Part of the oil used was obtained from the storage
tanks at various places along the line, either
directly, or by running small branch pipe lines
from these tanks to smaller ones nearer the areas
where the oil was to be applied. Part of the oil
supply was transported in two railroad tank cars
to places where connections could not be economi-
cally made with the trans-isthmian oil line. These
tank cars were filled at the Pacific storage tanks,
and discharged their cargo into small tanks at
various points. The aim was to have the tanks as
near the ultimate destination of the oil as possible,
but an adjacent location was sometimes imprac-
ticable, because of the absence of railroad and road
facilities. The tanks were properly covered, their
faucets kept locked, and sand was piled near them
for fire protection purposes.

From the small tanks the oil was transported to
the points where it was to be applied, through
pipes, in mule-drawn tank carts, in canisters, on
mule back, or rolled in drums or barrels. Where
none of the above methods was feasible, it was
carried by hand, or on the backs of laborers.

The Isthmian topographical conditions fre-
quently raised the cost of transportation and

166 The Control of Mosquitoes

application to several times the initial value of the
oil.

Heavy oils are best applied to water in the
form of a stream or spray. When applied in this
manner, the tendency to form a film is much
greater. A number of pumps on the market are
entirely satisfactory for this purpose. The pump
found most satisfactory in the Canal Zone was
called a ‘“‘barrel pump”’ and was especially strong.
The smaller types of hand boiler-pumps are also
serviceable. The requisites of a good pump for
heavy oils are simplicity and strength of construc-
tion, and valves without rubber.

It will be found occasionally that the heavier
fuel oils, those below 25° Baumé, are too thick for
the pumps ordinarily used. These oils may be
thinned by adding kerosene, two per cent. by
volume of crude carbolic acid, or five per cent. of
the ‘‘larvacide” described in this chapter. Other
compounds of a character similar to the “‘lar-
vacide”’ can also be used.

A “knapsack sprayer” of the type used for
spraying in orchards has given satisfactory ser-
vice. This knapsack is made of heavy galvanized
steel, or of copper; it contains a small powerful
pump and has a capacity of five gallons. It
pumps very satisfactorily the asphaltum base

Application of larvacide or oil by knapsack sprayer: Miraflores

Attack by Oiling 167

fuel oil used in Panama, about 20° Baumé, pro-
vided five per cent. to ten per cent. by volume of
larvacide is added.

The advantages of its use are that oil may be
applied properly on the water surface wherever
desired from one to twenty feet from the operator.

Regular Bordeaux. Bordeauz Special

Spray nozzles.

He pumps with one hand and directs the oil-stream
or spray with the other. The work is less tiring
than using a watering can, and while traveling over
rough ground the laborer has both hands free. No
oil is spilled or wasted by too rapid application.

It is not necessary to have the spray as fine or
its particles of liquid as minute as those used in the
application of insecticide. All that is desired is
that the oil should spread rapidly upon the water
and form a thin film.

168 The Control of Mosquitoes

Oil may also be applied by using a garden
watering can; this method is useful where a very
heavy layer of oil is desired, and where the places
to be treated are accessible.

In oiling shallow waters inaccessible from the
shore, a flat bottomed boat with an oil tank may
be used. The boat may be propelled by a small
motor, or by hand, and may have a motor or
hand-driven pump. A spray nozzle of the type
shown is very useful, and saves time.

Two and four-wheel horse carts of special
design are used on the Canal Zone for applying
oil. The carts were designed by Mr. H.R. Trask,
an inspector in the sanitary service of the Canal
Zone. They consist of iron tanks, holding two
hundred to five hundred gallons of oil, mounted on
suitable wagon gear. To the lower sides of the
tanks are attached three-inch pipes leading to a
transverse pipe mounted in front, at the feet of the
driver. This pipe is ten feet long, and the outer-
most three feet of it is perforated with three rows
of one-eighth-inch holes set an inch apart. The
pipe is mounted with a universal joint, and is
controlled by a leather-operated valve. It is
raised and lowered by a pedal. In addition, each
cart is provided with a large valve at the back, for
withdrawing the oil.

SAN DEPT.

Le anvacie

Field supply tank

Oil cart for applying oil to roadside ditches

Attack by Oiling 169

These carts are very useful where they can be
driven, and save much labor in transposting and
applying oil. The illustration shows their con-
struction.

They are especially useful in applying heavy
coatings of oil to roadside ditches, for inhibiting
vegetation, and preventing collapse and erosion.

They may also be used for transporting oil from
the storage tanks to service tanks and drip devices.

In the jungle covered country of the Canal
Zone, there are many places where the lightest of
carts cannot penetrate, and yet are accessible to
pack animals. These animals carried two fifteen-
gallon cylinders of oil fastened to the regulation
pack saddle. These little tanks are provided with
a two-inch screw stopper for filling, and a one-inch
spigot for discharging. The pack mule is especially
useful when filling drip barrels situated on streams
flowing through the jungle, where only trails can
be used. Trails had to be cut for the mule, but
the trouble was repaid. One good mule carried
as much oil as six porters on each trip, and in less
time. The photograph shows the pack mule
which has done valiant service in the cause of
sanitation for about five years. The equipment
shown is of the simplest type.

After the oil is sprayed on the water, the

170 The Control of Mosquitoes

problem of keeping the oil film in place arises, for
on water surface of more than a few square feet
the oil film tends to drift to leeward under the
impulse of a moderate air current, and leaves much
water uncovered.

The tendency to drifting is very difficult to
overcome. Various devices are in use, all being
modifications of the ‘‘boom” principle used by
lumbermen to catch log drifts. The simplest
method is to anchor planks at intervals of five or six
feet at right angles to the prevailing wind move-
ment. Where the winds are variable and strong,
a wooden grille may be employed, with spaces
inversely proportional to the wind force.

Stumps, brush, grass, and stones projecting
through the oil film tend to form oil-free rings
around themselves by their oscillation, and by the
breaking of the ripples against these obstructions.
All such obstructions should be removed from
areas to be treated with oil.

The heavier oils in conjunction with alge form
a tangled heavy mass, and while some of these
masses sink, many do not, but remain on the
surface, and produce defects in the film by drifting
with the surface current. It is well to remove all
algee from water to be oiled.

When heavy oil is poured repeatedly and liber-

Pack mules, for oil transportation in jungle trails

Attack by Oiling 171

ally on the common grasses, their growth is in-
hibited and in time the grasses die. This property
of the oil is used to advantage on the sides of
ditches and in shallow depressions where water
stands during rainy periods, when draining and
filling is not economically feasible.

Oil in any form is not an entirely satisfactory
larvacide for the following reasons:

Its effectiveness depends on a perfect film.
Films are unreliable. Perfect films can be ob-
tained only under exceptional circumstances, and
even a perfect film frequently develops defects.

The larvacidal action of oil is slow. In some
instances several hours must elapse after the oil is
applied before the larve of Culex are killed.
During this time many opportunities may arise
for the development of defects in the film. The
light oils are expensive, and the heavier oils do not
always spread well, unless treated in some manner
to decrease their specific gravity and viscidity.

A relatively large amount of oil is needed to
cover a given water surface, thus increasing the
cost of handling, especially where the oil has to be
transported a long distance by hand.

There is always more or less danger from fire,
while the oil is in storage.

The application of oil in the quantity necessary

172 The Control of Mosquitoes

to kill mosquito larve spoils water for use for
domestic or industrial purposes.

For certain purposes, however, oil is very useful,
and in the absence of a more acceptable larvacide,
the sanitarian must depend largely upon it.

The defects of oil were soon recognized in
Panama, and persistent efforts made to find a
more desirable larvacide. The product named
“larvacide” in Panama was thus introduced, and
although short of perfection, is under many condi-
tions superior to oil.

CHAPTER XI

ATTACK ON PROPAGATION AREAS BY LARVACIDES

HE requirements for a good larvacide are:

1. That it shall be of high toxic power, so
that a small quantity may suffice for a large volume
of water. This requirement is important for its
economic use in water lying far from roads and
trails.

2. That it shall kill rapidly, preferably in less
than ten minutes, that rain intervening, and con-
sequent dilution and weakening, may have as little
effect as possible.

3. That it be uniform in its toxic power and
capable of standardization.

4. That it shall mix freely with brackish and
alkaline waters.

5. That it be harmless to man, and domestic
animals, when in the dilution necessary for larvaci-
dal action.

6. That it shall not be susceptible to rapid
deterioration through age, and exposure to the

173

174. The Control of Mosquitoes

air or light, ze, it must have good keeping
quality.

7. It must be inexpensive.

A number of commercial products now on the
market satisfy these conditions to a large extent.
Many of these have been tried and a few have
proved equal to some of the requirements. Others
were frauds. None fulfilled all the requirements
desired. All commercial products were quoted
at high prices. The larvacide we finally adopted
cost one half less than commercial products.

While working out the problem of a satisfactory
larvacide, Mr. Jacobs, the chemist of the Board of
Health laboratory in Panama, suggested the fol-
lowing formula for a larvacide:

Resin, 150 to 200 pounds
Soda, 30 pounds
Carbolic acid, 150 gallons

This product is a black liquid resin soap, that
freely emulsifies with fresh water. In brackish
or alkaline water emulsion does not take place.
The active ingredient of the larvacide being
carbolic acid, to secure uniformity in the larvacidal
power of the product it is necessary that the active
constituent of crude carbolic acid, phenol, shall be
uniform in quantity. It is found that the best

woouy ye yuryd sproearey ayy,

Attack by Larvacides 175

results are obtained with a content of not less than
fifteen per cent. phenols. In order that the
finished product may neither float nor sink
rapidly to the bottom, but diffuse through the
water uniformly and with fair rapidity, the crude
carbolic acid should have a specific gravity of
about 0.97.

The process of manufacture is as follows:
One hundred and fifty gallons of the carbolic acid
is heated in a steel tank fitted with a steam coil.
When the acid is steaming hot, two hundred
pounds of powered resin is added and the mixture
continuously stirred by means of a paddle agitator,
until complete solution is effected. Thirty pounds
of caustic soda (sodium hydroxide) is dissolved
in six gallons of water, and this is added to the
resin-carbolic acid mixture. The heating and
stirring is kept up for about five minutes, and then
a sample of the product is withdrawn, and poured
into water. If complete and rapid emulsion
results, the larvacide is ready and is withdrawn
from the mixing tank into shipping drums. If
emulsion does not occur, or is incomplete, the
heating is continued until a sample emulsifies
satisfactorily.

Each lot of carbolic acid should be assayed for
phenol content. Carbolic acid containing less

176 The Control of Mosquitoes

than fifteen per cent. of phenols, or of a greater
specific gravity than 0.97, will not make a satis-
factory larvacide.

From time to time, preferably from each lot, a
specimen of the larvacide should be tested for its
larvacidal powers. If a 1 to 5000 emulsion does
not kill full-grown Anopheles larve in ten minutes,
the product is unsatisfactory.

This larvacide, however, loses some of its effi-
ciency when exposed, in emulsion, to the action of
the air, and still more when exposed to contact
with alge and other organic matter. Under
the latter conditions its efficiency is lowered con-
siderably after twenty-four hours’ contact, and
after a few days’ exposure its larvacidal power is
practically extinct.

The advantages of this phenol-resin soap larva-
cide are:

1. High toxicity to mosquito larve. A I to
5000 emulsion kills full-grown Anopheles larvee in
from three to ten minutes.

2. Concentration. Being effective for practi-
cal use in a I to 5000 emulsion, only a relatively
small quantity of the larvacide need be transported
to a given body of water.

3. Uniformity of toxic power. This proauct,
when carefully made, is uniform in toxicity.

Attack by Larvacides 177

4. Simplicity of composition. The manu-
facture of this larvacide requires neither compli-
cated apparatus nor highly skilled labor.

5.. Low toxicity to higher animals. It is
practically harmless in ordinary dosage or in
dilution to cattle, poultry, etc.

6. Rapidity of toxic action. When used in the
field, it killed all Anopheles larve and pupe in
ten to twenty minutes.

7. Cheapness of the product. In Panama, the
cost is about eighteen cents a gallon.

8. Absence of danger from fire. The con-
centrated larvacide is inflammable, but not easily
ignited. In dilution it is not inflammable.

g. It is useful in the rapid determination of
the presence of mosquito larve and kills those
at rest embedded in the mud.

10. In addition to its toxicity for mosquito
larve the phenol-resin larvacide is also highly
toxic to protozoa and alge, as well as most of the
varieties of the common grasses encountered in
Panama. The algacidal and herbicidal properties
of this larvacide are of frequent use in mosquito
eradication.

The disadvantages of this larvacide are:

1. It does not emulsify and is inert in brackish

water. This is a serious disadvantage because
12

178 The Control of Mosquitoes

many Anopheles breed in brackish water and
Culex breed in salt-water marshes and pools.
This defect, however, is shared by all the com-
mercial larvacides tested on the Isthmus.

2. The pure larvacide deteriorates upon expo-
sure to the air and must be kept in drums, barrels,
and other tightly closed containers.

3. It rapidly loses its toxicity after mixing
with water containing alge and other organic
matter. After twenty-four hours its toxicity is
so far diminished that it is practically non-toxic
from the standpoint of field practice.

The ideal mosquito lavacide should, in addition
to possessing all the desirable qualities of the
phenol-resin soap described above, possess none
of the disadvantages enumerated. We have not
yet found such a product, either on the market or
by experimenting with various mixtures.

The toxic action of the pheno-resin larvacide
upon mosquito larve is probably due to the action
of its phenol content upon the protoplasm of the
larve, probably intensified by the fact that the
phenol is in emulsion.

We have had practical field experience with
Pyrocresol. In composition and action this pro-
duct closely resembled the pheno-resin larvacide

Attack by Larvacides 179

manufactured in Panama. Its greatest dis-
advantage was variability of toxicity. Some
samples were practically inert. The cost of this

larvacide is considerably higher than the cost of
the product we used.

CHAPTER XII

ATTACK ON PROPAGATION AREAS AND ADULT
MOSQUITOES BY NATURAL ENEMIES

PROPAGATION AREAS

MALL top-feeding fish which prefer to obtain
their food at or near the water surface are

of great value in reducing the number of mosquito
larve that would otherwise pupate. It was
frequently observed that they seemed to prefer
the full-grown larve and pupe. Many different
kinds of these fish are found in all parts of the
Isthmus, in rivers, streams, and drainage ditches
connected with them. On the arrival of the dry
season they were often left stranded in small
puddles and died as these became dry. While
fish remain in small bodies of water they greatly
assist in reducing mosquito possibilities, but if the
water in the pool contains much green alge the
small fish destroy only a portion of the larve.

Where larve can hide easily, fish catch the fewest.
180

Natural Enemies 181

In ditches, streams, ponds, and at the edges of
lakes and rivers the less the amount of debris,
grass, alge, or other obstructions, the more useful
the fish become. It follows that in countries
where rank vegetation and alge are produced
rapidly, fish are less reliable as destroyers of
mosquito larve than in more northern climates.
During the first American anti-malaria cam-
paign at Havana, fish were of greater assistance,
and reduced the mosquito propagation more than
in the Canal Zone. Large amounts of finely
divided debris with bits of twigs and leaves are
washed down the streams during heavy downpours
of rain and collect in the lakes and quiet parts of
rivers. The constant winds collected this material
and concentrated it. At these places small fish
were nearly always to be seen, and darted about
catching the larve as soon as the sheet of debris
was stirred up or disturbed. By dipping out and
stirring a small portion of it in a white enamel
pail Anopheles larve in all stages of development
were seen as well as large pup; it was evident
that the fish caught but few of the larve so hidden.
When portions of green alge are detached from
the stream bank, fish invariably follow the float-
ing mass and work hard for the few larve they
catch. They cannot penetrate the mass, nor

182. The Control of Mosquitoes

pass through the small openings into which the
larve dart. When larvacide is applied at the
edges of streams, mosquito larve sometimes escape
beyond the treated zone and are immediately
snapped up by fish. When rainstorms remove the
alge, debris, etc., and break up the hiding-places
in streams and ditches, the fish are most actively
employed and have a feast. In quiet waters
they pick up the larve that venture far from their
hiding-places.

In one instance two solid embankments were
placed across the Corundu River and extended
from Diablo Hill to Balboa. Between the embank-
ments was a large flat area of about five hundred
acres. During the rainy season much water col-
lected in it and was followed by a change in the
character of the vegetation.

One of the growths was a moss-like weed that
had not been seen before. It grew in the water to
within an eighth of an inch or less of the surface
but did not quite reach it. There was enough
water above the growth to support Anopheles
larve, but not sufficient for the fish to swim in.
This area was about a mile south of Corozal and
was not treated until it had produced thousands
of full-grown larve. This happened before the
flight of Isthmian Anopheles was understood as it

Natural Enemies 183

is to-day, but the adults that traveled against the
breeze to Corozal were tracked back to this source.
Small fish were useless in this case, they could not
reach the larve.

In certain parts of the Bas Obispo River during
the dry season, shallow water varying in depth
from an inch to a foot ran over a bottom covered
with stones and gravel. A thin film of oil was
generally present, but seemed in no way to interfere
with the numerous small fish, We knew that
mosquitoes were still there in spite of oil and fish.
The film was apparently inadequate to suppress
many of the larve, probably owing to free air
spaces on the downstream side of some of the
partially exposed stones.

In the camps affected by this river the malaria
fever rate had been higher in the dry than in the
wet season, At all other camps in the Canal Zone
we had more cases of fever in the wet season. When
a heavy application of larvacide was given to this
moving water, numerous Culex and Anopheles
larve immediately appeared at the surface. Some
of the fish affected jumped out on to the banks and
the remainder were killed. Since that time the
local fever rate during the dry season has not
exceeded that of the wet season, and year after
year the adjacent camps have had few mosquitoes

184 The Control of Mosquitoes

and a very low fever rate. These instances in
which fish alone have not proved satisfactory are
not intended to give the impression that fish are
of small service in combating mosquitoes, but to
show that on the Isthmus recently developed
methods of control are more swiftly effective than
nature’s methods of more limitedcontrol. Fish
are most useful under many conditions, but the
places in the tropics where their control reaches
perfection are limited. Undoubtedly if all fish
were removed the number of adult mosquitoes
including Anopheles might become so great as
to be unbearable and many places would be
uninhabitable.

We believe that the introduction of the proper
species of top feeding minnows into lakes and
large ponds would be of decided advantage for
purposes of mosquito control in countries less
favorable to mosquito propagation than Panama,
and that the introduction of fish should be accom-
plished and the species selected with care, after
consultation with those competent to give advice
regarding fish propagation.

Probably in the near future artificial control of
mosquito propagation by means of top feeding
minnows will be used extensively. Many miles of
drainage ditches on the seacoast meadows of the

Natural Enemies 185

State of New Jersey are now being kept entirely
free from mosquito larve by small fish.

In Isthmian practice we examine places for
larve. If fish are able to give us satisfactory
results so much the better, if not, then fish must
be temporarily sacrificed to prevent human
suffering. To profit as much as possible by
fish control in general mosquito reduction, the
sanitary authorities introduced the Guirardinus
pecillodes of Barbadoes into the Isthmus. This
is a small fish commonly called a ‘top min-
now” which feeds at the water surface. The
female is about an inch and a half long and is a
rapid breeder. It was bred successfully in tanks.
The young were liberated in the reservoirs, rivers,
and ponds and at various parts of Gatun Lake,
but not in places where fish were unable to control
the situation and larvacide and oil were used.

Tadpoles were found in cow hoof-prints when
water was present, but the Isthmian species did
not reduce the larve in the depressions and obser-
vation failed to prove their value as mosquito
destroyers. The larve of dragon flies and water
beetles were of great value and were found in
places of a temporary character in which fish
could not live long and were not found. Probably
other aquatic insects were of value.

186 The Control of Mosquitoes

DESTRUCTION OF ADULT ANOPHELES

Spiders are useful as destroyers of mosquitoes.
Because mosquitoes prefer to rest on dark sur-
faces, a black band two feet wide was painted on
the walls of the barracks at Balboa between three
and five feet from the floor that they might collect
where they could easily be reached by the men
employed to catch mosquitoes. The walls above
and below this band were white. It was noted that
after the black band was applied spiders collected
on it while they were not to be seen on the white
paint.

Certain insects are caught and held fast when
they rest on spider webs, but it is not definitely
known that Anopheles and other mosquitoes on
the Isthmus are destroyed in this way. They
may be caught by spiders while at rest on or near
the webs.

Anopheles appear to have no difficulty in leav-
ing the strand of a spider web which they have
selected for a resting place. Thousands of spiders
and millions of ants are to be seen on tall grass
and weeds growing in shallow water on the
Isthmus and they probably destroy many newly
emerged mosquitoes.

The small lizards of different colored markings,

Natural Enemies 187

found in Cuba and on the Isthmus, are constantly
catching mosquitoes. They collect insects out
of doors on the patio walls as well as indoors,
and after watching them at work, we were satisfied
that they should be propagated or at least pro-
tected in every way possible. Near Havana in
the afternoons they come out on the whitewashed
walls and never miss a mosquito that dares to
alight within fifteen feet. They often take four
moves forward accompanied by rests, before
making the final rapid dash at the mosquito.
Sometimes they patrol the walls from about four
o'clock until dark, and are hard at work again in
the morning when the mosquitoes come out of the
rooms and settle on the walls.

In the Canal Zone they were found in the old
French barracks. Cunnette camp, near Empire,
held more than other camps. The buildings are
on posts about six feet from the ground and one or
more of the little brown lizards with brownish
orange colored heads can be seen on the house
walls at any time. One of these little fellows if
kept in a screened room would take care of any
mosquitoes or flies entering when the door was
temporarily opened, or carried in on clothing.

Small ants destroy mosquitoes whenever they
have the opportunity; they even interfered with

188 The Control of Mosquitoes

our mosquito trap experiments and caged mosqui-
toes at the experiment station. As soon as they
find a mosquito trap there is a constant stream of
them going to and from it and when they have
disposed of the dead mosquitoes they begin on
the live ones. One ant catches a mosquito by a
leg and almost immediately others come to assist.
In one instance they were seen catching a mosquito
larva in the cup of water at the base of a banana
leaf.

The “night jar” is the most interesting of the
numerous birds that feed on mosquitoes while in
flight. At Gatun, these birds invariably appeared
just before the evening flight began. And in the
morning they could be heard at the settlement,
and followed the returning flight from the settle-
ment back to the breeding place. They dis-
appeared when the morning return flight stopped.
After dusk it was too dark to see how late they fed.

Bats destroy large numbers of mosquitoes near
houses. Before the houses on the Isthmus were
screened they passed back and forth through the
balconies, and the flight range extended only
a short distance beyond the house. Since the
balconies were screened they have been more
numerous between half-past six and seven o’clock,
which is the time the Anopheles assemble on the

Natural Enemies 189

screens. They invariably fly back and forth
in wind-shaded gullies containing brush, where
mosquitoes are more numerous than on the
adjacent higher land. In these sheltered places
when Culex were swarming about the observers
and biting them, the bats approached closely,
while when they were not being bitten by mosqui-
toes the bats were fully ten feet away.

CHAPTER XIII

ATTACK ON PROPAGATION AREAS BY CLEARING
BODIES OF WATER

Y removing vegetation, alge, and drift from

bodies of water, much of the food and most

of the protection of Anopheles larve are withdrawn.

Under normal conditions these larve prefer to

remain at the surface of the water where they find
their food.

Where vegetation in the water is plentiful, it is
difficult to use oil effectively, and in order not to
leave any air holes large quantities must be used.
Vegetation interferes with the spreading of the oil
film; it also interrupts the action of the larvacide.
When rank grasses come up through the water it is
not easy to see if the oil film is satisfactory or not.
In moving water, such as ditches and streams, where
the vegetation is removed the current velocity is in-
creased, and tends toward washing the larve away;
and by leaving no hiding-place during rainstorms,

the stream or ditch can be swept clear of larve.
190

The Clearing of Water 191

Another advantage is that fish and aquatic
insects that prey on the larve have no difficulty
in capturing them and under these conditions
are extremely valuable in eradicating them. In
large ponds and at the edges of lakes where larve
are frequently found it is often necessary to remove
all vegetation in the water, and along the shore,
if it will ultimately reach over into the water.
Once this is accomplished the fish will keep down
mosquito propagation until the vegetation again
becomes rank, or until alge gives the required
protection. Theapplication of small quantities of
copper sulphate along the shore or periodical appli-
cations of larvacide to the small area infested with
alge will destroy it. Lagoons and many streams
of slow current contain water surface vegetation,
such as leaves of lilies, etc., which affords excellent
protection to the larveof both Culex and Anopheles.
Culex are seen in large patches under these condi-
tions and may be so close together as to form a.
black mass of thirty or more square feet. This
takes place even when there are fish. When all
vegetation is removed, additional fish arrive and
the larve soon disappear. To sum up the advan-
tages to be gained, the clearing of water may in-
crease the stream velocity, destroy the food supply,
remove the hiding-places, enable fish to become

192 The Control of Mosquitoes

more useful, and produce conditions that cease to
attract the mosquito.

Most species of Anopheles will not deposit their
eggs in bodies of water that would be unfavorable
to their complete development. They shun locali-
ties that expose them to natural enemies, lack of
food, drying up of water, etc. The difficulties in
clearing bodies of water on the Isthmus are the rap-
idity with which certain grasses grow up through
the water and along shore, and the rapid formation
of alge in shallow water exposed to the sun.
Where the bottom of a pond is soft some of
the grasses may be pulled up by the roots by
means of long-handled potato hooks, and give
little future trouble. On the Isthmus various
attempts have been made to prevent the growth
of vegetation, but no satisfactory economical and
permanent method has yet been devised. Pre-
parations containing arsenical compounds were
tested both in the dry and the wet seasons, in
ditches, ponds, and depressions on low flat lands
that became dry. The results were never perma-
nent, although in some cases vegetation was
retarded for several months. Certain grasses
that run flat along the ground and throw out roots
at each joint grew over ground recently treated
with arsenical compounds without forming roots,

Removing vegetation from overgrown streams:
Machetes are used

Screened verandas: Ancon

The Clearing of Water 193

but as soon as they reached the water in a ditch
their roots penetrated the soft mud. Experience
indicated that the results obtained by applications
of grass and weed destroyers in the dry season
were better than similar treatment in the wet
season. In some instances Bermuda grass was
planted, or encouraged, in order to replace and
kill other grasses and plants that grow taller and
are more troublesome and costly to control.

When it was purposed to use lands near settle-
ments in the Canal Zone for reservoirs, etc.,
the edges were cleared of vegetation and debris,
before the water rose, and the grass was cut very
short just above and below the proposed water
surface elevations. All dead vegetable matter on
the ground to be flooded that might float on the
water surface was collected and burned.

As a result we had bodies of water that were
relatively free from floating timber, sticks, leaves,
debris, etc., and with edges in a condition that
allowed oil or larvacide to be rapidly and economi-
cally applied whenever and wherever necessary.
Near small or temporary camps where conditions
did not warrant much expenditure of funds, the
trees in lagoons and flooded areas were not removed.
Under the above conditions it was found best

to remove brush, grass, and: floating sticks and
13

194 The Control of Mosquitoes

debris, which afford better protection to mosquito
larve than is given by standing trees. When
trees are not standing close together, unless their
branches reach the water surface, collections of
debris are not apt to gather and remain near them.

In the absence of an oil film small fish devour
most of the larve close to the tree trunk. After
the tree decays and falls, its twigs and branches
may tend to collect floating debris and afford
protection to larve. In the tropics vines and
plants may grow on floating logs and become the
nucleus of floating islands.

CHAPTER XIV

ATTACK ON PROPAGATION AREAS BY REMOVAL OF
JUNGLE

ANY square miles of jungle in the Canal
Zone have been removed since the American
occupation, including all camp and town sites
with their surrounding cleared areas, and charged
against sanitation. In 1904 some of the houses
could not be reached without cutting a way through
the intervening jungle. This first clearing was
made by using the ‘“‘machete,’’ and the native
laborer is at his best in using this instrument.
Until this clearing was made, it was impossible
to locate the breeding places near settlements, or
to follow up streams or small natural watercourses.
The seepage areas, pools, and water-holding depres-
sions were frequently covered and hidden.
Clearing uncovers many propagation areas,
and when the sun and wind act on the cleared
spaces, evaporation is greatly increased, and
numerous small ones become dry before a brood of
195

196 The Control of Mosquitoes

mosquitoes matures. It also facilitates inspection
and enables the location and extent of the seepage
outcrops to be determined. It shortens the effec-
tive length of the wet season so far as mosquito
production is concerned, and makes it possible to
locate all new small temporary production areas
that may exist during excessively wet periods.
We found it advisable to clear the ground before
locating ditches, to ensure their correct placing.
The jungle was so thick that on relatively flat
lands it was impossible to determine the position
of the low places by inspection, for when wander-
ing around in the brush it is easy to lose all sense
of direction. Adult Anopheles and other mosqui-
toes rest in the shade, and the removal of the
jungle reduces the number that enter dwellings.
Clearing made it impossible for the negroes to
throw containers into the tall grass or brush near
their houses without detection. They were accus-
tomed to throw them away and would’ carry
unserviceable containers a long distance to dispose
of them rather than put them in the nearby
garbage cans! We cut pathways along the edges
of streams and ditches, cleared the edges of ponds,
and removed grass from puddles and wet lands to
facilitate the application of oil and larvacide.
Much of the clearing in lagoons had to be done

Removal of Jungle 197

from boats. In wet periods certain grasses and
plants grew an inch or more per day, and the work
had to be done repeatedly over the same territory.

Toward the end of the dry season, spaces were
cleared by fire, whenever they could be b-rned
over without danger to property because of the
trade winds that assist in spreading flames. Near
all permanent settlements where the topography
permitted, the ground was prepared for using
mowers, and on some steep hillsides heavy spikes
were used on the wheels to prevent the machines
from sliding downhill. This was especially neces-
sary where the hillside terminated at the abrupt
canal banks. At times, when the tall grass needed
cutting, the ground was so soft that the use of
horses meant the creation of many new breeding
places by hoof-prints, etc. If the grass remained
uncut, evaporation was retarded and mud breed-
ing aggravated conditions.

The day’s work had to be arranged according
to weather conditions.

CHAPTER XV

SCREENING AND PRACTICAL DESTRUCTION OF ADULT
ANOPHELES IN HOUSES

INCE Sambon and Low with their two com-
panions lived for three months in their
screened hut in Ostia, Roman Campagna, where
malaria is usually prevalent, without contracting
it, attention has been directed to the possibility
of securing protection from mosquitoes by pro-
tecting the dwellings with mosquito-proof material
and a variety of fabrics have been used in various
places and many details of technique have been
elaborated. After many experiments, both in the
laboratory and in practice, the methods of mos-
quito-proofing described are now used in the
Canal Zone, where screening of dwellings has been
carried out on a scale never attempted elsewhere.
Because of the atmospheric humidity and the
salt laden air it was found that only the best
copper gauze will resist corrosion and subsequent

deterioration. The copper gauze used in the
198

Screening and Destruction —_199

Canal Zone must have not less than ninety per
cent. of pure copper, and not more than one half
of one per cent. of iron. The cost of this screening
is approximately fifty cents per square yard for the
18 mesh, No. 31, B. W. G. gauze, 7.e., gauze having
eighteen strands of wire of one one hundredth
of an inch diameter in each linear inch, which
is the number of strands necessary to reduce the
apertures to a size that will not permit the passage
of Aédes calopus and the smaller specimens of
Anopheles and Culex, and yet retain reasonable
strength of material and open space area.
Verandas are usually screened as shown on the
photograph. In the tropics the veranda is practi-
cally the living room; people usually spend their
evenings in it, and it must be screened. Aside
from the fact that screened verandas contribute
very largely to the comfort and probably to the
health of the occupants‘of the dwellings, screening
on verandas is easier and cheaper to maintain than
the screening of many doors and windows that
would otherwise have to be screened. And
because of their larger air and light admitting
area, screened verandas do not exclude quite as
much air and light as screened doors and windows.
A decided advantage is gained by having but
one door through which mosquitoes may enter.

200 ‘The Control of Mosquitoes

Unless near a wall mosquitoes do not congregate
about the door of a screened veranda as they
do about a screened door in a solid wall. They are
too busy examining all the screened surface to
concentrate at its entry.

If there is a prevailing wind direction during
the mosquito breeding season it is advisable to
have the balcony door on the windward side of
the house.

Where windows have to be screened, the sashes,
shutters, and frames are so arranged that a per-
manently fixed screened frame can be fastened
in place. Sliding screen frames are very unsatis-
factory in practice; they warp and refuse to shut
tightly, they cost more to install and maintain,
and are often left open.

A light solid door, well designed and made,
and rapidly self closing is preferable for a mosquito-
proof house. In damp countries such doors
should be made of close-pored lumber, well dried
and carefully filled and varnished, or painted,
and designed to reduce to the minimum the
tendency toward sagging, warping, and swelling.
The doors should be provided with efficient self-
closing devices, and they should open outward to
frighten away any mosquitoes resting on or near
the door or hovering in the vicinity. Occa-

4 AFTER
| APPLIED

A good device to prevent the sagging of screened doors

Screening and Destruction 201

sionally, when mosquitoes are very numerous on
hospital buildings, especially yellow fever wards,
it is desirable to double-guard the entrance by
providing a screened vestibule. Such a vestibule
should be at least six feet wide from door to door,
and should be arranged so that one door must be
closed before the other can be opened. A number
of devices will readily suggest themselves. One
that has been used with considerable success is
a pulley and rope arrangement.

If screened doors are to be used, the frame should
be made of light material that will resist warping
and swelling, and should be provided with a brace
to prevent sagging. A good contrivance for this
purpose is two rods connected by a turnbuckle.
One free end is fastened at the upper corner on the
hinge side of the door, and the other free end to
the corner diagonally opposite, as shown on the
diagram opposite. Tendency to sagging is cor-
rected by twisting the turnbuckle.

The screen panels should be protected from
injury by hands and feet by fastening over the
gauze on the outside of it, so as not to interfere
with any mosquito catching that may be required,
one quarter inch or three eighths inch galvanized
wire netting made of wire sufficiently thick to
resist denting by a push of the hand or foot against

202. The Control of Mosquitoes

the panel. Screen panels should not be more than
three feet wide and eight feet long. Wider or
longer panels resist wear less satisfactorily.

The screening should be fastened with copper
tacks to prevent galvanic action, and the edges.
overlaid with wooden strips three-eighths of an
inch by one inch, and fastened with galvanized
wire nails. The tacks should be placed at least
one half inch from the edge of the screening.
The screening should be drawn as tight as possible,
without the aid of special stretching devices; i.e.,
it should be stretched tightly enough to prevent
denting and consequent breaking in cleaning, but
not so tightly as to put immoderate strain upon
the strands bearing against the tacks.

It was found advantageous to protect in a
similar manner the lower panels of verandas, the
lower parts of the panels above hand rails, panels
immediately adjacent to doors, and other panels
subjected to hard usage by the inhabitants.
Strips of wood have not proven satisfactory in the
protection of screening. They have the inherent
defect of interfering seriously with the success of
mosquito destruction within the houses.

When screening is kept clean, by frequent
removal of the dust and the products of corrosion,
the amount of light and air excluded by the eigh-

Screening and Destruction 203

teen-mesh gauze used on the Canal Zone is negli-
gible. The best proof that screening does not
materially interfere with the comforts of the house
occupants is the fact that not a single complaint
has ever been heard on that score from the people
on the Zone.

It was occasionally necessary to use tents and
railway cars for quarters. The method of making
tents mosquito-proof was as follows: A matched
lumber floor is laid, and on this are built up the
sides of a framework of two by fours, with the
necessary doors. These frames are screened. The
ridge pole is formed by two vertical two by fours,
and one horizontal member, properly braced.
Brackets of the right length and height are fas-
tened to the wall frames at intervals. The tent is
fastened to these brackets by pulling it tightly over
the upper horizontal members of the wall frames,
and fastening it to the members by clamping
strips of wood to form mosquito-proof joints.
Over the tent roof a fly is drawn, and its edges
fastened near the ends of the brackets to leave a
space of at least fifteen inches between the tent
and the fly edges.

' In the railroad construction camps, screened
cars are used. The cars are screened in the same
manner as the dwellings. The illustration shows

204. The Control of Mosquitoes '

some of these cars. A camp consisting of screened
cars, and a tent hospital screened in the manner
outlined above, have been in use for about four
years. Although Anopheles were very numerous,
as shown by the daily catches in the cars, and
although very little oiling, etc., was done in this
locality, the screening, plus daily catch of mos-
quitoes in the cars, kept the malaria incidence
to almost the average incidence of the Canal
Zone.

The crucial point about screening is the thor-
oughness with which the work is done, and the
constant vigilance and care that must be exercised
in speedily detecting and remedying defects.
Mosquitoes, and particularly Anopheles, will
readily find a very small aperture in their endeavor
to enter a house to find blood, but once within a
house, the mosquito very seldom, if ever, finds
its way out again unless doors or windows are
open. A screened house with rents in the screen-
ing, cracks in the floor, openings between the plate
and roof, or any of the many imperfections of
mosquito-proofing, is a veritable trap, and from
the point of view of the protection of its inhabitants
from mosquitoes, more dangerous than an un-
screened house.

A perfectly screened house may yet offer in-

Types of screened houses: Colon Hospital grounds

Screening and Destruction 205

numerable avenues for the entry of mosquitoes.
To enumerate a few of these: cracks and knot
holes in the floor; ill-fitting doors; spaces between
a corrugated iron roof and the plate left uncalked;
cracks in the siding of an unsealed frame building;
open spaces around plumbing, stove pipe, etc.
In the Canal Zone, with the extreme variation
in humidity between the dry and wet seasons,
defects develop constantly in the woodwork of the
light frame constructions employed. The screen-
ing is also exposed to rather severe strain. To
maintain efficiently the mosquito-proofing of
the dwellings, a weekly inspection was made of all
the screened houses, and the needed repairs noted,
and made at once. Since 1908 the repairing of
screening has been delegated to the Quarter-
master’s Department for administrative reasons.
From the standpoint of sanitation, it is wiser to
maintain a screen-repairing force under the
immediate control of the sanitary authorities, to
insure this important work being done by a force
specially trained to detect defects in mosquito-
proofing, and to render dwellings mosquito-proof
unhampered by the interference of other duties.
After experience with both systems, the writers
earnestly recommend that all questions of mos-
quito-proofing should be entrusted to the sanitary

206 The Control of Mosquitoes

authorities, who are more competent to judge
the importance of the defects deemed unworthy
of attention by the average layman, builder, and
architect.

In a malarious locality the screening of dwellings
is a necessity. With a view to determining as
nearly as possible the difference in malaria infec-
tion between the inhabitants of screened and
unscreened dwellings at Gatun, a study was made
by. Dr. Orenstein. The conclusions reached were
that screened dwellings reduce the malaria inci-
dence by at least one third.

It is almost certain that the protection afforded
is greater than this study showed, for the reason
that while the people residing in the screened
quarters of Gatun were obliged to consult a govern-
ment physician when incapacitated for duty by
illness, the residents of the unscreened section
were not required to adhere to this rule, and con-
sequently there is no record of a probably large
number of cases from these houses.

INFORMATION CONCERNING METALLIC MOSQUITO
SCREENING

16 mesh copper alloy gauze will permit the
passage of Aédes calopus only under stress of
circumstances.

Screened cars in railroad construction camps

Screening and Destruction 207

17 mesh and 18 mesh will exclude this species.
16 mesh will exclude Anopheles.

16 mesh—Area of aperture 0.00235 square inch.
Wire No. 28 B. W. G.—Diameter 0.014
inch.
Weight per square foot 0.2224 lb.
It is twice as strong as 18 mesh.
60.16% of open space.
95% increase in area over No. 31 B. W. G.

17 mesh—Area of aperture 0.00219 square inch.
Wire No. 30 B. W. G.—Diameter 0.012
inch.
Weight per square foot 0.17339 lb.
It is 50% stronger than 18 mesh.
63% of open space.
43% increase in area over No. 31 B. W. G.

18 mesh—Area of aperture 0.00208 square inch.
Wire No. 31 B. W. G.—Diameter 0.01
inch.
Weight per square foot 0.12835 lb.

67.39% of open space.

All copper alloy used in screening should contain
not less than ninety per cent. copper and less than
one half of one per cent. of iron.

DESTRUCTION OF ADULT ANOPHELES

In addition to the screening, a further decrease in
the possibility of infected mosquitoes transmitting
malaria can be accomplished by daily destroying

208 The Control of Mosquitoes

the Anopheles found in the dwellings. This has
been systematically carried out in all the barracks
in the Canal Zone, and is considered a very valu-
able measure when painstaking men are employed.
Systematic mosquito catching in dwellings as a
prophylactic measure against malaria originated in
the Canal Zone. Briefly, it was the adaptation to
sanitary purposes of the method employed by the
entomologist in securing specimens. It was first
tried in 1908 by W. R. Proctor, sanitary inspector,
in a temporary camp known as Cocoli. This
camp was in close proximity to extensive Anopheles
breeding areas. Because of the camp’s temporary
character, more thorough methods of eradication
were not used. Malaria soon developed among
the laborers in the camp, and the incidence of this
disease became very high.

Although there was much doubt concerning the
efficacy of mosquito catching as a means of reduc-
ing the malaria incidence, it was tried as other
measures could not be applied. To the surprise
of all, the systematic destruction in the camp
speedily reduced the malaria incidence. The
catching was done at sunrise and sunset, by a
negro laborer.

Mosquito catching in dwellings was then adopted
as one of the regular prophylactic measures,

Screening and Destruction 209

During I909 a number of camps were located
along the new line of the Panama Railroad,
officially known as the Panama Railroad Reloca-
tion. This is to be the permanent railroad after
the Canal is completed. Four of these camps,

1909, 1910. 1911.

Months. Con- Con- Con- Remarks.
struc- | Entire | struc- | Entire | struc- | Entire

tion Zone. tion Zone. | —tion Zone.
Camps. Camps. Camps.

1.55% | .96% | .67%| .72% | Dry Season.
67% |. .11% | .42% | .77% | Dry Season.
1.44% | 1.04% | .28% | .85% | Dry Season.
2.50% | .81% | .56% |} .68% | Dry Season.
2.80% | 1.16% |-1.94% | 1.91% | Rainy Season.
«| 2.41% | 2.42% | 5.06% | 3.59% | Rainy Season.
3.29% | 3.89% | 2.84% | 4.14% | Rainy Season.
1.34% | 2.81% | 1.36% | 1.82% | Rainy Season.
| 96% | 1.76% | 2.64% | 1.40% | Rainy Season.
2.59% | 1.25% 2.58% | 1.00% ‘| Rainy Season.
-50% | 1.16% | 2.58% -82% | Rainy Season.
-11% | 1.16% | 1.21% -76% | Rainy Season.

Averagepermonth| .94% | 1.50% | 1.53% | 1.55% | 1.84% | 1.54%

Table of malaria incidence at construction camp.

with an aggregate population of about 1200, were
located along twelve miles of the road, at intervals
of three to four miles.

It would have been costly to attempt drainage
and oiling operations of sufficient magnitude to
protect all these camps from malaria, and it was
decided to house the employees in screened cars,

and to carry out daily mosquito catching in all of
14

210 The Control of Mosquitoes

them. The number of Anopheles caught in the
cars, which were surrounded by mosquito breed-
ing places, reached 1800 a week. The malaria
incidence in these camps was but slightly above
the incidence for the Canal Zone shown in the
table. It will be observed that in 1910 the
malaria incidence was two one hundredths of one
per cent. less for the camps, and in 1911, three
tenths of one per cent. more, as gauged by admis-
sions to the hospital. It must be remembered,
however, that the percentages for the whole Zone
are based on the final diagnoses made by the exam-
ining physician. Experience has shown that the
examining physicians sometimes give a provisional
diagnosis of malaria in a somewhat larger number
of cases than the final diagnosis sustains.

Another instance of the application of this
prophylactic measure under more favorable condi-
tions occurred near Corazal. In this case the
camps were remote from native habitations. In
June, 1908, several hundred United States marines
were quartered for two months on Diablo Hill.
During that period the malaria incidence among
them averaged fourteen per cent. a week. No
mosquito catching was done. Some cars in which
railroad laborers lived were located at the foot
of the same hill. From the early part of May to

Screening and Destruction 211

the end of November, in the rainy season, when
malaria incidence is high on the Isthmus, only
four cases of malaria occurred among the forty
laborers occupying these cars—a weekly incidence
of three tenths of one per cent. The difference
was due to having a man devote half an hour a day
to the destruction of the Anopheles found in these
cars. The work cost five cents a day.

The technic of hand catching is as follows: A
glass tube about four and one half inches long
and one inch in diameter, of the variety used for
packing small camel hair brushes, is used. An
inch layer of small rubber bands is packed into the
bottom of the tube. They are held in place by a
plug of absorbent cotton, which in turn is covered
by a disk of blotting paper to facilitate the removal
of mosquitoes from the tube, by preventing their
entanglement in the cotton. A few cubic centi-
meters of chloroform are poured into the tube,
which is then covered, and the chloroform allowed
to become absorbed by the rubber bands. A tube
thus prepared will be lethal to mosquitoes for
several days. The chloroform tube has the ad-
vantage of being safer than the cyanide tube in
the hands of a more or less careless laborer. To
catch a mosquito with the tube, the cork is re-
moved, and the mouth of the tube quickly placed

IGE

212. The Control of Mosquitoes

over the mosquito while it is at rest on some
object. In afew moments the mosquito drops to
the bottom of the tube.

Equipped with one of these tubes, a “slapper”
made of a six-inch square of wire gauze fastened

Slapper, chloroform tube, acetylene lamp, and pill box; used
in catching Anopheles in houses.

to a two-foot stick, and, if necessary, a bright
search-lamp or portable electric light and cord,
an ordinary laborer will soon learn to catch a
surprising number of mosquitoes. That mosquito
catching in dwellings may be facilitated, prefer-
ably white or light colored walls should be insisted
upon. Anopheles rest during the day in the

Screening and Destruction 213

darkest nooks in the room; in remote dark corners,
behind various objects on the wall, and under
window ledges, etc. Careful search is required
to find them and on a dark colored wall they are
almost invisible.

In the early morning hours soon after daybreak,
and in the twilight hour of the evening, Anopheles
usually collect on the screening of verandas,
doors, and windows, and are more easily caught
than when indoors. The chloroform tube does
not work very well on screening and the ‘‘slapper”’
is employed.

The cost of mosquito catching in dwellings is
very slight. We employed common laborers at
ten cents an hour. One man can cover a score
of barracks in two or three hours. If the houses
are in a group, an experienced mosquito catcher
will examine as many as twenty in three hours.
Of course, the number of houses covered depends
on their size, the number and interior arrangements
of the rooms, the number of mosquitoes found,
and the distance from one house to another.
The statement made above regarding the number
of houses a man can examine, refers to the standard
barrack of the Canal Zone, a one-room, one-story
building about forty by sixty feet.

In the early part of 1911 hand catching was

214

The Control of Mosquitoes

Wire Screen High Ridge
Frame of White Pine

. 4 —————
i ¥
= 1 1 D L
ee
cee
f Wire Screen Low
43) Ridge ;
sh Frame of White
5A) Pare perueiibw watpe uy re LA Hla | i Pine
Na
/
r

Mosquito trap, with details of construction.

Screening and Destruction 215

supplemented by mosquito traps. The evolution

ge
gs
8 §
Ze
B%3
a Oo
&
ba
ed
8
Ee
Fd
S3
a
Bs

twisted & turned down

d
2
3
My §
a
wi 8
ae oS
aT 43
Sel; 6g
wi] Og
¥
Le 3
% e
re &
3 5
2
a 3
be F
gm
22
“CE
3%
i
oa
a
Es
Ras

of the mosquito trap into the model used at present
has been gradual. The first trap constructed,

216 The Control of Mosquitoes

while ingenious, failed in its purpose—mosquitoes
simply would not enter it. Gradually by develop-
ing the original ideas and plans of the senior author
and by constant experimenting under his immedi-
ate direction, Messrs. Bath and Proctor, sanitary
inspectors in the Canal Zone service, evolved
the Anopheles trap now in use.

This trap is essentially a labyrinth built of wire
gauze on a wooden framework. The illustrations
on pages 214 clearly show its construction, and
method of application to a screened window. As
a mosquito catcher, the trap is of great service
when properly used. It was found that in order
to catch Anopheles the traps must be attached
to the lee side of buildings, and that usually more
Culex are caught if the trap is fastened on the
windward side. The reason for this is yet to be
determined, but the phenomenon is a fact, and
should always be remembered when installing the
traps. The possibility of some species of Anopheles
being attracted by scent was considered. When
two adjacent buildings were supplied with Anophe-
les traps, one inhabited and the other vacant, no
Anopheles entered the traps attached to the vacant
building.

It is found that more mosquitoes are caught
when they are attempting to enter a building.

Interior of laborers’ barracks, where Anopheles were caught daily

Screening and Destruction 217

Therefore traps are installed with their openings
turned outwards and are not used for collecting
the Anopheles which have already gained access
to the building.

Traps are of little use on unscreened houses and
are never used on them.

The number of mosquitoes caught in these
traps has been surprisingly large and it is not
uncommon to catch several hundred per night in
one trap. .

In order to determine whether any mosquitoes
escape, a trap in which a number of mosquitoes
were confined was fastened to an empty trap.
None passed from one trap to the other. If they
could have found the openings in the trap in which
they were confined they would have collected in
the other trap.

Mosquito traps, built after the model used on
the Canal Zone, are of distinct value as a prophy-
lactic measure in anti-malaria campaigns, as an
auxiliary to screening, and the destruction of mos-
quitoes by hand.

These mosquito traps cost about $1.25 each
when made by hand by the dozen.

CHAPTER XVI

THE RESULTS ACCOMPLISHED BY THE ANTI-MALARIA
CAMPAIGN

HE methods of malaria control applied and
developed on the Isthmus were put to a
severe test and gave successful results. The
topography, meteorological conditions, and con-
stant changes due to the construction work, to-
gether with the character and constant moving of
the population and their dwellings, and social
conditions, were peculiarly unfavorable to the
control work undertaken. The natives and em-
ployees, infected or well, were at liberty to live
where and how they chose. We had no control
over their movements or methods of living.

The area covered by anti-malarial operations
was about fifty square miles, with a shifting
population of many nationalities. It is now evi-
dent that even under such adverse conditions,
malaria in the tropics may be kept down to a

minimum rate without prohibitive cost.
218

Results 219

Much valuable experience has been gained,
that will aid future anti-malarial campaigns, and
these newly devised methods of procedure and
the important discoveries made as the work
progressed will now make it possible for many
sections of the tropics to develop their agricultural
and natural resources.

Until now Aédes calopus and a few species of
Anopheles have prevented the Caucasian race
from settling and developing the tropics. In some
of the richest parts of the United States more
than fifty per cent. of the agricultural population
is infected with malaria. In some parts of the
tropics the malaria rate exceeds seventy-five per
cent. of the rural population, and the transmission
of infection often continues throughout the year.
Under such conditions can progress or social
development be expected?

In South America there are parts of the sea-
coast, and even cities, to which laborers from the
hill-country cannot be induced to go. They
refuse a high rate of wage, and their reasons
and judgment are sound. In recent railroad
construction in Brazil, European laborers
would not remain in the company’s employ
although paid an exceedingly high rate of wage.
Where eighty per cent. of the laboring force is

220 The Control of Mosquitoes

sick at the same time their actual earnings are
small.

The rich lands in malarial sections of the tropics
cannot be economically developed without syste-
matic malaria control. It is no longer a question
of whether a corporation can afford to pay for
the necessary anti-malaria measures; it is now
admitted to be foolish to attempt development
without reasonable sanitary precautions.

The preventive measures taken in the Canal
Zone cost less than one cent a day for each
person.

During the reconstruction of the Panama
Railroad, in many camps we housed laboring
forces in outfit-cars surrounded by extensive
swamps, where drainage or economic control of
mosquito life was impossible. By screening the
cars, and destroying Anopheles in them once a day,
the infection of laborers was prevented. Live
mosquitoes caught daily in the cars were examined
for parasites, but no infected specimen could be
found. These preventive measures had not been
attempted before, but can be applied again, and
the efficiency of our troops on the Isthmus, and
elsewhere, may be kept up to the standard.

Large forces of men employed in commercial
enterprises may now be kept in good health where

221

sickness formerly made their day’s work uncertain.

Tales = Bblh

_Llorathly Tikes wt S47

1912
AvGEsO G34,

LQ

¥Ge © 0.64%

MALARIA CHART
FOR 1911, 1912 ann 1913
Showing the number of malaria cases each
month among employees, expressed as a per-
cenlaGe of the entire working force.
ay,
37
2%
wl
3!
i
;
:
N
1%, :
0%

The item of high labor cost, because of malaria,
has been a controlling factor in preventing progress

in the South.

222 The Control of Mosquitoes

It has been ascertained that marked differences
exist in the habits and life history of different
Anopheles, and between Anopheles and other
mosquitoes. We learned how to observe their
flight, and that the malaria-carrying species can
be stained, and by this means it can be determined
how far and in what direction they travel; and
by intercepting them in the path of their flight,
they are made to register its general direction.
It was established that some species of Anopheles
travel farther than others.

It was observed that while Culex, and possibly
other mosquitoes, enter a residence on the wind-
ward side, Isthmian Anopheles seldom do this.
The importance of placing screened entrances
and verandas on the windward side is now under-
stood, but had not previously been considered.
A method of making Anopheles trap and destroy
themselves has been perfected; and it has been
determined that some species avoid bright artificial
light.

New methods of collecting specimens have been
devised.

The migration of one species across the Isthmus
was closely observed. Previously it was thought
that malaria increased in proportion to the number
of Anopheles in a malaria-infested community,

Results 223

but Isthmian observations show several instances
to the contrary. Where Anopheles tarsimaculata
and Anopheles albimanus traveled long distances
from the breeding to the feeding grounds their
number at the settlements greatly increased, but
the malaria rate did not rise. It is of great
importance to know if these facts hold true in
regard to other malaria-conveying species, as such
facts determine the right areas for malaria control
operations.

Due to factors not fully ascertained the same
species do not always fly long distances, or even
a quarter of a mile, to reach a house or village.
Since learning how to observe and follow a flight
we can now decide correctly which location to
drain or treat for the protection of a town or
settlement, and leave out the unessential.

The varying distance of mosquito flights of
which so much, assumed and untrue, has been said
and written, no longer appears impossible to
solve. At present the important point to decide
is the length of flight of different species of Anoph-
eles in so far as it affects malaria transmission,
and this may differ from the flight length probably
essential for the propagation of the species.

The decidedly brackish marshes and the edges
of shallow tidal flats, which were once the chief

224 The Control of Mosquitoes

sources of Anopheles at Gatun, Christobal, and
Colon, presented conditions that may not be
limited to the tropics, and afford precedents for
future anti-malarial work near seacoasts farther
north.

Although the Anopheles is usually considered a
night-biting mosquito, Isthmian observation has
proved that some species bite people in the day-
time, both in houses and in forest shade; and when
very numerous these insects appear less afraid of
capture or exposure to natural light. In one
instance Anopheles tarsimaculata and Anopheles
albimanus bit freely all day in the sunlight, but
this was the sole exception noted during more than
twelve years of observation.

Other new facts of importance ascertained and
improvements devised relating to control measures
that resulted in reducing the cost and increasing
the efficiericy of anti-malaria work were:

Concentrated larvacide, effective when and
where oiling is useless, is now made and trans-
ported economically.

Practical tests were made to discover the cause
of the decay of copper alloy screening. It was
found that monel metal mosquito wire, sixteen or
eighteen mesh, remains intact after the best grade
of copper bronze wire has decayed.

Results 225

Devices for making wire screening continuously
effective, together with systematic inspection
and repair.

The complete eradication of mosquito-producing
conditions in hillside ditches and the elimination
of maintenance costs by the introduction of a
thin reinforced concrete lining of special design.

The elimination of seepage water breeding areas
by the use of special methods designed to suit
local conditions, where standard methods of sub-
surface drainage were valueless.

The use of automatic oil drips to control mos-
quito breeding in running water in streams and
ditches, and that of oil-soaked waste for small
see page outcrops.

In describing the use of oil on the Isthmus Dr.
Malcom Watson, who has had extensive experi-
ence in measures for malaria eradication, and
directed the successful control of malaria in the
Federated Malay States, after inspecting the
methods used at Panama, reported to the Royal
Colonial Institute as follows:

As the greatest sanitary achievement the world has
seen, it has a lesson for us. I went there because
I wanted to study the details of their methods, in
particular how much of their excellent results was
due to drainage, and how much to screening, oiling,

15

226 The Control of Mosquitoes

and quinine. From my visit I concluded the results
were mainly from oiling, which was done for practi-
cally half a mile on each side of the canal, or at least
the inhabited portions of the Zone. The - great
majority of the population do not live in screened
houses and very few take quinine. Drainage is
constantly interfered with, and there is no agriculture.
Colonel Gorgas had therefore everything against him
except the determination to win and money to back
him. Depending on oiling the organization has to
be, and is, perfect, for failure for a single week would
allow mosquitoes to develop. Their position is as if
they had a wild beast by the throat, but were not
allowed to kill it; yet if for a moment their grip were
to be released, the beast would be on them. No
greater sanitary work has been done: I doubt if we
will see as great again. It is perfect work, and its
organization is the only kind that would have suc-
ceeded under the circumstances. Some of the methods
I saw are already being put into action in the Fed-
erated Malay States on my advice.

The health officials had no control over the
selection of camp sites, yet were expected to main-
tain the standard of good health of the laborers.
In some cases the camps were placed near prolific
Anopheles production areas, and no extra funds
provided for their treatment. Similarly difficult
problems and conditions were brought about by
unnecessary obstructing of natural watercourses
and flooding hundreds of acres of jungle near

i
TTT TT TT tir

| a
4+ 1 7
Lt _| 7 t ra ag
| & | |
7 a | rest “30
a a t -2-, 8 -S tv
mA 3 fii | 1 | al" al Aine
ris 4 ae aabe
2 fo ee ail +
a b=] - t 7 [au
qt 2 Cl | aa s\ | ih
a ies —+ Ya ad
A fe 2 Se
T T
28 gett een a8
So Sel ia |_| | | [is An Be
a 3 eC Pro TH btarave
88 427 TTT Ls
»g—2#ttrrtit
P 5 “co 9.2 eS Ss ae
=) | {i=}
aes So Bmn| tH
m3 Ae ao poh
B= oOo gaLl ll
poe acer a
oo n
ja = 8 841
Boo ee] L | |
RReris foe Tey
deer 3s) Pokey |
S208 8 8a
28 sete
oe as co r
peegs [| ! 2.
; ay
fee 23Ll TTT 428
2 3 Petes: 6) al | y ] | eet
ee ea A al | iz
|_| iE Ss
= zo/ ||] CI oe
eg 5 }—} + __1 t—t far
<g¢ 4 ee | jh 4
| {| | L_| mE
aa & — mi
s 5 s | | i
a S te ey. a)
4 2 Ss a a +e
S @ tte
| a
ESSSEReCESeRteR
T | TI | | Ls
rt Tr Poe boree |
| rt | = B| LT liz
al ek
=. 4 a
T an | ) t
j T
pier + —— t =
| r | l_] | ie
. Toh aoe a
seen [te
| | | Lie fet
ae Cote
EE pany mei
|_| | al
|

Malaria in the Canal Zone: Isthmus of Panama

Results 227

populated districts. Had the necessary topo-
graphical changes been considered or foreseen, the
advantage of selecting other camp sites would have
been obvious.

The proximity of proposed camp sites to exten-
sive Anopheles producing areas, or native villages
in which many cases of malaria exist, merits careful
consideration.

The old idea of locating a house on a small hill
to escape ‘‘malaria miasma”’ is of little importance.
The slight advantage gained is only that of in-
creased air motion causing temporary protection
from mosquitoes.

The selection of poor locations for camps, and
the creation of new and extensive Anopheles pro-
duction due to topographical changes, and other
causes, produced fluctuations in the malaria sick
rate, many of which might have been avoided.

By studying each problem separately and
devising methods to solve it, and concentrating
effort on the most important points of attack in
order to obtain the largest net returns for each
dollar expended, we were able to reduce, from year
to year, the malaria sick rate of the force employed.

The malaria charts show the results accom-
plished.

80%

20%

60%

30%

40%

30%]

202) T

10%

2%

0% =
es a a2 gf g@ g
Ss 8 &§ € &§ & & &

HosPiTal CASES oF MALARIA AMONG EMPLOYEES -1906-19/3
Compiled by adding monthly percentages of force sick with malaria

PART II

The Yellow Fever Campaign

229

CHAPTER I
THE CAMPAIGN IN HAVANA

HE measures adopted to eradicate yellow
fever from the Isthmus were based upon
the knowledge gained from the work done by
Colonel Gorgas and his associates in the eradi-
cation of yellow fever in Havana, Cuba. Havana
was the first place in which anti-yellow fever
measures were carried out with success, working
along lines suggested by the theory of Carlos Fin-
lay, and observations of H. R. Carter, which
greatly aided and were confirmed by the experi-
ments of Reed, Carroll, Lazear, and Agramonte,
that the Stegomyia fasciata, now known as the
Aédes (Stegomyia) calopus, was the only trans-
mitting agent of this disease.

In order that the methods adopted in Havana
may be better understood, it may be advisable to
review the status of the knowledge regarding the
transmission of yellow fever at the time the work

of its eradication was inaugurated in Havana,
231

232 The Control of Mosquitoes

under the direction of Colonel Gorgas, in February,
IgOl.

Colonel Gorgas has treated this subject so
comprehensively in his final report to General
Leonard Wood that we quote from it:

For over 200 years this disease had, at short inter-
vals, devastated the Atlantic and Gulf coasts of the
United States, causing great loss of life, and still
greater financial loss, due to the entire cessation of
commerce which occurred during the epidemic of
1878, which affected particularly the lower Mississippi
Valley, amounted to $100,000,000, and in years when
there was no epidemic, quarantines had to be kept up
against the infected regions around the Gulf of Mexico,
which stopped almost all travel and greatly interfered
with commerce. The United States had come to look
upon Havana as the particular point from which
infection was spread. Yellow fever had been con-
tinually present in this city since 1762. Every
month in every year during that time there have
been some cases. In all other localities of North
America where yellow fever occurred, it occurred
epidemically; that is, the locality was free from the
disease for a longer or shorter time. In places above
the frost line, winter always puts an end to the dis-
ease, and in localities in the tropics it always termi-
nates after a greater or lesser period of years from the
exhaustion of the non-immune material.

Briefly stated, on February 1, 1901, the date of
the commencement of anti-yellow fever work

The Campaign in Havana — 233

based upon the theory of mosquito transmission,
the following facts were known concerning the
method of the transmission of this disease, and
concerning the transmitting agent, the Stegomyia
(Aédes) calopus, then known as Culex fasciatus
(Fabricius) :

The Army Medical Board had proven:

1. That yellow fever is capable of being trans-
mitted by the female of the mosquito now known
as the Aédes (Stegomyia) calopus.

2. That this mosquito must have previously
fed on the blood of an individual sick with
yellow fever, within the first three days of the
disease.

3. That an interval of twelve days or more,
after feeding on a yellow fever patient, is necessary
before the infected mosquito is capable of convey-
ing the infection.

4. That yellow fever can be produced by the
subcutaneous injection of the blood of a yellow
fever patient, taken in the first and second days
of the disease..

5. That the period of incubation of the disease
is between forty-one hours and five days and
seventeen hours.

6. That “fomites’ cannot convey the disease.

7. That the Stegomyia fasciata (Stegomyia

234 The Control of Mosquitoes

calopus; Aédes calopus) is a house-haunting mos-
quito. Its breeding habits, however, were not known.

After the conclusions of the Army Medical Board
were made known, the next problem of importance
was the application of this knowledge to the best
advantage. Whether it would be possible to
eradicate, or reduce to a non-infective minimum,
the Stegomyia calopus was then not known, and
did not appear practicable. It was noted that the
Aédes calopus entered and left buildings, and that
its larvee were mostly found out of doors. It was
generally thought that it oviposited in the open,
under conditions similar to those suitable for
various other species of Culicide; and that it
could fly long distances to reach habitations.
The limits of its flight were not known. It had
been observed that some Culicide travel with air
currents, and it was supposed by many that the
Aédes calopus did likewise; there being no evidence
to the contrary.

These theories brought into the project of eradi-
cating the Aédes calopus the question of drainage
or filling of all the tidal and swampy land near
Havana Bay, a very extensive area.

Further investigation and research was carried
on to determine the possibilities of immunizing
the non-immune or foreign population, by subject-

The Campaign in Havana =. 235

ing them to the bites of infected Aédes calopus, in
the hope of producing mild attacks of yellow fever,
and thus conferring immunity. It was soon found,
however, that the severity of the attack could not
be controlled, and this work was abandoned.

The board of army surgeons who verified Dr.
Carlos Finlay’s theory of the transmission of yellow
fever by the mosquito, and made use of Dr. H. R.
Carter’s observations on time of incubation, had no
opportunity to study the natural life history or the
possibilities of controlling the Aédes calopus; it
was therefore natural that they did not suggest
the methods of procedure needful for yellow
fever control.

Colonel Gorgas decided to ascertain at once
the possibility of Aédes calopus destruction on a
large scale throughout the city, to determine what
could be accomplished along this line. Work
of this nature had not been attempted before, and
all facts of importance relating to Aédes calopus
destruction were collected as the work progressed.
He believed that the propagation of this mosquito
could be controlled, but few other people at that
time believed as he did. It was the general
opinion, with few exceptions, that an impossible
task had been undertaken.

The possibilities of the use of oil for purposes

236 The Control of Mosquitoes

of mosquito destruction and control were dis-
covered by Dr. L. O. Howard and published by
him as early as 1892. This information was of
great value to the American officials in Cuba and
has since been used extensively in many other
parts of the world. In 1892 he published an
article in Insect Life on the use of kerosene
against mosquito larve; in 1896 in Bulletin No.
4, Bureau of Entomology, Washington, D. C.,
several pages were devoted to the subject of
remedial measures against mosquitoes; and again
in 1900 in Bulletin No. 25, he published ‘‘ Notes
on Mosquitoes of the United States, Giving some
Account of their Structure and Biology, with
Remarks and Remedies.’”’ In the latter publi-
cation he summarized everything which had been
done up to that time on the remedial question.
The first systematic work of mosquito destruc-
tion in Havana was started on March 27, I9ol.
Before this time a considerable number of adult
mosquitoes had been destroyed by fumigation
directed against fomites. Operations were at
first confined to the area lying between Monserrate
Street and Havana Bay (see map). This portion
of the city was known as the yellow fever district.
In this district were the commercial houses, and
the hotels and boarding-houses the immigrants

~o THE City OF ~

HAVANA

MAP OF

aa OE

“7 g
Co

AUGUST 1[8T1899.

Scale in yards.
300 400 500

ODN

);

Map of Havana: Showing the yellow fever district

The Campaign in Havana = 237

occupied while looking for employment. It is the
custom in Havana for clerks and employees to live
and sleep in the stores or establishments where
they are employed. This arrangement helped
to make a concentration of non-immunes, and
hence the yellow fever cases in the locality men-
tioned.

In the preliminary mosquito destruction cam-
paign one inspector and two oilers were employed
to visit all houses and yards, and to destroy all
mosquito-producing containers. The length of
time necessary for the thorough inspection of the
district above mentioned was soon determined, and
it was then divided into sections, of such sizes that
one inspector and two oilers could destroy or con-
trol any mosquito breeding places in every house
and yard in their section once a week. Later this
system was extended throughout the entire city,
which was divided into thirty districts. The first
inspection of the city showed a total of 26,000
water containers with mosquito larve. Most of
these were producing Aédes calopus, whith were
very numerous in the residences.

The method of procedure was as follows: A
list of all houses was kept at the main office. The
district inspectors turned in each day a report
of containers found, and an entry was made on

238 The Control of Mosquitoes ,

a card kept for each house, thus giving a history
of the Aédes calopus producing status of every
house in Havana.

The entire clerical work was accomplished by
two men. Abbreviations were used to save time
on reports: thus, ‘‘Real No. 3, 2B, 3T, 1Tna”’ re-
corded the fact that at No. 3 Real Street were
two barrels, three tanks, and one tnaja, in which
mosquito larve werefound. Similar abbreviations
were used for all classes of water containers.

The inspectors’ report forms were designed
to save time, and were arranged as follows:

District No. 5

Date——______
Street No. B T Tna (Aljibe) Latas Pozos Misc. Name of
House
Tenant
Jose
Real 1 Gomez
Juan
a: a ee Perez
Rosa
ar me I Eave Mora
trough
Ant Alez
Guards Rito
sg in Garden
Manuel
Moreno

(Inspector)

The Campaign in Havana = 239

The reports were read at a glance, and rapidly
transferred in the office to the house card show-
ing the names of owner and tenant, with dates
showing when mosquito larve were found, and
in what class of containers. These reports were
of great use, in showing progress made in Aédes
calopus control and preventing the continuance
of mosquito propagation, besides furnishing valid
evidence in courts adjudging fines.

A system of checking the daily inspection was
employed. Each day a trustworthy man made an
independent inspection of several houses in every
district, both before and following the visit of the
district inspector. It was thus known to the
office, beforehand, what the district inspector
should find and report as well as whether several
reports of the previous day’s work were reason-
ably correct. In the early stages of the campaign
this checking was very useful, and kept the in-
spectors up to the mark. Later, however, the
inspectors worked more thoroughly, and the
preinspections and reinspections seldom differed
from the district inspectors’ reports. In addition,
the inspectors were transferred from one district
to another at irregular intervals, and thus a check
was obtained, and the force kept very much alive.
The chief sanitary officer and his assistant made

240 The Control of Mosquitoes

frequent inspections, and kept in close touch with
the daily work. :

It will be readily seen that such a system made
the possible mosquito producing water containers
disappear speedily. The number of adult Aédes
calopus in houses quickly decreased, and to this is
attributed the rapid disappearance of yellow fever
from Havana.

When the work of Aédes calopus destruction was
commenced, it was not known how far they could
travel or migrate, nor was it then known that they
were exclusively container breeders. Culex with
banded legs were seen at a distance from dwellings,
and were reported as Aédes calopus. Little had
been written concerning the habits and life history
of mosquitoes up to that time, and it was vaguely
supposed that all mosquitoes were more or less
alike in their habits, and that all traveled long
distances. It was not known that Aédes calopus
do not breed in pools and puddles, or that the larvee
of this mosquito are never found far from houses,
unless conveyed by man. As a matter of fact, we
occasionally find a few larve in streams and
ditches, but it was not known at that time that
they had been washed out of containers.

As the larve destruction in houses and yards
progressed, fewer adults were observed in dwell-

The Campaign in Havana = 241

ings. The floors of all rooms were swept after
each fumigation, and definite evidence obtained
of the decrease in the number of Aédes calopus.
Having definitely established the possibility of
greatly reducing the number of these mosquitoes
in Havana, it was easy to foresee that by con-
tinued reduction very few of them need exist in
the city; and that it was within the bounds of
possibility to eradicate them.

It was found that the tenement house districts,
locally called Casas de vicino, were the most
difficult to control. In these houses several fami-
lies used in common a yard, or patio, in which
their water-holding containers were placed. It
frequently happened that the owner of a container
holding mosquito larve preferred not to claim
ownership through fear of arrest, and at times it
was very difficult to prove who was the real owner.

At one time many tenants hid the red earthen-
ware jars (finajas) containing their day’s water
supply. The coming of the inspector was signalled
along the street and the water containers disap-
peared as though by magic. It even became neces-
sary to use a list of all water containers in some
districts to avoid missing any having larve, and
to save time during inspection.

The experimental work on the prophylactic
16

242 The Control of Mosquitoes

inoculation of yellow fever proved unsatisfactory,
but fortunately at that time the destruction of
Aédes calopus had progressed far enough to show
what could be accomplished in a few months of
effort and persistence.

The discovery that yellow fever propagation
could be arrested by destroying, by fumigation,
the Aédes calopus in the dwelling where the case
occurred, and the houses immediately adjacent,
showed that they do not fly long distances. This
also indicated that even if these mosquitoes bred
in the country, they did not find their way into the
city in dangerous numbers. This discovery was
very useful, as well as encouraging.

The scheme of the work being carried out was
strongly supported by General Leonard Wood,
military governor of Cuba, who had full confidence
in the ability of the chief sanitary officer. The
people of Havana, being mostly immune, took
small interest in yellow fever eradication, but
submitted with little objection to the enforcement
of the sanitary regulations. They thought that
the Americans did not know how to control the
situation, and had seized upon something new to
occupy their time and make a showing. There
was more opposition, in proportion, from foreigners
than from natives. It was most fortunate that

The Campaign in Havana = 243

so little opposition interfered with the campaign,
for it must be remembered that this was all pio-
neer work. There was much ridicule of our
methods at that time, but fortunately this spirit
of raillery and its pernicious influence in encour-
aging active and passive opposition to sanitary
methods is now decreasing. Sanitarians need
the interest and coéperation of the general public.

In July, 1901, while the sanitary department
was busy with yellow fever in Havana, six cases
were reported from Santiago de las Vegas late
one afternoon. Before daylight on the following
morning a complete sanitary brigade was on its
way to that town. In six weeks yellow fever was
eradicated, and it became exceedingly difficult to
find Aédes calopus there.

At the instigation of Colonel Gorgas, a law was
enacted, making it compulsory for physicians to
report every case of yellow fever; and also all cases
of fever, of whatever nature, occurring among
foreigners. In locating yellow fever cases, the
sanitary department also received assistance from
the hospitals in Havana connected with the
societies known as Centros.

The larger part of the foreign population is
Spanish, mostly unmarried men. The majority
join one of these Centros, each of which has a

244 The Control of Mosquitoes

well-equipped hospital, where the members receive
medical and hospital treatment without extra
cost above the annual dues. Because of this
custom, the hospitals in Havana receive most of
the sick Spaniards, and the detection of yellow
fever cases was greatly facilitated. All American

’

“‘yellow fever suspects,’’ as well as yellow fever
patients, were sent to Las Animas Hospital, then
under the direction of the chief sanitary officer,
and as a rule such cases were reported in the
earlier stages. All cases were advised to take
hospital treatment, and most of them did so.
When a patient preferred to remain at his resi-
dence, it was allowed; his room was screened, and
the screen door was kept locked and continuously
guarded. Only persons authorized by the chief
sanitary officer could enter. Non-immunes were
not allowed in the patient’s rooms under any
circumstances. As soon as a “yellow fever sus-
pect’”’ was reported, he was conveyed to the hospi-
tal, or put in a screened room. Next, the history
of his recent visits was traced, recorded, and
verified as thoroughly as possible, in order to
determine the source of his infection. Then
his house and the adjacent houses were fumi-
gated. When the patient preferred treatment at
home, his room was not fumigated until he re-

EVER

CURVA DE MUERTES POR FIEBRE AMARILLA.

CURVE LINE OF DEATHS BY YELLOW F

JUL

MAY

SOT | 981

APR

ee O ALIO SHL Ni

9bF | Loe

Il
HG
OF
R90
oor
RIL
69
itd

9

au

6

SI
cau
88
oo
19
88
86
8—

OAON

9O] 40q0990,
doquagdag

8383”
“SLHNOW

‘SUVA

| ‘W061 = VNVEVH V7 3d GVdNIO VIN VTITINVWY JYaald YOd SALUINW = 068}

Chart showing prevalence of yellow fever in Havana, Cuba

Ll,

La a uae
SEIN ST |

is |

The Campaign in Havana =. 245

covered, but all the other rooms were fumigated
immediately, together with all places visited by
him within a week before the onset of the disease
and during three days following. Wherever he
might possibly have been bitten, fumigation was
ordered to destroy the mosquitoes which had
infected him, or might have become infected by
biting him. It was found important to investi-
gate his history very thoroughly, and to verify
and check all obtainable information. Some-
times the history proved inexact, and certain foci
of the infected mosquitoes were left undestroyed.
It may also be true that walking cases of yellow
fever occurred, which would account for so-called
sporadic outbreaks of yellow fever. As the cam-
paign progressed, and the mosquitoes were slowly
but surely reduced in number, these instances
became of less importance.

The storehouse for all materials and apparatus
necessary for fumigation operations on a large
scale was established near the stables where the
wagons and ambulances of the sanitary depart-
ment were kept. The fumigation brigades, each
consisting of a foreman and ten men, reported every
morning for duty at the storehouse. In addition,
each foreman knew where to find his men at any
hour in case of emergencies, either during or after

246 The Control of Mosquitoes

working hours. Material was kept in such readi-
ness that required quantities could be measured
out and issued rapidly, and the brigades relied
on for quick service at short notice. Time was
also saved by transporting men and supplies in
ambulances.

When a suspected case of yellow fever was
reported, the yellow fever commission at once
saw the case, and if fumigation was decided upon,
the tenants were immediately notified of the
proposed fumigation, verbally, and by written
notice posted on the house, stating the time work
would be started, and the probable time of com-
pletion. At the date and time arranged for, the
fumigation brigade commenced operations. First,
openings such as doors and windows were closed,
to retain all mosquitoes within the building. As
little disturbance as possible was made, in order
that they should remain at rest in the rooms.

The doorway through which supplies had to
pass, was kept closed as much of the time as cir-
cumstances would permit. One or two laborers
were assigned to each room, to paste with paper
strips all cracks, and other communications with
the outside air, to make the room air-tight. Large
openings were subdivided by wooden strips so
arranged that single sheets of manilla paper would

The Campaign in Havana = 247

reach from one piece to the other. On windy days
a center support for each large sheet of paper was
used. The paste was made of flour and water.
Paper strips of different widths were kept in stock
to expedite the work.

The cubic contents of each room was ascertained
to determine the number of fumigation pans
required.

Just before a room was finally closed, the pans
containing the sulphur or pyrethrum were placed
in position on the floor. Each pan was placed
inside a larger shallow iron vessel containing
water as a precaution against fire, in the event of
the iron pan containing sulphur or pyrethrum
breaking during the progress of the fumigation.
Small boxes holding exactly one and two pounds
of sulphur or pyrethrum powder were used for
measuring, and it was the duty of thesanitary
inspector in charge of the fumigation to supervise
the placing of the fumigants in the pans, and the
correct number of pans in each room. When as
much of the sealing was accomplished as could be
performed without closing the door, or other exit,
the inspector arranged the fumigation pans in the
most advantageous way, sprinkled the surface
of the sulphur or pyrethrum powder with a small
amount of wood alcohol, and then ignited the

248 The Control of Mosquitoes

fumigating material. After igniting the contents
of the pans, the inspector remained in the room
uutil he was sure that the fumigating powder was
burning properly, and that the fires were not likely
to go out before the fumigants were entirely con-
sumed. A man was ready to paste up the exit
the moment the inspector left the room. Room
after room was treated in this manner, and hall-
ways were treated last. The laborers engaged
in pasting work remained at the doors long enough
to detect escaping fumes, and repair defects in
the sealing.

The amount of pasting varied with the nature
of the building treated. In some cases the entire
wall of a room had to be covered with paper. All
ceilings or roofs, walls, and floors were carefully
examined, and every opening closed. Each room
was dealt with as a separate unit, to prevent the
formation of air currents and the traveling of fumes
to a possible avenue of escape. During the entire
fumigation period in Havana absolutely nothing
was burned by the fumigation operations, although
wooden houses might easily have caught fire.
Every precaution was taken against accident.
It was always necessary to make sure that no
cats or dogs were hiding in out-of-the-way places.

In well-built stone houses we used one pound of a

The Campaign in Havana 249

superior grade of pyrethrum powder per thousand
cubic feet of air space. Sulphur, in the form of
“flowers of sulphur,”
portion. Where a probable leakage of fumes was
seen to be more or less unavoidable, the quantity of

was used in the same pro-

fumigants was increased proportionally, and the
inspectors became quite expert in gauging the
amount required; as a matter of fact, they usually
erred on the side of safety.

In loosely built houses, the wind velocity had to
be considered. Sometimes it was not advisable to
paper the entire wall area of a house to prevent the
leakage of fumes. Time was of value, and too
much moving around of laborers and the manipu-
lations incident to the papering might cause the
mosquitoes to fly away, so that it became neces-
sary to avoid any commotion that would result in
sending out of doors the very infected mosquito the
fumigation was intended to destroy.

After all the sulphur pans were burning, and the
last door of the house was locked and sealed, a
notice was posted upon it stating the hour of
closing and that of reopening by the sanitary
department, and also that any person, other than
the inspector in charge, who opened the door
would be guilty of a misdemeanor and treated
accordingly. A laborer was detailed to watch

250 The Control of Mosquitoes

the house, and attend to any additional pasting
that became necessary. When the fumigation
period had elapsed (or after three hours), as many
doors and windows as possible on the lower floor
of the leeward side were first opened to let out the
fumes. Those on the windward side were opened
next. The inspector entered the building as soon
as the fumes would permit, and inspected the
pans, to see that none had been prematurely
extinguished, and ascertain that the fumigation
was entirely satisfactory. A repetition of fumi-
gation was seldom necessary. Floors were then
swept and mosquitoes carefully collected. These,
although nearly always dead, were at once burned,
to obviate danger from their revival.

We had established by experimental fumigation
that apparently dead mosquitoes revived after
several hours’ exposure to the air.

Immediately on the return of tenants to their
rooms, they were requested to examine their
effects, and verify their good condition.

All traces of paper and paste were removed after
each fumigation, by hot water and scrubbing
brushes where necessary. Before the fumigators
left a house, the inspector obtained a signed state-
ment to the effect that no articles belonging to the
tenants were missing. Great care was taken to

The Campaign in Havana = 251

select honest laborers, and practically no thefts
occurred. As soon as the brigade completed
one fumigation, the next was started. Occasion-
ally the removal of paper, cleaning up, etc., was
delayed for a day or two and the men were em-
ployed at this work while other houses were closed
during fumigation.

In Havana houses were being fumigated before
February, 1901, and where pyrethrum was used
at that time it was bought from local merchants.
The results varied considerably. Sometimes the
mosquitoes were killed, and at other times they
were not. The quality of the pyrethrum varied
greatly. Samples were obtaind from the dealers,
and tests made to determine the quality on which
the quantity needed for a given volume of space
depends.

A room at Las Animas Hospital was used for the
fumigation tests. The cubic content was about
1500 cubic feet. A glass transom over the door
served as an observation window to note the effect
of fumes on mosquitoes subjected to fumigation.
It so happened that house flies were in the room
during the first series of tests, and after the killing
of Aédes calopus it was noted that the flies still
lived. The amount of pyrethrum per 1000 cubic
feet was increased until it was sufficient to kill the

252. The Control of Mosquitoes

flies. When all flies were killed, no mosquitoes re-
mained alive. An hour longer was required to kill
flies. Cages containing both were hung near the
windows, close to the glass, that they might be ob-
served through the fumigation period, and it was
realized that the pyrethrum on the local market
varied so much in quality as to be useless for prac-
tical fumigation, and samples for experimental
purposes were obtained from the United States.

On receiving a satisfactory quality of powder,
larger quantities were purchased, with the privilege
of rejection if there was variation in quality as to
generation of fumes and culicidal powers. The
contractors kept the quality up to standard, and
good results on a large scale were obtained.

The experiments determined the fact that in a
room with air-tight walls, floor, and ceiling, one
pound of this particular grade of pyrethrum pow-
der per thousand cubic feet and a three-hour
exposure to the fumes was sufficient to kill all mos-
quitoes in the room. Subsequent experiments
proved that with any possible chance of leakage
of the fumes, the quantity of pyrethrum used must
be increased to allow for the leakage. Where the
room was not reasonably air-tight and the leak-
age was large, pyrethrum could not be relied upon,
unless all walls, ceilings, roofs, etc., were covered

The Campaign in Havana = 253

with paper to prevent the escape of the fumes.
Where the leakage was extensive it was not
effective.

In the experimental work it was observed that
there were most dead mosquitoes per unit of
area near the source of light. It was also noticed
that before they were overcome by the fumes of
pyrethrum many of them pulled off one or more
of their legs, but that this did not occur when
fumigating with sulphur, tobacco, or formalde-
hyde. Pyrethrum powder deteriorates rapidly
when stored in a humid tropical atmosphere.

The proprietors of tobacco warehouses and
shops objected to pyrethrum fumigation of their
premises because of the odor, and experiments
were made to determine the value of tobacco
stems as a culicide for Aédes calopus. It was
found that the results from its use were equal to
sulphur, and the tobacco merchants reported no
harmful effect or deterioration of value in the
stored tobacco and tobacco products.

It-was found that sulphur fumes injured all
ordinary metals, unless they were protected by
vaseline or some other greasy substance; and that
it bleached and stained many fabrics, and stained
certain painted surfaces. Pyrethrum only stained
some materials slightly, but left an odor.

254 The Control of Mosquitoes

The results of experimental tests with the three
substances, as obtained at Las Animas, were as
follows:

Pyrethrum, burned at the rate of one pound per
thousand cubic feet in layers not greater than one
inch in thickness, destroyed all mosquitoes in a
well sealed air-tight room within three hours.

Tobacco, at the rate of two pounds of stems
(waste from Havana factories) per thousand cubic
feet, if it all burned, killed them in three hours.

Sulphur, at the rate of one pound per thousand
cubic feet, was effective with an exposure of three
hours or less.

It is almost impossible to find a satisfactory
grade of pyrethrum powder for fumigation pur-
poses on the market at the present time.

In 1906 Kendall experimented with campho-
phenique as a fumigant; his methods and con-
clusions are quoted:

This substance first used in the Gulf States, has
been tried with great success here. Berry, experi-
menting with this mixture, found that a shallow pan
placed upon a short piece of stove pipe formed a good
apparatus for volatilizing the campho-phenique, and
his method was tried here. The results were dis-
appointing, and except for a fortunate trial of the
fumes of this substance, they would have been dis-
continued. It was observed that the fumes killed

The Campaign in Havana =. 255

mosquitoes with great rapidity, and that the insects
were actually dead, not merely stupified, as is the
case with pyrethrum. With this observation as a
basis, a piece of apparatus, which will be described
in great detail later on, was evolved, which has given
excellent results in our hands.

The apparatus was constructed with the idea of
removing every possible source of danger which would
occur with the mixture, as well as to provide for the
fact that unskilled labor would have to be employed
in practical fumigation. Cheapness is the first consid-
eration, because it is frequently necessary to use
large numbers of pots in fumigating large buildings.
The use of galvanized iron, which is comparatively
inexpensive, not only reduces this factor, but makes
the apparatus light and not easily broken.

To provide for the notorious lack of skill shown by
native workmen, steps have been taken to make it
absolutely necessary for the apparatus to be set up
correctly. It is practically impossible to make it work
in any other way than the correct one. Each and
every part must occupy its correct relation to every
other part, and of necessity this must be carried out in
order to evolve fumes.

In order to prevent any possible accident due to
the catching fire of the camphor, a special fumigating
pan has been constructed—this pan is circular in
shape, with the edges turned downward and outward,
making a total diameter of seven inches. In section
it is concave, so that in reality it is a shallow bowl
having a diameter greater than that of the stand
upon which it is supported, with the edges turned
down at an angle of 45° with the perpendicular.

This shaped pan is especially adapted to the evapo-

256 The Control of Mosquitoes

ration of any liquid; the concavity causes the last
remaining portions of liquid to remain in the center
of the apparatus, where it is exposed to the hottest
part of the flame, and the convexity presented tothe
flame permits the maximum effect of the latter for
evaporation. Inasmuch as it is made of galvanized
iron, hammered out of one piece, there are no joints
to leak, or solder to melt, and the form is eminently
suited for stacking a large number, one inside the
other, for transportation.

The stand of galvanized iron, riveted into a cylindri-
cal shape, 6 inches in diameter, and 514 inches high.
Near the top is a row of holes through which the hot
air from the lamp escapes, forming a natural draft.
This hot air acts upon the campho-phenique that
may be near the edge of the pan, away from the direct
action of the flame, causing rapid evaporation at
this point as well as in the center of the pan, which
sets into the top of the stand. The stand is supported
on three iron legs two inches long in the clear, which
is turned outward at right angles at the bottom one-
half inch. This is done so that they may fit into
appropriate straps in the bottom of the safety pan, in
which the stand sets, to assure the correct placing
of the latter.

The safety pan is also made of galvanized iron
eleven inches in diameter. In the bottom are three
straps into which the “turned out” part of the legs
of the stand fit. In the center of the safety pan there
is a ring one-half inch high, three inches in diameter,
into which the alcohol lamp fits. With the stand
confined in its proper place by straps, the lamp like-
wise placed in its ring, it will be seen that there is
no possibility of any portion of the apparatus occupy-

The Campaign in Havana =. 257

ing an incorrect position. The safety pan not only
holds the stand in position, but when half-filled
with water forms an effective protection in case the
alcohol lamp should begin to leak. Any lateral spurt-
ing of the alcohol is prevented by the sides of the
stand, and the excess alcohol would immediately be
extinguished by the water.

It should be mentioned in passing that the edge
which is turned down on the fumigating pan is to
prevent the flame of the alcohol lamp passing over
the edge of the pan, and thus setting fire to the
camphor and carbolic acid.

To summarize: ’

We have an apparatus consisting of four parts, a
fumigating pan, designed to evaporate campho-
phenique, formaldehyde, or other similar liquid, with
an especially moulded edge designed to prevent con-
tact of the liquid with the free flame, and to concen-
trate the fluid as it is evaporated in such a manner that
the last portion shall be distilled off.

The stand is designed for steadiness, protection of
flame from drafts, and to give rigidity to the evapo-
rating (fumigating) pan.

The safety pan is intended to support in their
proper relations both the stand with its pan and the
lamp which supplies the heat for evaporation, as well
as a safeguard against leakage of the lamp.

The lamp found to give the best flame, everything
considered, is No. 15 metal spirit lamp manufactured
by Whitall, Tatum & Co.

The whole apparatus is made as cheaply, durably,
and easy of transportation as possible. It is so con-
structed as to permit its use by unskilled labor, with
a minimum chance for damage both to itself and

258 The Control of Mosquitoes

to the house or other structure in which it may be
used.

The fumigating pan will use 8 ounces of campho-
phenique when made in the dimensions given, namely,
6% inches in diameter exclusive of the overhanging
edge, and 134 inches in depth in the center. With
the alcohol lamp one-third full, ample time is given
for complete evaporation of the liquid, with about five
minutes to spare to allow for conditions which slow
down the rate of evaporation. Ina fairly tight house,
with corrugated metal roof, tightly papered, campho-
phenique has given practically perfect results. Four
ounces of the camphor-carbolic acid mixture and sul-
phur in amounts varying from one to three pounds per
1000 cubic feet have been used for comparative tests.
In several instances considerable leakage, due to the
wind passing over the metal roof carrying off fumes,
has occurred, but even when small leaks of known
size have been made purposely no change in the results
have occurred. We feel justified in saying that cam-
pho-phenique in moderately well-built tight houses
will give as good results as sulphur, and considering the
extra time needed for sulphur fumigation, at an equal,
or more probably, a lesser cost.

In order to test the action of pyrethrum, sulphur,
and campho-phenique upon fruit, a series of two ex-
periments have been made under the same conditions.
Bananas, quite ripe, mangoes, and oranges have been
used in these experiments. The mosquitoes were not
only placed in the customary positions in houses, but
were also carefully hidden as much as possible in
bunches of bananas to simulate conditions which would
occur on ship board. The results showed that cam-
pho-phenique is less harmful than sulphur or pyre-

<A TT

The Campaign in Havana = 259

thrum to the fruit. There was very little darkening of
the skins of the bananas or other fruit when exposed to
this substance, and the bananas did not taste bad.
Several observers made these observations simultane-
ously, tasting fruit before and after fumigation, and
also carefully preserved controls. The indications are
that campho-phenique is less harmful than sulphur.
With the campho-phenique these insects were killed
in every case; in the first trial, where the time of
exposure was only three-quarters of an hour, one
mosquito which was situated near the roof, where the
fumes were practically removed by the action of the
wind, survived for nearly half an hour, although
it could not walk or fly. It was the only survivor
of about eighty Aédes calopus placed in the building
at that time. The mosquitoes hidden in the fruit
were all killed. It is interesting to note that the
first experiment carried out in a strong breeze, also
restilted in a large number of survivals of mosquitoes
in the house fumigated by sulphur, and that they
too, were near the roof, where there was loss of fumes
due to wind action.

Sulphur is decidedly objectionable in many ways;
it kills mosquitoes, however, and may be looked upon
as a nearly perfect fumigant for these insects. Pyre-
thrum, if it were more efficient, would do very well,
although it leaves a decided coloration on light colored
paint and is objectionable from the point of view of
time consumed in fumigation, three or four hours
being necessary, asa rule, for its combustion. Cam-
pho-phenique in well-furnished houses (which in the
vast majority of cases are also well-constructed) is
very much less objectionable than either of the above,
and it combines this factor with the comparatively

260 The Control of Mosquitoes

short time necessary for its evaporation, and the fact
that it leaves a very slight residual odor, which
quickly disappears, so that the houses are comfortably
habitable as soon as the cleaning-up process is com-
plete. In such cases it has given quite as good
results as sulphur, and, considering everything, is
much the cheapest fumigant of the three.

CHAPTER II

THE SITUATION ON THE ISTHMUS BEFORE SANITARY
WORK WAS STARTED

N his Report on the Isthmian Canal, 1904,
Colonel Gorgas stated:

In my opinion the sanitary problems are grave.
The question of eliminating yellow fever from an
endemic focus has only been once before successfully
managed, and that was at Havana. And from this
successful work, to argue that it is easy and simple, I
think is not warranted. The malarial work on the
scale at Panama is entirely new. It has never been
attempted elsewhere, and any health officer who has
any experience in enforcing measures of individual
prophylaxis such as will be required at Panama, can
understand how great the difficulties of administration
will be. Personally, I believe it can be done, and I
approach the work with great hope of success, but I
know that it will be neither easy nor simple, that we
will meet with many disappointments, and have to
modify our plans many times.

YELLOW FEVER

From the knowledge gained in the anti-yellow

fever campaign in Havana, Colonel Gorgas rightly
261

262. The Control of Mosquitoes

concluded that, in so far as yellow fever was con-
cerned, the cities of Panama and Colon were the
danger points for the Panama Canal. Both of
these cities; Panama, with a population in 1904
of about 25,000, and Colon, of about 4000, are
seaports situated within a few days’ travel from
ports where yellow fever is endemic.

Panama and Colon were infested with Aédes calo-
pus, the conditions for the propagation of this
species of mosquito being exceedingly favorable,
as the climatic conditions permit mosquito breed-
ing throughout the year. The people of Panama
and Colon depended principally on rain water for
their water supply. This was peddled by water-
men, and kept in numerous vessels by house-
holders.

Wells, cisterns, barrels, and many other varie-
ties of water containers, were not protected in any
way from the female Aédes calopus seeking a
favorable place for ovipositing. In addition
to the various wells and water vessels, a large
number of the houses had eave-troughs, to catch
the rain water, and conduct it to various storage
devices. These troughs, or gutters, many of
them sagging, and holding water, were also favor-
ite ovipositories of this mosquito.

There was no systematized garbage collection.

Conditions before Sanitation 263

Back yards were littered with refuse of all sorts,
as were vacant lots, alleys, and even the streets.
This rubbish included many articles capable of
holding water, and serving as breeding-places
for mosquitoes.

The difficulty of obtaining an ample and cheap
supply of water resulted in extreme economy of
water consumption by the people of Panama,
especially among the poor. This economy found
expression in the widespread custom of leaving
dregs in the water vessels, and refilling them when
the water ran low. To appreciate the full signi-
ficance of this custom, it must be recalled that
larve of Aédes calopus remain for the greater
portion of the time near the bottom of a vessel;
they descend, also, with great rapidity, when
in any way disturbed. A five-gallon bucket
may be nearly emptied by tipping, and if
the water is not poured out very rapidly, most
of the Aédes calopus larve will remain in the
few teaspoonfuls of water left in the bottom of
the bucket.

Ten years ago decorative shrubs in the grounds
of Ancon Hospital, and in private grounds in Pan-
ama were protected from the ravages of the leaf-
cutting ants, by surrounding each shrub with
shallow ring-shaped dishes filled with water.

262 ‘The Control of Mosquitoes

concluded that, in so far as yellow fever was con-
cerned, the cities of Panama and Colon were the
danger points for the Panama Canal. Both of
these cities; Panama, with a population in 1904
of about 25,000, and Colon, of about 4000, are
seaports situated within a few days’ travel from
ports where yellow fever is endemic.

Panama and Colon were infested with Aédes calo-
pus, the conditions for the propagation of this
species of mosquito being exceedingly favorable,
as the climatic conditions permit mosquito breed-
ing throughout the year. The people of Panama
and Colon depended principally on rain water for
their water supply. This was peddled by water-
men, and kept in numerous vessels by house-
holders.

Wells, cisterns, barrels, and many other varie-
ties of water containers, were not protected in any
way from the female Aédes calopus seeking a
favorable place for ovipositing. In addition
to the various wells and water vessels, a large
number of the houses had eave-troughs, to catch
the rain water, and conduct it to various storage
devices. These troughs, or gutters, many of
them sagging, and holding water, were also favor-
ite ovipositories of this mosquito.

There was no systematized garbage collection.

Conditions before Sanitation 263

Back yards were littered with refuse of all sorts,
as were vacant lots, alleys, and even the streets.
This rubbish included many articles capable of
holding water, and serving as breeding-places
for mosquitoes.

The difficulty of obtaining an ample and cheap
supply of water resulted in extreme economy of
water consumption by the people of Panama,
especially among the poor. This economy found
expression in the widespread custom of leaving
dregs in the water vessels, and refilling them when
the water ran low. To appreciate the full signi-
ficance of this custom, it must be recalled that
larvee of Aédes calopus remain for the greater
portion of the time near the bottom of a vessel;
they descend, also, with great rapidity, when
in any way disturbed. A five-gallon bucket
may be nearly emptied by tipping, and if
the water is not poured out very rapidly, most
of the Aédes calopus larve will remain in the
few teaspoonfuls of water left in the bottom of
the bucket.

Ten years ago decorative shrubs in the grounds
of Ancon Hospital, and in private grounds in Pan-
ama were protected from the ravages of the leaf-
cutting ants, by surrounding each shrub with
shallow ring-shaped dishes filled with water.

264 The Control of Mosquitoes

Each of these dishes was a mosquito incubator
of a most efficient type.

The result of all these factors favoring mos-
quito propagation was, that the two cities har-
bored vast numbers of these insects. The
Aédes calopus was the commonest mosquito in
Panama, and the mosquito bars used by some
of the natives were intended particularly as a pro-
tection against their annoyance.

Turning to the records of the four years im-
mediately preceding the inauguration of sanitary
work on the Isthmus, the available records of
deaths from yellow fever in the city of Panama
show:

109 deaths in 1900

5 “oc ce 1901
202 “ ‘ 1902
sr “ “ 1903

These figures, however, indicate the yellow
fever mortality in a population including few non-
immunes. The accuracy of these statistics may
also be challenged, as there was practically no
sanitary organization in this province of Col-
ombia, nor was the reporting of yellow fever
compulsory.

With myriads of Aédes calopus on the wing, with

68sl OL 188! WOYUS SNWHISI AHL NO SASVO YSA34 MOTI1SA ONIMOHS ‘LYVHO 'E
orv
‘OO IVNVO VWYNVd HON3Ys AO SAQHOOSY WOU

Y3SAa4 MOTNISA GNV NOILVINdOd ANNWIWI-NON

68sl sesl Z881 988l Sssl vss €s8sl Z8sl IS8l YVSA

T
Bk NOILVINSOd = 0921

0061 0003~ IL S3SvO
ee

Zor x,
Y
&
€Zl
Oost — S008!
oosst
00S9l oo00gt
08!
QZ

806

Conditions before Sanitation 265

countless breeding places in and about the houses,
with climatic conditions particularly favorable
to rapid and continuous development of insect
fauna, and yellow fever foci at the very threshold,
what would be the expectation for rapid spread of
yellow fever were a large non-immune population
introduced into Panama, Colon, and their vicini-
ties? A glance at the curves on the chart will
furnish a basis for an estimate.

These curves were plotted from the figures
published by Colonel Gorgas in a pamphlet issued
in 1906, entitled: ‘Population and Deaths from
Various Diseases in the City of Panama, by Months
and Years, from November, 1883, to August, 1906..
Number of Employees and Deaths from Various
Diseases among Employees of the French Canal
Companies, by Months and Years, from January,
1881, to April, 1904.” The number of employees
given is nearly correct, but the number of deaths
from yellow fever is probably much below the cor-
rect number. The only mortality statistics availa-
ble are those of Ancon Hospital, and many cases of
yellow fever never found their way into thehospital,
either dying before they could be sent there, or re-
maining outside because the contractors were more
or less reluctant to pay for the treatment of their
employees. After carefully investigating various

266 The Control of Mosquitoes

sources of information, the authors believe that
the number of cases given in this table approaches
the true mortality more nearly if multiplied by
two. Basing the estimate on a case mortality rate
of 33% with an average population of 15,000
in 1886, the number of deaths given in the statis-
tics is corrected to the more probable number 616
(308 X2); the cases developed during that year
must have exceeded 1500 or ten per cent. of the
entire working force.

The year 1886 was the fifth of the French occu-
pancy, the year during which the work was at its
height, and when the number of non-immunes
was comparatively high. This year can be taken
as a fair basis for an estimate of the mortality from
yellow fever that would probably have had to be
faced by the American forces on the Canal. The
average force employed by the United States
on the Canal was 35,000 to 40,000 mostly non-
immune to yellow fever. With a mortality of
four per cent. per annum, as was the case with
the French force in 1886, we should have lost
more than 1400 men each year from yellow
fever alone!

The curves on the chart show a very interesting
phenomenon, in the rapid rise of yellow fever with
the influx of non-immunes; this rise taking place

Conditions before Sanitation 267

at a swiftly accelerating rate; then the decline as
the non-immunes either die or recover from
the disease; and the subsidence of the epidemic
as the non-immune material is exhausted.

CHAPTER III

GEOGRAPHY, METEOROLOGY, ETC., AND THEIR
BEARING ON THE PRESENCE OF AEDES CALOPUS

HE Canal Zone is a narrow strip of land ten
miles wide and approximately forty-eight
miles long, situated entirely within the torrid zone.
Its northern boundary is in latitude 9° 24’ 4o” N.
(vicinity of Punta Mala Remo). Its southern-
most latitude is 8° 54’ 40” N. (Punta Bruja).
The seasons are well marked: a rainy season of
nearly eight months’ duration and a dry season
of about four months, January to April.
The mean annual temperature is approximately
80° F.
The extreme range of temperature. is between
68° and 97° F.
The climate and temperature are favorable to
mosquito propagation the year round.
In 1904 when the Canal Zone was ceded by the
Republic of Panama to the United States, it con-

‘tained a number of villages and small settlements;
268

Geography and Meteorology 269

two of these were at the ocean terminals of the
Canal, Cristobal on the Atlantic, and La Boca, re-
cently renamed Balboa, on the Pacific. Inaddition
to these were the cities of Colon and Panama, which
although not in the Canal Zone politically, are yet
geographically so, and were placed within the san-
itary jurisdiction of the United States by treaty.

Colon and Cristobal are situated on Manzanillo
Island, a flat marshy piece of land the elevation of
which is approximately four feet above mean tide.
This island is largely a coral formation covered
with a layer of soil. There are no rivers or springs
on the island, and no wells. Formerly the only
source of water supply for Colon was rain water
and fortunately the precipitation in this locality
is very great, averaging 120 inches annually.
Some water was brought into Colon by rail and
stored for use. The towns in the Canal Zone de-
pended for their supply on adjacent rivers, streams,
and springs, which are plentiful inthe Zone. Pan-
ama, Ancon, and Balboa (La Boca) are situated
at the southern extremity of the Canal. The
rainfall there is far less than on the Atlantic slope,
averaging about seventy inches per annum. The
stored rain water was therefore supplemented by
wells and aqueducts carrying surface and subsoil
water, with tanks for storage.

270 The Control of Mosquitoes

The owners of the aqueducts and wells sold the
water to carriers, who in turn sold it to the con-
sumer. The price of water sold from the carts was
usually five cents for a five-gallon can and some-
times doubled toward the end of the dry season,

It will be seen from the foregoing that climatic
and topographical conditions are very favorable
to Aédes calopus in the Canal Zone, and espe-
cially in the cities of Panama and Colon. In the
adjacent settlements of Balboa, Ancon and Cristo-
bal, where scarcity of water compelled each house-
hold to husband to the utmost any supply
obtainable, the result was that these cities and
their environs harbored thousands of ideal mos-
quito breeding places.

CHAPTER IV

THE FIRST SANITARY WORK DONE AT PANAMA—ITS
PURPOSE

N July of 1904, when Colonel Gorgas and his
assistants undertook the sanitation of the
Isthmus, it was not definitely known that yellow
fever was present. The authorities were not aware
of any cases. At least none had been reported in
the months of February, March, or April, 1904,
and there was reasonable doubt regarding the
diagnosis of the two cases reported in May and
June. The number of non-immunes was relatively
small in proportion to the native population. It
was understood, of course, that as soon as the
percentage of non-immunes to total population
increased, if yellow fever were present, a number
of cases would develop, unless the chief sanitary
officer had the necessary organization, police
power, supplies, appropriations, and power to
enforce the sanitary regulations; in brief, that

unless the proper means were used to combat
271

272. The Control of Mosquitoes

the disease in its incipience, conditions would be
favorable for an epidemic as soon as enough
American non-immunes arrived.

Colonel Gorgas earnestly endeavored to have
a properly qualified representative of the sanitary
department on duty in the United States, so that
all supplies necessary for the control of yellow
fever and operation of the hospitals might be
properly selected, expeditiously purchased, in-
spected, and rushed to the Isthmus without delay.
Delay, he foresaw, might result in many deaths
among Americans, and cause a longer and more
expensive yellow fever campaign. His recom-
mendations regarding this matter were, unfortu-
nately, not approved.

The treaty between the United States and the
Republic of Panama contained the following
clauses:

The Republic of Panama agrees that the cities of
Panama and Colon shall comply in perpetuity with
the sanitary ordinances, whether of a preventive or
curative character, prescribed by the United States;
and in case the government of Panama is unable or
fails in its duty to enforce this compliance by the
cities of Panama and Colon with the sanitary ordi-
nances of the United States, the Republic of Panama
grants to the United States the right and authority to
enforce the same.

First Sanitary Work 273

In accordance with the terms of the treaty, the
Republic of Panama agreed that the sanitation of
the cities of Panama and Colon should be directed
by the chief sanitary officer of the Isthmian Canal
Commission. At the request of the latter, the
Mayors (Alcaldes) of the cities of Panama and
Colon issued the following ordinance:

~ Ordinance Number 6. Under the authority granted
by decree No. 25, of July 8, 1904, issued by the Presi-
dent of the Republic of Panama, the following sani-
tary regulation is issued:

Breeding of mosquito larve (wigglers) is prohib-
ited within the limits of the city of Panama and the
occupants of premises will be held accountable for
violation of this regulation.

All cisterns, water barrels, and deposits of fresh
water must be made mosquito-proof, and all occupants
of premises must see that other deposits of water are so
arranged that mosquitoes cannot breed in them.

Violation of this ordinance will make the offenders
liable to a fine of five dollars gold.

(Signed) W. C. Goreas,
Chief Sanitary Officer.

Previous to sailing for the Isthmus, the chief
sanitary officer called the attention of the Isthmian
Canal Commission to the advisability of immedi-
ately obtaining experienced assistants and sani-
tary supplies. He submitted an itemized list

showing what was required. Although the
18

274 The Control of Mosquitoes

estimated cost was not large, the amount allowed
him was below the sum asked for.

The first house to house inspection of the city of
Panama showed that Aédes calopus larve existed
at practically every house in town. Adults of the
species were plentiful everywhere, and everyone
who could afford it used mosquito bars. Foreign-
ers found it impossible to sleep comfortably
without a mosquito bar, and one could not sit
down even in the daytime without being annoyed.

The ice supply was limited and ice was expensive,
five dollars per hundred pounds, when obtainable.
The natives cooled water by keeping a supply
indoors in semi-porous earthenware vessels, called
tinajas. In these larve thrived in great numbers
inside the houses, as well as in the barrels, etc., in
the patios.

After the first inspection of the town was com-
pleted, and proper notices published in the press
concerning the proposed campaign against mos-
quitoes, a second inspection was made, and at
all places where mosquito larve were found, the
owners of the water containers were requested to
assist the sanitary officers by complying with
Ordinance Number 6. The number of containers
with mosquito larvee was reduced. A written notice
was next sent out, inviting the attention of those

First Sanitary Work 275

who were violating the ordinance to the fact that
they would be liable to prosecution, in accord-
ance with the ordinance; and in every way
possible, a strong effort was made to obtain the
people’s support.

Subsequent inspection showing many delin-
quents, their names and addresses were forwarded
to the mayor of the city, who had authority to
impose fines. Practically no fines were imposed,
and the people of the lower class, finding the ordi-
nance not insisted on by the mayor, gave it little
attention. Each week a list of citizens violating
the ordinance during that week was duly for-
warded to the mayor. It is not known if any na-
tives were fined although rumor had it that some
foreigners were.

The number of American employees and foreign
population increased as time passed. Every
means was used to induce the Mayor to enforce
the law, and render the work effective, but without
result.

The first case of yellow fever occurred at the San
Tomas Hospital on November 21, 1904. The
patient was an Italian, and came from a restau-
rant at Santa Ana Plaza, near the center of the
town. ‘The case was at once isolated in the mos-
quito-proof ward of the hospital. During De-

276 The Control of Mosquitoes

cember six cases developed. There were no deaths.
No cases appeared among Canal employees in
1904. In January, 1905, eight more cases were
reported, two resulting fatally. Three cases of
yellow fever developed on board the steamship.
Dora, en route to Havana from Colon. They
were Italians, members of an opera troupe, and
contracted yellow fever while sojourning in Colon.
Two of these cases terminated fatally.

An outbreak of yellow fever occurred on the
U.S. Cruiser Boston, while she was at target prac-
tice at Chame Point,in Panama Bay. The Boston
was in Panama harbor during November, 1904.
Six cases developed on board the ship. The vessel
was thoroughly fumigated, and no further cases
followed.

It was evident that strenuous measures were
needed to control the situation; but little help
could be counted on from the local native authori-
ties, and with the constantly increasing number
of non-immune arrivals, yellow fever cases would
increase in number, and assume the form of a
serious epidemic. The inspection force was
therefore increased to enable the inspection of all
houses once a week. All containers found with
mosquito larvae were washed out. Tanks and
larger vessels that could not be freed from larvee

First Sanitary Work 277

in this manner were treated with oil. Sufficient
oil was used to cover the water surface com-
pletely. As the force became more experienced,
better control was established.

In the early part of 1905 there was a large
number of non-immunes in the city of Panama,
and it became necessary to work with increased
vigor, because, even with a greatly reduced num-
ber of Aédes calopus, the chances for the spread of
yellow fever increased.

The fumigation work proved the marked effects
that elimination of the containers had made on the
eradication of Aédes calopus by the decreased
number of dead insects found after fumigation.
Ultimately we controlled the situation to the
extent that but one per cent. of the houses were
producing larve, and inspection work caused a
large percentage of larvee to be destroyed. One of
the difficulties was in locating hidden containers.
A few of these in a single block of houses were
sufficient to enable Aédes calopus to reach other
houses and decrease the efficiency of the work.
As the ordinance was not properly supported, it
became evident that the sanitary department
would be obliged to undertake the task of screen-
ing, and making inaccessible to mosquitoes, all
water barrels, tanks, cisterns, wells, etc. This

278 The Control of Mosquitoes

work had to be done thoroughly, or was of practi-
cally no value.

The city was divided into eleven districts for
purposes of house to house inspection for locating
and eradicating breeding places. An inspector
was assigned to each district. He rendered daily
reports of the status of each house visited, with
regard to water containers. The force employed
was composed largely of natives, who did not at
first realize the importance of their work. They
were fully instructed as to their duties, and their
work was checked. When doubt arose as to the
quality of the inspector’s work, half a dozen of the
houses in his district were inspected the day before
he visited them, and his reports were checked
against the special inspection reports.

Then, again, the districts were changed at
frequent intervals, which served as a check on
each man’s work, so that before long each inspector
either became efficient, or was replaced by a better
man. The reason for reinspection and changing
of inspectors was to determine the amount of
time required for inspecting the districts, and to
teach the men not to miss a single container.
Every man was checked, and had to work with
about the same rapidity, cover as much ground,
and show as thorough results as the most efficient

First Sanitary Work 279

man on the force. The scheme worked out well,
although at first, in the weeding-out process, many
changes occurred in the force.

Reinspection was continued throughout the
entire campaign, to prevent unsatisfactory condi-
tions in any district. A record of every house
was kept, and those in which mosquito larve
were apt to be found most frequently became
known to the department, and were watched more
closely. These house records showed the number
of wells, tanks, cisterns, barrels, etc., within each.
In order to eradicate these sources of mosquito
production, a force of carpenters was employed
to make mosquito-proof covers for them, and
another gang to attach the covers to the containers.
After they were placed, it became the duty of
the inspectors to report any that had been detached
or tampered with. When a cover was taken off
a barrel or well, the water container was considered
a menace to health, and was removed or destroyed.
All containers not absolutely necessary were re-
moved, and properly disposed of. The exami-
nation of the larger water containers, finajas,
open-top stone filters, etc., in dark corners of the
houses, consumed much time. These vessels
were listed, that none might be missed, and
inspectors had to report the condition of each

280 The Control of Mosquitoes

one at every inspection. House tenants hid
containers, and could never understand why an
inspector who had missed a container one day
always came in search of it the following day. A
small acetylene lamp or an electric flash lamp
was used to determine the presence of mosquito
larvee in large containers in dark places.

The non-immune residence sections were located
in three limited areas, and especial care was taken
in these localities. Most of the Americans in
Panama in 1905 were employed by the Isthmian
Canal Commission, and the greater part of them
worked in what was then known as the ‘‘Canal
Building,’ which occupies an entire block and is
three stories high. Naturally, this building, if
it contained infected mosquitoes, would become
an important focus for the spread of yellow fever.
To prevent such an occurrence, two inspections for
water containers were made weekly, and the entire
building was fumigated once every fortnight. It
was found that the screened doors leading onto
the verandas were propped open or held open by
the janitors, and that mosquitoes actually gained
access to the building through these doors. In
order to prevent this, the department of sanita-
tion placed a lock on each veranda door, and kept
them locked, with the keys in charge of one re-

First Sanitary Work 281

sponsible man. Only one entrance door to the
building was allowed, and a watchman was
stationed continuously at that door to see that it
was not left open a moment longer than necessary.

It was unfortunate that greater power was not
given to the chief sanitary officer, to enable him to
prosecute the yellow fever campaign more rapidly.
Under the organization then in power, he was often
limited to making recommendations, which were
not always carried into effect. Those inexperi-
enced in matters pertaining to yellow fever control
cannot and should not be expected to take part
in the work of its suppression, unless they are sub-
ject to the orders and under the direct control of
the officer charged with the eradication of yellow
fever.

A case in point was that of the so-called Canal
Building in Panama, the original headquarters of
the Canalforce. There was much delay in having
the building properly screened. The architect who
had charge of this matter was overworked, and
busy with plans and estimates for the repair and
construction of buildings all along the line of the
Canal. Work was waiting for all the men avail-
able, and there was not enough room to accommo-
date the needed force. Of course, the architect’s
first consideration was to provide housing facil-

282. The Control of Mosquitoes

ities. The sanitary department thought it of
prime importance to have the building, which
was located in an infected area, properly screened,
as it was used almost exclusively by non-immunes.
The importance of neglecting some other work, if
need be, and attending immediately to the screen-
ing of the Canal Building was brought to the atten-
tion of the architect time and again, but he thought
the matter of secondary importance compared with
the work he was doing elsewhere, and the screening
was delayed. The architect even joked about the
fuss made over the screening. He had not lived in
tropical America before, and had but little faith
in modern ideas pertaining to yellow fever trans-
mission and control. This able and efficient
young man was among the first Americans who
died of yellow fever, contracted in the building he
had not screened, and other employees in the
offices in the same building also contracted the
disease. It cannot be said that the architect
was responsible for his own death and the death of
the others: the source of the trouble was higher.
When the power of the sanitary officer is limited,
the efficiency and speed in obtaining results will
always be limited proportionally, and often with
very serious results.

The non-immune population lived in Panama

First Sanitary Work 283

in three well-defined areas: the Italian colony
on Malambo Street, the Americans near Central
Park, and a considerable number near Plaza
Santa Ana. Many Americans lived in the Ancon
Hospital Reservation. The buildings in these
areas most likely to harbor yellow fever cases
were fumigated every two weeks, to eradicate any
infected mosquitoes that might be present. It
is remarkable that in Ancon Hospital, where most
of the yellow fever cases were cared for, no cases
of yellow fever were contracted. The reason for
this was, that the hospital grounds were under
the direct control of the chief sanitary officer,
and the few adult Aédes calopus remaining there
were not allowed to enter the buildings after
they were once screened. Everybody in the
hospital codperated to prevent them from getting
a foothold, and all were interested in the results
of the anti-mosquito campaign.

It was often difficult to locate the house where a
yellow fever suspect had been bitten by an infected
mosquito. In instances of yellow fever ‘walking
cases,’’ not infrequent in the early stages of the
disease, the source of the infection may never be
located.

When a case or a suspected case of yellow fever
was reported we obtained from him as promptly

284 The Control of Mosquitoes

as possible a list of the places he had frequented
(so far as he could remember), and a list of his
friends or relatives. We questioned them in turn
to make the history of the patient’s previous move-
ments as exact as possible. We often received con-
flicting testimony, and the patient in many cases
was not in a condition to remember where he had
been at the time of infection, or since. It was
very important to obtain this information, so that
effective measures might be taken to destroy by
fumigation the infected mosquitoes that had
inoculated the patient, and that the Aédes calopus
which had bitten him during the first three days
of the disease be destroyed, before they in turn
should become infectious and bite non-immunes.

In some cases the patients were delirious, and
gave no clue to their history. In a few cases
false information was given and the much-needed
facts withheld. The missing of one link in the
chain of evidence meant leaving some infected
mosquitoes alive, with the possibility of secondary
cases. As some cases occurred which could not
be traced to previous known sources of infection,
it was decided to fumigate all houses in the city
of Panama in succession as rapidly as possible.
The fumigation brigade was increased, and the
work pushed as rapidly as conditions would per-

iT

Part of the Panama fumigation brigade

First Sanitary Work 285

mit. All the houses in one or more blocks were
fumigated in a single day. Block after block was
fumigated from one end of the town to the other.
At this time we had a number of cases of yellow
fever, and though the myriads of mosquitoes
were greatly reduced, the comparatively few
remaining, together with enough cases of yellow
fever and a large non-immune population, made
transmission easy.

The following measures were adopted to insure
knowledge of every case as rapidly as possible.

Reporting of suspected cases of yellow fever,
and of all fevers among non-immunes, was made
compulsory by the following section of the sani-
tary rules and regulations for the cities of Panama
and Colon:

Contagious Diseases.

Section 10. Every physician, druggist, school
teacher, clergyman, midwife, nurse, and every head of
a family, having knowledge of any case of the follow-
ing named diseases, shall immediately report the same
to the Health Officer: Asiatic cholera, yellow fever,
typhus fever, smallpox, plague, dysentery, diphtheria,
and membranous croup.

All hotels, etc., were required to keep a register,
and a physician visited every hotel and boarding
house daily to ascertain whether any non-immune

286 The Control of Mosquitoes

residents were sick. In order to prevent the
concealing of cases, patients were not compelled
to go to hospital, but could remain at their resi-
dences and be treated by their family physicians
under conditions specified by the chief sanitary
officer. In addition, a reward of $50.00 was paid
to the first person reporting a case of illness which
was subsequently diagnosed by the Yellow Fever
Board as yellow fever.

When a suspected case of yellow fever was
reported, the patient and his relatives or friends
were interviewed, and an effort made to transfer
him to Ancon Hospital. Almost all Americans
acceded. Where the patient preferred to remain
at his home, his room was at once made mosquito-
proof. The windows and all doorways except
one were screened with immovable screens. The
‘single entrance door was fitted with a vestibule
provided with two screened doors. Both these
doors were fitted with strong springs to make
them self-closing. The outer door was kept
locked, and the key was in charge of a guard
employed by the sanitary department. These
guards were on duty in eight-hour shifts, that
the door might not be left unguarded until the
patient was removed from the room. At the
termination of the case his room was fumigated,

ibule door

Screened vest

First Sanitary Work 287

the adjacent rooms and houses having already
been so treated.

No non-immunes were allowed in the patient’s
room. His physician, nurse, and possibly one or
more immune friends, with the approval of the
chief sanitary officer, were allowed to enter the
screened room. No one could pass the guard
without written authority.

In Ancon Hospital two wards and several pri-
vate rooms were used solely for yellow fever sus-
pects. To enter the wards it was necessary to
pass through a vestibule provided with three
screened doors. A pan of pyrethrum powder
was kept burning constantly between the last
two doors, and a guard was always present to see
that the doors were opened and closed rapidly.
The nurses and physicians learned to pass through
the doors quickly and inspected the wards daily
for Aédes calopus. The windows were screened
on both the inside and outside and examined
twice daily for defects. The buildings and the
grounds in the vicinity of the hospital were fre-
quently inspected for possible water containers.

All of the nurses and many of the physicians
were non-immunes, but no case of yellow fever
developed among those caring for the patients and
living in the hospital grounds, although at times

one

288 The Control of Mosquitoes

there were as many as 200 Americans living there.
All suspects arriving at the hospital were placed
in beds under large screened cages located within
the screened wards, and the yellow fever cases
were kept in these cages during the first five
days of their illness. These precautions made it
necessary for a mosquito to pass through a cloud
of smoke and four screened doors before it could
reach a patient, and then escape through a door
into the ward, where it must survive a twelve
days’ search for mosquitoes before attacking a non-
immune. It was almost impossible for such a
combination to happen, and it never happened.

When the yellow fever wards were first fitted
up, no metal screening was available. Cotton
mosquito bar had to be used for screening the
buildings, and with care and the precautions taken,
it gave satisfactory results. Its fragile nature
renders it unserviceable for extensive use, but in an
emergency it can serve, as at Ancon.

With the installation of water and sewerage:
systems in Panama, it became unnecessary to use
water containers for storage purposes, and as fast
as water was installed in each house, the tanks,
barrels, cisterns, etc., were removed or destroyed.
This assisted to a very large extent in eliminating
mosquitoes. But public water and sewer systems

A poorly constructed eave trough

A clogged eave trough

First Sanitary Work 289

alone do not rid a Central American community
of them. Rain water is soft, and people were
accustomed to use it. The widespread custom

\ eeufnces IRON ROOFING

WA:

| SUPPORTED
EAVE TROUGH

HEAVY ia IRON SUPPORT

A firmly supported eave-trough.

of collecting and keeping it was difficult to
eradicate.
Aédes calopus breeding also occurred in the

eave-troughs, which sagged, and were difficult to
19

290 ‘The Control of Mosquitoes

inspect. Eventually they were allremoved. The
tinajas are still used, but as water costs much
less than formerly, they are emptied more fre-
quently, and are less dangerous.

In localities where there is no available water
supply other than rain water collected by roofs
of buildings, it may be absolutely essential to use
eave-troughs to obtain all the water possible for
storage and future use.

If eave-troughs are used for that purpose they
should be made of galvanized iron: of sufficient
thickness to prevent sagging, have a good slope,
and be securely attached to the roof. It is neces-
sary to have the points of support as close together
as possible. They should be arranged to permit
inspection without leaning heavy ladders against
them. Eave-troughs are convenient but nearly
all towns where the yellow fever mosquito is
found could do without them.

In an anti-yellow fever campaign, it is not safe
to rely on fumigation only, for we cannot be sure
that all the places in which infected mosquitoes
exist have been fumigated. It is only possible
to accomplish a certain amount of fumigation in
one day, and meanwhile mosquitoes in the adja-
cent unfumigated buildings may return. It .is
evident that fumigating on a wholesale scale, 7.e.,

Mosquito-proof water-barrel

A firmly supported eave trough with good slope

First Sanitary Work 291

fumigating all houses in a town or village, will not
necessarily kill all the mosquitoes present in and
about the houses, though probably the greater part
may be destroyed. It issurprising how soon they
will disappear, if all the water containers and
places in which they oviposit are removed. We
believe that the thorough and rapid destruction
or protection of all possible containers and breed-
ing places is the most effective method of yellow
fever control, and that where the health officer is
supported by the residents, Aédes calopus eradi-
cation is quite possible. The campaign at San-
tiago de las Vegas, Cuba, proved this beyond
doubt.

It is not absolutely necessary to eradicate all
Aédes calopus in order to eradicate yellow fever.
It is only necessary to reduce the number of adults
sufficiently to preverit their presence in the vicin-
ity of the places visited by non-immunes.

The matter can be stated thus: In order to
eradicate yellow fever from a given community it
is necessary to reduce the number of Aédes calopus
to such a minority that there is no opportunity for
their biting an infected case during the first three
days of the disease, and surviving twelve days
after. Of course, the nearer the reduction reaches
complete eradication the better, especially in re-

292 The Control of Mosquitoes

gard to the reintroduction of the disease at some
future time. .

When the reduction of Aédes calopus has
reached a certain stage, the chance of a person
being bitten becomes so small that transmission
of fever stops. Yellow fever was eradicated
-in the epidemic at Havana, Santiago, and other
towns and villages in Cuba, as well as in the Canal
Zone and the cities of Panama and Colon, when the
number of Aédes calopus present was much higher
than now. Since the last case of local origin in
1906, on several occasions imported cases have
gained access to, and actually been at large in
Panama and Colon, without producing secondary
cases; because the number of adult Aédes calopus
is so small that there are almost no chances of
completing the chain necessary for the transmis-
sion: a patient being bitten, this particular insect
surviving twelve days, and then biting a non-
immune. In some settlements in the Canal Zone
Aédes calopus are now seldom seen.

It must be remembered, however, that but a
short time is needed to reintroduce this species
of mosquito, and that if conditions are allowed to
become favorable, it will both multiply and spread
over a populated area with great rapidity.

CHAPTER V

THE ANTI-YELLOW FEVER CAMPAIGN AND ITS
RESULTS

FTER the arrival of Colonel Gorgas and his
assistants on the Isthmus, preparations for
the campaign were begun before the first case
of yellow fever was reported. As rapidly as
possible all barrels used for storing water were
made mosquito-proof. Wooden barrel covers
with wire-gauze protected openings were made
and shipped to the different villages. These
were nailed to the barrel tops by the sanitary
department, to leave no doubt of their remaining
mosquito-proof. An inch or so below the top
of the barrel an augur hole was made, and mos-
quito wire nailed over it. This hole was intended
for an overflow, to prevent the water from coming
in contact with the wire-gauze covered opening
in the top, and thus inviting ovipositing.
The covers were made by fastening together two
boards about twelve by twenty-four inches, leaving
293

294 The Control of Mosquitoes

an opening four inches square when the boards
were fastened together. This opening was covered
with a square of eighteen mesh copper wire
gauze, protected by placing above it a square of
half-inch mesh galvanized iron wire, and the
whole fastened in place with wooden strips. A
wooden spigot was inserted near the bottom of the
barrel.

It was necessary to see that the edge of the
barrel was ‘‘true,” and when not, to correct it;
otherwise the cover would not fit snugly and the
entire work would be useless.

Prompt results in mosquito control may be
obtained by making water barrels non-accessible
to mosquitoes in the following manner. Muslin
or sacking is fastened securely over the top of
the barrel, and also over a one-inch overflow hole
six inches below the top. The water is obtained
from a wood or metal faucet.

The use of this scheme on a large scale neces-
sitates much careful inspection and also codper-
ation on the part of the house tenant. When
careless people remove and replace this barrel
covering, the barrel may be left temporarily
uncovered and mosquitoes propagated and liber-
ated unknown to those conducting the anti-
mosquito campaign. The cloth should be kept

Campaign and Results 2905

free from holes and be fastened to the top of the
barrel with strong wire. The twisted portion of
the wire must be so sealed that it cannot be
unfastened or replaced without the fact being
easily detected. Although this method of control
is often not satisfactory, it may sometimes be
helpful. It is rapidly installed, inexpensive, and
may serve temporary purposes.

At first the natives took off the covers, and did
not replace them properly. When taken to task
about this, they professed a desire to clean the
barrels. Observation proved that before the
barrels had been covered they had seldom been
washed out; then it appeared that the people were
so acctistomed to dipping out the water, they did
not want to use the spigots. This practice was
soon stopped.

When the sanitary work was started, the houses
in the villages were mostly surrounded by jungles
or high grass. All bottles, cans, and other poten-
tial water containers and rubbish were thrown out
of doors, and hidden by the vegetation. It was
therefore necessary to do much clearing of vege-
tation and collecting of water containers. Within
a week or two after this work was completed
the grass was high again. Much effort was
required to make tenants place useless water

296 The Control of Mosquitoes

containers within the garbage cans. Where the
houses were close together it was difficult for the
inspectors to find the guilty individual. Appar-
ently it was not easy for the people to change their
habits. It was found necessary to keep the grass
short near houses, so that water containers could
be easily found.

The interiors of all houses were inspected for
water containers, which disappeared as the inspec-
tor approached. Usually the tenants emptied
water vessels at the back door, while the inspec-
tor was entering at the front.

In 1904 houses in the Canal Zone were not
screened. Each American employee was furnished
with a mosquito bar, but they were not used ina
way to make biting impossible. Even bars
that were intact, when improperly tucked in
around the mattresses, proved ineffective. Many
people roll in bed, and thus come in contact with
the mosquito bar while asleep; and mosquitoes
bite through the bar and the sleeper may be un-
aware of it. It was also found that bars were
easily torn, and seldom promptly repaired. In
order to make the bars effective, the sanitary
department undertook to keep those of canal
employees in repair, free of charge.

After the houses were screened, and the number

Campaign and Results 297

of adult Aédes calopus became very small, no more
bars were used. At the present time they are
unnecessary and are not used by people in the
Canal Zone.

We learned by experience that when mosquito
bars must be used, the following precautions should
be carefully observed.

I. The bar should be ample in size, and made
of durable material of considerable tensile strength.

2. It should be carefully tucked under the
mattress, or if no mattress is used, under three
sides of a piece of matting laid on the bed or cot.
Just before entering the bed, the interior of the
bar should be carefully searched for mosquitoes,
and those within destroyed. When a mosquito
is hidden within a net, a small electric flash lamp
helps to discover it.

3. Enter the bed quickly, disturbing the bar
as little as possible.

4. Tuck in the bar carefully on the side through
which you enter.

5. Sleep ina wide bed, and cultivate the habit
of sleeping quietly in the middle of it.

We knew that yellow fever would occur at
Panama and Colon as soon as the percentage of
non-immune population was sufficiently high to
favor. its propagation. As it was proposed to

298 The Control of Mosquitoes

concentrate the newly arriving Americans at La
Boca, Ancon, Culebra, and Gorgona, a strong
effort was made to control the Aédes calopus
production at these villages as rapidly as possible.
It was expected that the employees quartered
in these settlements would visit the terminal
cities, and that some of them would contract the
fever. Every effort was made to prevent trans-
mission of yellow fever from any of those who
might have become infected in Panama or Colon,
to others living at the villages mentioned. The
first cases of yellow fever in the Canal Zone de-
veloped at Culebra, Balboa, and Corozal. All
Americans living at Corozal worked in the city
of Panama and none of the residences of the
employees were screened when yellow fever first
appeared. When the first cases occurred, the
history of every patient was rapidly obtained.
All places visited by them since their infection,
and where they might have become infected, were
fumigated with sulphur, as were also all adjacent
houses. The patients whenever possible were
removed to the yellow fever wards at Ancon
Hospital.

The fumigation was at first a difficult task to
accomplish. The sanitary department was allowed
to pay only the lowest rate of wages current, and

Campaign and Results 299

consequently only poor and unintelligent labor
could be obtained. Any intelligent laborer could
find employment elsewhere at a higher rate of
pay. Even the men employed for scavenging and
ditching were used as a fumigation squad. Some
of them had not enough sense to hurry while
working in sulphur fumes. It was not unusual
for them to leave us while fumigations were in
progress, for working in sulphur fumes day after
day at the lowest rate of wages was not attractive.
At that time in the Canal Zone a laborer could
find work anywhere else for the asking and often
at higher wages. The sanitary inspectors deserve
great credit for the manner in which they won
success in spite of the numerous difficulties they
faced. Apparently many of the laborers had
never moved with celerity before in their lives,
and some of them could not even climb a ladder.
Otheis who went into the rooms to open-windows
after fumigation did not understand enough to
return quickly, and were dragged out in a state of
collapse. This was the only class of labor aliowed
the sanitary department. We had to use it, and
succeed or fail.

Many of the native houses were so loosely built,
and so full of cracks, that it was necessary to paste
paper over the outside of the greater part of them,

300 The Control of Mosquitoes

and contend with leaks in the roofs also. At
times ten pounds of sulphur per thousand cubic
feet of air space was used, to obtain a maximum
of fumes in a minimum of time, and allow for
leakage. ,

Non-immunes stationed in the Canal Zone were
advised not to visit the terminal cities except
when absolutely necessary, because yellow fever
was more prevalent there than in the Zone. Even
the train schedule was changed on Sundays, to
make it inconvenient for American employees from
the Zone to remain long in the city of Panama.
The object was to keep non-immunes away from
the infected mosquitoes as much as possible.

Two cases occurred in the Zone that were of
special interest. The first was at Paraiso; the pa-
tient was a locomotive engineer, and he had not left
his home district except for about an hour anda
half during the period in which he could have
contracted yellow fever. On that occasion he had
traveled three miles from Paraiso to Culebra in his
engine cab, and arriving at Culebra, stepped out
of his machine and into the pay car, where he
remained only long enough to collect his pay.
He at once returned to Paraiso with his engine.
There was such a crowd moving about the pay car
and its vicinity, it was not probable that any

Campaign and Results 301

Aédes calopus hovered in the vicinity, and there
was no house near.

No case of yellow fever had occurred at Paraiso.
What was the source of infection? The histories
of all cases that might be the source of the infection
of the last case were reconsidered, and in point of
time occurrence only one case seemed important.
It was that of a peddler in the city of Panama, ill
with yellow fever in the hospital. His history
showed that he had not been to Paraiso, and on
being questioned, he confirmed the statement.
His friends corroborated it. The Paraiso case
was still a puzzle. The patient’s quarters at
Paraiso, and adjacent houses, were fumigated to
prevent secondary cases, and then the following
action was taken. It was assumed that the
possibility of infection at the pay car was too
remote for serious consideration, and that some
unknown infected person must have been to
Paraiso before the engineer became sick. We
visited every house, seeking for information, and
received much confusing testimony. A second
trip was made to determine if any stranger had
been there to sell anything. We found that a man
had been there selling trinkets similar to those in the
above-mentioned peddler’s box. It was finally

proven that the Panama peddler had not only

302. ~=©The Control of Mosquitoes

been in Paraiso, but had rested in the shade on the
porch of the house in which the American engineer
slept. The peddler acknowledged this fact later,
but refused to give any reason for his previous
story.

Another case of interesting history occurrcd in
the city of Panama. A man was arrested for
intoxication. While arguing witha policeman, he
was seen by a sanitary officer, and sent to the
hospital. Soon after arrival he became delirious;
he proved to be a yellow fever case, and died that
night, leaving only a fragment of his history.
His statement, as far as it could be understood, was
that he was a native of Greece, had no friends in
Panama, and lived at a certain hotel. The hotel
proprietor said he had never seen or heard of the
man. Fora day and a half the city was ransacked
for additional information, without result. It be-
came known that the hotel proprietor was not
sure if the man had stayed at his establishment.

Several men were employed to look up all the
Greeks living in Panama. One of these learned
that the case had been seen in an Italian café.
The proprietor denied all knowledge of the fact.
He was told that it would be necessary to close the
café for a day for fumigation purposes, and he then
volunteered the information that he knew the

Campaign and Results 303

deceased and knew that a fellow-countryman of
his was a bartender at the old theater. A day’s
search located this person. He was in bed, sick
with a pronounced case of yellow fever. Hestated
that he and his fellow-countryman had slept in the
barroom of the theater. It was then too late an
hour to fumigate the theater. It was closed and
no performance permitted. The same night the
daughter of the prima donna came down with
yellow fever. She died a few days later. The
data obtained showed a strong focus of infection
at the theater, and this information prevented
further spread of the disease. The theater was
particularly dangerous, because non-immunes
gathered there in numbers, and had infected Aédes
calopus remained in the building they would have
continued the infection.

The last case of yellow fever in the Canal Zone
occurred at Matachin on September 28, 1905.
The patient was the station agent. The case had
a number of interesting features. For two months
previous to the attack he had never left the village.
No other case of yellow fever had occurred recently
in the Canal Zone, although a few had been found
in Colon. This was the first and only case that
occurred at Matachin. The only case that it
corresponded with in point of infection was that

304. The Control of Mosquitoes

of a man at Colon who had never been to Matachin.
However, the agent remembered distinctly having
been bitten about the ankles and wrists while
checking freight in cars that arrived from Colon.
This is a characteristic biting habit of Aédes
calopus. Subsequent investigation showed that
these cars had been located on a siding at Colon
near the house in which the case of yellow fever
had occurred, corresponding exactly as to time of
infection with the case at Matachin.

At the time the yellow fever epidemic was
coming to an end, the sewer system, water supply,
and paving of the city of Panama were well under
way. The old theory of the upturning of the soil
being responsible for yellow fever was still believed
by many. There was not a single case of yellow
fever in the houses on or adjacent to the streets
being torn up, nor in streets where trenches
were being dug. Although the main water sup-
ply distribution system was installed before yellow
fever was eradicated, it had but little effect upon
the suppression of this disease. Few houses
were at that time supplied with plumbing fix-
tures, and the native population preferred
rain water, to which they were more accus-
tomed. In subsequent years, as the houses be-
came supplied with plumbing, less rain water

Campaign and Results 305

was collected, and rain water barrels were not
allowed to exist. Of course this prevented the
continuance of breeding places favorable to mos-
quito development, and has been an important
factor in keeping yellow fever out of Panama.
Attention is invited to the fact that the city of
Havana had an excellent source of water supply,
and nearly all the houses in its yellow fever dis-
trict had water taps in houses and yards long
before the first American occupation. But a
water supply alone, or even one supplemented
by a proper sewer system, is not sufficient to
control yellow fever or Aédes calopus propagation.
In Havana, there was as much mosquito propaga-
tion in houses occupied by Americans as in those
occupied by natives. Systematic inspection by
capable and energetic inspectors, working under
a health officer especially trained for this work, is’

essential for yellow fever control.
20

CHAPTER VI

MEASURES TAKEN TO KEEP THE ISTHMUS FREE
FROM YELLOW FEVER

re houses in the Canal Zone occupied by

American and foreign employees are not
only screened, but the screening is kept in effective
condition, inspected at frequent intervals, and
repaired whenever defects occur. The lower parts
of doors and all other screens subject to abuse are
properly protected. All houses near settlements
and adjacent territory are inspected each week
for possible mosquito-producing containers. Any-
thing that can hold rain water is removed, or so
placed that water cannot remain in it. Foremen
of gangs are ordered not to leave any material
exposed that may collect rain water. Water in
or on objects that cannot be moved, is treated
with oil or larvacide. Tin cans and other waste
material are buried. Where water is collected in
barrels or tanks at isolated houses, these containers

are made non-accessible to the mosquitoes. Tall
306

Preventive Measures 307

grass is not allowed near settlements, as water con-
tainers may become hidden under it. Interiors of
houses are inspected to detect mosquito-producing
containers, and tenants are invited to complain
to the district sanitary inspector when mosquitoes
are noted indoors. All houses are inspected for
Anopheles at frequent intervals, at which times a
keen search is also made for Aédes calopus. When
adults of this species are noticed, very thorough
inspection is made of all houses and yards in the
vicinity. Each week the district inspector reports
the number of Aédes calopus breeding places found
during the regular weekly inspection, and notes
their location in the station diary. As his district
is inspected by the division inspector, the assistant
chief sanitary inspector, and the chief sanitary
inspector, any neglect of duty is soon detected.
As a matter of fact, Aédes calopus are now very
scarce near the settlements where the non-immunes
live or spend their time.

Attention is also given to movable containers.
Whenever any large machinery or new supplies
are moved into a district, they are inspected for
parts capable of containing rain or other water.
It has been found that certain types of tanks on
locomotives and parts of certain dump cars hold
rain water, and are capable of breeding Aédes

308 The Control of Mosquitoes

calopus. Such traveling containers can do much
harm, as they may release mosquitoes at many
places in a relatively short period. Many types
of barges and canal boats contain and retain rain
water. On one or two occasions they were known
to have transported Aédes calopus twenty miles
down the coast, from Colon to Porto Bello; adults
appeared in large numbers in the settlement
adjacent to the wharf to which the barge was
tied, and in which this species had been absent
for a long period.

Although every precaution is taken to keep
yellow fever out, it is always assumed that a case
may enter at any time without the knowledge of
the health authorities, and precautions are taken
accordingly. The light cases known as “walking:
cases’’ are the most dangerous type.

In order to keep a record of everyone coming from
possibly infected areas, the chief sanitary officer
keeps in touch with all infected seaports north
and south of the Isthmus. Boarding houses and
hotels are compelled to keep a correct register of
guests, showing port of departure, and date of
arrival of each, so that the history of any case
may be rapidly obtained, and non-immunes
associated with a case located at once and kept
under observation.

Preventive Measures 309

Yellow fever is kept out of the Canal Zone and
the cities of Panama and Colon by the following
measures: Inspection of arrivals, quarantine sus-
pects, and of all who have been less than six days
in transit from ports where yellow fever is present,
in order to complete the period of incubation of
yellow fever under observation. For quarantine
purposes this period is considered to be six days.
In the ports of Panama-Balboa and Colon-Cristobal -
the quarantine officers board all vessels, and in
addition to the usual examination and quarantine
measures, detain those who appear at all likely to
be suffering from yellow fever, and also all those
who have boarded the ship at yellow fever ports
less than six days from the day of the inspection.

The careful exercise of quarantine regulations
by vigilant, conscientious, and competent officers
makes it unlikely that a case of yellow fever can
land in a port, and be allowed to go at large.

The three weak links in the chain are:

1. An unrecognized case, in the early stages,
might be allowed to go at large.

2. A person in the early stages of yellow fever
might land on the coast from some small vessel,
and find his way overland into the Canal Zone, -
Panama, or Colon. This is possible, but ex-
tremely improbable because it would be a mat-

310 The Control of Mosquitoes

ter of considerable difficulty to find a suitable
landing place and then travel overland to some
center of population.

3. A ship might touch at a yellow fever port,
and either take on a passenger there, and show
him, by false entry on a passenger list, as having
boarded at another port; or allow some passengers
or members of the crew to go ashore for awhile ata

. yellow fever port, and not make this fact known
to the quarantine officers at Colon or Panama.
While this would be a dangerous thing for the
commander of a ship to do, one such case occurred
at Colon. A ship of a well-known steamship line
touched at a certain yellow fever port before com-
ing to Colon. At Colon the captain subscribed to
the statement that no one was allowed ashore
while the ship remained in the yellow fever port.
This statement was accepted by the quarantine
officer at Colon, and the passengers allowed to
land.

Among the passengers was a young Englishman.
Four days after his arrival he had an attack of
yellow fever. Eight days later he was taken to
Ancon Hospital where he died on the thirteenth
day of the disease. Before he was taken to the
hospital, throughout the period of infectibility, he
lived in an apartment house in the city of Panama,

A short length of eave trough used over a doorway

Preventive Measures 311

boarded in a hotel, and mingled freely with people,
many of whom were not immune to yellow fever.
After the patient’s removal, the usual precautions
of fumigating were taken in the house in which he
resided in Panama, as well as the house where he
visited his sister, her boarding house, and a factory
in which he had worked a few days. No other
cases developed, because aksolute reliance is not
placed on quarantine alone, but on quarantine
service supplemented by continuously sustained
efforts for Aédes calopus destruction, which is the
second and most important line of defense.

As means of prevention of the reintroduction
of yellow fever the following measures are adopted:

Each house, garbage dump, cistern, well, water
barrel, water-jug (tinaja), and other breeding
places of Aédes calopus are inspected at least once
a week.

Breeding places that can be destroyed are
immediately abolished, or notices are served upon
the owners to destroy them within twenty-four
hours.

Wells, cisterns, water barrels, etc., which are
more or less permanent, are protected against
mosquitoes by screening with mosquito-proof

metal screening.
Troublesome eave gttters are cared for by

312, The Control of Mosquitoes

frequent inspections to detect sagging, etc., which
tends to collect water. In practice it was almost
impossible to maintain existing roof gutters on
private residences in a self-draining condition
except those placed over entrances and not exceed-
ing eight feet in length.

‘From time to time a wooden vessel holding a
little clean water is exposed in a position convenient
for ovipositing. The vessel is observed daily for a
week or two to detect the presence of larve in a
given locality and reach a fairly definite con-
clusion regarding their presence or absence.

The methods adopted to keep the number of
Aédes calopus on the Isthmus below the infective
minimum and the results that have been accom-
plished are due to the thorough and painstaking
application of now well-known measures. It has
been asserted that the eradication of such a pro-
lific insect species as the Aédes calopus is impossi-
ble. The senior author believes it possible; but
for the practical purposes of the prophylaxis of
yellow fever, entire eradication of Aédes calopus
is desirable but not a necessity. What is needful,
is to keep their number in a given locality below
that required to propagate the disease. It is dif-
ficult to determine what this number is. It would
differ according to the number of fever cases

Preventive Measures 313

present; to the efficiency of the measures taken
by quarantine for the exclusion of yellow fever;
and to local conditions of housing. Each locality
must establish its own rules in this respect.

Discussion has arisen at intervals regarding the
advisability of destroying water-holding plants
such as banana, alocasia, etc. We are inclined
to believe that danger from these plants has been
greatly overestimated. It is possible that larve
of Aédes calopus may have been found in them, but
careful observation in Cuba and on the Isthmus
has convinced us that fully developed larve or
pupze of this species seldom occur in these water-
holding plants, and that the banana plants are
unimportant in producing them. The destruction
of ornamental and useful plants causes opposition
and we would urge that, before the wholesale
destruction of: these plants is undertaken, very
careful observations be made to determine their
true importance in the propagation of this
mosquito.

CHAPTER VII

THE VALUE OF YELLOW FEVER ERADICATION IN THE
CONSTRUCTION OF THE PANAMA CANAL

SIDE from the humanitarian aspect, the

value of the work lies in its practical

aid to the problem before the people of the

United States: the construction of the Panama

Canal. Was the eradication of yellow fever

essential to the solution of the problem of digging
the canal?

In reply three points must be considered:

First, the loss of employees through the mortal-
ity of yellow fever.

Second, the effect of the presence of yellow fever
on the morale of the corps of employees.

Third, the effect of the presence of yellow fever
on the canal on the attitude of the people and
press of the United States and other nations
towards the canal construction.

The average number of employees on the canal

314

Importance of the Campaign 315

has been about 40,000, practically all of whom are
non-immune. Their families residing on the Zone
and in Panama and Colon would add approxi-
mately as many more. If the yellow fever rate
of the French company’s employees is used as a
basis for estimating the probable yellow fever mor-
tality among the canal employees between 1905
and 1915 a mortality of about 1400 per annum
among the employees would not err on the side of
exaggeration. To this should be added at least
an equal number for the employees’ families, were
their families there. At this rate, during the ten
years estimated for the completion of the canal,
mortality from yellow fever would reach the
frightful total of 14,000 employees,—300 deaths
for each mile of the canal. Even this estimate is
based on the assumption that yellow fever in-
creases in direct ratio with the increase of non-
immune population, and that the number of
cases of yellow fever progresses along a definite
level, while in fact, the curve of yellow fever
ascends much more rapidly than the curve of non-
immune population; as knowledge of the mode of
the transmission of yellow fever would premise.

It is impossible to estimate even approximately
the probable mortality from yellow fever had its
devastation been left unchecked.

316 The Control of Mosquitoes

The effect on the morale of the employees. The
presence of yellow fever in a community has a
peculiarly depressing effect upon the non-immune
inhabitants. Yellow fever is one of the diseases
for which the human race may be said to have an
inherited phobia. It is well known that in 1904-
1905 special inducements were required to secure
desirable men for work on the canal and even
many of these soon left. We can testify from
personal experience to the panic which cases of
yellow fever occasioned among the American
employees, many of whom returned to the United
States within a few days after arrival, frightened
away by the presence of yellow fever. It is
obvious that the class of men who go to a yellow-
fever-ridden country might not be the best from an
economic point of view.

The effect on the people and press of the United
States, etc. The bulk of the laborers were re-
cruited from Barbadoes, Jamaica, Trinidad, Spain,
Italy, and Greece. Would these countries have
permitted the emigration of a large number of
their non-immune citizens to a country where
yellow fever was rampant? Witness the action
taken by these governments to prevent emigra-
tion to the Manaos region, and Brazil region. The
skilled labor, professional and clerical employees,

Importance of the Campaign 317

are American. Many people in the United States
regarded the project, if not with hostility, then with
indifference. The general mental attitude was
that of judgment reserved, but with an underlying
expectation that for one reason or another the
-Canal might not be completed. With several hun-
dred cases of yellow fever among the employees,
how long would the people of the United States
have countenanced the construction of the Canal
at such a price? We think that had the method
of yellow fever transmission not been discovered
before 1904, and had it been impossible to eradi-
cate it from the Isthmus, the Panama Canal would
not be completed, as anticipated, in 1914.

Nor should another phase be forgotten. The
work accomplished in Panama proved to the world
the possibility of eradicating yellow fever from
any locality where the work is properly conducted.
Colonel Gorgas expressed his opinion regarding
the eradication of yellow fever as follows:

I look forward in the future to a time when yellow
fever will have entirely disappeared as a disease to
which mankind is subject, for I believe that when
the yellow fever parasite has once become extinct,
it can no more return than the dodo or any other
species of animal that has disappeared from the

earth.

318 The Control of Mosquitoes

At what cost can this work be accomplished?
Many factors must be taken into consideration
before even an approximate estimate can be
attempted. The size of the community; the
geographical and topographical features; the me-
teorological data; the character of the water
supply; the method of sewage and refuse disposal;
the architectural characteristics of the dwellings;
and the customs and habits of the natives, must
all be considered.

If it is desired to permanently rid a city
of yellow fever, a complete water and sewer
system should be installed, and all houses
connected with them, to remove the necessity
of having water containers in or near habita-
tions.

It is obvious that the cost of such a system must
vary, and depend on local conditions, such as the
size of the community, the distance from avail-
able water supply and sewage disposal points, the
cost of material and labor, etc. It is scarcely fair,
however, to charge the cost of water and sewerage
system to -anti-yellow fever work exclusively,
because these systems will benefit the community
in many more ways than in aiding yellow fever
eradication; some of the most obvious benefits
being the reduction, if not the entire elimination,

Importance of the Campaign 319

of water- and fly-borne diseases, of the cost and
trouble to the individual householder in obtaining -
and maintaining a private water supply, and
refuse disposal plant, and also in protection from
fire with all the gain the last item carries with it
in the way of reduced insurance rates, business
development, etc.

The cost of maritime quarantine, if necessary in
a given community, should also not be charged
entirely to anti-yellow fever work. Maritime
quarantine will benefit the community in many
ways, by excluding undesirables and carriers of
various infectious diseases. Only a portion of this
expenditure, say one fourth, can be justly charged
to anti-yellow fever measures.

Assuming that an anti-yellow fever campaign
were to be inaugurated in a city similar to Panama,
a city of 40,000 inhabitants, covering an area of
approximately one square mile; with a well-
marked rainy season of eight months’ duration
and Aédes calopus breeding all the time; with
yellow fever foci within six days’ travel, and being
in constant communication with the city; with no
communal water and sewer system, and with yel-
low fever cases present, the following preliminary
organization would be desirable for an anti-yellow

fever campaign:

320 The Control of Mosquitoes

Preventive Organization

Per Month

t healthy WMG. vsenes eae uddesegawnewas $ 300.00
I assistant health officer................. 250.00
3, CLOTS oc wiecaca.te vantage ha tetlesWiacisharmlaue asevend i 250.00
BF INS PeCLOPSi: waist ancy ud Sudwer eaten taioree 1,000.00
TOMOLEMED 0% 34-2casurhu dracrars eins siae A ilaroedere 350.00
TOO: ADOLETS), ws sssahcad xan cma oie wee Re 2,500.00
7 WAGONETS es winee sewage cestode ees 225.00
$4,875.00

7 WAROUSs jnewepeny Kowa wR iahaN Ree S $ 250.00
TAMU eS sc vais ve Mana neukaee een hee 2,000.00
14 sets of harness............0.000ceeeee 450.00
$2,700.00

Material necessary: Ladders, sulphur, pyre-
thrum, alcohol, paper, one-inch strips of wood in
lengths of six to twelve feet, tacks, tools, oakum
or cotton waste for calking, canvas for tentage,
cords, pans for fumigation, brushes for pasting,
flour for paste, scrubbing brushes for removing
the adherent paper, soap, wire gauze, spigots, and
barrel tops.

The hospital organization: A fifty-bed hospital
of small wards preferably. Double-screened
vestibules, with the doors arranged so that only
one door can be opened at a time and a watchman
constantly on duty to see that both doors are

Importance of the Campaign 321

never opened at the same time. The hospital
personnel should consist of:

B GAVSICIGNE i062. dh bua ews ee ews $450.00
15 nurses (1 chief nurse)................. 1,100.00
25 orderlies, etc............ 00... 400.00

6 3 (2525 i ne 150.00

Clerk ce ccd bee oiM eae age kare 125.00

2 COOKS ore 5 cae Pane as niahente eens Oea aso Satien aca 100.00

‘ambulance driver... 0.6. .aeae esas 50.00

$2,375.00

I ambulance with appurtenances......... $1,200.00

Subsistence of patients at $1.00 per day per patient.

The Quarantine Organization

I quarantine officer........ $ 300.00
I assistant quarantine officer 250.00
1 launch engineer.......... 75.00
1 launch hand.............. 30.00
I custodian of quarantine

Station. .oyecsey organs 125.00

$ 780.00 X % = $195.00

and the necessary attendants, waiters, etc., re-
quired for the detention house, which should pay
a considerable portion of its maintenance from
the fees collected from the passengers.
Calculating the cost of this organization on the
basis of salaries paid in Panama by the United

al

322, The Control.of Mosquitoes

States, the cost chargeable to the anti-yellow fever
campaign would be as follows:

For salaries and wages about $7500 per month.

For material, etc., $3000, inclusive of cost of
maintenance of hospital and one quarter the cost
of the maritime quarantine, but not including that
of water and sewer installation.

The campaign in Panama lasted approximately
fourteen months. It is the opinion of the chief
sanitary inspector, who conducted the campaign,
that this time could have been reduced to six
months had supplies been more readily available,
together with active codperation on the -part of
the civil authorities. Assuming that the prelimi-
nary work could be completed in nine months, the
cost of the

Preventive Organization would total $57,400.00
Hospital Organization would total 34,875.00

$92,275.00

After this period, only a small fraction of the
time of the health officer and his organization
need be given to anti-yellow fever work, and the
organization could be reduced or ‘ts activities
directed into other channels.

The hospital could be converted into a general

Importance of the Campaign 323

hospital, or closed. The quarantine organization
should remain permanent.

It is to be understood that the figures given are
merely approximate, and based on the cost in the
city of Panama. This estimate is subject to wide
variations under varying local conditions.

Yellow fever was eradicated in towns in Cuba
without the installation of sewer and water supply.
In the absence of these conveniences, it will
depend upon the conscientious work of those in
charge of the mosquito-control campaign whether
fever will be reintroduced. At Santiago de las
Vegas, Cuba, a town of about 10,000 inhabi-
tants of which ten per cent. were non-immunes,
yellow fever was eradicated and mosquito breed-
ing places eliminated within three months at a
cost of less than $5000.

In that case the chief sanitary officer was well
supported by the governor and town officials.

In 1911, in reply to a question relating to the
world’s greatest events, the Right Honourable
James Bryce, Ambassador to the United States,
said: In modern times most of the events of high-
est ultimate significance have been discoveries in
the realms of nature or inventions in the realm of
industry; and their magnitude is seldom known
at first. Little was said of the discovery that

324 The Control of Mosquitoes

mosquitoes are the carriers of yellow fever and
the intermittent fevers, yet what immense conse-

quences are already seen to flow from the determi-
nation of that fact!

INDEX

Acetylene lamp for mosquito
catching, 91, 211

Aédes calopus, biting experi-
ments, 234

biting habits of, 304

— collection of dead, 241,

277,

—— container breeders, 55,
240

control of, 235

—— detection of, 312

eradication of, 234, 241,
291, 312

—— flight of, 234, 242

habits of, 234, 240

— infection of, 245

larve, habits of, 263

—— period of incubation, 235

time of biting, 274

Agramonte, Dr., experiments
in Havana, 231

Algz, copper sulphate applied
to, 45

larve found in, 47

matted with crude oil, 61,
71, 170

—— penetration by larve, 65

protect larve, 26, 65, 180

rapid development of, 26

retards currents, 60

—— warmth increases growth

of 65
Ancon Hospital, before 1904,
16, 18
— mortality statistics, 265
— preventive measures
adopted at, 283, 287
Anopheles, avoidance of light
by, 22, 51, 84, 94

325

Anopheles, biting experiments
with, 21, 46, 89

—— biting in sunlight, 78, 90

—— in brackish water, 74

—— carriers, animals, 92

—— ——— boats, 308

cars, passenger and

freight, 92

travelers, 50, 51, 92

—— caught after staining, 114

—— characteristics of larve,
5,222

— collection of larve and
pupe for laboratory, 45,

4
——in cracks of hydraulic
fills, 120
deprivation of air, 152
—— destruction by hand, 52,
186, 202, 208, 220
—- evaporation fatal to, 46
flight, against the wind, 95
—— —— at Gatun, 76, 98
—— —— long distance, 47, 96,
98, 100
—— —— in relation to rain-
fall, 38
—— —— time of, 188
hiding places of larve, 62
in hospital wards, 20
hours of biting, 21, 85,
89, 224
—— identification of larve, 48
lack of data concerning,
43, 44) 73
—— larve, methods of detec-
tion of, 78
— long distance transporta-
tion, 50, 92

326

Anopheles, methods of stain-
ing, I10

—— prevalence of, before sani-
tation, 20

— prevalence of, at Gatun,
1913, 98

—— propagation areas, 43, 64

—— sheltering habits, 86

—time of collecting on

screening, 51, 213

Anophelines of the
Zone, 51

—— habits of, 47, 91, 196

— persistence of, 52, 204

Ants, natural enemies a: mos-
quitoes, 186

Apparatus to register Hight of
mosquitoes, 108

Aquatic insects, natural ene-
mies of mosquitoes, 185

Aquatic plants, 10, 40, 97,
129, 182, 190

—— ‘change in character of,

Canal

190
—— in Gatun Lake, 99
—— protect larve, 40
reduce stream flow, I0, 29
—_— Podge surface velocity,

Army “Medical Board,
clusions of, 233, 235

con-

Bath, Mr., mosquito trap, 216

Bats, natural enemies of mos-
quitoes, 188

Bermuda grass, 193

Birds, natural enemies of
mosquitoes, 188

Birds, ‘‘night-jars,’’ 102, 104,
106, 188

Birds, water-fowl, 41

Blood samples, taking of, 18

Boats as carriers of malaria,
308

ve for clearing rivers, 64,
I

— Bay potianiney for oiling,

Benois (see fills) *

Index

Breeding places, of A. albi-
manus, 47

——— in borrow pits, 70

— clearing promotes evapo-

ration of, 195

—— crab holes, 72

—— excavations, 40, 70

—— puddles, 25

—— railroad tracks, 38

-—— rivers in dry season, 64

—— stream-beds, 62, 63

—— streams on steep hillsides,

25

—— temperature of water in,
72

Bryce, Right Honorable James,

323,
Busck, A., of Smithsonian
Institute, 44, 53

Cages, for Anopheles transpor-
tation, 46

—— for "fumigation tests, 252

—for immersing larve, 50,
152

—— screened, for yellow fever
patients, 288

—for staining Anopheles,

110
Campho-phenique (see Fumi-
gants)
Camps, near Bas Obispo River,
183
— celeron of sites, 226
—— for troops in tropics, 220
“Canal Building,” screening
of, 282
—— yellow fever focus, 280
Carroll, Dr., experiments in
Havana, 231
Cars, screening of, 203

Carter, Dr. R., observa-
tions on time of incubation,
231, 235

Case histories, 300

— to determine sources of
infection, 244, 308

—— Panama, 285

Catch basins in sewers, 157

Index

Chloroform tube for mosquito
catching, 91, 211

Clothing for rainy season, 81

Collecting Anopheles larve and
pupee for laboratory, 45

Colombia, Jack of sanitary
organization, 264

Concrete (see Ditches)

Containers, the breeding places

of Aédes calopus, 55, 240

dregs in, 263

experimental wooden, 312

flower-pots at Ancon
Hospital, 18, 263

—— hidden, 241, 280

mosquito producing in
Havana, 237, 240

mosquitoes disappear on
removal of, 290

movable and traveling,
307,

resembling natural breed-
ing places, 55

—— in rubbish heaps, 263

screening of, 311

Coral formation, 7, 269

Costs of Canal Zone prevention
measures, 220

Crab holes, 72

Culverts (see Drainage)

Darling, Dr. S. T., biting
experiments by, 46 ~
identification of larve by,

8
— ab ics of Isthmian Ano-
pheles, 45
Diary, station, 307
Ditches, alignment of, 146
branch, 126, 134, 141, 143
comparative cost of lin-
ings, 137, 145, 148
concrete lining, 136
conditions producing mos-
qutitoes, 71
— cost of spoiling grade, 147
factors in construction
of, 149
—— fieldstone lined, 19

327

Ditches, grading of, 126, 128,
132, 134, 147

—— herring bone, 7, 126

—on hydraulic fills, 120

—— inspection of, 129, 144,

145

—— intercepting, 9, 16

—— key walls, 140

location of, 125, 196

—— maintenance of, 125, 144

cost of maintenance of,

15, 137) 145

—— masonry lined, 17

—— narrow bottomed, 125

—— natural water-courses, 149

—— near seepage outcrops, 136

—— open, 125, 129, 130, 136,
149

—— permanent lining of, 8,

137
pot-holes in, 61, 125, 128
—— regrading of, 130
reinforced concrete lining,
15, 138
reinforced concrete lining,
cost of, 138
— scouring of, 128, 130
——slope of, 125, 136, 138,
140, 144, 148
—- soft bottomed, 56, 148
—— storm water, 127
in truck gardens, I1
—— undermining of, 41, 143,
144, 146
— width of, 123, 143, 148
—— weep holes in, 136, 139,
141
Dragon flies, 185.
Drains, outlet of intercepting
tile, 135 |
Drains, storm water, 127
“Drainage, Anopheles,”’ ree
quirements for, 125
Drainage, badly planned, 124
—— channels, blocking of,

39
—— in coral formation, 7
—— culverts, 41, 70, 150
—— use of dynamite in, 118
—— grade, change of, 131, 149

328
Drainage, grade, to fit topog-
raphy, 141
—— —— flatness of, 72, 125
—_— steepness of, 128,
132
—_ taking levels for, 126,

132
—— open, 125
—— pipe lines for, 149
—— subsurface, cost of main-

tenance, 134

— tile, 130, 132 4
—— intercepting tile, 131, 13
“Drips’’ (see Oil) i

Eave-troughs, 262, 289, 311

Employees, mortality of, 314

Enamel-ware saucers and dip-
pers, 45, 79

Evaporation of cleared spaces,
12, 195

Excavations, 70

Fills, on absorbent ground, 115

borrow pits for, 116

—— heavy, 118

—— hydraulic, cost of, 121

cracks in, 119

—— —— filling cracks in, 120

———— shallow ditches on,
120

— porous material for, 116

— settling of, 39

—— shallow, 116

vegetation on, 118

Fines, sanitary regulations, 239,

275
Finlay, Dr. Carlos, theory of
yellow fever transmission,
231, 235,
Fire, clearing by, 197
danger from stored oil,
165, 171
—— precautions
tion, 248
Fish, Barbadoes, 185
— used after clearing and
training, 10
—— notes on, 64, 184

in fumiga-

Index

Fish, small, cannot penetrate
alge, 65

—— value as eradicators of
larve, 182, I91, 194

—— value as eradicators in
New Jersey, 185

—— in wells, 8

Flight, apparatus for register-

ing, 108

——solution for registering,
108

—— (see Aédes calopus and
Anopheles)

Floating islands, drifting of,

99, 194
Fomites, yellow fever, 233, 236
Foot-prints, holding water, 60
Formula for drip oil, 159
—— Panama larvacide, 174
—— pumping oils, 167
—— stain, III
—— stain testing solvent, 111
French occupancy of Canal
Zone, statistics, 23
villages and camps, 17
Fumigant, campho-phenique,

254
—— formaldehyde, 253
—— pyrethrum, 254
—— pyrethrum, deterioration
of, 253
—— sulphur, 254, 300
—— tobacco, 253, 254
Fumigation, action on fruit
cargoes, 258
—— Ancon Hospital Reserva-
tion, 283
—— brigades in Havana, 243,
245
brigades in Panama, 284
—— city of Panama, 284
— experiments in Havana,

250

—— metheds of, 246

—— pans, 247, 258

—— system of, 236, 244, 252,
286

tests, 251

— of theater, 303

—— uncertainty of, 290

Index

Fumigation of vestibule at
Ancon Hospital, 287

——after yellow fever cases,
244, 286

Geography of the Canal Zone,
268

Gorgas, General W. C., anti-
malaria work before 1904, 3

appointed by President
Roosevelt, 16

—— Aédes calopus destruc-
tion in Panama, 235

on disappearance of yel-
low fever, 317

eradication of yellow fever
in Havana, 231

—— Ordinance 6, 273

pamphlet, 1906, 265

—— physicians’ reports, 243

preparations for cam-
paign in Panama, 272,
oe

rapid selection and de-

livery of supplies, 272,

273
—— report to General Leonard
Wood, 232
report on Isthmian Canal,
1904, 261
Grade (see Drainage
Ditches)
Guard, at door of ‘Canal
Building,’’ 281
——at door of yellow fever
ward, 320
—at residences of yellow
fever patients, 244, 286,
287

and

Hand catching, (see Anopheles)

Hoof-prints, 19 Aas

Anopheles propagation in,

10, 71, 146 .

—— difficulty of treating with
oil, 56

— filling, 115

—of horses with mowers,
197

329

Hoof-prints, sanitary regula-
tions, for 57

Hospital organization, a plan
for, 320

Hotels and boarding houses,
sanitary regulations for, 308

House flies, fumigation effects
on, 251

Howard, Dr. L. O., suggestions
and advice in Panama cam-
paign, 44

——on oil for mosquito de-
struction, 235

writings of, 236

Hydraulic fills (see Fills)

Infection of Aédes calopus, 284

from seaports, 308

Inspection, checking system,

239

—— of ditches, weekly, 129,
144

—— of fumigation, 247, 250

— house to house for Aédes
calopus larve, 237, 274,

278

—of houses weekly for
Aédes calopus larve,
276

— of oil drips, 161

——of quarantine suspects,
3099,

—— of screening, weekly, 205

—— systematic, 305, 307

—— system of, in Havana, 6,
236

— yellow fever campaign,
Panama, 274

—— of yellow fever wards, 287

Inspectors, efficiency of, 59,

305

—— report forms, 238

—— responsibility of foreman,
147

Jennings, A. H., collections of
Zone mosquitoes, 44

Jungle, cleared from camp and
town sites, 195

330

Jungle, cleared from abandoned
French camps, 22
—— unexplored, 22, 125

Kendall, on fumigation, 254

Key walls (see Ditches)

Knab, F., determination of
species of Isthmian mosqui-
toes, 45

Labor, class allowed Sanitary
Dept., in Panama, 298, 299
— high cost of, prevents
progress in the South,

221
—— intelligent, at Havana, 6
—— non-immune, infection of,

39
——- shortage of, in malarial
districts, 219
—transportation of, in
ambulances, 246
—— transportation of in cov-
ered wagons, 5
— ultimate cost of infection
of, 42
Laborers, barracks, 213
Larvacide as a detective, 79
Panama, advantages of,
172, 176
—— —— cost of, 174, 177
—— —— formula for, 174
—— — process of manufac-
ture, 175
toxicity of, 152, 176,
178
—— requirements for, 173
Larvacides, comparative values
of, 151
Larvee (see Aédes calopus and
Anopheles)
Lazear, Dr., yellow fever
experiments, 231
Legislation, Panama, 243
Le Prince, J. A., discovery of
mosquito flight, 103
—report to International
Congress of Hygiene,
1912, 95

Index

Light, effect on mosquitoes,
94, 97, 253

Light and dark, effect on
mosquitoes, 186

Lizards, 186

Low, Dr. G. C., observations at
Ostia, 198

Machetes, use of, 195
Malaria developed by excava-
tion and settling of fills, 39
— control, effect of topo-
graphy and climate on

cost of, 165
— control, avoidable costs,
7O

——cost of preventive mea-
sures in the Canal Zone,
220
—— effect on labor cost, 219,
221
infection, effect of screen-
ing on, 206
—— infection of United States
agricultural population,
219, 221
—— mortality statistics in
Havana, 13, 14
— mortality statistics, city
of Panama, 23, 24
in native population,
Panama before 1904,
16, 18
—— at old Ancon Hospital, 20
—— parasite, 47
—— pernicious, 8
—— prevents development of
the tropics, 42, 220
—— prevention, need for sani-
tary regulations for, dur-
ing engineering work,
38, 70, 116
—— rate in the tropics, 219
—— transmission of, 51
— United States Public
Health Service reports,

24
Materials for a plan of cam-

paign, 320
Materials, storehouse for, 245

Index

Monel metal, mosquito wire,

224

Mortality statistics, city of
Panama, 23, 24

—— under French occupancy,

23

Mosquito bars, 20, 46, 89,
112, 274, 288, 296

Mosquito catching in dwellings
as a prophylactic measure,
202, 208

—— in screened cars, 210

destruction in Havana,

236
Mosquitoes, artificial producing
areas, 41
biting experiments, 809,
91, 92
carried on clothing, 187
change of species, 72
collection of dead, 250
— collecting and staining,
112
—— flight observation, 76
— hiding places of, 36
—— hidden producing areas, 9
hours of biting, 21
—— infective minimum, 312
—— Isthmian climate favor-
able to, 268
larve, destruction by
aquatic insects, 185, 191
— natural enemies of, 185,
19g
—— sudden influx at camps,

155
trap for, 96, 214

Oil, application of, 171

——rule for application of,
163

carts, 168

cost of distribution, 164,
165, 166

low cost of crude, 153

drips, barrels, 159

for heavy and light
oils, 160

inspection of, 161

notes on, 154, 225

—

331

Oil drips on rafts, 156

—— —— on streams, 155, 156
—— films, 154, 162, 171

—— — breaking of, 71

—— — drifting of, 170

—— —— spreading of, 190
—— heavy, inhibits vegeta-

tion, 171

—-how it kills larve and
pup, I51

—-and larvacide for pumps,
166

— methods of distribution,

164

—— pipe lines for, 164

—— spraying, 166

—— tanks, 164, 168

—-viscidity in lower tem-
perature, 157

Oiling, from boats, 168

continuous, 154

costs of, 171

— cotton waste in drip can,
158

—— application by
waste, 161

——- drips for light oils, 160

—— intermittent, 162

—at irregular intervals, 6,

cotton

12
—— by “knapsack sprayer,”
166

—— methods of, 154
—— precautions for floods, 161
—— simplest form of drip, 158
—— treatment at Havana, 5
with watering can, 168
Orenstein, Dr. A. J., prophy-
lactic value of screening, 206
Outlet of tile drain (see Drain)

Pack mules, for transporation
of oil, 169

Panama population near ende-
mic sources of yellow fever,
262

Preventive organization, a
plan for, 320

Pipe lines for distribution of
oil, 164

332

Pipe for drainage, 149
—screening entrances of,

149

Plan of campaign, estimated
cost, 321

Plants, water holding, 72

Porto Bello, reintroduction of
Aédes calopus, 308

Pot-holes (see Ditches)

Potato hooks, to uproot aquat-
ic plants, 192

Proctor, Mr., mosquito trap,
216

Prophylactic inoculation of
yellow fever, 241

Pumps for oiling, 166, 168

Pyrethrum, fumigant, 247, 254

Pyrocresol, larvacide, 178

Quarantine, cost of maritime,
319

——a plan for organization,
319, 321

precautious, 308

—— suspects, 309

Quimby, E. F., apparatus for

registering flight, 107

Rainfall, charts, and topo-
graphy, 27
continuous, effect of, 23,
58
—— continuity compared with
volume, 33
—— heavy showers remove
larvee, II
— effect of sudden, 61
—— variations of, 59
Real estate values affected by
mosquito control, 3, 10, 121,

319 : :
Reed, Dr. W., experiments in
yellow fever transmission,
231
Reinforced
Ditches)
Research required, 69, 223
Reservoirs, clearing of, 68
— preparation of sites for,
193

concrete (see

Index

Rivers, blocking of, 70

Rivers, mosquito control on
the Rio Grande, 64

Rock, decomposition of, at
Culebra Cut, 118

Roosevelt, President, appoint-
ment of Colonel Gorgas, 16

Ross, Sir Ronald, anti-malaria
work in the East, 3, 5

—— Prevention of Malaria, 151

Sambon, Dr. L. W., observa-
tions at Ostia, 198
Sanitary regulations, compul-
sory reports of fevers in
Panama, 285
—— fines, 239, 275 .
—— hotel and boarding house
registers, 308
—— impounding cattle, 57
—— need for, during engineer-
ing work, 70, 116
—— Ordinance 6, 273
—physicians’ reports at
Havana, 243
—— in Panama treaty, 272
Santiago de las Vegas, cost
of yellow fever eradication,
BP os
—— eradication of
fever, 243, 291
Screening of cars, 203
—— city of Panama, 277
— copper gauze, 198
—— crevices in perfectly
screened buildings, 204
—— doors, 87, 199, 201, 280,
306
—— imperfect, 204
—— inspection by Health
Dept., 205
—'metal, composition of,
199, 207, 224
—— metallic, 206
—— position for doors, 87
—— protection of, 201
repairing of, 205, 306
—— of sick-rooms, 244
— of tents, 203
——— verandas, 199

yellow

Index

Screening vestibules, 201

water-barrels, 279, 293

windows, 200

Seasonal changes, effect on
mosquito production, 44
68, 72, 81, 112, 126

Seepage, location of, 126

notes on, 65
—— outcrop, Ancon, 19
—_ Havana, corrected
by gravel and cin-
ders, 8
—— outcrops between lines
of tile, 135
covered by vegeta-
tion, 196

on hill side, 37
—— permanent and in-
termittent, 131
—— on stream banks, 36
in unusually rainy
years, 127

planes, 116

sharp changes of grade, 66

Sewerage, catch basins in, 157

, installation in Panama,
288

Slapper, for mosquito catching,
212

Species, change of, Anopheles,

2

L iaaih 6

determination of, 45
Spiders, 186
Spraying, methods of, 166
Spray-nozzle, 167°
Staining of Anopheles, 110

Station, experimental, for ma-

laria transmission data, 25

Stations, meteorological, 31
Stream-beds, erosion of, 35
Streams, branch, 36

and drains, blocking of,

39.
—— “drips” on, 155

—on hillside, carriers of
larve, 35
on hillsides, changes of
source, 62

as oil carriers, 155, 161
pot-holes in, 36, 61

333

Streams, regrading of, 130

—— as sources of mosqui-
toes, 35, 60 ,

training of, 7, 63, 141

Subsurface dees has j
Drainage)

Sulphur, fumigant, 253, 254
Supplies, importance of rapid
delivery, 21, 272, 273, 246
system of preparation,

245, 246
torehouse for, 245
Surveys, preliminary, 16, 22,
25,126
Susceptibility to bites, 85, 91,

93
System of checking inspections,
239

Tadpoles, 185

Temperature of water, effect
of, 65, 66, 72

Tent hospitals, 204

Tents, screening of, 203

Tides, low lands near coasts
affected by, 25, 75

Tides in Rio Grande valley,75

Tile drainage (see Drainage)

Tile drains, inspection of, 135

Tobacco, fumigant, 253, 254

Topography, changes in, 6,
38, 145

Topography, cause of difficulty
in mowing, 197

Trask, H. R., cart for oiling,
168

Treaty, United States and
Republic of Panama, sanita-
tion clause, 272

Trees, in flooded areas, 63, 72,
751 193 is vee anta

Truck gardens, irrigation of,
II

United States Public Health
Service reports, 24

Vegetation breaks oil film, 190
—— burning of, 118, 197

334

Vegetation, changes in, 182
— clearing of, from water,
63, 191
— cutting on edges of
streams, 10, 64
—— grass cutting, 295
— effect of dead, 79
— on fills, 39, 118
—— inhibition of, 171, 177,
192
— retards current, 129, 130
——retards surface velocity,
10
— on seepage outcrops, 196
— shelters mosquitoes, 43
toxic effect of Panama
larvacide on, 177
— uncut, retards evapora-
tion, 37, 197
Verandas, screening of, 199,
202
Vestibule, on houses of yellow
fever patients, 286
—— yellow fever ward, 287

Walking cases (see Yellow
fever)
Water barrels, mosquito proof-
ing of, 293
—— impounding of, 150
—— near sewer outlets, 81
—— subsurface, 116
—— supply, effect on yellow
fever in Panama, 262,
269, 288, 304
— in French camps, 17
——-——in Havana before
yellow fever cam-
paign, 305
—— — notes on, 318
—— ——-in towns in Cuba,
323 : ‘
— temperature of, in relation
to Anopheles produc-
tion, 65, 66, 72
— velocity of, effect on
larve, 144, 149
Watson, Dr. Malcom, report
to Royal Colonial Institute,
225

Index

Wave action, effect on mosqui-
to larve, 68

—ripples break oil films,

163, 170

Weep holes (see Ditches)

Wells, Anopheles in, 55

in coral formation, 7

Wheel-tracks, filling of, 58, 115

Wind action, loss of fumes by,

259 .
— moves and collects debris,

68

—— Anopheles travel at right
angles to, 105

—— Anopheles rest
houses during, 86

—— breaks, 170

—— drying effect on mosqui-
toes, 46, 113

— effect on larve in lakes,
etc., 68

—— evaporation of flat areas,

under

12
—— velocity of, 86, 90
Wood, General Leonard, sup-
ported mosquito control
measures in Cuba, 242
Yellow Fever Board, Panama,
286
—a plan of campaign
against, 320
—— cases, 243, 275
—— investigation of cases and
suspected cases, 283
—— case histories, 300
—— chain of transmission, 292
Commission, Cuba, 246
— cost of eradication, 318
— effect on emigration, 316
— effect on morale of em-
ployees, 314, 316
—— eradication, 291, 314, 317
— eradication without sewer
and water supply, 323
— foci, 265
—— fomites, 233, 236
—— history of, 232
—— imported cases, 292
—— infection, 245
—— mortality curves of, 315

Index

Yellow fever, mortality statis-
tics in Panama, 264

old theory of, 304

— panic, 316

parasite, 317

period of incubation, 233,
309,

—— phobia, 316

—a plan of campaign with
costs, 319

prophylactic inoculation
of, 241

reintroduction of,
3Ir

— subsidence of, with ex-

291,

335

haustion of non-im-
mune material, 267

Yellow fever, ‘ ‘suspects,” 24I,
246, 283, 286

—— transmission,
208, 323

—on United States Cruiser
Boston, 276

— “walking cases,” 245,
283, 308

—— wards, Ancon Hospital,
287

232, 292,

Zetek, J., marking and staining
mosquitoes, 107, 110

Pe S

Putnam’s
Science Series

The Study of Man. By A. C. Happon, $2.00
The Groundwork of Science. By St. Gzorcr Mrvart. $1.75
Rivers of North America. By IsraEt C. RUSSELL. $2.00
Earth Sculpture; or, The Origin of Land Forms.

By James GEIKIE. $2.00
Volcanoes: Their Structure and Significance.
Revised Ed. By T. G. Bonney. $2.00
Bacteria. By Grorce NEWMAN. $2.00
A Book of Whales. By F. E. Bepparp. $2.00
Comparative Physiology of the Brain, etc. By
Jacques Lors. $1.75
The Stars. By Simon Newcoms. $2.00
The Basis of Social Relations. By Danret G. BRINTON. $1.50
Experiments on Animals. By STEPHEN PaceET. $2.00
Infection and Immunity, By GEorGcE M. STERNBERG. $1.75
Patigue. By A. Mosso. $1.50
Earthquakes. By CLARENCE E. Dutton. $2.00
The Nature of Man, By Evie MeTcHNIKOFF. $1.50
Nervous and Mental Hygiene in Health and
Disease. By Aucust Foret. : $2.00
The Prolongation of Life. By ELIE METCHNIKOFF. $1.75
The Solar System. By Cuar_es LANE Poor. $2.00
Heredity. By J. ARTHUR THompson, M.A. $3.50
Climate. By RoBerr DECourcy Warp. $2.00

Age, Growth, and Death. By Cartes S. Minot. $2.50
The Interpretation of Nature. By C. LLtoyp Moraan. $1.25
Mosquito Life. By EveLyN GRoESBEECK MITCHELL. $2.00
Thinking, Feeling, Doing. By E. W. Scripture. $1.75

The World’s Gold. By L. pe Launay. $1.75
The Interpretation of Radium. Revised Ed. By

F. Soppy. $2.00
Criminal Man. By Cesare Lomsroso. $2.00
Social Evil. By E. R. A. SELicman. $1.75
Microbes and Toxins in Nature. By E. Burnet. $2.00
Problems of Life and Reproduction. By M. Harrtoc. $2.00
Problem of the Sexes. By J. Finor. $2.00
‘he Positive Evolution of Religion. By F. Harrison. $2.00
The Science of Happiness. By J. Finor. $1.75
Life and Death of the Globe. By A. Bercet. $1.75

Genetic Interpretation. By James MARK BALDWIN. $2.00

THE FLY

Some Important Books on this Subject.

The Subject of the Fly, the Mosquito, and other insects as carriers
of disease is receiving more attention at the present moment than
it ever has before. Knowledge of recent investigation in this field
is essential to every physician in both practice and research work.

Flies in Relation to Disease
Non-Bloodsucking Flies. $3.50 net. Revised and Enlarged
By G. S. GRAHAM-SMITH, M.D.

24 Plates and 32 Text Figures. (Cambridge Public Health Series)

Flies in Relation to Disease
Bloodsucking Flies. $3.75 net
By E. HINDLE, B.A., Ph.D.
(Cambridge Public Health Series)

The House-Fly

Musca Domestica, Linn. Its Structure, Habits, Development
Relation to Disease, and Control. $4.50 net

By C. GORDON HEWITT, D.Sc., F.R.S.C.
With 104 Illustrations and a@ Map. (Cambridge Zodlogical Series)

House-Flies and How They Spread

Disease
By C. G. HEWITT, D.Sc. 40 cts. net
No. 46 of the Cambridge Manuals of Science and Literature

Insolation Hospitals

By H. FRANKLIN PARSONS, M.D., D.P.H.
Formerly First Assistant Medical Officer of the Local Government
Board. $3-75 net. With 24 Plates and 32 Text Figures

The Bacteriological Examination of
Food and Water

By W. G. SAVAGE, B. Sc., M.D, D.P.H.
County Medicai Officer of Health, Somerset. $2.25 net
With 16 Illustrations

Complete Catalogue Sent on Request

G. P. Putnam’s Sons

American Representatives of the
University Press, Cambridge, England

see etarenal Bia
=

ara