^ A HANDBOOK

OF THE

GNATS OR MOSQUITOES

GIVING

THE ANATOMY AND LIFE HISTORY

OF THE

CULICID^

TOGETHER WITH DESCRIPTIONS OF ALL SPECIES NOTICED
UP TO THE PRESENT DATE

LIEUT.-COL GEO. M. GILES

I.M.S (Retd.), M.B.Lond., F.R.C.S.
Formerly Naturalist to the Indian Marine Survey

Second Edition, Rewritten and Enlarged

3Lon&on

JOHN BALE, SONS & DANIELSSON, Ltd.

83, 85, 87 & 89, GREAT TITCHFIELD STREET, OXFORD STREET, W.

1902

PREFACE TO THE SECOND EDITION.

It is now well-nigh a year ago since my publisher
telegraphed to me in India, requesting permission to reprint,
on account of the entire first edition being "sold out," and
I feel that some explanation is required of what might
appear undue delay in meeting the demand for further
copies. I felt at once, however, that a mere reprint was
out of the question, though I had little idea of the magnitude
of the task before me. The first edition had served its
purpose in providing workers on the subject with a handy
compilation of existing literature, for it neither was, nor
professed to be, in any sense original ; but a year's work in a
very malarious and much mosquito-beridden country had
furnished me with material on which to found personal con-
clusions on many points, and so much had been written by
other students of the subject, in all parts of the world, that
it was clear that no mere compilation would now serve the
purpose. The statements as to the anatomy of the adult
insect, reproduced from various authors, in the first issue, I
had already found were inaccurate in many points, and the
chapter on that subject is now the result of a couple of
months of constant work with the microtome, and by dis-
section. So much had been said and done on the question
of malarial prophylaxis, that it was clear that a separate
chapter on the subject was absolutely essential, and the net
result of these, and other changes, was that when the
scattered notes came to be put together, hardly a paragraph
of the old matter of the first, or general, portion of the book
remained in its original form, and it had been, for all
practical purposes, rewritten.

I hoped, however, that the second, or systematic, por-
tion would only require bringing up to date by inserting in

vi. Preface

their places any new species that might have been described
in the interval ; but, on reaching England, I found that, in
this portion, practically none of the old material could be
utilised. By the timely energy of the authorities of the
British Museum, and of the Eoyal Society, an enormous
collection of mosquitoes had been brought together; and
the work of preparing a monograph of the family, based
on the material so collected, had been entrusted to my
friend, Mr. Theobald, who met me with the news that,
to say nothing of some 160 new species, almost all the
old descriptions were hopelessly inadequate, even where
they were not positively misleading, and that he had found
it necessary to revise the entire classification of the family.
To adhere to the original plan of quoting original descrip-
tions in extenso was obviously out of the question, as the
text of Mr. Theobald's descriptions of new species would
alone have filled more than the entire space of my first
edition, and the only alternative was to redescribe, as
concisely as possible, every member of the family, as
illustrated in the splendid collection now at our disposal.

In the case of the genus Anopheles, and of some other
important genera, and generic types, the course adopted has
been to prepare a new description from the actual speci-
mens, but for the rank and file of the still enormous genus
Culex, it proved more convenient, though I have seen and
handled most of the species enumerated, to epitomise the
descriptions of the monograph. In both cases the plan
followed has been to carefully describe the same structures
in all, and to append to each description, in smaller type,
a few detailed characters ; emphasising those that separate
the species from its neighbours ; but to save space, the
conventional plan of repeating, in the detailed notes on a
species, the characters already given in the short descrip-
tion, has been deliberately avoided. With this object, too,
certain numbers and signs have been employed in the
descriptions, the explanation of which will be found in
the introductory remarks to the systematic portion and
in those on the genus Anopheles.

From what has already been said, it follows that many

Preface vii.

of the descriptions embody, verbatim, entire paragraphs of
Mr. Theobald's work, but the necessities of condensation
have so frequently involved alterations of verbiage and
sequence, that it has been found impracticable to indicate
these by inverted commas, and I trust that this acknovi^-
ledgment will suffice to show how fully I realise my in-
debtedness to the Monograph for much of the matter of
the following pages ; and I desire here to express my deep
appreciation of the courtesy and kindness with which my
work has been in every way facilitated by Professor Kay
Lankester and the other authorities of the Museum, and
in especial to Mr. Theobald, but for whose exceptional
generosity in placing at my disposal early proof sheets,
and in personally helping me in every possible way, the
appearance of this edition must necessarily have been
delayed for several additional months.

There are some hundreds of drawings in the new plates
and figures which, with a few exceptions, have been drawn
by the writer himself from camera lucida outlines, and
it is needless to say that all these changes have taken
much time, but were inevitable, if the new issue was not
to be rendered almost immediately obsolete by the appear-
ance of the Monograph.

To facilitate comparison of neighbouring species, the
figures illustrating them have been, as far as possible,
grouped into plates, instead of interspersing them in the
text.

As to the question of which of the older names are mere
redescriptions of one and the same species — the synonymy of
the group — I have followed implicitly the new Monograph,
and it must be understood that, in relegating any given
name to this category, I merely reproduce, and desire to
imply no personal opinion as to the justness or otherwise
of the conclusions involved.

Saving only, then, those of some few species, of which
no examples have come to hand, all the descriptions in
the present edition are drawn up from actual observation,
and no attempt has been made to reproduce the original
descriptions.

viii. Preface

The bibliography of the subject is akeady so extensive
that the inclusion of any at all complete list of publications
would add some fifteen or twenty pages to the book, with-
out being of any real use to the majority of my readers,
who must needs work far from libraries and museums ;
while those more favourably situated can easily refer to the
very complete lists included in Mr. Theobald's Monograph,
as well as those appended to Professor Grassi's " Studi di
uno Zoologo sulla Malaria," B. Acad, dei Lincei, Kome, 2nd
edit., 1901 ; to Nuttall and Shipley's " Studies in Eelation
to Malaria," Joium. of Hygiene, vol. i., Nos. 1, 2, 3 ; and
to Dr. Edmonston Charles' " Letters from Kome, with
Notes and Postscript by Major Konald Ross, F.E.S."
Liverpool, 1901. Fairly complete systematic references will
be found included in the descriptions of each species, in the
second part of the book.

My thanks are also due to Professors Celli and Grassi,
for placing the resources of their laboratories at my disposal
during a short visit to Rome, and to friends and correspon-
dents in all parts of the world, too numerous to mention,
who have helped me by sending collections, and by con-
tributing even more valuable observations.

In conclusion, I may mention that I shall always be
grateful for specimens of mosquitoes, as well as of ticks,
biting insects, and other pests obnoxious to man and
animals. The small boxes and materials for collecting, that
I used to send to friends good enough to collect, can now
be obtained from Messrs. Baker, of 244, High Holborn, but
I shall always be glad to do my best to identify specimens
sent me, and to help other workers as far as lies in my
power.

GEO. M. GILES.

Byfield, Mannamead,

Plijmoiith, December 'lijt/i, 1901.

CONTENTS.

PAET I.— GENERAL.

PAGE

Chapter I. — On the Position and Terminology of the Culieidce ... 1
Chapter II. — On Collecting, Preserving, and Appliances for Obser-
vation ... ... ... ... ... ... ... 19

Chapter III. — The Anatomy of the Larva 40

Chapter IV. — The Generic Characters of the Larvae of the Culieidce 58

Chapter V. — The Anatomy of the Pupa 66

Chapter VI. — The Anatomy of the Adult Mosquito ... ... 76

Chapter VII. — On the Life History and Seasonal Prevalence of

Mosquitoes 112

Chapter VIII. — On the Conditions Influencmg the Prevalence of

Mosquitoes, and on the Prophylaxis of Malaria... 152

Chapter IX. — On the Distribution of the CwZicitite 237

PART II.— SYSTEMATIC.

Chapter X. — On the Classification of the Family
Chapter XI. — The Megarliinina Sub-family
Chapter XII. — The Anophelina Sub-family ...
Chapter XIII. — The Culicina Sub-family ...
Chapter XIV. — The Mdomina Sub-family ...
Chapter XV. — The Corethrina Sub-family . . .

254
265
280
333
475
500

LIST OF ILLUSTRATIONS.

PAGE

Photographs of Living Mosquitoes ... ... ... ... ... 4

An Indian " Sand Fly " 5

Head of Female (7ttZe.r 8

Wing oi Culex 14

Leg of a Mosquito 17

Ficalbi's Trap Bottle 22

To Illustrate the Method of Pinning a Mosquito ... ... ... 26

To Illustrate the Method of packing Pinned Insects for Trans-
mission by Post 27

Apparatus for Breeding-out Insects ... ... ... ... ... 29

Box with Glass Front and Back, for Photographing Living Mosqui-
toes in profile... ... .. ... ... ... ... ... 31

Case for Mosquitoes 32

Ventilated Box for forwarding Mosquitoes 32

Plate I. — To Illustrate the Anatomy of the Larva 39

The Median and Angular Bristles of the fore Margin of the

Clypeus of five Species of larval J.«q2j7ieZes ... ... ... 42

Larva of Anopheles Bossii ... ... ... ... ... ... 49

To Illustrate the Anatomy of the Eye of the Larva... ... ... 55

Plate II. — To Show the Characteristics of the Larvse of Different

Genera ... ... ... ... ... ... ... ... 57

Plate III.— To Illustrate the Anatomy of the Pupa 65

Scales and Other Chitinous Appendages of Mosquitoes 77

Mouth Parts of an Indian Sand Fly ... ... ... ... ... 82

Diagram to Illustrate the Mouth Parts of a Female Mosquito ... 83
Semi-diagrammatic Representation of Serial Preparations of the

Anterior Part of the Head and Proboscis ... ... ... 84

To Illustrate the Regional Anatomy of the Thorax 88

Semi-diagrammatic drawing of a Transverse Section of the Thorax

of C. pipiens (L.) ... ... ... ... ... ... ... 89

Transverse Section through the Posterior Part of the Abdomen of

a Hibernatmg Female ... .. ... ... ... ... 91

Plate IV.— To Illustrate the Anatomy of the Digestive System

of the Imago... ... ... ... ... ... ... ... 95

Transverse Section of the Head of An. Culicifacies ... ... 97

To Illustrate Certain Points in the Anatomy of the Respiratory

System, and of the Pneumatic Vesicles 104

List of Illustrations xi.

PAGE

Diagram of the Nervous System ... ... ... ... ... 106

Generative Organs of a Gravid Female Mosquito ... ... ... 107

Sagittal Section of last Abdominal Segment of a $ Culex 109

Male Generative Organs of C.fatigans 110

Various Forms of Mosquito Eggs 123

Larvie and pupae of C.fatigans 128

Plate V. — Living J.TCqp7ieZes Larvae ... ... ... ... ... 128

Larva and Nymph of AnojjJieles ... ... ... ... ... 130

Caudal Extremity of Anopheles Bossii infested by a Parasitic

Stalked Infusorian 150

Plate VL — Photographs of Living J.?iop /ieZes and CttZe.^! ... ... 139

Typical Garden Tank 194

View in an Indian cantonment ... ... ... ... ... ... 198

" Borrow pits '' beside an Indian Railway ... ... ... ... 198

Collections of Water in the Surface Drains of an Indian Cantonment 200

Levelling up the Site of an Actual Anopheles Tank 203

Railway Servant's Cottage in the Roman Campagna 227

Graphic Key to Generic Distinctions, based on Scale-characters ... 257
Plate VII. — To illustrate the differences between the sub-families,

Anophelina and Culicina ; reproduced from the Jour, of the

Bombay Nat. Hist. Soc 264

Typical Wing of Genns Meg arhina 266

To Illustrate the Genus Afe^ar/itnci 269

Wing of An. maculipalpis 3' (diagrammatic) ... ... ... 297

Wing of An. Pharcensis ? , Theob. var. alhofimhriatus, Giles ... 303

Anojjheles Lutzii ? ... ... ... ... ... ... ... 304

VJing oi An. leiocojjhyruSf'Domtz, J (diagrammatic) ... ... 312

Plates VIII., IX., X., XI. — Illustrating the Genera Anopheles and

Cycloleppteron ... ... ,.. ... ... ... facing 332

Plate XII.— Illustrating the Genera Mucidus and Psorophora facing 346
Plate XIIL— Illustrating the Genera Janthinosoma, Panoplites,

&nd TceniorliyncJius... ... ... ... ... ... facing 358

Plate XIV. — Illustrating the Genera Stegomyia and Armigeres

facing 384

Venation of Wing in Culex ... . • ... ... ... ... 387

Plate XV. — CitZices with Spotted Wings facing 390

C. pipiens, Li., ajid C.fatigans, Wied., Contrsisted 440

Plates XVI., XVII. — Illustrating some other Culices ... ... 468

Deinokerides cancer, Theoh. J... ... ... ... ... ... 473

The Wing oi j3^doniyia veniistipes ... ... ... ... ... 479

Drawings to Illustrate the genus t/ra,n.oi«?i4«. ... ... ... 490

Wing oi Corethra Asiatica, sp. n. ... ... ... ... ... 507

Wing oi Mochlonyx velutinics, Biuthe ... ... ... ... ... 508

Figures illustrating recently described Anopheles and the New Genus

Limatiis ... ... ... ... ... ... ... ... 509

LIST OF ABBEEVIATED QUOTATIONS.

The following quotations recui- so frequently that it appears more
convenient to employ abbreviated references in their case, m place of
citing them in full : —

B. M. J British Medical Journal.

CM Angelo CeUi, "Malaria," translated by J. T. Eyre.

London, 1900.
Desv. Essai Robineau Desvoidy, " Essai sur les Cuhcides, Mem. de la

Soc. d'Hist. Nat. de Paris," III, p. 390, et seq. (1827).

D. S " Diptera Saundersiana."

D. Sc Zetterstedt, "Diptera Scandinaca."

F. A F. A. Schiner, "Diptera Austeriaca."

Fab. E. S Fabricius, " Entomolog. Syst." IV. (1794).

Fab. S. A Fabricius, " Syst. Antliat." IV. (1805).

F. R Eugenio Ficalbi, "Revisions delle Specie Europei della

Fagmilia deUe Zanzare," Bull. Soc. Ent. Ital. 1896,

p. 239, et seq.

F. V. S Eugenio Ficalbi, " Venti Specie de Zanzare Italiane,'

Bull. Soc. Ent. Ital. 1890.

G. S. Z "Studio di uno Zoologo, suUa Malaria." Memoria

del Socio Baptista Grassi, R. Acad, dei Lmcei.

Rome, 1900.

I. M. G Indian Medical Gazette.

J. T. M Journal of Tropical Medicine.

L. A Felix Lynch Arribalzaga, " Dipterologia Argentma," I,

p. 345, et seq. " Revista del museo de la Plata" (1891).

Macqt. D. E Macquart, " Diptera Exotica," I (1838).

Macqt. H. D Macquart, "Hist, des Dipteres," S. a Buff. I (1834).

Meig. S. B Meigen, " Syst. Beschreib. Europ. Zweifl. ins." Thl. VI

(1830).
Memert Fr. Meinert, "De Eucephale Myggelarver." Vidensk.

Selsk. 6. " Raekke, Natiirvidensk. Og. Mathem

Afd." Ill, 4 (1886).
R. S.,M. C Reports to the Malaria Committee of the Royal

Society. London : Harrison and Sons, 1900.
S. A. C Skuse, " Monog. Aust. Culicidae," Proc. Lin. Soc.

N.S.W. (1877).

Walker, List List of the Diptera m the British Museum, I (1848).

Wied. A. Z. I. ...Wiedemann, " Aussereurop. Zweipflug. Insec." (1828).
Wied.D. E Wiedemann, "Diptera Exotica," I (1821).

HANDBOOK

GNATS OR MOSQUITOES

PART I— GENERAL.

CHAPTER I.
On the Position and Terminology of the Culicidae.

Without entering into minute detail, it may be well to
premise that the Gulicidce belong to the Order of Diptera,
or two-winged insects, in which the hinder of the two pairs
of wings of the typical insect are absent as such, and are
represented only by a pair of small club-like bodies, the
halteres or balancers. All the members of this Order
undergo a complete metamorphosis, i.e., they are hatched
as worm-like larvce, and after attaining, as to size, if not as
to form, the dimensions of the adult insect, and under-
going several changes of skin, they cease to eat, undergo
profound anatomical changes, and become nymphs or pupce,
and, finally, by a last change of skin, they emerge from the
pupa case as the externally entirely different imago, or
adult insect.

The Diptera are divided into two Sub-orders, the Orthor-
rhapha and the Cyclorrhapha, according to the method by
which the pupse escape from the larval skin. In the former,
the rupture is in the form of a T-shaped rent, and the larva
is " encephalous," i.e., has a more or less perfectly developed
chitinous head ; in the latter, the pupae escape by a circular
opening and the larva has no definitely separated anterior
division of the body, formed by the blended cephalic somites,
1

2 GNATS OR MOSQUITOES — CHAPTER I

and the appendages and jaws are so little specialised, that it
is commonly spoken of as being acephalous.

It is obvious that this division, though associated with
fundamental structural differences, is of little value to the
observer who possesses only the perfect insect, but none
of the Cyclorrhapha are at all likely to be confused with
Mosquitoes, and the sub-divisions of the Orthorrhapha are
marked by characters of a very obvious sort in the adult
insect ; the Sub-order being divided in a very natural
manner, into two sub-divisions, by the characters of the
antennae. In the Nematocera, to which the Culicidce
belong, the antennae are large and prominent organs, con-
sisting of more than six joints, and the palp of four or five
joints, while in the Brachycera the antennae are of in-
significant dimensions, consisting of but two or three
apparent joints, and the palpi are also but one or two
jointed, Osten-Sacken further sub-divides the Nematocera
into the true and the anomalous groups of families. The
true Nematocera, which mclude the Cecidomyida, Myceto-
philidce, Culicidce, Chironomidce, TipuUdcB, PsychodidcB, and
possibly the Dixidce, have the following characters : —

(1) The eyes are never blended into a single mass, and

there is little or no difference in the size of the
head and eyes in the two sexes.

(2) Eyes round, oval or lunate; they may meet but

never blend.

(3) Antennae very large in proportion to the small head.

(4) Legs long and weak, not fitted for walking.

(5) Generally slighter, and more slender.

(6) Inhabit damp, shady places, and prefer twilight.
The anomalous Nematocera, which include the Simtdidcs

or sand-flies, the Bihionida, dc, on the other hand, are
characterised as follows : —

(1) Head generally holoptic in both sexes, nearly always

so in the male.

(2) Eyes often bisected, the upper facets being the

largest.

(3) Legs well adapted for walking and often thick.

(4) The sexes generally differ considerably.

POSITION AND TERMINOLOGY OF THE CULICID^ 8

(5) Have a peculiar and often sporadic geographical
distribution. ,_.

The Culicidm, or sfleas.) are now included in the anoma-
lous Nematocera by the majority of authorities.

Confining our attention to the true Nematocera, the
Culicidm may, for practical purposes, be easily distinguished
from the other families by two very obvious characters.
The first of these is the possession of the long, suctorial
proboscis, which differs markedly from the mouth parts of
any insect likely to be confused with them ; and the second
is that in all the veins of the wings are fringed with scales
like those of butterflies and moths. It is true that the
wings of certain genera, such as Molopheles and Bypha-
Gophua have the veins of the wings scaly, but even in the
former the scales are very elongated, while in the latter
they are more of the character of hairs, and in both the
general arrangement of the scales is of a shaggy and
irregular character as compared with that of the CulicidcB,
apart from which they present unmistakable differences in
the venation of the wings.

The family that is most easily confused with the Giili-
cidce is the Chironomidce or midges, which not onh^ frequent
very much the same situations, but in general form so
closely resemble the gnats that they can scarcely be dis-
tinguished by the naked eye ; but this family has neither
the long proboscis nor the scaly wing veins, and a moment's
examination with a lens suffices to distinguish them.

Those who wish to follow more closely the question of
the classification of the Nematocera are recommended to
consult Mr. F. V. Theobald's " Account of British Flies,"
which is not only very plainly written, but appears more
up to date than most of the accessible works on the subject
in the English language. Although as yet it is, unfortu-
nately, not completed, it contains a most handy synopsis of
the genera of the Order, and will, therefore, be most useful
to anyone commencing the study of any group of Diptera,
even in tropical regions, for it must be remembered that
the Dipterous fauna of India, and most other tropical parts,
remains to be written, so that a knowledge of the general

4 (^NATS OR MOSQUITOES — CHAPTER I

principles of the classification can only be gained from
European and American works.

Hitherto I have treated the terms Culicidce, Mosquitoes,
and gnats as synonymous, and the present work is confined
to the consideration of the GulicidcB alone, but it must not
be supposed that every insect that bites and is annoying to
man necessarily belongs to this family. As far as we at
present are aware, however, it is the GulicidcB alone that
are concerned in the transmission of malaria, and as this
handbook is mainly intended for the use of those who may
be working on this problem, it does not appear worth
while to include the Siviulidce and other obnoxious insects
that attack man in the same way.

The word Mosquito is a diminutive of the Spanish and
Portuguese " mosco " fly. A variety of insects of the Guli-
cidcB and other families are known under this name in
various localities, the only common characteristic being the
power of annoying man by their bites. It is not uncom-
mon to see in the press, notices of the occurrence of
" Mosquitoes" in England. When investigated by compe-
tent entomologists, the insects always turn out to be one of
the common indigenous English gnats, generally G. pipieiis,
L. ; and in point of fact this species has as good a claim as
any other to the name, and is quite capable of inflicting
as much annoyance as any other, the tropical species sur-
passing our English gnat rather in numbers and persist-
ence than in their individual capability of annoyance.
Something of the same sort may be noticed in the case
of the common fly, which even where fairly common, rarely
exhibits in England the same dire determination to sit on
one's nose that it does in India and other hot climates, and
which it will do, even in England, when the weather is
sufficiently hot. In short, the question whether gnats will
earn for themselves the dreaded title of Mosquito or not is
rather a matter of temperature than locality, or in other
words, it is only in hot weather that gnats show any strong
tendency to attack human beings in place of being content
with their more usual vegetable food. At any rate, no one
species is in any way entitled to the name.

f-

Pig. 1.

Photographs of living mosquitoes (C. impellens,
profile ; in the middle ventral aspects of the same ;
views. About twice natural size.

Face p. 4.

Walk.;. Above <? and '^? in
and beneath 9 and S dorsal

POSITION AND TERMINOLOGY OF THE CULICID^E 5

Oil this point, Mr. F. V. Theobald remarks: "Much
ditterence exists between the so-called Mosquitoes of various
parts of the world. Some are true gnats, others sand-flies
{SimulidcE), and yet others midges, or Ghironomidce, of the
genus Ceratopogun, such as the American so-called ' grey-
gnat.' American Mosquitoes belong to the genera Culex
and Simulium ; that of Cuba,. according to Desvoidy, is a
Culex, while in Brazil the Mosquito is a small Simulium,

Fig. 2.— An Indian " Sand Fly," common in Naini Tal. x 15 Diams.
a, The left half of the insect is shown denuded of its tomentum, so as to
indicate more clearly the details of the venation of the wing ; b, antennse of
the same, more highly magnified.

sharing the honour with a true gnat {Psorophora molestus),
according to Pohl and Kollar. From the West Coast of
Africa I have had several kinds of Mosquito sent, including
one or two Ctdicidce, but some were midges. It will thus
be seen that the term is no guide to the family of insects to
which any given ' Mosquito ' belongs. In any case, they
all belong to families which are found in greatest abundance
in swampy places."

The term "sand fly" is, however, at least as loosely
applied as the word Mosquito, so that it must not be

6 GNATS OR MOSQUITOES — CHAPTER I

supposed that all of them are Siviulia, and some of the
most common and vicious of Indian species approach so
closely in general structure, and even in the general appear-
ance of the venation of the wing, to the anomalous gnats
of the genus Corethra, that they are difficult to distinguish
without close examination, as may be seen by comparing
the figure given on previous page, of a species (probably new)
of Phlebotomus, most pestilent in its attentions in Himalayan
hill stations during the rains, with that given of the wing
of Corethra in the systematic description of that genus, but
it will be seen that what looks like a gnat's hinder fork cell
is really not the fourth, but the fifth longitudinal vein, and
that the sixth, in place of being simple, is also forked :
distinctions that will be better appreciated after a perusal
of the following pages on the terminology of the group.
The insect, too, is clothed, not with scales but with hairs.
A more darkly tinted species of the same genus is equally
troublesome in the plains.

In almost all cases it is the female alone that attacks
man and animals, an exception being noted in the case of
C. salinus by Ficalbi in his description of that species ; but,
however bloodthirsty they may be, it is obvious that they
must mainly depend on other nourishment than blood, for
these insects are sometimes found in just the places where
air-breathing vertebrates are most uncommon.

The males of some Indian species, notably of Stegomyia
fasciatus (Fabr.), undoubtedly often settle on one, and
place themselves in position as if to bite, but I have never
seen one actually do so, although afforded every oppor-
tunity.

In England, and probably in other parts of the world,
gnats may be seen feeding upon the nectar of flowers, and
here it is certainly their usual food, a fact I have often
verified by personal observation ; but Mr. Theobald has also
seen C. ciliaris sucking the juices of small Diptera. The
peculiarity of the females alone being addicted to blood-
sucking is shared with the Tahanidce, or gad-flies, whose
males, too, live entirely on the juices of flowers.

Coming now to the question of terminology, it may first

POSITION AND TERMINOLOGY OF THE CULICID^ 7

be observed that the terms used by Dipterologists are
broadly the same as those used by entomologists generally,
with such modifications and additions as are required by the
special pecuharities of the Order. Unfortunately, neither
among the former nor the latter is there any agreement
as to the terms employed. It would be impossible, in any
moderate compass, to cover even the practice of the best
known authorities, and without expressing any preference
for Loew's terminology over that adopted by other
authorities, it represents, I thmk, better than that of any
other author what may be called the average practice in the
matter, and on this account the definitions given below are,
to a great extent, quoted verbatim from his " Monographs
of the Diptera of North America," published by the Smith-
sonian Institution, Washington, 1862.

In the systematic portion of this book, except where
the original descriptions have been transcribed or translated
as they stand, Loew's terminology is adopted, except for
the wing, for which Skuse's nomenclature is followed.

As in all typical msects the body of the dipterous imago
is divided into the three regions of the head, thorax and
abdomen. In the first two of these the component somites
are so soldered together that their limits can hardly be
distinguished, but in the abdomen the segments are clearly
separated.

The entire external surface is covered with a chitinous
investment, which is in all sufficiently dense to afford
efficient protection to the internal organs and a firm attach-
ment to the muscles that move the various parts, and in
many families forms a veritable coat of armour.

This chitinous coating is always, to a greater or less
extent, covered with an armature of chitinous processes in
the form of hairs, spines or scales, and where these are so
closely arranged as to form a more or less uniform covering,
they are spoken of as a tomentum.

In the CulicidcB all three of these forms of armature
are well represented, but its characteristic feature is the
abundance of scales, which in most genera thickly cover the
entire body, and in almost all cases the decorations of gnats

8

GNATS OR MOSQUITOES — CHAPTER I

are due to variations in the coloration of this scaly coating ;
and as these are very easily rubbed off, caution is necessary
in comparing them with descriptions.

The tinting of the chitinous body wall itself is spoken
of as " ground colouring," and in insects which are provided
but scantily with hairs, this may be arranged in sharply
defined and brilliant markings, as, for example, in many
midges, but in the Culicidoi it is usually of a nearly uniform
dull tint, and at most shares only in the production of their
markings by showing through, between tracts of scales, in
the form of symmetrical bare lines and patches.

Fig. 3.— Head of Female Culex.

a, thorax ; b, neck ; c, uape ; d, occiput ; e, vertex ; /, eyes ; g, frc
fe, clypeus ; i, antennae ; fc, palpi ; I, proboscis.

In descriptions of Mosquitoes it is not uncommon to
find the coloration of parts qualified as " when denuded,"
and where this is the case it must be understood that it is
the ground colour that is referred to.

In Anophelea, as a rule, there are few or no scales on
the abdomen, and its coloration is mainly ground colour.
Their appendages, on the other hand, are exceptionally
densely scaled. In Corethra, again, the appendages, though
very hirsute, are so mainly through the abundance of hairs,
but in all genera of gnats the armature of the wings is
entirely composed of scales.

POSITION AND TERMINOLOGY OF THE CULICID^] 9

The head has a hinder plane, opposite the thorax,
called the occiput (occiput) ; that region of it lying over the
junction of the head is the nape (cervix). The part of the
head which reaches from the antennae as far as the occiput,
and is limited laterally by the compound eyes, is the front
[frons), the upper part of which is the crown {vertex), the
limit between the front and the occiput having the name of
the vertical margin {margo vertical-is). The middle of the
front, being often of a more membranous substance, and
sometimes differing in colour from its borders, is called the
frontal stripe (vitta frontalis). On the crown are the
simple eyes {ocelli), which, however, are absent in the
Gulicida', but in the majority of Diptera are usually three
in number, sometimes on a sharply defined triangular space,
the ocellar triangle {triangulum ocellce). Most of those
Diptera which undergo their metamorphosis within the
larval skin possess, immediately above antennae, an im-
pressed arcuate line, which seems to separate from the
front a small, usually crescentic piece, the frontal crescent
{lunula frontalis). The impressed line itself, which con-
tinues over the face nearly as far as the border of the
mouth, is called the frontal fissure (fissnra frontalis). It
owes its origin to a large bladder-like expansion, which
exists at this place in immature imagines, and which helps
them in bursting the pupa case. The frontal fissure is,
of course, the true anterior limit of the front, and the
frontal crescent in fact belongs to the face ; but from its
usual situation it is commonly considered as part of the
front. In many genera the eyes of the males meet on
the front, so as to divide it into two triangles, the upper of
which is called the vertical triangle (triangulum verticale),
the inferior, the anterior frontal triangle {triangulum
frontale anterium), or simply the frontal triangle. The
anterior portion of the head, reaching from the antennae
to the aural margin {yeristomium) is called the iiiGe{ fades).
In most Diptera it is divided into three parts adjoining
each other, the limits of which depend on the situation
which the frontal fissure, continued to the oral margin,
occupies in the developed imago. The form and mutual

10 GNATS OR MOSQUITOES — CHAPTER I

proportion in size of these parts are of the highest value
in the classification and distinction of species. Beneath
the antennae there are, in many Diptera, longitudinal
grooves for their reception, the antennal furrows {fovecc
antennales). The antennae lie in these while in the pupa
case, and sometimes even after escape. That part of the
head which lies on the side beneath the eyes is called the
cheek (gena). The compound eyes are sometimes encom-
passed in a greater or less portion of their circumference
by a ring, somewhat swollen, and separated, more or less
distinctly, from the rest of the surface of the head, and
named the orbit, the successive parts of which may be
called the anterior (orbita anterior, sive facialis), the inferioi'
{inferior s. genalis), the posterior {posterior s. occipitalis),
the superior {superior s. verticalis) , and the frontal {frontalis)
orbits. An orbit is also often spoken of when there is no
distinct ring ; but in this case there is some difference of
colour or structure to mark it off.

The oral parts of Diptera, destined for sucking, are
called the sucker or proboscis. They are either inserted
at the end of a more or less prolonged, cylindrical portion
of the head called the snout {rostrum), or project from a
wide aperture, often occupying a great part of the under
surface of the head, called the mouth hole {cantus oris).
The common, fleshy root of the oral parts is connected by
a membrane with the border of the mouth. This mem-
brane often has a fold of sometimes quite a horny con-
sistence, and is then called the clypeus {clypeus s. prce-
labrum). It is either entirely concealed by the anterior
border of the mouth, and is then usually movable, or it
projects over it as a ridge, and is then usually immovable,
as in the Culicidce, in which it is usually bare, but is
covered with crescentic scales in the genus Stegomyia, and
is densely hairy in Trichoprosopon. The largest of the
oral parts in most Diptera is the fleshy under lip {labium),
consisting of the stem {stipes) and the knob {capituluin
labii), formed by the two suctorial flaps {labella), and
modified in the CidicidcB into the membranous, scaly
sheath which encloses the other mouth parts. Besides

POSITION AND TERMINOLOGY OF THE CULICID^ 11

the under lip the palpi are most perceptible, and must
be noticed in the description of species. The remaining
oral parts are generally rather small and stunted, having
the form of bristles or horny lancets. They are considered
as being the tongue {lingua), under jaws {maxillcB), upper
jaws {mandihulce) , and upper lip {labrum), the latter shut-
ting the under lip from above, and certain prolongations of
the chitinous pharynx such as an epi- or hypopharynx.
These parts are not particularly valuable in the distinction
of species.

As in other insects, the thorax of Diptera consists of
three segments, the prothorax, the mesothorax, and the
metathorax ; but in this Order the mesothorax is so much
more developed than the other segments, that it forms by
far the largest part of the thorax, and in the description of
Diptera is exclusively designated by that name, while other
names are given to the pro- and metathorax when some
particular part of them has to be characterised. The pro-
thorax, being generally very little developed, sometimes
forms a neck-like prolongation which bears the head, and
is then called the neck {collum). Sometimes the fore-
corners of the mesothorax or shoulders (humeri) are covered
by a lobe of the prothorax (lobulus prothoracis humeralis),
distinctly separated from the mesothorax ; but it is not
uncommon for this to be soldered to the mesothorax so
that no distinct limit is perceptible, except by differences of
colour and armature, in which case it is called the shoulder
callosity (callus humeralis). Sometimes also the prothorax
is closely applied to the anterior border of the mesothorax,
and has then the name of the collar {collare). The meso-
thorax frequently also has a transverse furrow {sutura
transversalis) crossing the middle of its upper side, and
ending on each side a little before the base of the wing ;
its presence or absence, as well as its form, furnishes
characters of importance in classification. On each side
of the pleura or " breast-side " there is, beneath the
shoulder, a spiracle belonging to the prothorax {stigma
prothoracis). To the back of the mesothorax applies the
scutcheon (scutellum), separated from it by a furrow.

1'2 GNATS OR MOSQUITOES — CHAPTER I

Beneath the scutellum a part of the metathorax is to be
seen called the metanotum, general!}^ descending obliquely,
often very convex, and on each side a more or less inflated
space, called the lateral callosity of the metanotum. In
most mosquitoes this portion of the thorax is nude, but in
Mr. Theobald's new genera Wyeomyia and Trichoprosopon,
is provided with certain characteristically arranged groups
of bristles ; and care should be taken not to confuse these
metathoracic tufts with somewhat similar but much more
uniformly present tufts on the scutellum, which overlie
them in the usual position of the parts. The poisers (hal-
teres) have their origin beneath this callosity, and before
each is placed the stigma of the imetathorax. In many
Diptera these are protected by membranous covers placed
above them, and called covering scales {tegulce s. squamcB).

The abdomen is the third of the principal parts of the
body, but the word often is applied to the dorsal side only,
the under side being called the belly {venter). Its seg-
ments, as in other Orders, are counted from before back ;
but the anterior ones are often soldered together, while
the posterior ones are stunted and concealed. Much caution
therefore is desirable in counting them. The statements
as to their number are often rather arbitrary and conven-
tional, and must be taken to refer simply to such as are
plainly obvious under a simple lens.

A careful observer will generally find they are really
more numerous ; and it must also not be assumed that the
entomologist's " segment " is necessaril}^ equivalent to the
"somite" of the comparative anatomist. In many mos-
quitoes the scales clothing the abdommal segments are
so coloured as to produce a conspicuous alternation of
darker and lighter bands ; and the presence or absence
and relative position of these bands is of the greatest
importance in distinguishing the various species.

This banding may be produced in three different ways.
Either the lighter coloured scales may be confined to the
front of the segment, in which case it is said to be " basally
lighter banded"; or they may be located on its hinder
portion, when the lighter banding is said to be "apical" ;

POSITION AND TERMINOLOGY OF THE CULICIDJE 13

or, lastly, the coutiguoas portions of consecutive segments
may be lighter or darker, and the middle portion of each
of the contrasting tint.

Another term, frequently recurring in entomological
descriptions, especially m those of earlier date, is to speak of
the fore parts of the segments wher6 they are connected with
that before them as the incisurce abdominis ; and where, as
in banded insects, this part of the segments is coloured in
contrast, the term " with lighter incisurce " may be taken
as equivalent to " basally lighter banded," and the con-
verse ; though often the term is loosely employed, and it
is therefore preferable to employ the modern equivalent.
At the end of the abdomen we see in the male the claspers
{hypopygiiLm), in the female the ovipositor. If the claspers
have the form of pincers, and are not bent under the belly,
they are called forceps. The Ovipositor may be either a
borer (terehra) or a style (stylus), or, as in the family with
which we are mainly concerned, may take the form of
paired, lobular organs, w^iich, though wanting the hook of
the male clasper, resemble it closely in general appearance.

Both organs are very important in the distinction of
species in many families, and their structure being generally
very complicated and varying much in different families,
deserves a most attentive study.

So far the terms employed are in such universal use
among systematists that they may be considered as definite
as those employed in human anatomy, but it is when we
come to the wmg that the confounding of confusion com-
mences ; for it may almost be said that the systems of
nomenclature are as numerous as the authors who employ
them.

This has come about from the natural desire to devise a
terminology applicable to the entire Order, combined with
the widest diversity of opinion as to the correspondence of
parts in the various families ; so that probably a rigidly
numerical system would be really the most convenient, in
spite of the names affording no clue to the correspondence
of the veins ; but an exclusively numerical nomenclature
is, as a matter of fact, adopted by no entomologist who

14

GNATS OK MOSQUITOES — CHAPTER I

has written on the Culicidce, and, for our purposes it will
be best to confine ourselves to the wing of the mosquito
and the names that have been most commonly used by
those who have written on the family.

Commencing with the anterior border of the wing, we
find that it is formed of a strong chitinous rib, which is
called the costa. It tapers off externally, but in the Guli-
cid(B is continued round the entire border of the wing,
though this is not the case in all Diptera.

2. AB a A 5 h A f

i. — Wing of C. concolor, B,. Desv., ^ to illustratk the
Terminology of the Wing in the Culicid^.

C, costa.

a, auxiliary vein.

1 to 6, first to sixth longitudiual

veins and their branches.
7, seventh, or false longitudinal

vein (unsealed).
VI, unsealed vein, between fifth

and sixth longitudinals.
h, humeral transverse vein,
s, supernumerary transverse vein.
m, middle transverse vein.
p, posterior transverse vein.

A, costal cells.

B, sub-costal cell.

c, marginal cell.

D, anterior fork cell or first sub-

marginal.

E, second subuiarginal cell.

F, fii-st posterior cell.

G, hinder fork cell, or second pos-

terior.
H, third posterior cell.
I, first basal cell.
J, second basal cell.
K, anal cell.
L, axillary cell.
31, spurious cell.

Looking now at the root of the wing, we find that the
remaining longitudinal ribs, or veins, spring from three
main trunks, which are connected together near their base
by a somewhat elaborate arrangement of cross bars. From
the anterior of these spring two veins, the front one of
which is usually called the auxiliary vein ; on account of
its being absent in many Diptera, though Theobald calls
it the sub-costal and by others it is termed the mediastinal.

POSITION AND TERMINOLOGY OF THE CULICID^ 15

It runs into the costa considerably before the tip of the
wing, and the relative position of this point is useful in
distinguishing species. The other of the anterior group
o£ veins is the first longitudinal, which runs, without
branching, to join the costa, close to the tip of the wing.
From it, well out towards the middle of the wing, springs
the second longitudinal, which divides near the tip of
the wing into two branches, though in most Diptera
this vein remains single. Two veins spring also from
the second of the vein-trunk. These are the fourth and
fifth longitudinal veins and both of them become forked,
the fourth near the tip of the wing, and the fifth further
in. In the outer part of the wing, between the forked
second and fourth, may be seen the third longitudinal vein,
which is distinct, and fringed with scales only in this outer
part of the wing, and often appears to take its origin from
a cross vein ; but in many cases, as in C. concolor, B. Desv.,
it can be distinctly traced as an unsealed vein right to the
root of the wing, and, in the majority of cases, can be so
followed for a considerable distance, while it is not uncommon
to find a short length of the vein, internal to the cross veins,
provided with scales. From the hindmost of the three
root-trunks there springs only a single scaled vein, the
sixth longitudinal, which runs without branching to end
in the internal margin of the wing beyond the middle. On
either side of this, however, is an unsealed vein, one of
which joins the tip of the hinder branch of the fifth
longitudinal, and is spoken of by Mr. Austen as the sixth
longitudinal (the sixth becoming his seventh), and the
other being his eighth longitudinal.

As a rule, these unsealed veins are left unnoticed in
descriptions and figures of mosquitoes.

We come now to the transverse veins or vemdce, of which
there are four in the CuHcidce.

Between the costa and auxiliary, close to the root of
the wing, is the humeral transverse ; and the spaces into
which it subdivides the interval between those veins are
known as the costal cells.

Between the second and third longitudinals lies the

16 GNATS OR MOSQUITOES — CHAPTER I

supernumerary transverse vein, which is considered by some
as merely the angulated origin of the third longitudinal.

Doubtless in some cases, as in Mucidus, the third
springs from the second longitudinal very far out, but in
these cases the supernumerary is absent, and the middle
transverse, w^hich m most mosquitoes connects the third
and fourth longitudinals, spans the entire space between
the second and fourth. The remaining cross vein, or
posterior transverse, connects the fourth with the anterior
branch of the fifth longitudinal.

The relative length and position of these three last-
named cross veins is of great value in distinguishing species.
The branched portions of the second and fourth longitudinal
veins and their unbranched portions, outside the transverse
veins, are often spoken of as the anterior and posterior
foicrchettes and their stems, and the spaces enclosed within
the forks are commonly called the fork-cells. The relative
proportions of these are also valuable aids in classification.

Some authors entirely eschew the use of numbers in
describing the longitudinal veins, and employ in their stead
certain special names.

Where such are used, the first is commonly called the
"post-costal," the second the cubital, the third the sub-
marginal, the fourth the marginal, the fifth the brachial,
and the sixth the anal, while Mr. Austen's eighth longi-
tudinal would be spoken of as the axillary.

Ficalbi speaks of the third longitudinal vein as the
*' vena interposita."

The names of the remaining spaces, or " cells," can best
be followed by a careful study of fig. 4.

For the rest, the incisura axillaris, a retiring angle in
the outline of the wing in its axillary border near the
base, and the alula, a lobe appended to the wing between
the axillary incision and its base, alone require special men-
tion. The latter, however, should not be confused with
the covering scale that lies above the halteres, as is a not
uncommon mistake.

The legs of Diptera, like those of other Orders, consists of
four principal parts, the hips or coxce, the thighs or femora,

POSITION AND TERMINOLOGY OF THE CULICID/E 17

the shanks or tihim, and the feet or tarsi. Of these, the
Coxce consists of two joints, the smaller second joint being
called the trochanter ; the femur and tibia each of a single
joint ; and the feet generally of five joints, of which the
first is sometimes called the metatarsus. At the tip of
the last joint are two claws {imgues), and under each of
them there is generally a membranous appendage called the
pulvillus ; many families having, in addition, between the
pulvilli, a third appendage of similar structure called the
epipodium , while in other families this appendage is bristle-
like or altogether wanting. These claws vary greatly in
form in the Gidicidce, and afford valuable specific characters.

Fig. 5— Leg of a Mosquito.

The only remaining point that requires notice is the
meaning of the conventions that are adopted with respect to
the names of species. Two names are always employed ;
the first of these indicates the Genus, always given the
complement of a capital initial ; the second the species, of
which there may be any number in a genus, and, unless the
name refers to a person or a place, commences with a small
letter.

Like the French Eepubhc, the species is " one and indi-
visible." Varieties there may be, though the practice of
making them is usually regarded with disfavour, except
among horticulturists, so that to speak of "a species of
pictus " is apt to raise a smile on the face of the non-medical
naturahst, much as a slip by the latter, in the use of a
2

18 GNATS OR MOSQUITOES — CHAPTER I

medical technicality, would amuse a medical man. Prac-
tically speaking, a third name, that of the authority who
first described the species, is always added ; so that the
system is really trinomial. Not unfrequently a species is
described again and again by different naturalists before it
is discovered that all are referring to one and the same
animal ; and in such cases, however incomplete the des-
cription or inappropriate the name, the custom is to employ
that adopted in the description which has the priority of
date, and the names used in subsequent descriptions are
spoken of as " synonyms," e.g., Mr. Theobald finds that
the species originally described by Fabricus as C. fasciatus,
and referred to in the first edition of this handbook as
G. tceniatus, Meig., has been redescribed under no less than
seventeen other names, all of which appear merely as
"synonyms" in the present edition. If, however, through
error of the original describer, or on account of subsequent
rearrangement, the species is referred to a genus other than
that used by the original authority, the name of the latter
is placed after the specific name in brackets, e.g., Anopheles
hifurcatus (L.), in which case the species was originally
described by Linnseus before the genus Anojyheles was
separated from Culex. I devote some space to these points,
because the failure to adhere to these conventions is a
frequent source of confusion.

19

CHAPTER II.

On Collecting, Preserving and Appliances for Observation.

In his delightful " On the Prowl," our genial Anglo-
Indian naturalist " E. H. A. " remarks : " The gun and net
I would gladly leave behind, but they cannot altogether be
dispensed with. Without a collection a man's knowledge
of natural history becomes nebulous, and his pursuit of it
dilettante. I am sorry it is so, for in spirit I am a Buddhist.
But alas ! every Buddhist is not a Buddha ! " I am glad to
say that "E. H. A." has lately decided to go on the "prowl"
after Mosquitoes, for there is no one more likely to profit-
ably pry out the secrets of their lives, and then, in quaintly
humorous fashion, to hold them up to mingled execration
and ridicule.

The training of medical men specially fits them for such
work, and although its usefulness may not be at once
apparent, sooner or later it becomes of value ; and the
insignificant worm or insect whose habits of life we have
traced, turns out perhaps to be the free stage of a parasite,
dangerous to animals or man ; or a " pest " destroying
crops of incalculable value. It is well nigh on a century
ago since Fabricus, Wiedeman, Desvoidy and Meigen laid
the foundation of our knowledge of the Mosquitoes, but
to-day we reap the fruits of their patient labours, and if the
last-named naturalist were alive to-day he could hardly fail to
congratulate himself on the rare intuition which led him to
denominate his new genus Anopheles by the Greek equiva-
lent for noxious, for we now know that the insects of this
genus are far more efficient checks on over population than
all the armies of the world, armed though they be with all
the products of modern science and ingenuity. Fragile

20 GNATS OR MOSQUITOES — CHAPTER II

though they be, many of Meigen's " types of species of
GuUcidoi are still preserved in very fair condition in the
Jarden des Plantes, though some must be seventy or eighty
years old. Few realise the enormous field of work open to
the collector in all tropical climates, and to say the least,
the pursuit forms a most fascinating hobby, and may be
recommended as a far better insurance against boredom in
old age than the most intricate knowledge of Hoyle and
Cavendish.

As regards the Gulicidce an infinity of work remains to
be done in identifying species with their larvae, in dis-
covering their habits of life, food, and natural enemies, as
well as in discovering how species tide over seasons un-
favourable to their free multiplication, for it is only by work
of this kind that we can hope to ultimately establish an
exact and rational prophylaxis against malaria.

Mosquitoes are rarely found far from water, as during
the first two stages of their existence they are aquatic
animals, and the imagines rarely fly far from the pool in
which they lived as larva?.

The adult insects are found, not only in houses, but in
groves, forests, and in any other situations where shade can
be obtained during the day, while the larvae and pupae are
common in all small collections of water where there is no
strong current. In the hills they are common in pools in
water-courses. They are to be found in all countries from
the Tropics to the Polar regions, and some species have so
wide a distribution as to rival that of man. In all countries
the adults may be found at all seasons of the year, the main-
tenance of the species being mainly secured by the survival
of impregnated females, which hide, and remain quiescent
during seasons unfavourable to the well-being of the larvae.

In searching for them in houses it should be remembered
that it is in the darkest corners that they are most likely to
be found, and that their favourite haunts are draperies,
clothing hanging up, the corners of cupboards, behind
furniture, and such like situations.

A very favourite hiding place, especially of Anopheles, is
the under surface of tables, and indeed these Mosquitoes

COLLECTING, PRESEKVING AND APPLIANCES 21

seem to have a special liking for suspending themselves
from the lower side of horizontal surfaces.

For collecting in the open, the best time is the early
dusk, but in shady gardens many species, notably Stegomyia
fasciatiLS (Fabr.), and S. scutellaris (Walker), are active and
vicious during the day, and in the autumn W\\\ even venture
on short flights across patches of sunshine ; but I have
never met with an Anopheles in the open during the day
in India; though like other Mosquitoes, they will always
leave the house if they can at sunset, and certain European
species are undoubtedly diurnal.

Jungle and garden species, however, commonly harbour
during the day in ditches and under bushes, so that it is
necessary to disturb them by beating the bushes with a
stick.

A number of pill boxes with glass bottoms should be
taken, as the Mosquitoes must be brought home alive, or
else portable setting materials must be carried, as delicate
insects such as these can only be satisfactorily set imme-
diately after killing.

In this sort of shikar nothing in the way of a call-bird
is required, as the method of procedure resembles rather
the plan of big-game shooting known as " sitting up over a
kill," only the sportsman officiates as his own kill, " in
addition to his other duties," as the Gazettes have it.

All that is required is to sit moderately still and the
Mosquitoes will come.

Mosquitoes may be captured : —

(a) By slipping over them a small wide-mouthed bottle,
as they sit on a wall or window, for which purpose a small
"killing bottle" is best, provided that each insect can be
set at once ; or the ingenious trap-bottle, devised by Ficalbi,
may be used. This is constructed on the principle of the
ordinary lobster pot, and consists of a wide-mouthed phial,
through the cork of which is passed the tube of a small
glass funnel, so as to project about finch into the interior
of the bottle. The funnel should be somewhere about 2
ins. long, and about the same width at the rim, and the
narrow part must be fairly wide (about i an inch), so that

22 GNATS OR MOSQUITOES — CHAPTER II

the Mosquito can be driven through it into the bottle
without injury. Once inside the insect is very unhkely to
find its vi^ay out, and the great advantage of the appliance
is that a large number of insects may be captured in it
without risk of the earlier takes escaping, but one would
have to get the funnel specially made.

Fig. 6. — Ficalbi's Trap Bottle, J natural size.

The insects can be killed by dropping through the funnel
a scrap of blotting paper saturated with dilute hydrocyanic
acid, and then covering its mouth with a piece of glass.

(b) By means of a net. — Bend 2 yards of stout iron wire
so as to form a ring 9 in. in diameter, with a handle about 2
ft. long, formed of the two ends twisted together. The net
is a bag 2 ft. deep, secured to the ring, and should be made
of fine silk gauze (chiffon) ; and a strip of cloth should be
wound round the twisted wire of the handle to afford a
more comfortable grip.

(c) By breeding out from larvae and pupae. — The larvae
are found in pools, and in domestic collections of water, and
when undisturbed generally remain at the surface.

Place a score or so of full-grown larvte and pupa?, in the
water in which they have lived, in a tumbler, and tie over
it a covering of gauze supported on a twig or piece of wire
bent into an arch.

In the course of a few days the adult insects will escape
from the pupae and be found in the gauze. They should
not be killed for a day or two, and it is better to introduce
a slice of banana into the net so as to enable them to feed,
and so fill out to their full size.

"When a sufficient number of specimens have appeared

COLLECTING, PRESERVING AND APPLIANCES 23

and been pinned the remaining larvae should be preserved
in a small phial, in rectified spirit, or in 4 per cent,
formaline solution, and marked with a distinguishing letter
or number in order to identify them with the adult pinned
insect. If spirit be used it is well to place the larvae in
dilute spirit for twenty-four hours, and then to change them
to rectified spirit.

The appliances required for collecting are simple and
inexpensive, and, as a matter of fact, are better improvised
at home than obtained from a dealer's, as the articles they
manufacture are, as a rule, unsuitable for such extremely
delicate specimens as those of the CidicidcB, their nets in
particular being rarely fit to deal with anythmg less robust
than the larger Coleoptera, while their cabinets and boxes
rarely have sufficient provision for storing the large amount
of camphor necessary for the safety of the collection in a
tropical climate, unless one is to be continually examining
them and replenishing the supply.

The first step in the preservation of collected specimens
is to kill the Mosquitoes, and for this the best plan is to
employ a " killing bottle," which any one can easily
manufacture for himself.

Those supplied by dealers are always far too large for
small Diptera such as the Culicidce.

Select a wide-mouthed phial about 3^in. high by 2in.
wide, fitted with either a well-fitting cork, or preferably,
with a metal screw-top. Most chemists store such bottles.

In the latter case the disc of cork in the top of the cap
should be removed and replaced with one of thick rubber,
which may be secured in position by means of ordinary
bicycle tyre-repairing cement. Mix equal bulks coarsely
powdered cyanide of potassium and dry plaster of Paris,
and put a depth of fin. in the bottom of the bottle; dust
over this a little dry plaster, and then pour over all ^in. in
depth of liquid plaster of the consistence of cream — when
the plaster has set the bottle is ready for use.

A bottle such as this is very handy for slipping over and
catching sitting Mosquitoes, as in a few seconds the insect
is stupefied, and drops into the bottle uninjured by attempts

24 GNATS OR MOSQUITOES — CHAPTER II

to escape. When the insect has been taken by the net
the bottle is passed into it, and it is easy to shp the bottle
over it as it sits on the gauze. The Mosquito should never
be left in the bottle for more than 90 seconds or it will get
too stiff to be conveniently set, and it should be pinned
immediately.

Another very effectual killing agent is tobacco smoke,
which may be applied by holding a lighted cigarette a few
inches beneath the net and letting the stream of smoke play
over the entangled insect — or by puffing smoke from the
lips into the pill box or bottle, if it has been caught in that
way.

Chloroform is useless for the purpose, as the insects
recover after setting, but a scrap of blotting paper moistened
with dilute hydrocyanic acid and slipped into the pill box
or bottle answers very well.

Practically speaking, the only satisfactory method of
preserving Mosquitoes for identification is to pin them in
the method described below.

It is of course very easy to mount Mosquitoes as micro-
scopic specimens in balsam, or to preserve them in bottles
in spirit, hut such specimens are absolutely useless for iden-
tification, as their coloration depends entirely on the re-
flection of light from the scales with which they are
clothed, and is lost if they be immersed in balsam or any
other fluid ; and on this account, although additional
specimens preserved in spirit are not without their uses,
carefully pinned specimens are the first essential. The
following requisites are required for the work.

^ (1) No. 20 Insect pins : (Obtainable from D. F. Tayler
and Co., New Hall Works, Birmingham). A quarter of an
ounce, costing about half a crown, will last a long time.

' To avoid the difficulty of seeing the point of the pin, with the disc already
on it, Dr. Adolf-Eysell, in the Archiv. fiir Schiffs-und Tropen-Hygiene, Nov.,
1900, p. 354, recommends the use of pins pointed at both ends. Such pins
would have the additional advantage of enabling one to remove the insect
from the disc to examine the ventral surface ; but, if adopted, it would be
necessary to first puncture the disc ; as the pin would certainly bend if forced
through the disc when held above the mosquito.

COLLECTING, PEESERVING AND APPLIANCES 25

(2) Card discs — cut from rather thin cardboard by means
of a 16 or 20 bore gun-punch, on a thick plate of lead.

(3) A small flat piece of cork, covered with white paper,
on which to place the insects while pinning them.

(4) Ordinary toilet pins of medium size.

(5) An insect box. — Any small wooden box, not less
than l^in. deep, may be utilised for the purpose b}?^ covering
the inside of the bottom with a sheet of " cork carpet,"
cork, or solah pith. If intended for transmission by post,
they must be very strongly made, and provided this be
attended to there is not the least need for enclosing them
in an outer box. Those I use are made of wood at least a
third of an inch thick. Their internal dimensions are 6iin.
X 3in. X l^in. The two end pieces are made of hard
wood, the better to hold the nails, but this would not be
necessary if one could get them dovetailed together. One
of these end pieces is narrower than the other so as to
leave a gap of about ^in. wide between it and the top,
and before nailing the latter on, a piece of muslin, about
Sin. wide, is glued across this end of the box, from side
to side and along the upper edge of this shorter end. In
this way, after the top has been secured in its place, a sort
of pocket is left between it and the muslin, the mouth of
which is formed by the gap between the narrower side piece
and the top ; and the inner end of the pocket is closed by
placing over it a thin slip of wood, secured to the inside of
the top of the box by a couple of tacks. This pocket is
then filled with a mixture of camphor and naphthaline, and
its mouth closed by means of a slip of wood forced into the
gap, and held only by friction, so that it can be at any
time readily prized out and the pocket refilled.

This box portion is made to lift off the bottom, which
consists of a simple piece of thin plank, on to which is
tacked a piece of " cork carpet " exactly fitting the internal
plan of the box. As there are no sides to get in the way,
it is far easier to arrange the specimens in position than
in insect boxes of the usual plan. The boxes may be
made deeper, so that they can be utihsed to carry a killing
bottle and other materials on the outward voyage, and some

26 GNATS OR MOSQUITOES— CHAPTER II

have returned from the uttermost ends of the world simply
wrapped in brown paper, and have landed their cargoes in
good condition, so that the plan can be recommended as
having stood the test of rough actual practice.

The insects should be pinned immediately after killing,
as if left for more than a few minutes they get so stiff that
it is difficult to arrange them nicely in position without
damaging the legs. As a rule, it will be found that they die
with the wings nicely spread out at right angles with the
body, so that the insect w^hen shaken out of the bottle on to
the setting cork lies naturally on its back. The process of
pinning is conducted as follows : —

Fig. 7. — To Illustrate the Method of Pinning a Mosquito: Twice
Natural Size.

(1) Take a disc and write on it date and place of collec-
tion — " House," " bites," " sylvan " — or other information ;
also a distinguishing letter if there be several species.

(2) Place the disc, writing upwards, on the piece of cork
and then take an insect pin in a pair of forceps close to the
point and transfix the disc near the middle.

(3) Place the Mosquito on the cork on its back.

(4) Take the pin, with the disc on it, in a pair of forceps
near the head, and holding it so pass the point through the
thorax of the insect between the roots of the legs from
venter to dorsum.

COLLECTING, PRESERVING AND APPLIANCES

27

(5) Pass a common pin through the disc, near the edge,
and force the point of this into the cork at the bottom of
the box.

(6) Spread out and arrange the legs and wings in suitable
position by means of a fine-handled needle.

For handling the pins it is best to obtain a pair of
specially-made entomological forceps, but if such be not
available the "ciliary" forceps used for extracting eyelashes
will serve very well.

In arranging specimens in the box for transmission by
post, care should be taken to force the pins well home into
the cork, and additional pins should be placed between the
contiguous discs so as to prevent their shifting laterally, in
the position shown in the figure.

Fig.

-To Illustrate Method op aeeanging Pinned Insects
FOR Transmission by Post.

a a a, pins carrying the discs ; b b b b, extra pins placed between the
discs to prevent lateral movement.

Another plan is to thrust the disc pins into the cork
obliquely, so that if the disc shift it will rest against the
cork instead of the next specimen ; but this method is not
so efficient as the first.

Insect boxes for keeping a permanent collection are best
obtained from regular dealers, as they require to be nearly
air tight ; but it is well to specify that they should be lined
with " cork carpet," as this holds the pins much better than
sheet cork, and a space ^in. wide should be partitioned off
one end by a slip of wood, to contain a long narrow muslin
bag full of naphthahne or camphor.

28 GNATS OR MOSQUITOES— CHAPTER II

After a few trials it will be found that pinning an insect
in the way above described involves far less trouble than
making it into a microscopic specimen ; but if materials for
pinning be wanting, fairly recognisable specimens may be
made by mounting the insect dry in a deep cell, or in one
of the slides recommended by the late Dr. Carpenter for
mounting foraminifera .

These consist of a slip of deal 3in. X lin. X iV^-> with a
hole |in. in the middle. This perforation forms the wall of
the cell and is closed on both sides with ordinary cover
squares, secured in place by perforated labels, so that the
specimen between the covers can be viewed from either side.
The sides of the perforation should be brushed with creasote
to prevent mildew, and the preparation dried as rapidly as
possible in the sun.

Wings mounted dry as microscopic specimens are, how-
ever, valuable, but when made, great care should be taken
to mark with corresponding letters slide and pinned speci-
men, without which latter such slides are valueless.

Specimens may also be transmitted fairly safely in short
lengths of glass tubing of a size just sufficient to admit the
insect, but too small for it to shake about easily. The tubes
should be simply tied up in a square of muslm, as if sealed
the contents are certain to mildew; but whatever plan
you adopt, ON NO ACCOUNT PACK INSECTS IN
COTTON WOOL, as it is impossible to extricate them from
it without breaking them. Tubes may also be made by
rolling a piece of gummed paper round a pencil and cutting
them to suitable lengths when dry.

Just as mature insects can be obtained from larvae, so it
is generally possible to get larvse from the former ; but a
somewhat larger apparatus is necessary. Take an earthen-
ware dish, at least 1ft. in diameter and 4in. deep, and
fill it with puddle water which has been strained through
muslin to avoid the fallacy of its already containing larvae.
A cover is made for this consisting of a square of thin plank
a few inches wider than the dish, with a large hole occupy-
ing the greater part of its centre. In the four corners are
small holes into which are fixed four small upright sticks

COLLECTING, PRESERVING AND APPLIANCES 29

about 18in. high, so as to form the supports of a miniature
Mosquito net made of gauze or the material known as
" leno," which is made close by means of tin tacks to the
edges of the plank.

The whole thing can be lifted off and on to the dish,
and when in position a Mosquito introduced into the net
is securely confined. The triangular corners of the board
can be utilised to carry banana or syrup as food, or may be
smeared with mud in order to ascertain if the species ever
deposit eggs in such situations. It is best to experiment

Fig. 9. — Apparatus for Breeding-out Insects.

with females that have had a feed of blood ; or in the case
of sylvan gnats with specimens taken in the open, as unless
fully fed they will rarely deposit their eggs. The form of
the egg boats, or groups in which the eggs are deposited,
should be carefully noted and the larvae preserved when
sufficiently grown.

It is rarely necessary to confine males, as most species
couple immediately after escape from the pupa.

The above appliance is also useful for obtaining from
larvae large numbers of individuals for use in observations
on malaria, filariasis, &c. A piece of cardboard is slipped
under the opening so as to close it, and in this way the
contained Mosquitoes can be carried without injury to the
subject of experiment, and liberated under his Mosquito net
by simply removing the card and inverting the net.

By means of a simple apparatus of this sort the life

30 GNATS OE MOSQUITOES — CHAPTER II

history of the genus Cidex can be followed out with great
ease, but this is by no means equally the case with
Anopheles, the species of which are very difficult to main-
tain in captivity, and I have not yet succeeded in rearing
any batch from egg to imago. Very often the females
refuse to deposit their eggs, or when they do so the larvae
rapidly die, and the experiences of numerous correspondents
have been similar. It is difficult to say why this is so, as I
have used vessels holding much more water than many
natural collections I have met with containing Anopheles'
larvae, have had them partly filled with mud from genuine
Anopheles' pools, and tried in every way to as nearly as
possible imitate natural conditions. At any rate, it is
useless to attempt to experiment in test tubes and such-like
small receptacles, and I am inclined to think that the best
plan will be to have made a frame of wire gauze large
enough to cover a natural pool. In the north-western
provinces of India, at least, the scale of such an experiment
need be neither unwieldy nor costly, as here, perhaps, the
commonest natural haunt of Anopheles' larvae are certain
small masonry tanks, one or more of which are to be found
in the garden of almost every bungalow, and these are often
no more than two or three feet square.

For rearing perfect insects from a comparatively limited
number of larvae taken from some natural source, a very
handy little apparatus may be improvised from one of the
ordinary prune bottles fitted with a screwed metal top.
The greater part of the middle of the cap is cut out, taking
care to leave at one point of the circumference a tongue-like
projection to serve as a support for food. A single loop of
wire is soldered on to the sides of the cap, of sufficient
height to serve as a support for the bag of gauze or " leno,"
which when bound with string to the rim of the cap
completes the apparatus. All appliances for observing
captive Anopheles should be well fitted, as they will discover
and creep through the smallest interstice, and much of the
ordinary Mosquito netting sold is quite valueless as a pro-
tection against them.

The study of the habitual resting position of living

COLLECTING, PKESERVING AND APPLIANCES 31

insects is an interesting one, and has attracted considerable
attention lately. With reference to the habitual attitude
of Anopheles as contrasted with Culex, and with the view
of more accurately studying this point, I have devised an
arrangement for photographing living, resting Mosquitoes
in profile. The appliance consists of a frame of thin wood
about lin. deep, and 4jin. x 3jin. in dimensions. This
frame is converted into a box with glass top and bottom
by means of a couple of the glasses of waste quarter-plate
negatives, kept in position by adjustable brass clips. In
each of the two vertical sides, is an oval opening, closed by
a small piece of gauze glued on to the inside of the frame.
When a living Mosquito is introduced into the box it
generally settles on the gauze, and if the box be then placed
in front of a long extension camera so that the edge of the
vertical side of the frame crosses the field of view, one can
generally obtain a satisfactory picture.

10. — Box WITH Glass Front and Back for Photographinc
LIVING Mosquitoes in Profile.

Mosquitoes make excellent sitters, and I have obtained
negatives showing most minute detail, although the exposure
has to be some ten to fifteen seconds, as they are uneasy
and restless in too strong a light. It is useless trying to
get too great magnification, as with any lens of less than
4^in. focal length, placed so as to enlarge about two and
a half times, it is impossible to get all the parts simul-
taneously in focus. For most purposes a sheet of white
paper is the best background, but it is quite possible to get

32

GNATS OR MOSQUITOES — CHAPTER II

negatives showing much detail by reflected light, in which
case a dark background should be substituted.

With the wet collodion process and subsequent enlarge-
ment, it would, I believe, be possible to produce useful
records of the markings, not only of Mosquitoes, but of
other insects.

For certain experiments it may be necessary to trans-
port living Mosquitoes to a distance, and for this purpose
Dr. L. Sambon contrived a method, described as below in
the Brit. Med. Journal, September 29, 1900, p. 949, from
which also are copied the subjoined figures.

Fig. 11.— Case for Mosquitoes. Fig. 12.

-Ventilated Box for forwarding
Mosquitoes.

"When the insects had fed. Dr. L. Sambon, who had
gone to Eome on Experiment No. 2, placed them in small
cylindrical cages made of Mosquito netting stretched on a
wire frame (fig. 11). Four such cylinders were packed in a
well-ventilated box (fig. 12) and forwarded to the London
School of Tropical Medicine through the British Embassy
in Eome. The box was 9in. in depth and 8|in. on the
sides. The wire openings were Sin. square on each side.

By the courtesy of the Postmaster-General they came for-
ward by the Indian mail, so that they arrived in London
some forty-eight hours after leaving Kome. A good many
of the Mosquitoes died on the journey or soon after arrival ;
but a fair proportion survived and appeared to be healthy
and vigorous."

W^ith respect to instrumental outfit, a good hand lens

COLLECTING, PRESERVING AND APPLIANCES 33

magnifying about 10 — 15 diameters is indispensable ; and
for dissecting one must have a simple microscope.

Nothing elaborate is required, and indeed a perfectly
practical dissecting stand can be improvised from a cigar-box
and a scrap of mirror, with a bent wire to hold the lens ;
but, in choosing an instrument, be careful to select one with
broad, gently-sloping wings on which to rest the hands. It
is a mistake to fit it with expensive lenses, as all such, work
very close to the object in proportion to their power, and
room to manipulate is of more importance than excellence
of definition. Two simple plano-convex lenses of lin.
and ^in. focus, are all that is required for the actual dis-
secting, but one modern achromatic lens of high power may
be added, not for working under, but for the purpose of
examining the preparation at intervals.

As regards the compound microscope, the practitioner
in the tropics requires something that he can conveniently
take to the bedside of the patient for the diagnosis of
malaria.

As myself the designer of a folding microscope, I can
with the better grace say, that none of them appear to me
sufficiently rigid for use with high powers ; and therefore
prefer a portable, but not a folding stand, and for the last
year have been using that catalogued by Leitz.

Messrs. Smith and Beck also showed me a somewhat
similar and equally excellent instrument. Both these stands
have sufficient weight and rigidity for use with the highest
powers. For diagnostic purposes the 3mm. Zeiss dry
apochromatic is a most useful lens, but for critical work
on the parasite the highest powers of the best makers are
required, though there is no need to spend additional money
on lenses of exceptional angular aperture.

I give these details, which may seem somewhat out of
place here, as one is often asked what should be recommen-
ded as the most convenient outfit for the tropical medical
practitioner.

Entire larvae are best mounted as microscopic objects in
4 per cent, formol solution, to which a very little glycerine
has been added, but I should doubt if such mounts would
3

34 GNATS OR MOSQUITOES — CHAPTEH II

be very permanent, especially in hot climates, so that it
will be well to preserve a few slides with Farrant's solution.
The slide should be first prepared by forming on it a
shallow cell of Hollis' liquid glue, of such size that the edge
of the cover rests on the middle of the cell-wall, leaving an
edge of the glue outside it. This should be allowed to set,
but not to dry hard before using, so that the edge of the
cover can be imbedded in the semi- solid material. When
all superfluous preservative has been drawn off with blotting-
paper, a ring of the glue is run round the edges of the cover
and the preparation set aside to dry. The sectionising of
perfect insects is even more difficult than that of larvae,
as unless the razor be exceptionally keen, it carries the
dense chitinous covering before it instead of cutting, and
so crushes the internal parts.

There are, however, many structures alike in the larva
and adult insect which can be demonstrated in no other
way than by the method of serial sections, and as already
mentioned, there are especial difficulties in applying this
plan to animals with a chitinous integument. Thin as it
is in most of the species with which we have to deal, it
yet is apt to resist anything but the sharpest of razors, and
what is even worse, is well-nigh impervious to the entry
of preservative and other fluids, so that I have found it
quite impracticable to adopt the plan of staining eii masse
either in the larva or adult insect, and this must hence
be done after the sections have been fixed on the slide.
After some experimentation I find that the following
method may be relied upon to yield satisfactory and well-
preserved preparations. For many of its details I am
indebted to suggestions from Mr. Allen, the Director of
the Marine Biological Laboratory at Plymouth, who has
recently been working on certain copepods, which are not
altogether dissimilar organisms, as far as consistence is
concerned.

The larva or adult insect, as the case may be, is killed
by immersion in a solution consisting of two parts of
alcohol (90 per cent.) to one part of aqueous solution of
perchloride of mercury (1 per mille.), in a test tube, which

COLLECTING. PRESERVING AND APPLIANCES 35

is then gently boiled for a minute or two so as to expel
the air contained within the trachea. As the fluid cools
it is necessarily drawn through the stigmata into the body
of the insect, and is thus at once carried to all its tissues.
It is left in this fluid for a few hours, and is then placed
first in 90 per cent., and finally in absolute alcohol. To
imbed it, it is first placed for at least twenty-four hours in
oil of turpentine and is then imbedded in the usual manner
in paraffin. As the various structures are very loosely con-
nected it is very important to choose a specimen of paraffin
with a melting point suitable to the temperature of the
air of the place in which one happens to be working, for
the least curling of the sections is fatal to the production
of really satisfactory sections, so that it is well to try a
sample of the paraffin in the microtome before employing
it for imbedding. In Europe, a paraffin with a melting
point of about 105" is not at all too soft for the ordinary
temperature of the laboratory, but in the tropics I have
found samples melting from 115° to 125° most generally
useful ; the former for the cold and the latter for the hot
weather. The specimens should be kept for at least six
hours in the bath of melted paraffin, and are then, with due
attention to orientation, placed in the microtome, which, it
is needless to say, should be one of a type constructed to
produce ribbands of serial sections, the ordinary pathological
instrument being quite useless for this purpose.

As the sections are to be stained on the slide, albumen,
and not creasote-shellac, must be used for fixation.

A single drop of Mayer's albumen mixture (equal parts
white of egg and glycerin, with 1 per cent, salicylate of
soda, well beaten up with an egg whisk and filtered) is
added to a watch-glass of water and the sHde is prepared
by brushing over it a liberal allowance of this very dilute
albumen, so that the sections rest on a thin layer of fluid.
AVhen as many of the series as the slide will accommodate
have been arranged in position, the slide is placed on the
warm plate of the imbedding apparatus and warmed just
sufficiently to flatten the paraffin and no more. It is then
placed aside to dry as far as the presence of the glycerin

36 GNATS OR MOSQUITOES— CHAPTER II

in the mixture will allow ; they are then placed on the
warm plate and the paraffin melted for a moment, after
which they are successively passed through batbs of turpen-
tine, absolute alcohol, and 40 per cent, spirit, and are then
ready for staining.

For this purpose I find no dye better than Manson's
methylen blue (borax 5 per cent., methylen blue 2 per cent.,
aqueous solution) ; this is allowed to act for several minutes
and then washed off with water, after which it is well to
give a ground staining of watery solution of eosine or
fuchsine. After staining the slides are passed successively
through baths of 90 per cent, spirit, absolute alcohol, and
turpentine ; and finally mounted in balsam. Gentian violet
also gives good, and Ehrlich's hsematoxylin fair, staining,
but I have not been able to get any result with borax-
carmine. Working in England I *have not been able to
test this plan on infected Mosquitoes, but the perfect way
in which the most delicate tissue elements are preserved,
and the fact that it is so well suited to the use of Manson's
stain, makes it a hopeful method for demonstrating the
parasites in the salivary glands. Great care must be taken
to lose none of the series of sections, as the salivary glands
are so small that they may easily be missed unless the
series be fairly complete.

As the majority of workers at tropical medicine must
necessarily conduct their investigations in places where
there is no gas supply, such as is required for the working
of the ordinary imbedding apparatus and other appliances
involving the employment of self-regulating appliances for
maintaining a constant temperature, it may not be out of
place to describe a simple piece of apparatus for the purpose
which I have used for many years in India. It consists
of a sheet of copper about 15in. long by Sin. broad and
at least xV^- thick. This is supported in a horizontal
position on two wooden feet sufficiently high to admit of
the chimney of a small paraffin lamp being placed under
one end.

In addition to the ordinary copper capsules for containing
the melted paraffin for imbedding, a special long narrow one

COLLECTING, PRESERVING AND APPLIANCES 37

is required. This is filled with a sample of the paraffin
which is selected for use, and is then placed near the middle
of the copper plate, not across, but parallel with the length
of the plate. If the lamp be now lighted and placed under
one of the projecting ends of the plate, its heat is conducted
by the copper to the narrow tray, and it will be found that
a greater or less proportion of its length will become melted.
At the point where the melted and solid portions meet it is
clear that the paraffin is just at its melting point, and
opposite this point are placed the small capsules of paraffin
in which the structures for imbedding are to be placed.
The long narrow tray, in fact, acts as a thermometer, and if
the plate be allowed to reach, so to say, a settled condition
before placing the capsules of tissues on it, it will be found
that the heat of the lamp is quite uniform enough to render
little or no close supervision necessary, and that in a still
atmosphere it may often be left for hours without touching.
I do not, of course, propose such an appliance as a substitute
for the self-regalating one where gas is available, but where,
as in most tropical countries, there is no gas supply, it will
be found to be something more than an inefficient make-
shift, and with a small amount of occasional attention will
yield as good results as can be desired.

With the exception of the wings, which are best mounted
dry, the cover being merely secured by a perforated label,
parts of Mosquitoes are best mounted in balsam, after pass-
ing through absolute alcohol and clove oil.

The subject of the methods of determining the relation
of Mosquitoes to blood parasites is rather outside the scope
of a book like the present, but a few words on the subject
may not be out of place. To infect a Mosquito all that is
necessary is to introduce a few of the species under investi-
gation under a Mosquito netting, beneath which the patient
harbouring the haematozoa is to sleep. The insects should
be females, and to avoid the possible fallacy of their having
been previously infected, it is better, if possible, to employ
for the purpose only insects that have been reared in
captivity from larvae. Although a certain number of con-
firmatory experiments of the infection of the vertebrate host

38 GNATS OR MOSQUITOES — CHAPTER II

by infected insects are desirable, there is no need of such
an experiment to estabhsh the biological relation between
Mosquito and parasite, for if it is found that the parasites
ingested by the Mosquito with the blood of the infected
vertebrate undergo developmental changes, it may be taken
as proved that the species used is capable of acting as
intermediate host, and is therefore dangerous to the verte-
brate from which the infected blood was derived. It is well
to make use of a fair number of insects in each experiment
as all may not bite, and in any case some ten or a dozen will
be required for dissection day by day in order to follow out
the changes undergone by the parasites within the Mosquito.
The method of dissecting out the stomach of the Mosquito
is described elsewhere, and it only ren:iains to add that
the best medium in which to examine is ordinary "normal
saline" solution. If the stomach be very full of blood it is
well to puncture it, and shake it to and fro in a watch-glass
of saline solution, with the point of a needle, or which is
safer, to irrigate it with drops of solution as it lies on the
slides.

For permanent preparations Major Boss found nothing
so suitable as formol solution, as balsam or glycerin render
everything so transparent that all details are lost. Such
experiments are, when negative, conclusive only when con-
ducted at the season of the year at which the parasitic
disease under investigation is rife, as at any other season
it may possibly be merely owing only to unsuitable climatic
conditions that the parasites fail to continue their cycle
within the Mosquito.

PLATE I.

PLATE I. — To Illdsthate the Anatomy of the Laeva.

Fig. 1.— Full-grown larva of Cidex anmdatus (Schrank) : a, respiratory

syphon ; h, swimming fan ; c c c c, anal papilla;.
Fig. 2. — An antenna, more magnified.
Fig. 3. — Respiratory syphon more enlarged, to show : a, the muscles ; b, the

valve-like terminal lobes ; c, the stigma.
Fig. 4. — Segmental respiratory apparatus : a, part of the main longitudinal

trunk ; h, lateral branch ; c, cord by which the cast-off lining of the

tubes is withdrawn during ecdysis.

Fig. 5. — Left mandible, seen from below.

Fig. 6. — Right maxilla, seen from below: a, internal lobe ; b, external lobe ;
c, maxillary palp.

Fig. 7. — Labium, with the lower part of the pharynx, seen from above:

a, the labrum itself.
Fig. 8. — One of the natatory compound bristles, much magnified.
Fig. 9. — Thorax of larva of C. pipiens, to show : h, the hepatic masses ;

i, intestine ; t, main longitudinal trachete.
It will be noticed that in this species the thoracic dilatations of the main

tracheae are by no means so marked as in G. anmdatus (Schi-ank), and

the whole respiratory system is less developed.
This plate is partly based on Meinert's figures, and partly original.

40

CHAPTEE III.

The Anatomy of the Larva.

From a purely anatomical point of view a good deal has
been written on this subject, but on the descriptive side
comparatively little has been recorded, so that except in
the case of a few of the commonest species, we are quite
without adequate descriptions whereby Ciilex larvae may
be distinguished among themselves. In by far the greater
number of recorded species the larvaB have never been
recognised, and still less described. It is obvious, however,
that as these insects can be much more easily destroyed in
large numbers in the larval stage, by insecticides, or by
the filling up of pools, than they can be as flying insects,
the accurate description of the larvae of noxious species
has become a matter of great importance ; but it is un-
fortunately one that can, as yet, be hardly said to be
commenced.

Observations on this subject, by "breeding out" adult
insects from larvae, require much time, and, above all, a
settled residence, and the work of the Sanitary Com-
missioner of an Indian province is of a kind that precludes
either requirement, so that my present visit to India has
been very unfruitful in results in this direction, and I still
can only offer a few general suggestions for future work.

As regards the genus Culex, I believe that the pro-
portional length of the breathing tube, as compared with
that of some comparatively fixed portion of tlie body, such
as the head or thorax, will be found to yield valuable specific
characters.

I was at first inclined to think that this proportion
varied greatly with the age of the larvae, but found subse-

' THE ANATOMY OF THE LARVA 41

quently that this was a mistake due to the occurrence of
several species in the same pool, and that a ratio of this
sort remains fairly constant throughout the growth of the
larvae. It is easy, for example, to distinguish between the
larvae of C. fatigans (Wied), and those of the species I
identify as C. impellens (Walker), which constantly are
found associated here in the cold weather, the breathing
tube of the former being of medium length, while that of
the latter is very short ; and I have met with larvae in
which the syphon is so short that they may easily be
confused with those of Anopheles, the more so as they
not unnaturally assume much the same position in the
water ; but I was unable to delay my departure in order to
ascertain the species to which they belonged, but I believe
they were those of a Stegomyia.

In Anopheles, on the other hand, apart from size, the
larvae of different species are most difficult to distinguish.
Their coloration is singularly alike, and as yet I could not
undertake to distinguish those of any individual species,
though they differ considerably in the matter of the size
of the anal tubercles.

Quite recently Grassi {G.S.Z., p. 81) has shown that
the bristles with which these larvae are provided, in various
situations, differ sufficiently in the various species to be
made use of to afford definite indications of the species to
which a larva belongs. These bristles are often of very
complex structure, and this is especially the case with those
on the dorsum of the thorax and head. Confining his
attention to those of the latter region, he finds that the
characters of two special bristles, viz., the sub-median pair,
in the middle of the fore border of the head, and those
placed at its external angles, suffice to easily distinguish
the four European species of Anopheles larvae known to
him, and on comparing his figures with the corresponding
bristles of the larvae of An. Eossii, I find that in that species
too these bristles again differ from any of those. It will be
observed that the difference between the bristles of An.
Eossii, as compared with those of An. superpictns, is very
marked, and is alone quite sufficient to set at rest any

42

GNATS OR MOSQUITOES — CHAPTER III

doubts as to the absolute distinctness of the species in
question.

In addition to the above, there are a variety of other
compound bristles, notably those of the swimming fans,
which might be made use of in distinguishing species, but,
with the above exception, we can go no further than the
recognition of genera, and of some even of these we are
quite ignorant of the characters of the larvae.

Mamlifiainis. ^t/unalus

Fig. 13. — Tbe median aud angular bristles of the fore margin of the
clypeus of five species of AnopJielcs. Copied from Grassi, with the exception
of the figure of the bristles of An. Bossii. (Original.)

Of the larvae of Megarhina, Pso7'ophorc( , and Mucidus we
know nothing. Those of Uranotrc?iia are said by Dr. Lutz
to have a very short syphon, and to look much like those of
Anopheles, resting in the water m an oblique position. In
Mr. Theobald's new genus Stegomyia, S. pseudotceniatus
and S.fasciatus certainly have very short syphons, and the
same was the case in a larva I met with in Allahabad in
August, and which, though I was unable to breed out, was
almost certainly that of S. gicbernatoris ; so that probably
the larvae of this genus are characterised by the possession
of short breathing tubes. The same, however, is the case
with certain Culices, notably of C. pulcliriventer, the syphon
of which is no larger than one of its anal tubercles.

The larvae of all the Culicidce are aquatic, and are
encephalous, i.e., have a well-developed head. When just
hatched, the larvae are of just sufficient size to be easily

THE ANATOMY OF THE LARVA 43

seen by the naked eye, and are at this stage so transparent,
that almost all details of their anatomy can be made out in
the living larva placed under the microscope in a little
water. If all superfluous water be carefully removed from
beneath the cover, by means of a strip of absorbent paper,
so as to slightly press upon without crushing the larva, its
naturally vivacious movements will be sufficiently restrained
to admit of its being observed at leisure.

The larvae are easily recognised by their vivacious
wrigghng movements, as minute worm-like bodies, with a
disproportionately large head with a pair of prominent black
eyes, and at first sight what appears to be a vertically bifur-
cated tail. Being quite unprovided with legs, they swim by
means of the strokes of a tail armed with large expansions
of bristles. When more closely examined, it is seen that,
like the adult insect, it consists of three well-defined
regions, the head, thorax, and abdomen. Commencing with
the head, it is seen to be not quite so wide as the thorax,
bat considerably wider than the abdomen, and forms a
truncated cone, wider than it is long, and separated from
the thorax by a distinct sulcus. It is ornamented on the
dorsum with several patches and lines of pigment, the
arrangement of which, however, is not very constant ; and
at the broadest part laterally are placed the two large
eyes. It is further ornamented ni various places with tufts
of strong bristles, the structure of which is often very
complex.

From two slight prominences a little in front of the
eyes spring the antennae, and projecting from the middle
of the anterior border of the head may be seen a complex
arrangement of bristles springing from the upper lip and
mouth parts.

The greater part of the upper surface of the head is
formed by the dorsal surface of the third metamere. In
front of this is the clypeus or dorsal plate of the second
metamere, a short but broad plate, with a shallow, curved
indentation in front, while laterally it is armed with a pair
of incurved tufts of bristles which form what Meinert (De
Encephale Mygelarver) speaks of as a whorl, or rotatory

44 GNATS OR MOSQUITOES — CHAPTER III

organ ; as he believes that it is by the vibrations of these
bristles that nutritive particles are directed into the mouth.
Anterior again to the clypeus is a small, rounded, median
prominence, the labrum or dorsal plate of the first
metamere.

The eyes are large and placed laterally, forming a sector
of about 150°, and close behind them may be distinguished
a minute ocellus, generally of oval form. The antennae
spring from the antero-external corners of the third meta-
mere, and form a pair of short, curved horns, fairly freely
movable on their basal articulation with the head, but
elsewhere rather rigid. The greater part of the organ
consists of a rather stout basal joint, which is provided at
the inner side of its distal extremity with a tuft of strong,
compound bristles.

The next joint is less than half the length of the basal,
and is distally armed with a few long, stiff bristles, while,
like the basal joint, it is beset throughout its length with
stiff, short spines. The last portion of the antenna or
flagellum is very minute, although there are indications
that it is in reality composed of three very short articula-
tions. Besides the flagellum there are also attached to the
end of the second joint two peculiarly-formed processes
or jointed hairs, which are almost certainly sense organs of
some sort, and are most probably olfactory organs.

In general form the thorax forms a sort of six-sided box,
and is somewhat larger than the head in all dimensions.

Although no sutures can be distinguished on its surface,
its division into its three component segments is sufficiently
indicated in its outline, and by three pairs of lateral tufts
of bristles, which are longer and stouter than those of any
other region of the body. The component hairs of these
tufts are all compound, each being clothed throughout its
length with filaments of considerable proportional length ;
and each tuft springs from a nipple-shaped tubercle, which
appears to be capable of a certain amount of voluntary
movement, though, from its position, it is obviously not
the rudiment of a leg. These tufts appear to act by way
of lateral keels, whereby the larva is maintained in any

THE ANATOMY OF THE LARVA 45

position without being made to revolve on its aixis under
the influence of the strokes of the tail. In addition to
these principal tufts, other smaller but also pedunculated
tufts spring from the sides of the terga of the segments,
the characters of which should be accurately noted in
describing species, as it is probable that their form, number
and position, may yield good specific characters. The great
preponderance of the mesothoracic segment is already well
marked, though not to anything like the same extent as
in the imago.

Proceeding with the examination of the living larva, the
first thing that catches the eye in the thorax are four pairs
of what at first sight appear to be pigment spots. A little
watching, however, suffices to show that these are not really
external markings, but internal organs seen through the
transparent carapace, and that their position is constantly
altering under the action of the surrounding thoracic
muscles, and of the pulsations of the dorsal vessel. Ex-
amined with a somewhat higher power, they are seen to be
glandular bodies of a sacular form, lined with secretory
cells, their cavities being filled with a clear fluid, holding in
suspension a quantity of deep brown granular matter. These
glands are connected with the upper end of the intestine,
and are probably hepatic in function. In the dead, or
preserved and mounted larva, they soon become invisible,
owing to the discharge of the contained brown matter into
the intestine.

These glands are arranged in two sets, an internal and
an external, the former of which consists of two glands
placed so closely together as often to look like a single
mass, and situated nearly in the middle of the thorax, close
to the intestinal canal. The other two pairs are placed at
a distance from the intestine, in the corners of the pro-
and metathorax respectively. In the middle line the dorsal
vessel can be seen pulsating, the action being rather that
of a peristaltic wave than a true systole, the heart being
here a long, valved tube, corresponding rather to an aorta
than to a heart in the usual acceptation of the word.

More or less in the middle line, too, its coarser image

46 GNATS OR MOSQUITOES — CHAPTER III

often obliterating that of the dehcate, superjacet dorsal
vessel, is the stomach, or chylific ventricle, its opacity
varying with the fulness of its contents ; and on either
side may be seen the dilated thoracic portions of the
respiratory trachese, which are easily distinguishable by the
spiral thickening of their chitinous lining, while branches
of the tracheal system to the various organs can be traced
in various directions.

The abdomen is between three and four times as long
as the thorax, but narrower, and consists of nine segments,
each of which is provided with a large tuft of bristles as well
as with numerous hairs, either single or grouped, on the
dorsal and ventral surfaces. The cardiac, intestinal, and
respiratory tubes can all be traced through the greater part
of its length, and the Malphigian tubes, and other intestinal
appendages, can all be made out, as the larva takes favour-
able postures. The nervous system, however, is completely
hidden, and owing to its transparency during life, cannot
be seen even if the animal be placed in the supine position.

In the genera Cidex and Mochlonyx, there springs from
the dorsum of the eighth segment a large process, at least as
wide and generally about two or three times as long as the
remaining segment of the abdomen, and into this the two
main respiratory trunks can easily be followed, and are seen
to open at its extremity by means of curiously guarded open-
ings. At the root of this breathing horn are a pair of rather
short but dense tufts of hairs. The last segment contains
the rectum and carries the anal tubercles, the anus being
placed almost at the very extremity of the body, but rather
towards its ventral aspect. Around the opening are two
pairs of delicate, leaf-like expansions, each furnished with
a branching twig of trachea, the lower pair being some-
what the larger. They probably act as gills, and subserve
respiration during the periods when the larva is completely
submerged, which, when the weather is cool, are often
somewhat protracted. On either side, too, but originating.
a little in front of these anal tubercles, are a pair of large
dense tufts of compound hairs, which are employed in
swimming much in the same way as a fish's tail, and are
so arranged as to form an expansion of similar shape.

THE ANATOMY OF THE LARVA 47

The internal organs are of typically insect plan, and
will be most easily understood by considering each system
separately.

Digestive System. — Apart from the gnathites or foot-jaws,
which, although of the ordinary masticatory type are rather
complex organs, the intestinal canal is of the simplest form,
consisting of a narrow oesophagus, which leads into a wide,
perfectly straight tube, commonly spoken of as the stomach,
which extends to the end of the sixth abdominal segment.
Into its thoracic end open the ducts of the hepatic glands,
which, when examined in section, are seen to be broad
crypts communicating with the • stomach by so wide an
opening as to almost appear as diverticulse ; and into its
distal, where the tube contracts before the commencement
of the rectum, open the five Malphigian tubules, which are
slender tubular glands, differing in no way from those of the
adult. Lying close to the oesophagus in the head are the
two salivary glands, but they are not easy to demonstrate,
except in section, and the same remark applies to the rectal
glands, which lie beside the rectum in the last segment.
The mouth parts consist of an upper lip or labrum, in the
form of a convex crenated plate, armed with numerous
hairs, and a longer and narrower lower lip, or labium, of
somewhat similar structure. Between these are placed the
two pairs of foot-jaws, the mandibles above and the maxillse
below, feach mandible consists of a somewhat pyriform
plate, the wider end of which forms an articulation with the
lateral structures of the mouth ; while the narrower end is
formed into a rather complex set of dentations divided into
two groups, the anterior of which are small and claw-like,
and mainly adapted for holding the prey, while the hinder
set are better adapted for cutting and crushing. The
appendage is further provided with brushes of peculiarly
formed hairs, and has, about the middle of the anterior
border, two large jointed processes or hairs, which probably
are tactile or gustatory in function. The maxilla is a plate
of somewhat quadrangular outline, the anterior border
being curved and its corners rounded off. It is richly pro-
vided with hairs, some of which have a curiously compound

48 GNATS OK MOSQUITOES — CHAPTER III

structure, and from its inner and posterior corner there
projects a short, three-jointed maxillary palp in the form of
a truncated cone, the extremity of which is armed with
minute spines. The above description should suftice to
enable the dissector to recognise the various parts, but it
would be wasted labour to enter into minuter detail in a
general description, as the form and arrangement of the
various parts differ somewhat in the various species, and
the minutiae will best be studied by teazing out the parts
from the head of whatever larva happens to be available,
with needles, under the simple microscope.

Bespiratonj System.-^Thm is very highly developed in
these larvae, and presents many peculiarities which differ
widely in the different genera ; and hence the remarks
below must be understood to apply to Culex only. The two
main longitudinal tracheae are in all, however, of such large
size that they may be considered rather as elongated air
sacs than as mere tubes for the conveyance of air to the
tissues.

Doubtless they serve a double function; namely, as
receptacles for the storage of air for use during periods of
complete immersion, and as hydrostatic organs to secure a
proper degree of buoyancy. It is only the two main longi-
tudinal trunks that are so largely developed, the remainder
of the tracheal system being of no more than the usual
complexity in organisms of this sort. Each of these main
trunks commences quite abruptly in the pro thorax, and,
rapidly increasing in diameter, attains its largest dimen-
sions in the mesothorax, where, in some species, e.g., in C.
nemorosus, there is a marked local dilatation ; in G. pipieiis,
however, there is no mesothoracic dilatation, and the tube
remains of nearly the same diameter throughout the entire
length of the abdomen to its termination at the end of the
peculiar dorsal process of the eighth segment. The anas-
tomoses between the two main stems are trifling, there
being only two transverse connections placed in the pro-
and mesothorax respectively before the tubes have attained
their maximum diameter. The anterior of these two cross-
branches is considerably the larger, and immediately in

THE ANATOMY OF THE LARVA

49

front of it, each trunk splits up into a number of branches,
which enter the head and are distributed to the various
organs contained ia it. The large anterior cross branch
gives off no branches, but the hinder one supplies some
twigs to the wall of the intestinal canal, and to the hepatic
glands.

In the larvae of An. Bossii, however, and presumably of
other species of that genus, the communication between the
two main trunks is extremely free, as not only are they
continued beyond the end of the thoracic dilatation to unite
with each other in the middle line in the form of an arch of
nearly uniform diameter, but a smaller anterior branch
unites with its fellows to form a second concentric arch, of

Fig. 14. — Larva of Anopheles Rossii, seen from in front, to show the
tracheal arches, and their anastomosing loops.

but little less calibre. Moreover, as will be seen in the
figure, each arch gives off eight branches, which anastomose
with each other to form a series of loops, four on each side,
those of the larger arch being directed tangentially, while
those of the smaller run forwards ; so that, in addition to
the transverse branches described above, there are no less
than four channels of communication between the main
trunks of opposite sides.

From each of the main trunks in the thorax there
issue also three rather large branches, one for each seg-
4

50 GNATS OR MOSQUITOES — CHAPTER III

inent, which divide up into branches that pass to the con-
tained muscles and viscera. Between these lateral thoracic
branches I have not been able to make out any longitudinal
communication ; but like the lateral abdominal branches they
extend as pervious tubes nearly to the skin of the larva, and
are continued to the latter as an impervious chitinous cord,
surrounded by true dermic tissue, to actually blend with the
external chitinous covering of the larva. In the abdominal
region the arrangement is essentially the same, only here
there is a continuous longitudinal anastomosis between the
lateral branches, one of which issues from the main trunk
of all but the last segment, and after a short course divides
into three main branches — an anterior to anastomose with
the posterior branch of the lateral trachea of the segment
in front of it, a posterior to communicate with that behind
it, and a continuation, which after distributing branches
of air supply is continued to the skin, first as a pervious
tube and then as an impervious cord, exactly as in the
thorax. These cords represent the rudiments of the future
segmental stigmatic trunks of the imago, and it is by
their agency that the discarded lining of the portion of the
tracheal system belonging to the segment is withdrawn
from the body whenever the larva undergoes its periodical
change of skin. From these three principal segmental
branches there issue branchlets which carry air to the
various muscles and viscera of the segment. These lateral
stigmatic cords can best be studied in the abdominal region ;
and if a cast skin, such as will be found floating in abund-
ance on the surface of any vessel in which larvae are kept,
be carefully examined, it will be seen that the main longi-
tudinal air sac has broken up into segmental lengths, and
that each piece is attached by its cord to the skin of the
<".orresponding segment ; for it must be remembered that
not only the outer skin but the whole of the lining of the
tracheal system is thrown off at each ecdysis. The struc-
ture of the peculiar dorsal horn of the eighth segment
remains to be described. Each of the two main stems
passes up through it separately, to end in a pair of stigmata
at its extremity, whose openings are considerably smaller

THE ANATOMY OF THE LARVA 51

than the diameter of the tube, being placed in the centre
of a small circular plate supported by chitinous rays. The
mechanism by which these openings are protected from
the entry of water is rather complex and consists of a sort
of valve, formed by five flap-like lobes at the end of the
process. These flaps consist of two pairs and a small
unpaired lobe, the larger of the two pairs being placed
opposite the small unpaired one. Between the two air
tubes and around them are a number of muscular strands,
which originate in the eighth segment and extend through
the process to be inserted into the bases of the flaps, so
that, when they contract, the stigmata are drawn down
somewhat within the process, and the valves close over
them.

Vascular System. — This consists of a delicate, long, wide
tube, which extends along the dorsal aspect of the body,
from the hinder part of the head to the very extremity of
the abdomen. Its structure is so delicate that, except in the
living larva, where its shape and position can be followed by
its movements, it is extremely difficult to make out. It
consists essentially of a long, thin-walled tube, with valvular
constrictions opposite the incisurae, communicating freely
with the perivisceral spaces by means of cribriform aper-
tures, which, however, are anything but easy to demonstrate.
There is no system of peripheral vessels at all comparable
with those of the higher animals, the function of the organ
being to keep in motion the perivisceral fluid, which is
thus kept flowing over the tracheae, which lie free in the
perivisceral spaces between the various organs and tissue
elements. It acts, in fact, rather as a churn than as a
pump. In front of the thorax it extends as a narrower
vessel into the head, and this portion is sometimes spoken
of as the " aorta," but I look upon this term as a misnomer,
as, save in diameter, the extension differs in no way from
the rest of the dorsal vessel, is quite unprovided with
branches, and appears at its anterior end to lose itself in
the general interstitial lymph spaces of the head. The
circulating fluid is colourless and contains but few cellular
elements, those that are present being of the nature of

52 GNATS OR MOSQUITOES — CHAPTER III

leucocytes, so that while fulfilHng the functions of both,
the fluid is rather of the nature of lymph than blood.

Nervous System. — The demonstration of the nervous
system can only be made by dissection, as its elements are
too delicate to be distinguished from the overlying intestine,
even in the supine position, and for this purpose specimens
hardened in alcohol are best. Even thus it is not altogether
easy to follow out any length of the chain, and for those
who will be satisfied to examine a closely-allied larva, it is
recommended to dissect the larva of one of the Epherneridce,
in which the nerve cord can be separated with singular ease.
These larvae are very common in grassy pools and small
running streams, and may be recognised by their long
caudal bristles, which give them the appearance of possess-
ing a long forked tail. When one has become accustomed
to what to expect to see by practice on these easier subjects,
it will be found easy enough to teaze out the nerve cord in
Culex larvae. In the head, however, this is scarcely practic-
able by the method of teazing, and sectionising must be
resorted to. Taking the young larva as a type, a pair of
ganglia can be demonstrated for each segment, and on
emerging from the thorax the lateral cords separate to pass
backwards across the oesophagus, to combine behind it in
the large superoesophageal ganglion or brain. From this
mass filaments are given off to the eyes and antennae, and
from it, as well as from the ganglia of the segmental chain,
fibres pass to the corresponding muscles and to the peri-
phery, sensory filaments having been actually traced into
the bases of the hairs. At the time of pupation the location
of the gangha undergoes changes of startling rapidity.

Mr. F. V. Theobald notes that in a few minutes prior
to the escape of the pupa from the larval skin the first
abdominal ganglia come to lie in the posterior part of the
thorax, and during pupal life the changes are equally
rapid. In four days the fore-brain increases tenfold in
bulk, the first abdominal ganglia fuse with the three
thoracic pairs, and about the same period the eighth pair
shift forward and fuse with the seventh ganglia, and in
the $ , but not in the c? , the double mass so formed

THE ANATOMY OF THE LARVA 53

shifts into the sixth segment. This last alteration is per-
formed with remarkable rapidity during the few minutes
the imago takes in emerging from the pupa case.

Organs of Sense. — The eye in the encephalous larva is
usually stated to be of the " simple type consisting of a
group of ocelli with lens and retinal expansion."

This sort of statement has been copied from one text-book
to another till it has become stereotyped, but as a matter of
fact, there is neither lens nor retinal expansion, properly
so-called, and the eye appears to be rather a transition stage
in the development of the compound eye of the imago, than
a structure in any way like the ocelli of adult insects, or the
eyes of spiders or molluscs.

In full-grown larvae of Ciilex pipieus, the eyes, although
distinctly separated by an unpigmented gap, form for all
practical purposes a single visual organ, the thick but
perfectly transparent cuticle forming an unbroken spherical
curve over both eyes, and although this is somewhat thicker
in the middle line opposite the separation between the
deeper parts of the eyes, there is nothing whatever in the
form of a lens, at any rate in the optical sense of the word,
as not only is its internal limit formed by the irregular sur-
face of the pigment, but even taking this as a regular surface
there is so little difference in the depths of the anterior
and posterior curves that any image formed by it would fall
nowhere near the visual nerve endings, but somewhere in the
animal's thorax. If we examine a not too thin radial section
of the organ it will be seen that it consists of a number
of conical masses of pigment, the combined bases of which
form the inner boundary of the transparent cuticular layer
of the eye, and into the apex of each may be traced a fibre of
the optic nerve springing from a bilobed mass of ganglion
cells, almost in contact with, but yet distinct from, the large
lobes of the cerebral mass.

Selecting a thinner section we find that each nerve fibril,
shortly after starting from the optic ganglion, begins to
acquire a rapidly thickening covering of pigment granules,
and that as soon as this has become sufficiently bulky to
merge with the neighbouring sheaths of pigment to form

54 GNATS OR MOSQUITOES — CHAPTER III

the base of the great pigment mass of the eye, the fibre
expands into a spindle-shaped body provided with a distinct
nucleus. Beyond this the spindle-shaped body contracts
into a rod-like structure, the actual ending of which I have
not been able to trace, though I think it simply ends in the
midst of the transparent contents of the tubular visual
elements to be presently described. A further examination
of radial sections shows that the pigment-covered sensory
cones each consist of a deep portion enclosing the spindle-
shaped bodies, which may or may not be radial in direction,
and may even be curved, and an outer portion containing
the rod- like nerve end, which is always truly radial. If we
now turn to the examination of tangential sections, we find
that the superficial layer of the general pigmented mass
consists of a number of cylindrical prolongations of the
transparent superficial layer of the eye, each enclosed in a
layer of pigment, which is very thick at the surface and
becomes gradually thinner as the deeper layer containing
the spindle-shaped bodies is approached.

In all sections that are sufficiently truly tangential to
afford a clear image, the contained terminal rod can clearly
be seen as a well-stained dot exactly in the middle of the
transparent contents of the tube of pigment ; but whether it
extends through the entire length of the tube, or ends some-
where during its course, is more than I can say ; in any case,
however, it extends along it for some considerable portion of
its length. From what has been said it is clear that the
terminal rods can receive only rays the direction of which
is parallel to the radius of the sphere of the eye coinciding
with the transparent axis of the pigment-clothed visual
element, and that such rays will reach this end rod and
no other, so that although there be no dioptric apparatus,
such as is found in each visual element of the facetted eye of
the adult insect, for the purpose of concentrating on the con-
tained rod the pencil of rays entering each transparent
cyHnder, it is, nevertheless, like such eyes, suited only for
mosaic vision, and is for such a purpose only less efficient in
so far as in that a smaller portion of the entering pencil of
rays will actually reach the end-rod placed in the axis of the

THE ANATOMY OF THE LARVA

55

transparent cjdinder. Indeed, to complete the development
of the eye of the imago no great alteration of the deeper
parts is required, but only the modification of the trans-
parent superficial layer of the eye into the beautiful system
of miniature dioptric systems, one for each end-rod, which
forms the most striking characteristic of the facetted com-

...*^MfS i^.

a.

■ . :-^.^ -'

m^

'^^^

^j^dS'fii&'^^ijji^ii 1 i.ijyjfMf^Hf

•aSp

uBKKUr^''.^

>

'^m

Hii^

^

^^Sffl^^

WTe.

---.=JtT«

d.

Fig. 15. — To Illustrate the Anatomy of the Eye op the Larva.

a. Radial section through anterior part of the head of the larva of Culex
pipiens, cutting through both eyes and the optic ganglia ; b, semi-diagram-
matic representation of three of the visual elements of the above more highly-
magnified ; c, portion of tangential section of the outer layer of the pigmented
portion of the eye.

pound eye. I have gone into somewhat more detail than
usual in this point, because I have nowhere met with any
account of the visual, organs of these larvae which appears
to be based on actual observation of the family, most
authors being apparently content to assume that their eyes
must needs resemble the visual apparatus of other and
often very distant groups of invertebrata. It is clear that

56 GNATS OE MOSQUITOES — CHAPTER III

in such an eye the mosaic must be a very coarse one, and
that, except in the case of very near objects, the impression
gained of their form must be extremely fragmentary and
ill-defined.

Hearing. — As regards the organ of hearing it happens
that in one member of the family, viz., in the larva of
Corethra, an organ as to the auditory nature of which there
can be little doubt has been studied in some detail. In
this genus the larvae are so transparent that they are known
as glass larvaB, and hence can be studied with exceptional
advantage, and the organ which is situated in the eighth
segment is described and figured by Professor Lubbock in
his popular work on the " Senses of Animals." In this form
the ganglion which is placed in the anterior part of the
segment gives off a branch, the auditory nerve, which after
a short course outward expands into a small auditory
ganglion from which a sheath containing two or three
auditory rods passes outwards and backwards to the skin.

The auditory ganglion and end-organ are further sup-
ported and kept in a state of uniform tension by a ligament
which runs in an opposite direction from the ganglion to the
skin at the anterior part of the segment.

The organs of smell and taste are probably situated in
the antennae and maxillary palpi and other parts of the
mouth respectively, but I am not aware of any work on
this subject bearing on the CuUcidcc in particular.

PLATE 11.

PLATE II. — To Show the Characteristics of the Larv.e op
Different Genera.

Fig. 1. — The larva of Anopheles maciilipennis.

Pig. 2. — Hinder extremity of the same, more enlarged.

Fig. .3. — The larva of Anopheles nirjripes, to show the arrangement of the

respiratory system.
Fig. 4. — Pupa of Anopheles macidipennis.
Ffg. 5. — One of the natatory bristles.
Fig. 6. — The larva of Corethra pliimicomis.
Fig. 7. — The pupa of Corethra plumicornis.
Fig. 8. —Embryo of Corethra plumicornis within the ovum.
Fig. 9.— The larva of JJochlonyx culiciformis.
Fig. 10. — The pupa of Mochlonyx culiciformis.

58

CHAPTEE IV.

The Generic Characters of the Larvae of the Culicidae.

The number of species which have been followed
through their complete metamorphoses is, as already re-
marked, very small. In the case of the genus Megarlmia,
I can find no record of any observation whatever on this
subject, while in ^des the only note met with is one by
Osten-Sacken, on jS^des fascus, an American species, and this
is of the most cursory description, all that is said being that
they exactly resemble the larvae of Culex, except that they
are smaller. Putting then aside Megarhina, the larvae of the
various genera may be divided into two categories, according
as to whether they possess a dorsal respiratory process to
the eighth abdominal segment or not. In the first category
are Culex, J^des, and Mochlomjx ; in the second Anopheles,
and Corethra. Amongst those possessing the respiratory
dorsal process, the larva of Culex has already been suffi-
ciently described : of ^des nothing more can be said, and
of those sufficiently described, only the larvae of Mochlonyx
remain to be described. I have actually handled the larvae
of Culex, Stegomyia, and Anopheles only, and the notes
given below are almost entirely a precis of Fr. Meinert's
paper, " De Encephale Myggelarver " (Vidensk Selsk., 6,
Raekke, Naturvidensk, og mathem. Afd. iii., 4).

Genus Mochlomjx. — Meinert's remarks apply in especial
to M. culiciformis, which formed the subject of his in-
vestigations. The full-grown larva is of a light brown
colour, the tracheae and air sacs often showing through the
skin with a golden lustre. On the dorsum of the thorax
are several small, dull white spots, and the pleurae and
venter are whitish. After each change of skin the colour

GENERIC CHARACTERS OF LARVAE OF CULICID.E b'i)

is almost white, and this hghter coloration persists longer
on the head and anal tube than elsewhere, so that the
newly-dressed larva presents quite a distinctive appearance.

In many points these larvae are intermediate in form
between those of Culex and Corethra, but the head is most
like that of Anopheles, forming like it a truncated cone, but
differing in being pinched in m front of the eyes, so as to
present a pyriform outline ; and seen from the side the
difference is even greater, as the tergal plate of the second
metamere is bent downwards and then backwards, so that
the anterior part of the clypeus overhangs the mouth. The
tergum of the third metamere is cordate, being deeply in-
curved in front for the reception of the hinder part of the
clypeus. The eyes are placed well back on the broadest
part of the head, and are of oval form, with the long
diameter transverse. The antennae consist of a single
joint, are moderately long and thick, and when at rest are
directed downwards so as to be scarcely visible from above :
they are provided with several jointed bristles, some of
which are longer than the antenna itself. The labrum is
almost rudimentary.

The mandibles are strong and trenchant, and in addi-
tion to their teeth, carry a variety of simple and compound
hairs, some of which are of very specialised forms. The
maxillae are short broad plates, with a sinuous anterior
border and stumpy, conical palps. They and the lower
lip, in addition to ordinary hairs, are provided with several
ranks of peculiar flattened bristles with truncated dentate
ends.

On the maxilla, the inner face of the external sinuosity
is provided with a single row of these peculiar plates,
but on the labrum, which is transversely oval, there are
three, and on the lateral plate of the metamere, with
which the maxilla articulates, no less than five ranks of
different lengths. They look as if they were designed to act
as a sort of sieve to exclude too coarse materials from the
mouth, but, as the larva is predatory and subsists mainly
on small crustaceans, they are more probably retentive
organs. The thorax is relatively very large, somewhat

GO GNATS OR MOSQUITOES — CHAPTER IV

flattened, and no distinction between its component seg-
ments is visible. The lateral tufts of balancing bristles
which are so marked in Ciilex and Anopheles, are ill-
developed. The abdomen is cylindrical, slender, and of
nearly uniform diameter as far as the seventh segment,
the segments also progressively increasing in length up
to this point, while the last two are of insignificant
dimensions. The seventh, which contains the two large
posterior air sacs, is of very exceptional size, being at
least as long as any other two segments. The eighth
segment, as in Culex, bears on its dorsum the respiratory
process,, but this is quite small as compared with the
organ in most members of that genus, and is of conical
form. The ninth segment carries four small, slender anal
tubercles, and a pair of swimming fans composed of closely-
arranged, multifid bristles. The tracheal system is peculiar,
combining many characters of these organs in Culex and
Corethra. The main longitudinal trunks have the same
course and general distribution as in the former genus,
but are quite small, the hydrostatic function being fulfilled
by two large pairs of dilatations or air sacs, which are
situated in the thorax and seventh segment respectively.
All four, and especially the thoracic sacs, are so large that
they quite overshadow the rest of the tracheal system,
the main trunks being reduced to the rank of mere
communications between them.

The pupa closely resembles that of Culex, but the
abdomen is relatively shorter and stouter. It may be
most easily distinguished by the disproportionately large
size of the seventh segment, which retains in the nymph
the proportions it holds in the larva. The respiratory
trumpets are small and somewhat olive-shaped.

Genus Anopheles. — Meinert's description refers in par-
ticular to An. 7naculipennis, with shorter references to
An. nigripes. The ground colour of the larva is a light
yellowish green, with a dark brown stripe along the back,
which, however, is whitish in the middle line. There are
also four small dark spots on the hinder border of the
anterior abdominal segment, and six small oblique bands

GENERIC CHARACTERS OF LARY.E OF CULICID.T: 61

on the sides of the other segments. In all the Indian
species the larvae are of a dirty brown colour ; with a few
rather ill-defined patches of pigment on the head and
thorax. In Anojjheles Bossii mihi there are two patches of
pigment on the dorsum of the head, which, combined with
the eyes, give this region of the larva, when regarded from
above, a quaint resemblance to a human skull. The head
is less rounded than in Culex, and the constriction between
it and the head is deeper but less obvious than in that genus,
as it forms rather a truncated cone, with the broader part
behind, than a sphere. The tergum of the third metamere
is in the form of a lozenge, with the anterior corner cut
off and the point behind, and is ornamented in front with
six plumed bristles, the outer of which are the largest.
That of the second metamere is broad and short, and
carries at each of its outer corners a single plumose bristle.
As in Culex, it is provided at its outer part with a dense
whorl-organ. The labrum, which is tongue-shaped and
crenated on either side, is small and hirsute. It is,
moreover, overhung by the clypeus, so that it is little in
evidence when viewed from above. The eyes form a
band of pigment of somewhat pyriform outline, with the
narrower end backwards, and on their outer sides are the
small ocelli.

The antennoe generally resemble those of Culex, but are
armed with a row of short, stout. spines along the inner
border of the basal joint. They carry also certain fan-
shaped and other specialised bristles. The mouth parts
closely resemble those of Culex, but the whorl-organs are
larger. The lower lip forms an equilateral triangle, with a
few strong dentations on its sides, and is prolonged into
a peculiar dentated process, besides which, as seen from
below, it is partly covered by two plates which appear to
be connected with the ventral plate of the second metamere,
which last structures do not appear to be represented in
the other genera. The mandibles and maxillse also closely
resemble those of Culex, but the latter have a straighter
anterior edge, and are less cut off at the corners. They
are fringed with bristles, some of which are of a specialised

<52 GNATS OR MOSQUITOES— CHAPTER IV

character, and from their outer posterior corner there
springs the maxillary palp, which is larger and more
acutely conical, and also carries certain specialised bristles.
Proportionately to the head the thorax is a good deal
larger, and exhibits three rows of bristles besides the large
lateral tufts, the posterior of which are the largest in the
body, while those of the pro- and mesothorax are of
insignificant size. The abdomen is cylindrical, and is
very distinctly separated into nine segments, which increase
in length while they diminish in breadth from before back.
The lateral tufts of bristles of the three anterior segments
are exceptionally large, but the hinder ones are very small.

On the dorsal surface of the eighth segment are a pair of
simply-formed spiracles, which can be withdrawn beneath
a fold of skin when the insect requires to protect their
openings. At the same time, if examined in profile, it is
evident that the difference between the larvae of Gulex and
Anopheles is really one of degree rather than of kind, and
that although rudimentary in the latter, all the parts of
the long syphon of Cidex are really present, and are
especially evident in some species ; while, on the other
hand, as already remarked, in some CiiUces the syphon is
very short. The last segment carries the four anal tubercles,
which, as well as the tail-fans, are rather less developed
than in Ciilex. Although arranged on the same general
plan within the body, the tracheal system is much less
developed, the main longitudinal trunks being of very
ordinary size, and quite without hydrostatic dilatations
in any part of their course.

The pupa differs from that of Cidex only in the respira-
tory trumpets being shorter and more squarely cut at
the end. It may be distinguished from the pupa of Mo-
chlonyx by the fact that the eighth instead of the seventh
abdominal segment is disproportionately long. In the
particular species examined by Meinert the pupa, like the
larva, is grass-green, but this coloration is not universal
in the genus, and those of Indian species, like the larvae,
are brown. Working in Italy on the same species, Grassi
was unable to confirm Meinert's statement as to a green
coloration, finding the larva, as we do in India, brown. As

GENERIC CHARACTERS OF LARV^ OF CULICID^ 63

a matter of fact, the green colour, when present, depends
largely on the nature and stage of digestion of the intestinal
contents. Hence, under certain conditions, Meinert's state-
ment is correct, but it cannot be called coloration in the
strict sense of the word.

Genus Corethra. — The species examined by Meinert
were C. plumicornis, and C. pallida. The larvae of this
genus differ markedly from those that have been already
described, and resemble those of the GhironomidcB rather
than any of the CidicidcB. These larvae are well known
as favourite objects for microscopic demonstration of
" pond organisms," and are generally known as "glass " or
"crystalline" larvae on account of their extreme trans-
parency, which is broken only by the four darker, but
still transparent, air sacs placed in the mesothoracic and
seventh abdominal segments respectively. The head is
much smaller than in any of the preceding genera, being
narrower than any other part of the body, except the last
two abdominal segments. Its hinder half is bounded by
straight parallel sides, as seen from above, but is contracted
in front, so that as a whole it presents the outline of a
broad-nibbed pen. The eyes are small and round, and are
placed well back on the head, rather on the dorsal aspect
of their sides, and behind each of the large eyes is a single
separate ocellus.

The antennae are articulated, as it were, at the point of
the pen and are relatively small. Each consists of a single
joint with a constriction, followed by a small node just
beyond the base, and is armed at the end with five large
bristles, which spread out like the claws of a lizard.

The tergum of the third metamere is rudimentary, but
its ventral plate forms the greater portion of the under sur-
face of the point of the pen, and is provided about its middle
with a sort of tubercle, from which radiate five pairs of
large bristles, while behind this are a pair of peculiar fan-
like plates, the posterior border of which is fringed with
fine hairs. The ventral plate of the second metamere is
generally spoken of as the labrum. It forms a sort of
ridge in the middle, and is provided with a number of
flattened hairs ; on either side are a pair of projections

64 GNATS OR MOSQUITOES — CHAPTER IV

carrying a number of strong radiating bristles, the ana-
logue of the whorl-organ of the other genera. The
mandibles are very large and trenchant, and being capable
of very wide abduction are particularly well adapted for
seizing the larva's prey. The maxillae are two small, very
simple plates, and the lower lip can hardly be said to be
represented as such.

The thorax, which is much the stoutest part of the body,
is of fusiform outline, and shows little or no indications
of its component segments ; both this and the abdomen
are provided with only a few small compound bristles.
The abdomen is composed of nine segments, which pro-
gressively increase in length to the seventh. These are
of nearly equal width to the sixth, after which the body
rapidly tapers off. The anal tubercles and swimming
fans are small, but round the anus are four bristles of
much greater size. Bound the anus also are several ranks
of peculiarly-shaped hooks.

The respiratory system is peculiar. Apparently the
function must be entirely aquatic, as in the very young
larvae there are no signs whatever of tracheae, and even in
the fully-grown creature there are no stigmata or external
breathing apertures whatever. In young larvae all that
can be seen are the two pairs of air sacs, which are
situated in the same positions as in Mochlonyx, but pre-
sent a very different appearance as they contain no air,
but are full of serum and are lined with a large-celled
epithelium with prominent nuclei. Gradually with suc-
cessive changes of skin the main longitudinal trunks and
their branches appear piecemeal. At first they are full of
serum, but as development proceeds they gradually fill with
air, which, however, must be secreted from the blood, as
there is at no period of larval life any direct communication
with the exterior.

The pupa in this genus is distinguishable by the rela-
tively large size of the abdomen and the small dimensions
of the cephalo-thoracic inass. As far as its respiratory
arrangements are concerned, however, it closely resembles
the pupae of the other genera, the breathing trumpets being
well developed, with very oblique mouths.

PLATE III.

J^z.'a/

Fto^

PLATE III. — To Illustrate The Anatomy op the Pupa (after Hurst).

Fig. 1. — Side view of the male pupa.

Fig. 2. — Ventral view of the female pupa, partly extended-
Figs. 3-6. — Successive stages in the metamorphosis of the epithelium of the
hinder part of the stomach.

Fig. 7. — Sagittal section of a very young female pupa. Ant. antennae ; Ao.
aorta ; At. respiratory syphon ; B. buccal chamber ; CG. cerebral
ganglion ; D. gastric pouch ; F. caudal fin ; FeK femur of first leg ;
G. ganglia; Gn. outgrowth of "ninth" segment, within which the
gonapophyses develop ; Hr. balancer ; H. head ; Ht. heart ; In. intes-
tine ; Lb. labium; Lbr. labrum ; M. Malphigian tubule; M.Ap. its
opening into the intestine ; MS. mesosternum ; Mt. metasternum ;
Mx. maxilla (first) ; 3Jxp. its palp ; NC. nerve commisures and ventral
cord ; Oc. ocellus ; Od. medium oviduct ; Op. compound eye ; P. pro-
sternum ; R. rectum ; S. aperture of salivary duct ; SG. suboeso-
phageal ganglion ; Si. larval respiratory syphon introverted into the
eighth segment ; Sp. spermatheca ; St. stomach ; Ta\ To? proximal
Joints of the tarsi ; K' — ■' tibiae ; Tr. trachea ; W. wing ; I. II. III. &c.,
first to eighth segments of the abdomen.

After the Plate in the late Mr. Hurst's Paper in the Mem. Owens Col. ,
somewhat reduced.

66

CHAPTEE V.

The Anatomy of the Pupa.

The following account is derived from the late Dr. C.
Herbert Hurst's excellent paper on the subject, published
in the " Studies from the Biolog. Lab , Owens College," II.,
1890, pp. 47, et seq.

The paper is too full to be reproduced in full in an
introduction like the present, but is characterised through-
out by the most painstaking accuracy, and I have been
able to verify most of the statements reproduced.

During the latter stages of larval life, in addition to the
visible head appendages, there appear eight other pairs
beneath the larval cuticle. Of these six are thoracic and
two abdominal. The thoracic pairs are three of them
dorsal, the future pupal syphons, the wings, and the
halteres ; and three pairs ventral, the future legs. The
two abdominal pairs belong to the last two segments.
Those of the eighth lie in the larval syphon, and are to
form the fins of the pupa ; the hindmost pair form the
outer gonapophyses of the adult, which are accessory organs
of copulation.

All these eight pairs arise as foldings of the epidermis
(" hypodermis ") outwards, and are quite hidden under
the larval cuticle. The antennae, too, are much larger in
an advanced larva than they appear to be externally, as the
growing basal portion is folded, or even telescoped beneath
the unyielding cuticle. Towards the end of larval life the
animal becomes sluggish ; profound changes in the mouth
parts deprive it of the power of eating, and it floats with
its breathing tube at the surface. Shortly, the cuticle
bursts in the thoracic region, the pupal respiratory

THE ANATOMY OF THE PUPA 67

trumpets are protruded, the abdominal tracheae appear to
collapse, and the animal floats with the anterior end
upwards, the new syphons coming to the surface. The
soft parts of the old respiratory syphons are withdrawn from
the cuticle and invaginated into the eighth abdominal
segments while the lining of the tracheal trunks breaks up
into pieces, which in the abdomen correspond to the
segments, and is cast off, with the other larval exuviae, by
means of the mechanism that has been already described.
The pupa which thus escapes, differs greatly from the
larva. It is in the larger species a little under 1 cm, in
length when fully extended, and consists of a bulky,
laterally compressed mass, made up of the head and thorax
with their appendages, and of a slender flexible abdomen,
which when at rest is carried curled under the thorax.

In a specimen 9 mm. long the thorax was 2'5 mm. and
the abdomen 6"5 mm., but the thorax appears much longer
on account of the wings, which extend downwards and
backwards from its sides. The head adds nothing to the
length, as it is carried tucked down under the thorax. It
is broad from side to side, short from back to front, while
ventrally it is drawn out into a long process, which extends
backwards under the thorax as far as the anterior part of
the abdomen, where it curves upwards. This process is
made up of the mouth parts, and includes all the parts
represented alike in the adult and larva.

On throwing off their larval chitinous covering, the
parts retain their larval masticatory type, but during the
four days of pupal existence the various parts mould them-
selves and develop into the basis of the adult condition, so
that by the time the chitin of the adult is ready for indura-
tion, they have altered their form to that of the adult
mouth. From the sides of the epicranial region, the
antennae run outwards to the sides of the thorax, one
beneath the anterior margin of each wing. The head and
all its appendages are immovable during the pupal stage.

The thorax is rounded, but somewhat compressed from
side to side. From the sides of its summit arise the respira-
tory syphons, a pair of conspicuous organs whose position

68 GNATS OR MOSQUITOES — CHAPTEE V

and form has led to their being termed horns or trumpets.

The wings are nearly flat, oblong plates, arising behind
the bases of the syphons, and extending downwards and
backwards.

Immediately behind them are a pair of triangular plates,
enclosing the halteres of the future gnat. The legs are
mostly hidden by the wings, but the femur, tibia and first
tarsal joint of the first leg, and the tibia and first tarsal of
the second are visible. The respiratory syphons are nearly
cylindrical, narrowed at their bases, and curved forwards
to be attached by flexible membranes to shght prominences
on the sides of the prothorax. Above, they are obliquely
truncate and open, and the margin is slightly notched on
the inner side. The outer surface is marked so as to
resemble imbricated scales, each with a minute spine at
its apex. The cavity of the syphon communicates directly
with the tracheal trunk at its base. Palmen (" Zur Mor-
phologic des Tracheen systems," Helsingfors, 1877) has
denied the communication of the syphons with the tracheae,,
and imputed to them the function of "tracheal gills" ; but
apart from the fact that their dense chitinous structure
renders them entirely unsuitable for the performance of any
such function, the reality of their communication with the
tracheEB can easily be proved by watching the imbibition
of suitable fluids through the syphons into them. All these
appendages originate as protrusions of the epidermic layer,
enclosing mesoblastic tissue. Those of the dorsum are all
at first flat, wing-like plates, but those of the mesothorax
alone retain this form as the wings of the adult, while
the halteres become club-shaped, and the anterior appen-
dages become rolled up to form the syphons of the pupa,
only to disappear on attaining the adult form. The legs,
on the other hand, appear from the first as cylindrical
processes. They are at first unjointed, but by the end of
the pupal period have segmented themselves into the
various joints of the adult.

The abdomen is flattened dorso-ventrally, and when at
rest is curved under the thorax. It is jointed and flexible,
and forms with the pair of large fins, borne by the eighth

THE ANATOMY OF THE PUPA by

segment, the only locomotor organ of the pupca, the wings
and legs lying immovable, and even adhering to each other,
though they are easily separated in specimens preserved in
alcohol.

Nine segments are easily recognised in the abdomen,
and the last one, though it is probably composed* of no
less than three condensed and highly modified segments, is
the smallest. Each segment has a chitinous tergum and
sternum, and setae are sparingly distributed over them,
those present being mostly on the hinder part of the terga.
Of these a pair placed on the hinder part of the first seg-
ment alone require mention. Each consists of a triangular
basal plate, articulated to the tergum by a soft membrane,
and distally divided into a number of bars, which by re-
peated sub-division give rise to about a hundred setae, all
lying in one plane parallel with that of the median, of the
body. When at rest the pupa floats with the tips of these
setae and those of the respiratory syphons at the surface of
the water, and the setae probably assist in maintaming
equilibrium, as well as serving as sensory organs for the
perception of disturbances of the water.

The eighth segment bears the fins, a large pair of thin
oval plates about 1'2 mm. in length, each of which is
strengthened by a midrib, which projects as a spine beyond
its hinder border. Beneath and behind them is the ninth
segment, a small though probably composite segment,
which contains the anus, and is provided on either side
and in front of it with a pair of blunt processes, larger in
the male than in the female.

The digestive canal differs but little in any stage of the
insect, the main change being the casting off of the thick
large-celled lining of the stomach and the substitution of
the more delicate mucosa of the adult. The cast-off larval
mucous membrane appears to be disposed of by digestion.
During this period also is developed the peculiar chitinous
dilatation of the anterior part of the thorax, already described
in the adult. Mr. Hurst describes it as triangular in section
with incurved sides, to the concavities of whose sides are
attached muscular fibres originating from the sides of the

70 GNATS OR MOSQUITOES — CHAPTER V ^

head, the mechanism being specially well developed in the
female. The peculiar air-containing sac already described
is also developed at this time. It is obvious euough that
the former mechanism is capable of being actively dilated
by muscular action, and may therefore assist in suction, but
it is difficult to understand how any one can have fallen into
the error of ascribing such a function to the latter organ.

The circulatory system consists of a long dorsal vessel,
which is broad and actively contractile in the abdomen and
contracts into an " aorta " in the head and thorax. From
its sides membranes, the alee cordis, which serve to suspend
it, run out between the extensor muscles and the stomach,
to attach themselves to the tracheal trunks. Each ala
consists of a dorsal and ventral lamina, and the space between
them has been called the pericardium. It contains the
pericardial cells and communicates freely with the body
cavity by the spaces between the alse. There is no distinct
constriction of the heart into chambers, and the paired ostia
or slits, which put it in communication with the "peri-
cardium," open backwards in the first segment, and forwards
and inwards in segments three to seven.

In the space between the ahe cordis are also the peri-
cardial cells, which are brown in colour and arranged in
ovoid masses, of which there are four pairs in each abdominal
segment, two of which are in its anterior and two in its
posterior portion. The protoplasm of these cells is extraor-
dinarily spongy and contains numerous granules which stain
deeply with borax carmine. The nuclei vary in number
from 3 or 4 to 10 in each mass, but the boundaries between
the cells cannot be made out. The glandular character
of these cells has been shown by Kowalevsky (" Biolog.
Centralblatt," ix., 1889), their function being probably some-
what analogous to that of the lymphatic and other ductless
glands of the higher animals. I reproduce more fully tbe
histological characteristics of these masses, as, alike from
their position close to the walls of the stomach, from which
they are separated only by the ventral layer of the ala cordis,
and from their general appearance they might easily be
confused with the parasitic " coccidia " of malaria recently

THE ANATOMY OF THE PUPA 71

described by Major Koss, I. M.S., by an observer not per-
sonally conversant with the appearances of the two struc-
tures ; and I shall not be at all surprised to find descriptions
of these bodies appearing in the form of notices of the
occurrence of the parasites in the pupal stage. The possi-
bility of the communication of the disease among Mosquitoes
through infected ova has already been mooted, and investi-
gators working upon this line should be on guard against
this fallacy.

The so-called aorta runs from the anterior end of the
dorsal vessel forwards, above the stomach and oesophagus
to the head, where it terminates in an open end. In trans-
verse sections of the thorax the aorta appears as a laterally
compressed tube, and does not appear to give off any
branches.

The respiratory system, during pupal life, undergoes the
changes which prepare the rudimentary stigmatic trunks of
the thorax and abdomen to take on functional characters in
the adult. With the exception of the first abdominal pair,
however, none of the stigmata are open except the pro-
thoracic openings which form the respiratory syphons.
These first abdominal stigmata open into the air space
which exists under the pupal skin beneath the thorax, and
in which the legs are undergoing development. This cavity
must exercise a hydrostatic function, and the patency of
these stigmata must be in this case necessary for the con-
veyance of air to the cavity. From the base of each syphon
tracheae run to various parts of the body and head. Among
these may be mentioned specially one transverse trunk
running across the thorax between the alimentary canal and
the nerve chain, which puts the two syphons in direct com-
munication with each other ; and a pair of longitudinal
trunks running back to the hinder end of the body, and
giving off branches to the various organs, and also a branch
to each of the future stigmata.

The cuticular lining or intima of the chief trunks and
their branches is well developed even at the beginning of
pupal life, and has the usual spiral thickening. The trunks
connecting the stigmata with the main trunks are the only

72 GNATS OE MOSQUITOES — CHAPTER V

ones that undergo any marked change. These widen round
their separated and collapsed intima, and a new and strongly
thickened intima is developed. In the main trunks no new
intima is formed, and when the imago escapes no portion
of the intima is shed, saving the portions connecting the
syphons and the first abdominal stigmata with the main
trunks. These fragments are, in the case of the syphons,
well developed, and have a fully-developed spiral thickening.
The portions connected with the first abdominal stigmata,
though better developed than the other abdominal branches,
have the spiral thickening only slightly developed. The
terminal portion is beset with very numerous small spines.

The nervous system is particularly interesting. Within
the short space of four days, certain ganglia increase enor-
mously in size by the addition of cells, apparently derived
directly from the epidermis ; and other ganglia shift their
positions bodily and sometimes fuse with others.

In the larva each of the first eight abdominal segments
has a pair of ganglia ; and yet a pupa, only half escaped from
the larval cuticle, has four in the thorax and none in the
first segment of the abdomen. During pupal life these four
ganglia fuse into one compact mass. During the first two
days of pupal life the eighth ganglia migrate into and fuse
with those of the sixth segment. In the female the change
goes further. A pupa almost ready to burst and give exit
to the imago has still the arrangement already described ;
but an imago killed immediately after its escape is found to
have no ganglia in the seventh or eighth segment, but in the
sixth segment are two masses ; the first the pair properly
belonging to the segment, lying at its anterior end ; the
other a double mass, formed of the seventh and eighth
ganglia, lying in the hinder end of the segment.

In the male imago the arrangement is the same as in the
advanced pupa. In the head the supra-oesophageal ganglion
increases enormously in size. The epidermal ("hypo-
dermal") cells, especially those near the borders of the
eyes, proliferate freely, and the cells budded off from their
inner surfaces migrate inwards and form the new cells of
the ganglia. By this process the ganglia, which at the

THE ANATOMY OP THE PUPA 73

commencement of pupal life were comparatively incon-
spicuous, grow till they almost fill the head, and there are
places in the advanced pupa where ganglia and epidermis
appear continuous.

Dr. Hurst's paper also contains some very interesting
details as to the development of the sense organs, especially
of the eyes and antennae ; but for these the reader is referred
to the original paper, as their interest is mainly that of the
development of these organs in insects in general than that
of the Culicidcc in particular.

Beproductive System. — The male generative organs of
the adult consist of testes, vasa deferentia, "prostatic
glands," copulatory organ, with a common pouch at its
base and two pairs of gonapophyses. Of these last, the
outer ones are a pair of large forceps for holding the
female. Both pairs originate in the larva and are probably
the appendages of two segments now fused and indistin-
guishable. The testes are a pair of cylindrical bodies,
already present in the larva, at the sides of the intestine in
the sixth segment. They are chambered and the spermatic
elements in the hinder chambers are more advanced than
those in front. The length of each segment is that of the
segment in which they lie. The vas deferens of each side
is a direct continuation of the wall of the testis, and is a
very narrow tube running directly backwards, quite distinct
from its fellow of the opposite side, but the two are closely
bound together in their hinder parts and they open behind
into the common pouch. The prostatic glands are a pair
of elongated glandular tubes, apparently simple, but seen
in sections to be double, though the cavities connnunicate
behind before opening into the common pouch. This
latter is a dilatation of the ejaculatory duct at the base of
the copulatory organ, which last is perhaps derived from
one of the component somites of the last abdominal
segment, and represents its appendages. The hinder part
of each vas deferens is, in some Cidicidce, expanded to
form a vesicula seminalis of considerable size, but this is
not the case in Culex nemoralis.

The female generative organs are a pair of ovaries, the

74 GNATS OR MOSQUITOES — CHAPTER V

oviducts uniting behind to form a median oviduct, a median
copulatory pouch, and three spermathecse opening into the
last. They correspond in size and position to the testes.
The median oviduct is formed by the invagination of a
region v^hich Dr. Hurst takes to be the ninth sternum,
while the anus opens at the posterior end of what he takes
to be the eleventh abdominal somite, so that there is no
common cloaca. This invagination is already far advanced
at the beginning of pupal life, and during it it grows
forwards, keeping pace with the forward shifting of the
last pair of ganglia, and at all stages lying just behind it
till the final ecdysis, when the rapid shifting of the ganglia
leaves it behind. Its anterior end is, in the adult, near
the front of the seventh segment. In the youngest pupae
three flattened invaginations, the future spermathecas, lie
on the dorsal wall of this median oviduct. During the
pupal period the anterior end of each becomes spherical and
acquires a strong chitinous lining. The anterior ends of
these organs remain stationary in the eight segment
throughout. The bursa copulatrix is a dorsal outgrowth
of the invagination which gives rise to the median oviduct,
and is a small pouch lying just behind and above the median
aperture.

The pupa does not eat. It breathes air through the
apertures at the end of its syphons. It floats, thorax
upward, by virtue of the large air cavity lying under
the hinder part of the thorax and front of the abdomen.
This cavity is bounded in front by the legs, at the sides by
the wings, and in front by the mouth parts. It extends
up at each side of the abdomen, where it is covered by
the halteres, and into it opens the patent first abdominal
stigmata. The pupa is sensitive to light, and immediately
darts backwards when a shadow falls upon it suddenly.
The movements, however, though very rapid, are devoid of
anything like steering.

The larva has to search for food, but the pupa has
simply to get out of the way of danger, and the direction
of its flight is of little importance, though since the move-
ment is always backward with reference to the pupa, it is

THE ANATOMY OF THE PUPA 75

chiefly downward with reference to the outer world. A
sudden loud noise or a very gentle tap upon the vessel
containing the pupse causes those at the surface to dart
downwards, but as shght sounds produce no effect upon
them, Dr. Hurst concludes that the tremor of the surface
of the water and not the sound itself is recognised by them.
The setse on the first segment of the abdomen are probably
the organs by which this movement is felt.

76

CHAPTER VI.

The Anatomy of the Adult Mosquito.

In its anatomy, the Mosquito conforms to the usual
insect type, the body being divided into the three sharply
separated regions of the head, thorax and abdomen; the
first accommodating mainly the organs of sense, the second
those of locomotion, and the third the digestive and
reproductive organs.

The entire body is covered with a chitinous integu-
ment, which is thick and rigid on the dorsum, and over the
entire head, and thin and elastic where it connects together
the various rigid portions.

This chitinous " exoskeleton," besides covering the
exterior of the body, sends inwards plates and trabeculae
which form a beautifully intricate framework supporting
the bases of the legs and other appendages ; and in addition
to this, the entire intracephalic portion of the intestinal
canal, and the ducts of the salivary glands are lined with
the same material ; while in the form of the tracheae,
tubular involutions of this outer covering ramify minutely
throughout every tissue, much in the same way as the blood
vessels do in vertebrates.

The whole of the outer armour is covered with appen-
dages of various forms, the most characteristic of which
take the form of scales.

These scales consist of flat plates springing from a
narrow pedicle, and differ a good deal in outline, not only in
different species, but on different regions of the same insect ;
though in any given species and situation their form is very
constant, and affords valuable characters in classification.
Each scale consists of an anterior and posterior membrane

THE ANATOMY OF THE ADULT MOSQUITO

77

separated by an almost imperceptible interval, and is
minutely ribbed and striated in a manner that requires the
best lenses to properly resolve. Perhaps the commonest
outline is that of a racquet, but as w^ill be seen from the

Fig. 16. — Scales and other chitinous appendages op Mosquitoes.

a, Scale of frontal tuft of Atwplieles Rossii, x 375 diams. ; b, detail of
membrane of the same scale, x about 1,000 diams. ; c, to show form and
arrangement of the scales on one of the veins of the wing of Ciclex fatigans
( Wied.), X 75 diams. ; d, to show form and arrangement of scales on margin
of wing of C. fatigans (Wied.), x 75 diams. ; e, scale of nuchal corona of
C. hortensis, x 185 diams., after Ficalbi ; /, scale of thoracic tomentum of
C. pipiens, x 185 diams., after Ficalbi ; g, scale of alar margin of C. pipiens,
X 208 diams., after Ficalbi ; h, scale of thoracic tomentum of C. spathi-
palpis, X 185 diams., after Ficalbi ; i, scale from the leg of C. pipiens, x 208
diams., after Ficalbi ; j, k, I, various forms of wing scale, after Arribdlzaga;
amplification not stated ; m, last tarsal joint of C. pipiens, 9 showing
1, empodium, 2, tarsal claws, and 3, Pulvilli, after Ficalbi ; n, fore, mid, and
hind tarsal claws of C. pulchriventer sp. n. c? ; o, single claw of fore leg of
An. Rossii, $ .

above figure, they vary in shape almost as greatly as the
leaves of plants, and some are so long and narrow as to be
almost indistinguishable from the hairs, w^hich are found
everywhere, either interspersed among the scales, or, as in
the abdomen of many Anopheletes, constituting the entire

78 GNATS OR MOSQUITOES — CHAPTER VI

tomentum. The third class of chitinous appendages are
the tarsal claws and epipodium, the latter of which is
essentially a compound hair, while the former vary in form
almost as much as do the scales. In the females these
claws are usually of simple form, and of no great size ; but
in the male, those of the fore and middle legs and more
rarely those also of the hind pair, are usually propor-
tionally large and armed with one or more accessory teeth,
and the two claws of the same leg may differ in size and
form. Those of the fore legs are generally the largest,
and the difference in size and form in the sexes is sugges-
tive of the function of those of the male being to grasp
the female. In many species, at any rate of Anopheles,
there is only a single compound claw on the fore leg,
the missing claw being apparently represented only by a
small projection at one side of its base. Varying as they
thus do, there can be no doubt but that the form of the
claws would, if generally noted, be capable of furnishing
valuable specific indications ; but though valuable notes on
this point have been made by Ficalbi, and also by Arribal-
zaga, information on this point is wanting in the great
majority of species, and the circumstance that it is impos-
sible to properly make them out without mutilating a
specimen by mounting the legs, for the compound micro-
scope tends to render these characters not so useful as
others for the practical purposes of classification. The legs
are always thickly clothed with scales, and in the one or
two species of the genus Sabethes of Desvoidy, and of Mr.
Theobald's new genus Conclnjliates, the sides of the tibiae
or tarsi of certain legs are provided with lateral fringes of
long scales so as to form a sort of paddle-shaped expansion ;
while in Psorophora and Mucidus the entire legs are shaggy.
The head is rounded, but wider than long, and bears the
usual appendages, all of these being represented, although
those forming the mouth parts, being modified to form the
style-like proboscis, differ markedly from the ordinary insect
foot-jaws, and present perhaps an even wider divergence
from the simpler forms than the suctorial mouths of most
other Diptera.

THE ANATOMY OF THE ADULT MOSQUITO 79

In this region of the body no sign of segmentation can
be made out, nor is there any visible separation between the
dorsal and ventral chitinous shields, such as is found in the
other regions of the body.

The greater part of the sides and front of the head are
occupied by the facetted eyes. These are alw^ays large and
well developed and in certain species may even touch each
other in front aiod nearly so below. Their anterior border
is usually somewhat hollowed back to lodge the bases of
the antennae, so that they tend to a reniform outline.
There are no ocelli, or simple eyes, such as are found in
the majority of Diptera, and though some authorities assert
that they are present in a rudimentary form, I have not
been able to satisfy myself as to the existence of any trace
of them in the adult insect, and for practical purposes, at
any rate, they may be considered as absent.

Immediately in front of the eyes will be seen the
antennae. These are of the moniliform type, and although
of fundamentally similar construction, differ greatly in
appearance in the two sexes, owing to the organ in the
male being so richly provided with long, silky hairs as to
form a pair of singularly beautiful plumes ; while in the
female these hairs are less numerous and down-like, so that
the joints of the antennae itself are plainly visible, and are
the portion of the organ that catches the eye. In both sexes
the antennae is typically formed of fourteen joints, the basal
one of which is much the largest and of globular form, the
constriction at its base being almost fused with the cephalic
shield so as to be capable of little motion. The greater
part of the front of this large basal joint is occupied by a
soft but tightly stretched membrane, and into the centre
of this is articulated the base of the second joint, which, like
those that follow it, though but little shorter, is not one-
fifth of the diameter of the basal joint. The chitinous wall
of the latter is grooved on its interior for a richly-developed
system of nerve threads, which are symmetrically arranged
in its lining membrane like the wires of a birdcage. The
structure of this joint has been described in great detail by
Dr. Christopher Johnstone of Baltimore {U.S. Quart. Journ.

80 GNATS OR MOSQUITOES — CHAPTER VI

Micros. Science, iii., pp. 97-102), and he contends that the
entire antennae serves as an auditory organ, the atmospheric
vibrations being received by the long hairs of the antennae,
and so transmitted to the drum-hke membrane which, as
already described, closes the front of the basal joint, and
thence through its contained fluid to the nerves lining its
cavity. According to this theory, the anterior membrane
is an actual memhrana Ujmpani, and the fluid w^ithin
corresponds in function to the endolymph, contained in the
cavities of the internal ear of the higher animals. It has
been further pointed out that the hairs of the verticils of
the male Mosquito respond to the musical note given by
the vibrations of the wings of the female insect. Professor
Lubbock (" The Senses of Animals," p. 115) seems to
regard with some favour Johnstone's idea on this point, but
antennae having this form of basal joint are not very common
in insects, and if it be really a tympanum it is certainly a
very exceptional arrangement. The articulation between
the first and second antennal joints is capable of very free
motion, so that at this point the entire organ can be moved
to considerable angle in any direction, while the extent of
mobility between the latter and the succeeding ten joints
is much more limited. Except in the very aberrant Deinu-
kerides cancer, in which the second is as long as several
of the succeeding ones, from the second to the twelfth
pieces inclusive, the joints closely resemble each other,
forming a moniHform series of short cylindrical pieces of a
length but little exceeding their thickness. From the base
of each springs a verticil of hairs, numerous (about forty)
and long in the <^ , and shorter and fewer in the ? . It is
more than probable that these hairs in the <? are really
chordotonal, auditory organs, as there is no doubt that they
respond to the note of the female wings, and it is the func-
tion of the male in these insects to seek out the female, but
the acceptance of this does not involve that of Johnstone's
theory of a tympanic function for the basal joint. The
last two joints greatly exceed the others in length, forming
in the male, together, much more than a third of the entire
length of the organ. The basal verticil of the penultimate

THE ANATOMY OF THE ADULT MOSQUITO 81

is well developed, but that of the last is ill developed, and
both are closely covered with short, downy hairs, and exhibit
also certain pits and specialised hairs, which probably are
sense organs, olfactory, tactile, &c. For further details on
the sense organs of insects, the reader is referred to
Lubbock's work, already quoted. Projecting from the
middle of the head {vide fig. 13), below the antennae, is the
characteristic proboscis which, although it at first sight
appears to be merely a cylindrical, trunk-like projection, is
really a very complex organ, being composed of no less than
nine separate pieces. The organ springs from a sort of
groove on the lower aspect of the head, through the inter-
vention of a flexible membrane, which admits of a certain
amount of protrusion and retraction as well as of flexion and
extension in the vertical direction; the bases of both the
upper and lower lips being furnished with sets of longi-
tudinal and vertical transverse muscular fibres to effect
these motions.

In all the Diptera the mouth parts of the adult insect
are modified, almost beyond recognition, from the obviously
trenchant jaws which have been already described in the
larva, and which persist, with but little alteration, through-
out life in some orders, such as the beetles. During the
three or four days of pupal life a marvellous series of
changes takes place, whereby the masticatory mouth of the
larva is transformed into the suctorial apparatus of the
adult.

To understand the dipterous mouth it must be remem-
bered that, in most of the order, all that is visible to the
eye is the lower lip, the other parts being hidden, wrapped
up within it. Everyone who has handled a microscope is
familiar with the stock popular object of the " blow fly's
tongue." This "tongue" is the labium, and with the
exception of the maxillary palps, little else of the mouth
parts is in evidence, the other foot-jaws being but little
developed, as the fly licks rather than sucks its food, and
the " tongue " is provided with an armature of chitinous
rasps, not unlike those that stud the lingual rib, and of
gasteropod molluscs. If now we examine a horse fly, we
6

82

GNATS OR MOSQUITOES — CHAPTER VI

find that the upper lip is developed into a strong spine ; and
that, hidden between it and the lower lip are four short,
stout blades, each of which has exactly the form of a Roman
sword. These blades are the mandibles and maxilloe, and
are capable of inflicting a sharp bite or rather puncture, but
the lower lip remains practically identical with that of the
blow fly.

Fig 17.— Mouth parts of an Indian Sand ply {Phlebotomus), the Labrum

HAVING BEEN REMOVED AND PLACED SEPARATELY TO THE LEFT.

a, labrum; b, mandible; c, labium; 7c, " hypopharynx " ; tv, maxilla;
X, maxillary palp, that of the left side being shown bent ; as habitually held
during life when at rest.

If we now examine the further modified mouth of a
sand fly, we find that the lower lip, though obviously
approaching in form to that of the Mosquito, is still recog-
nisably the same organ as the " tongue " of flies. The four
miniature, sword-like blades of the mandibles and maxillae
are present, and almost identical in shape with those of the
horse fly. There is, however, now an additional median
organ lying in the midst of this bundle of weapons, in the
shape of a conical tube, sharp at the point and very broad
at the base, prolonged from the mouth cavity. This organ
is the hypopharynx, and with this addition, we have now

THE ANATOMY OF THE ADULT MOSQUITO

83

represented all the parts of the highly evolved, suctorial
mouth of the gnats ; the only difference being that on
the latter the parts are attenuated and elongated into the
far more efficient, though more delicate, suctorial organ
represented belovi^.

In the Mosquito the labium is a long, narrow, grooved
spine ending in a pen-like point, and the lower lip, though
bearing some likeness to that of the sand fly, has become a

Fig 18.— Diagram to Illustrate the Mouth Parts op a Female
Mosquito.

A, median mouth parts, and gnathites of the left side (their relative position
at the base is indicated in a purely arbitrary way) ; B, end of proboscis seen
from below ; C, end of proboscis seen from above ; D, the same with the hypo-
pharynx drawn away to the left— with the exception of fig. A, after Ficalbi.
In the above, a, labrum; b, mandible; c, labium; k, hypopharynx; w, maxilla;
.t, maxillary palp ; y, basal joint of maxillae.

mere sheath for the other organs and no longer bears any
resemblance whatever to a fly's tongue ; the mandibles and
maxillae have become simple slender lancets, formed on the
plan of the mediaeval rapier, with a comparatively thick
rod-like back, and a broader edged and pointed lamina ; a
mere golden-coloured wire of chitine supporting a narrow
blade of the same material, so thin as to be absolutely
colourless. The hypopharynx has undergone a similar
change and has taken the form of a delicate plate, pierced
throughout its entire length by a minute canal ; the whole
of the mouth parts, except the lower lip, being so delicately

84

GNATS OR MOSQUITOES — CHAPTER YI

fashioned as to practically defy draughtsmanship, so that they
can only be represented in diagrammatic form.

The upper Hp, or labrum, springs from a bulbar enlarge-
ment of the front of the face, placed between the bases
of the antennae. This bulb is flexible and elastic, and the

Fig. 19. — Semi-diagrammatic eepeesentation op Serial Preparations of
THE Anterior part op the Head and Proboscis, x 150 diams.

A, median sagittal section of root of proboscis, showing the entry of the
salivary duct into the floor of the base of the hypopharynx, with the valvular
fold and its muscles, by which the duct is opened and closed. B, transverse
section opposite the point of entry of the salivary duct, and the roots of
the palpi. C, transverse section of the proboscis, near its base. D, trans-
verse section of the middle of the proboscis. In the above figures : —
a, labrum ; b, mandible ; c, labium ; d, buccal bulb ; /, salivary duct ; k,
hypopharynx; mx, maxilla; m.at., muscle opening the salivary valve ; m.rt.,
muscle closing salivary valve ; m.pr. Ibr., protrusor muscle of labrum ; to. rt.
Ibr., retractor muscle of labrum; t.o., taste organ (?) ; q, tracheae; x, root of
palp.

whole of its median plane is occupied by a fan-shaped
muscle, taking origin from its upper surface ; the fibres
converging to be inserted into a thickening of the chitine
at the base of the actual labium ; while, on either side of
this retractor, a number of antero-posterior transverse fibres
fill up most of the remaining space. By the action of the

THE ANATOMY OF THE ADULT MOSQUITO 85

retractor muscle, the base of the labium can be invaginated
within the bulb, so that its point is withdrawn behind those
of the other mouth parts, while by that of the transverse
fibres, it can be forced out and extended once more to their
level, or be3^ond thein.

Firmness is imparted to this arrangement by a pair of
strong spines, which project from the lateral trabeculse of
the front of the head into the bulb. Beyond the bulb, the
labrum becomes a strong, rigid style of a form not unlike
that of a surgeon's " director," but with a more complete
groove ; the transverse section forming at least three parts
of a circle ; and this incomplete canal is converted into a
tube by the hypopharynx, which, throughout its length,
lies in contact with its lower surface. It is through this
channel that blood or other food reaches the mouth cavity.
The hypopharnyx springs from the anterior part of the floor
of the buccal bulb, and is a tubular organ with, at first,
a very broad but shallow lumen. This, however, rapidly
contracts to a minute tube, of about the same diameter as
the salivary duct, of which it is virtually a continuation, and
on either side of the tube are developed the membranous
wings, which serve to convert the groove of the labrum
into a tube. Into the floor of the broad cul-de-sac of the
commencement of this tube opens the salivary duct, the
mouth of the latter being guarded by a remarkable arrange-
ment, which serves as a valve. The floor of the buccal
valve at this point is strengthened by a strong chitinous
plate ; and this, though of course continuous with the
thinner membrane near it, forms a veritable articulation
with a separate plate in the floor of the base of the hypo-
pharynx, in which is placed the termination of the salivary
duct. As the rest of the lining of the cul-de-sac is delicate
and flexible, this plate can be moved forward and back, and
is provided with two special muscles for the purpose. In
the latter position the mouth of the duct is effectually
closed, while it is freely open when drawn forward by the
anterior muscle, which springs from the chitine of the
ventral wall of the labium. Grouped round the lower sur-
face of the hypopharynx and the sides of the labrum are

86 GNATS OR MOSQUITOES — CHAPTER VI

the four lancets, which are simple, rapier-like blades, simpl}^
pointed, and unprovided with teeth or serrations of any
kind, though a false appearance of the sort is lent to the
points of the inaxillse by the plate of their blades being
fibrillated. The niaxillse are usually stated to be the smaller
and more delicate, but, as a matter of fact, they are con-
siderably the stouter, and except in this the mandibles
and maxillae closel}^ resemble each other. Neither pair of
lancets appears to be provided with any special musculature,
and they are probably forced into the skin simply by the
thrust of the head ; the labrum being drawn back while
they make their puncture, and then forced into the wound
they have made. Lastly, the labium is a comparatively
stout, fleshy organ, soft, flexible, and clothed with scales
on its outer surface, which is wrapped round the other
parts so as to form a sheath, open only above. Between
its scaly ventral surface and its delicate dorsal lining is
interposed a considerable amount of loose connective tissue,
in which lie a pair of tracheae, and immediately beneath
them two large nerves. The labium ends in a pair of valve-
shaped, articulated lobes, capable of being moved by certain
special muscles. The function of these lobular organs,
which are said to represent the labial palps of certain other
arthropods, is probably to grasp the styles as they pierce
the skin, while the stem of the labium is being looped
down out of the way.

The above account differs, it may be noticed, from pre-
vious descriptions, inasmuch that it has hitherto been
assumed that both food and saliva passed through the
proboscis by the same channel ; but there can be no doubt
of its correctness, which I have verified by several series
of sections, in both Culex and Anopheles ; and I gather
from Professor Grassi that he holds the same view of
the functions of the various structures, who I trust will
understand that his priority in the matter is herein fully
recognised ; though he must not be held responsible for
the existence of the salivary valve, and the other structural
details of the above description.

A little reflection will, moreover, show that an arrange-

THE ANATOMY OF THE ADULT MOSQUITO 8/

ment of the sort is a physiological necessity, as it would
be quite impossible for the saliva to be forced down a tube
while the food is being drawn up through it.

When the Mosquito brings the proboscis into action to
pierce the cuticle of the plant or animal from which it seeks
its food, all these parts, except the sheathing labium, are
forced into it, while the labium is bent down into a sort
of loop which progressively narrows as the piercing organs
penetrate more and more deepl}'. On either side of the
base of the proboscis may be seen a pair of jointed appen-
dages, the maxillary palpi. These organs have been
described by several authorities as labial palps, but a closer
examination shows that they are really connected with the
base of the maxillary lancets through the intermediation of
a piece which, in the usual position of the parts, is hidden
within the base of the labium. In most cases the palpi
consist of five visible joints, but they present great sexual
and generic differences within the family, and require
especial notice in systematic work, as the classification of
these insects is largely based on their characteristics.

In some species they greatly exceed the proboscis in
length, while in the genus ^des they are quite rudimentary
in both sexes and appear to consist of only a single short
joint.

The head is connected with the thorax by a soft flexible
neck corresponding in function to the intersegmental mem-
branes of the abdomen, though by some authorities it is
considered as a part of the prothorax, which allows of
a considerable freedom of movement between these two
divisions of the body.

The thorax is the most bulky portion of the body, as,
though but little broader than the head, it is more than
twice as deep. It is composed of three segments, but the
great preponderance of the middle of the three, or meso-
thorax, and the fact that the three are fused together into
a single rigid mass, makes it not altogether easy to make
out the lines of division between the component parts.
Viewed from above, almost all that is visible is the tergum of
the mesothorax, but in some mosquitoes, as in Psorophora,

88 GNATS OR MOSQUITOES — CHAPTER VI

a portion of the prothorax projects in the form of a pair of
lateral protuberances, the shoulder callosities ; and on either
side of the scutellum, which appertains to the mesothorax,
can be seen, even in the extended insect, a portion of the
better-developed metathorax, while in flexion the whole
width of its tergum can be easily seen. Laterally, how-
ever, the three component segments are easily distinguish-
able by the fact that each gives origin to a pair of legs,
the coxse of which guide the eye to the corresponding

Fig. 20. — p, prothorax ; w, mesonotum ; h, metanotum ; k, scutellum
pp, propleura ; mp, lobes of metapleura ; 7ntp, metapleura ; s, s, s, thoracic
stigmata ; -lo, root of wing (after Arribalzaga).

portion of the thorax. Just in front of the root of the
coxa of the front leg can be seen, in the great majority of
species, an oblong or oval scale, which covers the stigma
of the prothorax.

The stigma of the mesothorax is also well developed,
and can easily be seen just in front of the origin of the
wings; but those of the metathorax, on the other hand,
are small or not recognisable. From the propleura springs
the coxa of the front leg, and between it and that of the
middle leg is a comparatively large area of mesopleura.
This latter is more or less divisible into three areas, one
of which carries the stigma, the second the middle leg, and
the third and hindmost the wings. From the metapleura
springs the coxa of the hind legs, and above it may be seen

THE ANATOMY OP THE ADULT MOSQUITO

89

the halteres or poisers, a pair of small, club-like organs,
consistiDg of a delicate stem terminating in a pear-shaped
head, which represents the rudiment of the hinder wings
of other insects.

There appears to he good reason for regarding these
modified wings as sense organs, for although mere rudi-
ments, as far as their original functions are concerned, they
are supplied by one of the largest nerves in the body, and
have been regarded by Keller, Hicks, and Bolles Lee as
auditory organs.

Pig. 21. — Semidiageammatic Dbawing op a Transverse Section of the
Thorax of C. pipiens (L).

a, alar muscles ; b, portion of chitinous framework supporting the wing ;
c, coxal muscles ; d, dorsal vessel ; e, CBSophagus ; /, flexor of abdomen ; 7i,
nerve cords ; p, pneumatic vesicles.

According to the last-named authority (" Les Balanciers
des Dipteres," Becueil Zool. Suisse, ISSo), the organ consists
of a varying number of rows of minute vesicles placed at
the base of the organ, each vesicle being perforated and
containing a minute hair. Sense organs have been dis-
covered in a variety of very unexpected situations in insects,
and it is quite possible that, while the verticillary hairs of

90 GNATS OR MOSQUITOES — CHAPTER VI

the antennae of male gnats may serve as chordotonal organs,
whereby they are enabled to localise the whereabouts of the
female, ordinary hearing may be subserved by these peculiar
structures at the base of the halteres.

Almost the whole of the interior of the thorax is occu-
pied with the powerful muscles that actuate the wings, legs,
and halteres, but the detailed consideration of their arrange-
ment hardly lies within the scope of a work like the present.

The general structure of the Dipterous leg and wing
have already been sufficiently alluded to in the remarks on
the terminology of the Order, but as regards their special
characteristics in the CuUcidce, it may be noted that the
former are always proportionally long and slender, and that
the hips, or upper sections of the coxae, take the form of
obconical processes, immovably cormected with the corres-
ponding pleurae ; while the trochanters are much smaller^
and take the form of short, globular, or oviform pieces. Of
the remaining joints, the femur, tibia, and first tarsal joint
are long and linear, and, as a rule, differ but little in length,
though their relative proportions furnish valuable specific
indications. Not unfrequently the first tarsal is the longest
of the three. The remaining four tarsal joints taken together
do not, as a rule, equal either of the above linear joints,
and generally progressively diminish in length, the last
being often scarcely wider than long, and carrying a well-
developed epipodium and pulvillus, and a pair of claws, the
characters of which have already been sufficiently noticed.

The third division of the body or abdomen is the longest
of the three, but is much more slender than the thorax. It
is more or less cylindrical, but depressed, being broader
than deep, and is composed of nine segments, none of which
are provided with locomotor appendages, although the last,
bears a pair of jointed appendages which serve as ex-
ternal organs of generation. Each segment is composed
of a dorsal and a ventral, rather rigid, chitinous plate, united
at the sides by a softer membrane, in which are placed the
stigmata or external respiratory apertures. The anterior
segments closely resemble each other, except the first,
which is a good deal shorter than the rest, and has its

THE ANATOMY OF THE ADULT MOSQUITO

91

ventral aspect devoid of any rigid sternal plate, being
composed entirely of elastic flexible membrane. The last
three, however, are shorter and diminish rapidly in width ;
and the last two are specially modified to accommodate the
apertures of the digestive and reproductive systems, the
anus being placed on the ventral aspect of the eighth
segment, while the tergum of the ninth is quite invisible
in the usual position of the insect, and all that can be seen

Fig. 22. — Transverse Section through the Posterior Part op the
Abdomen of a Hibernating Female.

dv, dorsal vessel ; ^jrs, perivascular space ;
m, Malphigian tubes ; g, last abdominal ganglion.

Of, ovary ; r, rectum ;
X about 100 diams.

of it are two lobed appendages, thickly clothed with hairs
and scales, and not showing any very promment difference
in the two sexes. In both they consist of a pair of olive-
shaped protuberances, but in the male, instead of ending
in a simply rounded, bristly extremity, they have articulated
to each a strong incurved claw, by which the female is
grasped during copulation, the whole organ forming a
clasper. In the female they are somewhat spatulate and
act as an ovipositor, being held together to form a sort of
channel along which the ova are passed on to the upper

02 GNATS OR MOSQUITOES — CHAPTER VI

surface of the similarly held together hind tarsi, hy which
they are guided to the surface of the water on which they
are to be launched.

If we dissect the abdomen by teazing with needles or
by sectionising, we find that immediately beneath the
chitinous exoskeleton there is a soft, cellular layer, the
true dermis ; and beneath this again, a very scanty and
interrupted layer of longitudinal muscular fibres, which
not only serve as flexors, extensors and abductors, but
can also, when simultaneously contracted, shorten longitu-
dinally and therefore transversely widen the entire region,
an action which cannot fail to have an effect in keeping
in motion the air contained within the tracheal system.
For although there is no continuous, rhythmic action of
the body comparable with the respiratory movements of the
higher air-breathing animals, movements of this kind are
sufficiently frequent and habitual to exercise a powerful
action in preventing stagnation of the air contained in the
respiratory tubes. These longitudinal muscles are arranged
in sets, each set corresponding to an intersegmental mem-
brane and serving to connect two adjacent segments. There
are also a few transverse fibres, arranged mainly in two
lateral groups, placed near the middle of the segments, but
they are less easily made out.

The best way to exhibit them is to carefully remove all
the scales from the abdomen, a preparatory step which
should be taken as a routine matter in all dissections, by
gentle brushing with a camel hair pencil. Dip the specimen
for a moment in spirit, and then by a stroke of a pair of
fine scissors split the abdomen into lateral halves. Ex-
amined in salt solution, such a preparation, after cleaning
away the viscera by means of the needles, shows well the
stigmata and parietal tracheae, and subsequently, after
appropriate staining in borax-carmine, will give a good view
of these parietal muscles. The precaution of dipping the
specimen into spirit for an instant, before placing in salt
solution, is a necessary preliminary to all dissection of these
insects, as otherwise, however carefully one may have
brushed, one is sure to be embarrassed by air bubbles,
obstinately entangled in the remaining tomentum.

THE ANATOMY OF THE ADULT MOSQUITO 93

Having now sketched the external configuration of these
insects, their internal organisation remains to be considered,
and will probably be best dealt w4th by taking separately
the various systems, digestive, respiratory and reproductive,
in their turn.

The Anatomy of the Digestive System. — In the first
edition of this book I had perforce to rely mainly, alike for
facts and illustrations, on the work of previous writers. My
supply of fresh material for verification was very limited,
and hence numerous inaccuracies were reproduced. A
notable example of this was Arribalzaga's illustration of the
general relations of the digestive organs. In this figure the
salivary glands are shown protruding from the back of the
head, which, as a matter of fact, is just where they are seen
in the ordinary course of dissection ; as they are drawn out
from their natural position in the soft neck connecting the
head and thorax by the act of pulling away the oesophagus
from the thorax. He appears to have then hastily sketched
in an outline of the insect, and hence these glands are made
to appear as if placed in the anterior part of the thorax.

The present chapter is, however, the outcome of personal
observation, and the illustrations, with a few duly noted
exceptions, are reproductions of pen and ink drawings from
camera lucida outlines, and so may at least be trusted as to
the relative size and position of the parts represented.

The digestive tube may be said to commence at the
extremity of the proboscis. By the alternate retraction and
protrusion of the labrum ; as regards its relative position to
the lancets, the whole of the styles of the proboscis are
gradually worked into the wound inflicted, slightly in
advance, by the mandibles and maxillae ; the hypopharynx,
which is a delicate, flexible tube, far too yielding to be
forced unsupported into any substance, lying along the
groove of the labrum. Imagine a surgeon's " director,"
with a piece of thin drainage tube carried against its groove,
and with four slender bistouries grouped round it, and you
will have a fair working model of the malaria-inoculating
apparatus of the mosquito.

From this it follows that it is only through the tube

94 GNATS OR MOSQUITOES — CHAPTER VI

of the hypopharynx that the sahvary secretion, with its
contained raahxrial germs, can be injected into the tissues
of the vertebrate alternative host.

In the case of filariasis the route of infection is different.
Taken as a family, the filaj'im are essentially parasites of the
intercellular lymph spaces, and hence are most commonly
found in such situations as the cellular tissues, and the
great serous sacs, such as the cavities of the peritoneum and
pleura, and that stage of the parasite life-cycle of the filaria
inhabiting as an adult the human cellular tissue, vv^hich is
passed in the mosquito, forms no exception to this general
rule.

Necessarily it is along the groove of the labrum that
the embryonic filarial find their way, along with the in-
gested blood into the mosquito's stomach ; but once there,
they hasten to reach a more congenial situation, by boring
through the walls of the organ into the cellular tissue around
it, and are then free to wander in its interstices all over
the body, including that of the sheath of the proboscis or
labium, which is a comparatively thick fleshy organ, con-
taining between its inner and outer walls a considerable
amount of cellular tissue supporting the tracheae, muscles,
and nerves, with which it is liberally supplied. Now, as
Grassi has recently pointed out, filarice, though not usually
very destructive parasites, do cause a certain amount of
irritation and effusion ; and the result of the intrusion of
filarice into the substance of the labium will be to raise a
tense swelling of the delicate inner covering of the organ.

As already pointed out, when the mosquito bites, the
labium is sharply doubled into a loop, and if the organ be
swelled, owing to the presence in its substance of filarice,
the result of this doubling up will be to rupture its delicate
integument, and so set free the contained filarial. Once
free they can scarcely fail to be carried by capillarity
among the blades of the lancets, and so introduced with
them into the cellular tissues of the new victim in which
they propose to grow to maturity.

Passing through an opening between the trabeculte
which form the framework of the head, and to which it is

PLATE IV.

PLATE IV. —To Illustrate the Anatomy op the Digestive System op
THE Imago.

A. Outline figure of a female Culex, to show the relations of the intestinal

canal and its appendages, x 15 diams.

B. The rectal glands lying in situ en the large intestine, with their tracheal

supply. X 75 diams.

C. Isolated cells of rectal glands, x 300 diams.

D. Chitinous pharyngeal bulb, x 75 diams.

E. Blind extremity of a Malphigian tube, x about 200 diams.

F. Semi-diagrammatic representation of the parts about the mouth, x about

76 diams.

G. The salivary glands and their ducts lying in situ on the pharynx and

oesophagus, x 75 diams.

H. Salivary duct of one side and its branches, the middle one of which,
though mainly denuded, has attached to it two clusters of salivary
cells. X 430 diams,

K. Normal salivary cells somewhat more highly magnified.

In the above figures : a, labrum ; h, mandible ; c, labium ; d, buccal bulb ;
e, oesophagus; /, salivary duct; g, nerve; k, " hypopharynx " ; m,
Malphigian tubes ; n, chitinous pharyngeal bulb ; o, spiral valve of
large intestine ; p, pneumatic vesicle (dotted outline of) ; q, trachea
supplying the cellular tissue of the labium ; r, rectum ; s, salivary
glands ; t, rectal glands ; v, stomach.
The chitinous framework of the head is indicated by coarse oblique

shading ; the outline of labrum in dotted, and that of the labium in broken

dotted lines.

96 GNATS OR MOSQUITOES — CHAPTER VI

connected by a delicate membrane, reflected on to them,
and thence to the floor of chitinous pharynx, the hypo-
pharynx, as already described, ends as a blind sac lying in
the root of the labium, just in front of the buccal cavity.
The anatomy of the latter is by no means simple, and so
far as I know, has hitherto remained unnoticed.

As I have endeavoured to indicate in fig. F, plate IV.,
the framework of the mouth springs from the strong
tergal plate running along the middle line of the head,
and from which springs the base of the labrum. From
this, on either side, runs down a strong lateral piece, which
shortly divides into two rods, enclosing an oval space,
the anterior of which appears to be really the basal piece of
the mandible. Reuniting below, these lateral pieces are
continued downwards to join with the strong but narrow
sternal plate that lies in the ventral middle line. This
sternal plate, however, is only single and median for a
short distance ; after which it divides into lateral rods,
united by a delicate membrane, which run back between
the lower borders of the ejes. Through this transparent
membrane, if we can successfully clear away the opaque
mass of structures lying above them, may be distinctly
traced the single median common duct of the salivary
glands, running forward to open into the fundus of
the hypopharynx. The pharynx, or intracephalic portion
of the digestive tube, is entirely chitinous and terminates
in a cup-shaped expansion, which forms the back of the
roughly spherical frame of the buccal cavity. This, as
already described, is in no way rigidly connected with either
the framework of the head, or the end of the hypopharynx,
continuity of the tubes being maintained only by a very
delicate membrane, and into the space between the two
structures a mass of mesoblastic tissue is pushed forwards,
especially below, and expands between the cupped end of
the pharynx and the delicate connecting membrane into a
thick mass composed mainly of pecuHar spindle-shaped
cells. It is along with this mesoblastic intrusion that the
salivary duct and two large nerves are conducted to the
interior of the buccal bulb, and the presence and large size

THE ANATOMY OF THE ADULT MOSQUITO

97

of the latter make it probable that the function of this
thickening which fills up the greater part of the chitinous
sphere, is connected with the sense of taste. The anterior
part of the lining membrane, especially below, is closely
armed with minute spines, the back of the cavity being
lined with the soft tissues of what is evidently a nervous
end organ, and is probably that of taste.

At the point where it springs from the cup-like expansion,
the pharyngeal tube is somewhat constricted, but after this

Pig. 23.— Transverse Section of the Head of An. Culicifacies, sp. n.
X 120 diams. , showing the pharyngeal bul,b, with its dilator muscles. Imme-
diately above is seen the mass of the brain, or supra-CBSophageal ganglia ;
/, salivary duct ; g, ventral nerve cords ; n, pharyngeal bulb ; q, trachese.

it runs for a short distance as a tube of uniform thickness,
through the mass of the oesophageal nerve collar, and then
expands into a fusiform dilatation of considerable size. As
usually seen (in fig. D) it is longitudinally furrowed so that
its transverse section would have the outline of a wallflower,
but when filled by imbibition of glycerine these infoldings
disappear.

This bulb unquestionably serves as the pump, by means
of which the insect draws up the fluids of the tissues into
which its proboscis is plunged ; as it is provided with a
7

98 GNATS OR MOSQUITOES — CHAPTER VI

special set of muscles, by the action of which it can be
actively dilated. Above, it is fixed by the connective tissue
around it being intimately connected with that of the
supra-oesophageal ganglia ; and on either side are inserted
bands of muscular fibre, which pass obHquely forwards and
outwards, to their origin from the chitinous wall of the
ventral aspect of the occiput, and a glance at the figure
above will make it clear that they must necessarily exercise
a powerful suctorial action.

Behind this, the pharynx once more contracts to its
previous diameter and just beyond the back of the head,
where it enters the neck of the insect ; it ends in the soft
oesophagus or true fore-gut, the walls of which are greatly
thickened at this point so as to form a bulb, rather broader
than that of the pharynx, and mainly composed of a
muscular thickening, which probably acts by way of a
valve, by means of which the contents of the mid gut are
prevented from entering the pharyngeal bulb while the
latter is exercising its suctorial function. It is on the
ventral aspect of this junction, but mainly on the oeso-
phageal bulb, that are placed the salivary glands which are
therefore contained within the narrow neck connecting the
head loith the thorax.

In the fresh insect, they extend but a httle way into the
front of the thorax ; but in serial sections, as a result of
the shrinking action of the various processes to which the
preparation has been subjected, they are squeezed out of
the soft and yielding neck, back into the cellular tissue
filling the more rigid box of the thorax, and so are made
to appear to extend backwards for a considerable distance
into it ; and the erroneous representations that have found
currency have no doubt resulted from a too complete
reliance on the method of serial sections, unchecked by
dissection of the fresh insect.

I can find no previous description of the salivary glands,
including that quoted in my first edition, which is not, as I
now find, full of inaccuracies. This has arisen from the fact
that their extreme dehcacy renders their demonstration by
far the most difficult piece of minute dissection that it has

THE ANATOMY OF THE ADULT MOSQUITO 99

fallen to my lot to attempt. It is comparatively easy to
obtain a portion of them for microscopical examination, but
to lay them bare intact, and still more to follow out the
course of their ducts, is a task that will try the most patient.
For often, as Dr. Daniels remarks, all one has to show for
one's pains is the intra-glandular portion of a duct, denuded
of its cells. As however, it appears probable that it is
through the infection of these glands and their secretion that
the germs of malaria are carried to man, their structure and
relations require to be carefully studied. Although their
duct passes forward to open into the fundus of the hypo-
pharynx, these glands are rather closely bound down to the
oesophageal bulb on which they lie by means of connective
bands, and hence are occasionally brought away with the
rest of the digestive viscera in the dissection of the stomach ;
but such an occurrence is merely accidental and cannot be
relied on ; and where the glands are required for investiga-
tion, the following is the best method I can suggest for ex-
posing them : —

Place the insect in some salt solution on a slide, and then,
with the point of an edged needle, lay open the narrow neck
connecting the head and thorax, keeping well to the dorsal
side. Next pass the needle into the midst of the thorax,
so that it must divide the oesophagus about the middle ;
then by a few gentle strokes, tear through the integuments
of the neck here and there, where it joins with the thorax,
still keeping away from the ventral side ; and finally, keeping
the left hand needle entangled in the thorax, gently tear off
the head with that held in the right and clear away the rest
of the insect.

It is well at this stage to wash well with fresh salt
solution from a pipette, and it facilitates the recognition of
the various structures to just colour the solution with methy-
lene blue.

If the separated head be now carefully examined there
will be seen, protruding from the torn neck, the tag of
oesophagus which has been drawn out from the thorax, and
round it several minute glistening lobular bodies, which,
under a sufficiently powerful lens, look much like tiny bunches
of grapes.

100 GNATS OK MOSQUITOES — CHAPTER VI

These are the blind ends of the lobes of the saHvar}'
glands, of which there are six in all ; each gland consisting
of three sausage-shaped lobes of nearly equal size ; but as a
rule only two or three of their number will come into view
at this stage and their subsequent separation from the
structures surrounding them is purely a matter of manipula-
tion, for which I can offer no better receipt than patience and
steady hands. Once one has learned to recognise their very
characteristic appearance, there is little difficulty in getting
piecemeal preparations, but I was well nigh in despair ere I
had cleared up my last doubts as to their exact form and
connections.

In structure these glands differ entirely from any with
which the student of vertebrate histology is familiar, as,
strictly speaking, each cell is a separate gland, there being
no true duct formed by a lumen bounded by contiguous
gland cells and supported by a fibrous layer, such as is
typical of vertebrate glands, as well as those that appertain to
the intestine proper in arthropods. Instead of this, the
salivary cells are arranged, like the flowers of a " spike," on
a chitinous tube prolonged from the lining of the buccal
cavity, and the singularly sharp outline and appearance of
rigidity of this tube, form a marked contrast with the
indeterminate appearance of the lumen of an ordinary
tubular gland. Apart from the fact that it is quite straight,
the intraglandular portion of this chitinous duct resembles
nothing so much as an elastic fibre from the ligamentuvi
nuchce, but is of much less diameter. The external diameter
of even the main duct is only 3 or 4 /u,, and the lumen of its
intraglandular branches can hardly exceed 0'5/a. Each lobe
of the gland consists of a single branch of the duct with the
salivary cells closely clustered round it, but in no other way
connected with each other, so that the least touch or the
pressure of a cover glass, at once separates them from each
other and from the duct, to which each cell is connected
only by a delicate thread of cell substance prolonged from
their smaller ends. Each is a typical gland cell of con-
siderable size and pear-shaped form, with large nucleus and
nucleolus, the smaller end tapering off to a delicate thread

THE ANATOMY OF THE ADULT MOSQUITO 101

which connects the broad-ended cell-mass with the duct ;
and if the latter be examined under a good apochromatic
lens, it can be distinctly made out that, minute though it be,
this chitinous wall of the tube is pierced by spirally arranged
minute perforations, each corresponding to the point of
origin of a cell.

In connection with the search for parasitic elements it is
well to remember that, there being no connective or other
elements to confuse the issue, any other structure than these
very characteristic gland cells must necessarily be abnormal.

Each gland consists of three lobes about three times
as long as they are wide, and the three ducts unite at the
same point into the right and left salivary ducts ; but
these have only a short separate course, and at a point
near the base of the pharyngeal bulb, they, in their turn,
unite into a common median duct which terminates in the
manner already described. The six lobes lie close together
on the oesophageal bulb, and together cover the whole
of its ventral surface and the greater part of its sides.
The only other point requiring notice is that the extra-
glandular portion of these ducts is strengthened by a spiral
or annular thickening of the chitine, so that they present
a close resemblance to a trachea emptied of air ; but there
is this difference, that whereas in the trachea the thickening
is on the inside of the tube, in the salivary ducts it is
distinctly placed on its outer surface. There is no separa-
tion or distinction whatever between the salivary and poison
glands, which hence are now commonly referred to as the
" veneno- salivary " glands.

Behind the bubar thickening in which it commences,,
the oesophagus is continued as a simple transparent tube,
through the whole length of the thorax, and the only point
that requires special notice is that it is not uncommon,
in dissecting, to bring away attached to it a pair of delicate
bags of air bubbles, the true nature of which will be referred
to in the description of the respiratory organs.

The dissection of the remainder of the digestive apparatus
is a simple matter, and the structures that compose it may
be easily demonstrated in the following manner : — Take a

102 GNATS OR MOSQUITOES — CHAPTER VI

specimen, and place it on a slide with a little salt solution,
and carefully remove the legs and wings. Next tear open
the thorax, so as to loosen the anterior attachment of the
tube, and then carefully partially detach the last two abdo-
minal segments by separating the delicate intersegmental
membrane. Now place one needle so as to fix the last two
segments, and with the other entangled in the thorax pull
steadily on the anterior portion, when the alimentary canal
and its appendages will be drawn out intact attached to the
hinder fragment. If the operation be successful even the
cesophagus and salivary glands may be included in the pre-
paration, but as a rule, the tube is torn just behind the
latter, so that they and the so-called aspiratory vesicle are
left behind.

Assuming, however, that, as is usually the case, only a
portion of the oesophagus is brought away, we find that,
immediately after entering the abdomen the digestive canal
expands into the chylific ventricle or stomach, which forms
by far the largest part of the intestinal canal. It is of con-
siderable width throughout, but especially so at its hinder
■end. Throughout its entire length it is thrown into deep
transverse folds, which recall somewhat the valvules conni-
ventes of the higher animals, and probably serve the same
purpose of increasing the secretory area. In insects that
have but recently emerged from the pupal stage some
remnants of the last meal of the larva may occasionally
be seen, but in the well-established imago plant juice,
pollen, or blood will alone be found. Lastly, succeeding
the broad, hinder part of the stomach, comes the hind gut,
and running into the junction of the two are seen opening
five long, convoluted, dark-tinted glandular bodies, the Mal-
phigian tubes. These are supposed to have an excretory
function, and as uric acid and other renal products have
been found in theui, the balance of opinion regards them
as renal organs. They are lined with a series of large
nucleated cells, the protoplasm of which is exceptionally
rich in pigment and granules. These cells are arranged
in a somewhat peculiar fashion, for as they are too large
to admit of their forming a complete lining, the cell of one

THE ANATOMY OF THE ADULT MOSQUITO 103

side projects into the depression formed between two con-
secutive cells of the other, so that in a tube that has
been a little flattened out by pressure, the lumen appears
zigzag, and the individual cells roaghly triangular.

The last part of the digestive canal is much narrower
than the stomach, its hning epithelium smaller celled and
of a less glandular character, and its wall thick and mus-
cular, while around it may be distinguished some strands
of voluntary muscular fibre springing from the wall of the
penultimate segment.

By a little judicious teazing of the last two abdominal
segments, there may be made out, lying on the ventral
aspect of the hind part of the intestine, four pear-shaped
follicular glands, richly supplied with a close net-work of air
tubes, and connected with the intestine by short ducts.
These are lined with large gland cells, and probably secrete
some fluid accessory to digestion. In the natural position
they are arranged not symmetrically in pairs, but are
" echelonned " in obliquely placed couples.

The respiratory system is, for the most part, quite of
the usual type, the main trunks from the stigmata giving
off communicating branches to those in front and behind
them, and ending in a tuft of branches for the supply
of the muscles and other organs contained within the
segment. There are no distinct stigmata in the head or
in the last abdominal segment, though I have observed
certain tracheae undoubtedly communicating with the ex-
terior near the base of the proboscis ; and the two
anterior thoracic stigmata are by far the largest and
most important in the body, the abdominal openings being
individually small and difficult to make out, as they are
completely hidden by the scales fringing the edges of the
sterna and terga. They are placed rather nearer the front
than the back of the segments, and are best demonstrated
by first carefully removing all scales by brushing, and then
crushing the insect as it lies on its side between cover and
slide, so as to pinch the abdomen from side to side and
bring the lateral surface into view beneath the microscope.

There is, however, in this connection, one structure that

104

GNATS OR MOSQUITOES — CHAPTER VI

calls for especial notice ; the " aspiratory vesicle," or, as I
prefer to speak of them, the pneumatic sacs. This structm'e
is not, as has been stated, in any way pecuKar to the gnats,
but is, I find, often even better developed in the midges, and
other allied insects ; moreover it is not a median, but a paired
structure and I believe that its size, which has hitherto

Pig. 24. — To Illustbate Certain Points in the Anatomy of the
Respiratory System, and of the Pneumatic Vesicle.

A, camera lucida tracing of the distribution of the main tracheae in the
abdomen of C. fatigans (Wied.) x 15 diams ; B, one of the abdominal stig-
mata and the tracheae springing from it, x 75 diams ; C, dissection of a
midge, to show the relations of the pneumatic vesicle; p, pneumatic vesicle;
V, mid-gut, X 15 diams ; D, base of one of the pneumatic vesicles nearly
emptied of air, x 200 diams ; E, air bubbles contained in fibres of the
pneumatic vesicles as seen under -jVth apochromat. objective, x about 1,000
diams. It will be observed that the air bubbles are of irregular and varied
forms, and not uniformly spherical, as they would be if simply suspended iu
a mucilaginous fluid.

been absurdly understated, is inversely proportional to the
size and power of the wings in the different species in
which it is found.

The reason it has been hitherto mistaken for a single
median sac is that, owing to the pressure of the contained
air, the two sacs come to He one behind the other. They
have no true organic connection with the CBsophagus and
the only reason why they have been supposed to be
diverticula of the digestive canal is that they are often

THE ANATOMY OF THE ADULT MOSQUITO 105

brought away attached to it, owing to the fact that the
fibrous base of the sacs, which connects them together
across the middle hne, is divided into two bands, leaving
between them an opening through which passes the
oesophagus, a little behind the valve in which the latter
commences. To the naked eye they look like clusters of
minute air-bubbles, and when intact, their walls rival in
tenuity those of a soap-bubble. At the least touch they
collapse into a bundle of fibres amongst which a few
residual bubbles remain entangled.

Instead of being as they have usually been figured,
barely larger than the salivary glands, they occupy, during
life, as much or more space than the digestive canal when
at its utmost distension. Moreover, the more gorged the
insect or the heavier it be with eggs, the larger will these
sacs be found, as together they occupy a large space and fill
out the entire ventral portion of the body cavity from the
front part of the thorax to the end of the fourth or fifth
abdominal segment.

It is difficult to understand how any one could ascribe
to such a structure as this any active suctorial function, and
so describe it as " aspiratory," and there can be no doubt
that its function is identical with that of the air spaces
in the bones of birds, namely, to lighten the work of the
wings by increasing the bulk of the body ; and further, I
have now no hesitation in affirming that, as too, is the case
in birds, these sacs are mere extensions of the respiratory
passages. Into the base of the sacs may be traced large
tracheae and these split up and become continuous with
a brush of dichotomously dividing fibres of which the
base of each sac is composed. These fibres, like the iutra-
glandular part of the salivary ducts, look much like those
of elastic tissue, and are undoubtedly composed of chitine.
Moreover, if we examine the collapsed " sac " under a
sufficiently high power, a number of minute air bubbles
will be found to remain, and close examination shows
clearly that these bubbles are not entangled between
the fibres, but contained in dilatations of their continuity.
In other words, these fibres are neither more nor less than

106

GNATS OK MOSQUITOES — CHAPTER VI

extremely elastic and distensible tracheae, which swell out
into bubble-containing dilatations, wherever their mutual
pressure permits of their doing so. Apart from a few loose
connective elements, the sacs consist of nothing else than
these curiously modified tracheae.

The circulatory and nervous systems also present no
peculiarities beyond the few points that will be found noted
in the section on the anatomy of the larva and its meta-
morphoses. In the imago", however, the fused cephalic

Fig 25.— Diagram of the Nervous System.
a n, antennal nerve ; o I, ophthalmic lobes ; s g, supra-oesophageal
ganglion, or brain ; s b g, sub-cesophageal ganglion ; t g, mass consisting
of the third thoracic and first dorsal ganglion ; a g, abdominal ganglia.

ganglia, or brain, is very large, and although they do not
appear to possess any specially great amount of insect
intelligence, its relative weight as compared with that of
the body, cannot fall far short of that of even the human
subject. One striking point, however, is the manner in
which almost the whole of the branches of the last ab-
dominal ganglion are supplied to the internal generative
organs. So much is this the case, that a successful pre-
paration of either the ovaries or testes almost always carries
away with it this ganglion.

THE ANATOMY OF THE ADULT MOSQUITO

107

The genital organs of the female consist of a pair of
more or less spindle-shaped bodies, which, in the miimpreg-
nated insect, are contained mainly in the last abdominal
segment, and lie obliquely in it on either side of the middle
line, their distal extremities being supported by a sort of
ligament springing from the common fibrous lining of
the segment. They contain a large number of soft closely
packed ova in various stages of development, those nearest
the opening of the oviduct being the most advanced. They

<^#.

Fig. 26.— Generative Organs of a Gravid Female Mosquito.

A. Generative organs of C. fatigans, ? x 18 diams. ; or, right ovary and
oviduct full of mature ova ; ol, outline of left ovary ; g, last ganglion of
abdominal chain ; d, ovarian ducts ; ocl, common oviduct ; st, spermo-
thecse ; op, ovipositor ; B, spermatozoa from the spermotheca, x 450 diams. ;
C, a mature ovum, x 45 diams. ; b, portion of coat of ovum proper, show-
ing its spinous structure.

are richly supplied with tracheae, supphed from the stigmata
of the penultimate segment, as well as with large nerve
threads from the last abdominal ganglion.

When further advanced, the ovaries and their ducts take
up a large share of the cavity of the abdomen ; forming a
pair of large, lobulated masses. At this stage, very little of
the true ovarian tissue can be distinguished, almost all that
can be seen being the enormously dilated funnels of the
oviducts. Each egg consists of the ovum proper, clothed
in a chitinous envelope closely covered with minute

108 GNATS OK MOSQUITOES— CHAPTEli VI

spines, within which may be distinguished an elongated
embryo, and a pecuhar sac at one pole containing a single
large polynucleated cell, which appears to be in some way
concerned in the bursting of the egg membranes ; the whole
bemg enclosed in a delicate, structureless, external mem-
brane, the total length of the whole egg being 0'71 mm.
Their width is but a fifth of their length, and they are much
broader at the end provided with the polar capsule, the base
of which is marked with fine radiating striae. In the
oviduct this broad end always lies forward, so that the
narrow pointed end is that which appears first in delivery.
The ovaries communicate with the common oviduct by
means of a short, transparent, funnel-shaped tube ; and this
latter is a short transparent canal, which commencing in
the junction of the funnels of the ovarian ducts, runs straight
backwards without convolution or deviation from the middle
line, to open between, and at the base of, the ovipositors.
Just before its external termination, it receives on either
side the ducts of three small glandular bodies, each con-
sisting of a spherical glandular portion about 0-5 mm. in
diameter and a short neck or duct. They are filled with an
opaline white fluid, which consists of a material identical
with that found in the receptaculce seminales of the male,
the peculiar fibrous matrix as well as the spermatic elements
being undoubtedly transferred en masse from the receptacles
of the male to those of the female, their contents differing
only in the fact that, whereas in the male only the earlier
stages of spermatogenesis are represented, in the female, the
contents consist of more advanced spermatic elements and
of fully developed spermatozoa, so that the later stages of
the process take place in the body, not of the male, but of
the female.

The spermatozoa are minute comma-shaped bodies with
a rather stiff one-sided tail, and a pear-shaped nucleus, the
long diameter of the head being about 1 ^\{. I see no reason
for thinking that these bodies serve any other function than
that of spermothecae, and their dull colour is due, not to
any opacity of their contents, but to the sooty pigmentation
of the sac- wall. Thev contain absolutely no glandular

THE ANATOMY OF THE ADULT MOSQUITO

109

elements, and therefore cannot also as Arribalzaga supposed,
secrete the gelatinous fluid that cements together the eggs in
the " egg-boats."

The ducts of the spermotheca^ are strengthened by a
spiral fibre of chitine, just as in those of the salivary
glands ; and just behind their entry into the common
oviduct will be seen that of a special accessory gland which
secretes a mucilaginous fluid which can hardly have any
other function than to furnish a protective coating to the
eggs, and to cement them together in those genera which
deposit them in boat-shaped masses.

Fig. 27. — Sagittal, Section op last Abdominal Segment op a 9 Culex,

SHOWING THE EnTEY OP THE SpERMOTHECA AND CeMENT GlAND INTO

THE Oviduct, (After Kulagnin.)

The male generative organs consist of a pair of small
yellowish bodies, the testicles, from which the sperm is
carried by the vasa deferentia, which are simple straight
tubes, to the ejaculatory duct which originates in their
union. Just before the termination of the vasa deferentia,
they receive the ducts of the receptacula seminales.

The function of these vesicles requires further investi-
gation, as their contents do not, as one would expect,
consist of spermatozoa floating in a simple fluid, but of
spermatic elements in various stages of the process of
spermatogenesis, anterior to the appearance of completely
developed spermatozoa ; and what is more remarkable,
these are not floating in fluid, but are entangled in a fibrous

no

GNATS OE MOSQUITOES— CHAPTER VI

mesh. Presumably not only the spermatic elements, but
detached portions of the fibrous basis of the testicle are
carried down the vas deferens and stored in the receptacula,
but the manner in which this is effected is difficult to
understand. The ejaculatory duct is short, simple and
straight, and ends in a short, fleshy penis, which is unpro-
vided with any chitinous armature.

Fig. 28. — Male Generative Organs op C. fatigans.

A, The male organs, the vasa deferenlia being represented as curved to
save space (they actually pursue a perfectly straight course) x 18 diams. ;
t, testicles ; v d, vasa deferenlia ; r s, receptaculcB seminales ; e d, ejaculatory
duct; c, claspers or external gonapophyses ; ig, internal gonapophyses ; p,
penis ; B, contents of receptacula seminales, x 450 diams.

All these points can be made out, if not at the first attempt,
at least after examining a few carefully made, teazed prepara-
tions, and the reader is strongly recommended to follow
the preceding descriptions with such preparations before
him rather than to trust to illustrations, which, owing to
the extreme delicacy of the objects which it is attempted to
represent, are at best mere diagrammatic representations,
and can do no more than serve as a guide to the recognition
of the parts sought for.

Apart from the fact that the ova differ markedly from

THE ANATOMY OF THE ADULT MOSQUITO 111

the above description, which applies only to Gulex, there is
no noticeable difference between the internal anatomy of
the different genera; and, putting aside of course the
generative system, the two sexes closely resemble each
other, so that I cannot confirm Prof. R. Blanchard's state-
ment {Bull. Acad, de Medecine, 3e. Ser., t. xliv., p. 21.) that
the buccal apparatus of the ? , is much the more developed ;
for I find that, for all purposes of anatomical investigation,
one may use either the 3 , or the ? , . as interchangeable
material.

The above account will, I think, be found sufficiently
detailed to serve as an introduction to those engaging in
work on the subject, but is necessarily a mere sketch of the
subject, for so many points force themselves on one's notice
that it is difficult to avoid being tempted into too great
detail. It would indeed be easy to fill as large a work as
the late Prof. Huxley's wonderful work on the cray-fish if
not to copy the luminousness and charm of the author, but
anything beyond an outline of the subject would be out of
place in a book like the present.

112

K

CHAPTER VII.

On the Life History and Seasonal Prevalence of Mosquitoes.

The life history of any species is necessarily mainly
determined by the nature of its food ; and in these insects
this differs entirely at different stages of their career ; for
while the larva is a purely aquatic animal, feeding on pond
organisms, vegetable and animal, the adult flying insect,
from the structure of its mouth, can subsist only on fluid,
or at most, emulsified forms of nourishment ; and as
naturalists are generally agreed that gnats cannot travel
to any considerable distance, it follow^s that food both for
adult and larva must be obtainable within a limited area,
for mosquitoes cannot and do not fly far. It is impossible
to fix any absolute limit to their powers in this respect, but
it may be safely asserted that few individuals ever stray
much more than a quarter of a mile from the pool in which
their larval youth was passed, and the great majority never
travel further than the nearest shady spot. Nor, in spite
of popular beliefs to the contrary, can they be carried far
by the wind. Mosquitoes, indeed, exhibit a well-founded,
instinctive dread of boisterous weather, and will not leave
shelter in a high wind. Those accidentally carried away
are, I am inclined to think, rapidly disabled.

When serving as Naturalist to the Indian Marine Survey
I was necessarily at sea for months together, and though
butterflies, dragon flies, and other strong flying insects often
boarded us far out at sea, I found no evidence of small
and delicate species being transported in this way.

A run up the Hooghly for coal would fill our cabins
with these unwelcome shipmates ; but, once fairly at sea,
their numbers rapidly diminished, though a certain number

LIFE HISTORY AND SEASONAL PREVALENCE 113

remained long enough to make it certain that the dissemina-
tion of species through the agency of steamships is by no
means impossible, and may be of not infrequent occurrence.
It was only, however, when the ship chanced to lie close to
the shore that she was invaded in this way. The myriads
of Mosquitoes that invade a ship moored alongside the bank
in Garden Keach must be seen and felt to be believed, but
when moored " in the stream " one is left comparatively
at peace ; and in open ports, ships anchored at the usual
distance from the land remain practically free from these
pests.

Another reason that makes it impossible for Mosquitoes
to be carried overseas any considerable distance by the wind
is that, whatever ixiay be the rate of travel that they can
bear without injury, the entire journey must be made at
night, for in tropical regions shelter from the sun during
the day is a matter of life and death to a Mosquito, as is
proved by the following observation.

In Oudh, during April, it is difficult to sleep in comfort
except in the open. My servant one night was careless in
the tucking in of the mosquito net, and when I woke in the
morning I found I should have done well to have imitated
the Irishman of the story and " crept out under the bottom
bar," for the net was aHve with happy, satiated dames of
the gnat community.

My bed, as a matter of fact, had no bottom bar, but
doing my best to imitate the astute Hibernian, I crept out
with infinite precaution, tucked in the curtains, and gave
directions that the bed should be left where it was. The
day turned out somewhat cloudy, the temperature inside
the net never exceeding 105° F., but by noon every Mos-
quito was not only dead, but bone-dry.

For these reasons, we may I think reject, as having no
foundation in fact, such popular beliefs as that the swarms
of Mosquitoes that sometimes appear on the Persian coast,
have been carried by the wind 200 miles across the Gulf
from the Arabian shore ; albeit you must be prepared to
hear this belief quoted as an established fact, even by
European residents.
8

114 GNATS OR MOSQUITOES — CHAPTER VII

Food then, alike for larval and adult insect must be
obtainable in close proximity. The pupa does not eat.
In the adult stage, the usual food of probably all species
of both sexes is the juices of plants, and although the
females of many attack and suck the blood of vertebrates,
food of this sort does not seem to be in any way essential to
their welfare, as most species will lay their eggs equally well
on a purely vegetable diet. Moreover, even in the case of
notoriously troublesome species, it is often difficult to induce
them to bite animals, although they will still feed on their
more usual vegetable food. A high atmospheric tempera-
ture appears to be the main determining condition of these
outbursts of sanguinary instinct, and this is probably the
reason why, in spite of gnats being quite common, we
are so rarely troubled by their attacks in England, for it is
noticeable that whenever there occurs a spell of excep-
tionally hot weather, we find in the press notices of an
invasion of England by Mosquitoes, which, however, always
turn out to be common indigenous species when submitted
to a competent entomologist. The same thing may be
observed in Northern India, where it will be found that
Mosquitoes cease to be troublesome, some time before they
have finally hidden themselves for their cold weather rest.
Speaking generally, only the females bite, and it is notice-
able that different persons differ greatly in their reaction to
the stings of these insects. When Mosquitoes are unable or
unwilling to obtain blood they suck the juices of plants.
They are frequently found on flowers, and especially in
England on the catkins of the willow. On hot days and in
places exposed to sunshine they remain at rest until the
evening, but especially in wooded localities they often are
active in the middle of the day, and may often be observed
perched on leaves and making a sort of balancing movement
of the body, by alternately bending and extending their legs
in the same way as do many of the TipulidcB.

In India too, the most usual recognisable constituent of
the contents of the stomach, of all species I have examined,
in both sexes is pollen. Now some pollen grains look not
unlike the printed representations of zygotes, and I have

LIFE HISTORY AND SEASONAL PEEVALENCE 115

met with an instance when they were mistaken for them.
No fairly expert microscopist is of course likely to be misled
in this way ; but it will be at least a harmless precaution,
for anyone who has to depend for guidance on figures, to
make himself familiar with a few of the commoner pollens
before starting zygote hunting.

In addition to the juices of living plants and animals,
many species will also feed on such articles of human food
as are suitable to the structure of their mouth parts and
they show a marked preference for sweet things. Hence
mosquitoes may often be seen clustered on the moist sticky
sugar exposed for sale in the shops of an Indian bazaar, and
a few will generally be found feasting on the preserved fruit
and sweets laid out for desert on an Anglo-Indian dining-
table, or on puddings put aside in the larder. Not indeed that
they confine themselves to such articles ; they are not
adverse to milk, and will also attack raw meat, or indeed,
any article of food out of which anything can be got by
suction. On one occasion I found some Mosquitoes un-
mistakably feeding on filth, but believe such a habit is
adopted only as a last resource, for at the time I observed
this, all plants were dried up and juiceless. As is well
known, many butterflies which also normally feed on plant
juices, exhibit the same incongruous taste.

Although one of our commonest Indian species is reputed
to do so, I have never actually seen a male mosquito bite,
and beheve that when they do so it is only an exceptional
indulgence. There is, however, no doubt that some few
species do so occasionally.

On this point Mr. Austen remarks : " While it is certain
that in a natural state only an infinitesimally small pro-
portion of all the Mosquitoes that come into existence can
possibly taste the blood of a warm-blooded animal, it is
reasonable to suppose that primitively all Gulicidcs fed upon
the juices of plants. Indeed it has been stated that at the
present day some species are still exclusively vegetarian in
both sexes ; that in others, while the males are vegetarian,
the females suck animal blood — in some species only
exceptionally, in others habitually ; and, finally, that there

116 GNATS OR MOSQUITOES— CHAPTER VII

are species in which both sexes suck animal blood (this is
said to be the case in an Egyptian species and in two
Itahan ones; it has also been observed in two species
found in Madagascar, and has been noticed at Bannu (N.
India)."

When in the humour to attack, it is not only man, but
also other mammals as well as birds that are laid under
contribution ; and it is propable that they do not even con-
jfine their attention to warm-blooded animals, for if, as is
most likely the case, they are the habitual intermediate
hosts for the majority of the class of protozoal blood-para-
sites, of which that of human malaria is the best known
example, it seems likely that tortoises and frogs are the
qhosen victims of certain species, as haematozoa closely
aUied to Laverans parasite are well known to occur in
animals of these classes.

Bovine malaria has been shown by Smith and Kilborne
to be carried by a tick {Ripicephalus annulatvs), but owing
to the amphibious habits of the former animals it is
unlikely that the alternative host of their malaria can be
any wingless arthropod.

' The staple food of the larva? of all the well-known genera
is undoubtedly of a vegetable character, such as monocellular
and filamentous algae, and in dissecting larvae I have never
met with anything else in the contents of their stomachs.
There can be no doubt, however, that they feed also on
animal matter, and may be not only predatory, but actually
cannibaHstic. The larvae of Coretlira are said to be espe-
cially remarkable for procHvities of this sort, but I have no
personal experience of this genus, and otherwise there seems
but little difference in the habits of food of the different
genera. The Indian species of Anopheles are no exception
to this. Quoting from a communication to the Bombay
Natural History Society, advance proofs of which have been
kindly sent me by the author, Mr. E. H. Aitken, one of
our best field naturalists, I extract the following graphic
description of their habits : —

" The favourite vegetable food is a soft fucus, very like
cotton wool dyed green, which is found in clear running

LIFE HISTORY AND SEASONAL PREVALENCE 117

water ; or that thick spongy growth which clothes the sides
of fountains in which there are no fishes, and gets detached
and floats like a thick, dark green scum. The principal
animal food is the cast-off skins and pupa cases of water
larvae, including those of its own kind, and the remains
of dead mosquitoes and other small insects. These often
collect in patches on the surface of a pond, and one dip
under such a patch will secure hundreds of larvae after you
have searched the rest of the pond in vam. Now the larva of
Anopheles, unlike that of Gulex, floats flat on the surface of
the water, and it is much more unwilling to go down than
Gulex. If green food is to be had not more than two or
three inches deep, it will go down and feed, but it comes up
again very soon, and would evidently rather not go down
at all. As it floats you will see two little organs on the
front of its head incessantly stirring the water. These are
the " whorl-organs." They are crowned with little brushes
of bristles, and their function is to keep up an eddy, by
which every little floating particle which passes by is sucked
in towards the insect's mouth. With a lens you can see
this quite plainly and observe it seizing the little particles
with its jaws, sometimes eating them and sometimes throwing
them away with an angry toss of its head. I do not think
that even living objects, if small enough, are refused, and,
in fact, I am almost sure that the larger larvae sometimes
eat the little ones. They are all very ill-natured and bite
savagely at each other when they get the chance. This way
of feeding explains why Anopheles likes a certain amount
of motion in the water, for it brings food, and why it must
starve in deep water unless there chances to be a great deal
of food, animal or vegetable, floating on the surface. The
larvae of Gulex dive much more freely and are more pro-
miscuous in their diet, and, since they float with their heads
down, they do not care much for floating matter. Hence
they dislike any motion. The more stagnant the water is
and the more dirty, the better it pleases them. Lastly,
Anopheles larvae must have sunlight, though they will hate
it more cordially than other mosquitoes when they come to
mosquitos' estate."

118 GNATS OR MOSQUITOES — CHAPTER VII

The necessity of sunlight is mentioned by Meinert, and
is also noted in a letter I had received from Dr. St. George
Gray, of St. Lucia, in the West Indies, and furnishes an
explanation of the failure of all our efforts to keep these
larvae in health in captivity. In speaking of the habits of
Gulices, Mr. Aitken refers to the specially domestic species,
for those of many sylvan gnats are less tolerant of dirty
water than any Anopheles of my acquaintance, and those of
Mr. Theobald's new genus Stegomyia, are the most dainty
larvae I know of in this respect, which probably accounts for
the fact that they are all essentially insects of the rains.
Species, such as C. fatigans, which live in reall}^ foul water,
subsist largely on the vibriones with which it swarms.

Waterless tracts, or such as are desert owing to the
character of the soil, are necessarily free from Mosquitoes,
except through the indirect agency of man, but in by far the
majority of situations, food for both their stages of existence
is obtainable, and gnats of some sort are included in the
fauna ; though the species harboured will vary with the
sort of accommodation offered.

In searching for larvae, it is little use attempting to do
so by a mere inspection, as owing to their universally pro-
tective colouring, it is easy to overlook them, even when
present in large numbers, if one merely stoop down and
peer into the puddle. The examination should be made
by dipping up some of the water from the surface of the
puddle in an ordinary tumbler, and examining it by trans-
mitted light. If care be taken not to disturb the mud at
the bottom, it will usually be easy to do this at once, but
should the water get stirred up, the specimen must be put
aside till it settles.

The habits of the larvae, not only of the different genera,
but also of individual species of the same genus differ
greatly, and much misconception has arisen from hasty
attempts to generalise ; for, as a matter of fact, it is impos-
sible to say that you will find the larvae of Anopheles in
certain situations, or in water of a given degree of purity
or otherwise, as those of some species frequent only fairly
pure collections, while others are hardly less particular than
many, even domestic species of Culices.

LIFE HISTORY AND SEASONAL PREVALENCE 1

Lynch Arribalzaga (L. A., p. 23) classified Mosquitoes,
as regards their habits, into domestic, field, marsh and
wood Mosquitoes; and Ficalbi (F. Y. S., p. 62) into
" foveal," which I take it, may be translated as domestic,
paludal, and sub-paludal species.

The word paludal, however, appears to be a misnomer
based on an unconscious remembrance of bygone notions of
the etiology of malaria, for all marshes of dimensions
worthy of the name, shelter too many of the greatest
enemies of the larvae to be regarded as the typical habitat
of any species ; and for practical purposes, it is more
convenient to divide them into domestic and field Mos-
quitoes ; though the true forest gnats, such as the genera
Megarhina, Sabethes, Mdes, Mdoinyia, and Ceranotoenia
mostly, which inhabit the depths of tropical jungles, far
from the haunts of man, form a class by themselves.
Psorophora and Mucidus are field rather than jungle gnats,
and Stegonujia and Armigeres, are rather field than
domestic, and are never, so far as I know, found in dirty
water. I know of no really jungle species of Anopheles,
all species being either field or domestic, but Culices are
found in all possible situations.

It would be a mistake to attempt to attach too rigid a
signification to terms of this sort, and I would particularly
caution against premature attempts at generalisation, but
broadly speaking, the domestic gnats are those which infest
habitations, while their larvae are mostly found in small
artificial collections of water, such as garden tanks, broken
crockery filled by rain, waste water puddles and so on ;
while the field species, though often found in and about
habitations, pass their larval stage in fairly clean, natural
pools, or at any rate in such as afford conditions similar to
those found in the open.

Some typically domestic gnats, such as C. fatigans, will
deposit their eggs in the foulest of water, and I have found
its larvae in apparently flourishing condition, in the small
cemented tanks, for the temporary storage of sullage water
in Indian towns, known as nahhddn, the contents of which
are filth, far more concentrated than the sewage of any

120 GNATS OR MOSQUITOES — CHAPTER VII

European town. On the other band, unless my suspicion
that however close their resemblance, they are really
distinct forms be correct, the same larvae will be found
during the rains, alone or in company with Stegomyia
larvae, in almost drinkable water. Culex pipiens is of
course another typically domestic gnat, and C. annulatus,
spathipalpis, nefiiorosus, viridiventer and pulchriventer may
be cited as examples of field species, though the last is
perhaps, strictly speaking, a forest form.

The habits of the different species of Anopheles larvae
also vary greatly An. Bossii is certainly essentially a domes-
tic species, and so too, I think, is An. Sinensis with us, but
An. Jamesii, which appears to be the species that swarms
in the Madras paddy fields, is probably a field species, and
so too, it may be, is An. culicifacies , sp. n., which is
probably the species referred to by Mr. Aitken, in his
above-quoted note.

During the past year I have observed the larvae of
An. Bossii in a variety of situations, but all essentially
"domestic." In the spring months, in upper India, none
are in evidence, and I first found them towards the end of
April, breeding in a pool beside one of the piers of the old
bridge across the Goomti, which flowed just beneath my
house in Lucknow. The pool was but a few yards long
by not more than six feet wide, and though it did contain
a certain amount of green filamentous vegetation, was
extremely foul. Still it is the nearest approach to the
"Anopheles pool" of the West African Malaria Com-
mission that I have met with inhabited by the larvae.
Typical pools of the sort I have indeed come across by the
dozen, but in no case have I met with them in such pools.

The horizontal posture of the larvae was, however, very
noticeable, and when placed in a tray under lens, I could
also verify their peculiar trick of screwing round their heads
so as to look upward. During the dry, hot weather, how-
ever, their appearance is with us exceptional, though in the
damp eastern and southern parts of India they may be
found at any time of the year ; and it is not until the rains
.-are fairly established that one can expect to find them in
any numbers.

LIFE HISTORY AND SEASONAL PREVALENCE 121

In July and August last, my duties involved an extensive
tour through the province during the rainy season. In the
early part of the period it was evident that though Ano-
pheles larvae were common, they had not long been so, as
even in places where they were plentiful, I could find
no adults. It was not indeed till the end of the month that
I began to find them at all commonly in bungalows. The
situations, however, in which I found the larvae entirely
upset all the notions I had gathered from recent writings on
the subject. I began, of course, by looking for the typical
Anopheles pool of Ross, but such as I found never held any
of the expected larvae, and the first place I met with them
was in the garden of the Meerut Club, in the small irriga-
tion tanks described in the next chapter Here they were
present in enormous numbers, sometimes alone, but more
frequently in company, and apparently on excellent terms
with the larvae of C. fatigans. It was, however, noticeable
that while the Culex larvae for the most part remained in the
middle of the tanks, those of Anopheles generally kept them-
selves floating with their tails touching its side walls, and
so might easily be overlooked. In my subsequent wander-
ings, I met with Anopheles larvae in a variety of situations,
but always these small irrigation tanks were the " surest
find," and further I never met with them at any distance
from human habitations, so that I am inclined to suspect
that females are unable to mature and deposit their eggs
until they have had a feed of blood. I have also met with
Anopheles larvae in muddy pools of some size in brick-fields,
in the overflow from stand-posts in large cities supplied with
a regular filtered water supply, and even in a very shallow
depression in the concrete surface of the platform of a bust-
ling railway junction, also fed by a stand-post.

In Hong Kong, Dr. J. C. Thomson writes, in The
Government Gazette of November 24th, 1900. "The usual
habitat of the larvae of the Anopheles Mosquito is the natural
water courses, and their favourite locations little breaks in
the rocky surface by the side of the stream, where the
merest trickle from the stream itself prevents entire stag-
nation, and where there is no thi-ough-wash." This, it will

122 GNATS OK MOSQUITOES — CHAPTER VII

be noted, is exactly the same sort of situation in which they
were found by Drs. Stephens and Christophers, in Sierra
Leone, and there, both these gentlemen and Koss never
found them in really dirty water, so that presumably, both
An. costalis and funestus, the species concerned, confine
themselves to fairly clean water ; but, as we have seen»
several species can adapt themselves to less nice conditions,
so that no generalisation as to the sort of water in which
Anopheles larvae will or will not be found is possible, and
all attempts to formulate such are simply misleading for
the genus, and where recorded, should be accepted exclu-
sively for the species to which the observations refer, and
to no other.

How ditiicult it is to make any definite statement as to
the habits of even a single species, may be judged by the
fact that even in Italy, where more attention has been
devoted to such points than anywhere else, the larvas of
one and the same species are stated to be " paludal " in
habit by one authority, while the imagines are said to
be domestic by another observer. The fact is that it is
impossible to lay down any hard and fast rule, and all that
can be done at present is to point out the desirability of
always recording the exact species referred to, and in noting
the habits of either larvae or adults, so that materials for a
more complete statement of the case may accumulate in the
future.

The females of, I think I may say, all genera of Mos-
quitoes lay their eggs in water. During the past year, a
large number of observers have been noting these and
kindred points, and no one records any fact that in any
way confirms Boss's suggestion that they may sometimes
deposit them on the dried-up ground where puddles have
previously existed. A gravid gnat must deposit her eggs
somewhere, and in such surroundings as the interior of a
test tube, which is about the most unsuitable vessel I know
of wherein to confine gnats with the view to observe points
in their life history ; there is nothing really suggestive of
usual habits in her being driven to deposit them on the
walls of the tube.

LIFE HISTORY AND SEASONAL PREVALENCE

123

The attitude of the females in depositing their eggs is
pecuhar. Thanks to the air entangled in the pulvilli, with
which the feet are provided, they can rest on the surface
of still water, as easily as upon dry ground, and are able to
alight upon or take flight from its surface at will. When
they have settled themselves in this way on the surface of

Fig. 29.— Various forms of Mosquito Bggs.--(l) Egg-boat of Culex, seen
from above ; (2) the same, side view (after Sambon) ; (3) separate Culex eggs ;
(4) eggs of Panoplites (after Daniels); (5) eggs of Stegomyia; (6) the same
more highly magnified (after Theobald) ; (7) groups of Anopheles eggs, as
they float on the water (after Sambon) ; (8) egg of Anopheles maculipennis,
showing lateral floats, seen from above, x 30 diams. ; (9) the same, viewed
laterally, (after Nuttall).

the water chosen for the purpose, they spread out the
four anterior legs as a support, while the posterior legs
are crossed behind them so as to form an acute angle, and
it is into this angle that the eggs are dropped side by side,
until an elongated, boat-shaped mass is formed, somewhat
raised at each extremity. The hind legs which carry them
are extended little by little until the boat is finished, when

124 GNATS OK MOSQUITOES — CHAPTER VII

the insect allows it to drop into the water at the mercy of
the wind.

In speaking thus of " boat-shaped masses," it must be
understood that Ciilices only are referred to, as Ano})heles,
Mucidus, PanopUtes, Stegotmjia, and some drop their eggs
separately, and in these genera the eggs are provided with
separate arrangements to secure flotation.

In Culex, the eggs are glued together, as they are laid,
into canoe-shaped masses, which, in size and colour look,
as they float on the water, not unlike caraway seeds. Each
boat consists of from '250 to 300 eggs, which latter are
oblong, more pointed at their upper extremity as they lie
in the boat, larger and more rounded below, and ending
abruptly in a bordered edge, much like that of certain
liqueur flasks ; the opening of which may be said to be
closed by a thin membrane, by the rupture of which the
larvae escapes. The lower or immersed end is furnished
with a curious opening, or rather thinning of the egg-shell,
and is surrounded with some curious radiated markings.
The object of this arrangement is probably to enable the
embryo to obtain oxygen from the water in which the
little raft floats.

The egg-boats must needs float on the surface of the
water, as the embryos perish if they become submerged.
Only their neck comes in contact with the water. When
just hatched the eggs are entirely white, but they soon
become shaded green, and in less than half a day they be-
come grey. Usually they are laid between five and six in
the morning, but in England Mr. Theobald finds that the
evening is also chosen for the purpose, and the larvae escape
in two or three days, but the interval varies a great deal,
according to the temperature, and the particular species.
Those of Mucidus apparently take a long time, as Major
Close, I.M.S., at Moradabad, kept some, laid by a confined
female for a month without the larvae appearing. As, how-
ever, the experiment was conducted in a test tube, and it is
possible the embryos may have perished for lack of oxygen,
this observation must be taken with reservation. As soon
as the larvae of Culex have escaped from the eggs, the egg-

LIFE H [STORY AND SEASONAL PREVALENCE 125

boats break up, and the empty shells form very pretty
microscopic objects, their surface being closely set with
minute spines, the form of vi^hich varies in different species,
and so might possibly be made to serve as a means of
identifying the species to which they belong.

C. pipiens ova measure 0"9 mm. in length, C. nigritulus
6, and according to Howard those of C. pungens (= C.
fatigmis, Wiedemann) 0*7 in length and 0"16 in diameter.

Anopheles eggs are quite different in form and are de-
posited separately. Sometimes, they lie on the surface of
the water in stars or groups of other forms, but as they
are in no way connected together, I regard this as quite
accidental, and of no moment. As regards the form of
the egg, the following excellent description, extracted from
Dr. G. H. F. Nuttall and A. E. Shipley's paper " On the
Structures and Biology of Anopheles " {maculipennis) in
the Journal of Hygiene, I., p. 49, may be quoted: —

" When first deposited the eggs are white, but they soon
darken. Each ovum measures 0'7 to I'O mm. in length
and is at its greatest breadth about 0"16 broad. The egg
is boat-shaped and one end is sHghtly deeper and fuller
than the other. The surface which, were the egg a boat,
would be the upper, is flattened but slightly convex. It is
marked by minute reticulations (fig. 2). The under surface
of the boat is characterised by much larger and more
regular reticulations, which divide the surface into fairly
equal hexagonal areas. The rim {a) of the boat is thickened
and very regularly ribbed. Along the centre of each side,
extending over a space of rather more than one-third the
total length, this rim is much thickened, the ribbing is
more marked and the whole forms a very conspicuous and
characteristic feature of the egg. This thickening recalls
the rounded float which runs along the edge of a life-boat
(fig. 29 ; 7, 8, 9). It serves the same purpose, being com-
posed of air chambers and is used to keep the boat-shaped
egg with its flat surface uppermost. Howard refers to the
membrane we are about to describe as the ' clasping mem-
brane,' notes the reticulated surface exhibited by the eggs, as
also the presence of 5 — 7 minute dark circular spots at the

126 GNATS OR MOSQUITOES — CHAPTER VII

ends. His measurement of the egg is given as only 0"57 mm.
As in other insects the egg doubtless varies in size.

" The colour of the egg soon after it is laid is greyish
black. If the eggs are subject to much attrition a delicate
membrane splits off which gives the surface of the intact
egg its reticulated appearance. Stripped of this membrane,
which desquamates in irregular whitish fragments, the egg
appears with a glistening black surface comparable to that
of patent leather. One end of the egg is slightly blunter
and more rounded than the other, and this contains the
head end of the embryo. It is an interesting point that
when the egg, as frequently happens, is drawn by capillary
action a little way up from the water on to a leaf or some
other half-submerged object the head or blunt end always
points downwards, and thus should the hatching take place
whilst the egg is in this position the larva emerges into
the water, and not into the air."

I doubt if these ova can resist more than a very
temporary dessication, but it is very difficult to judge of
this point by laboratory experiments, as not unfrequently,
they refuse to hatch out at all. The ova of some Stegomyice,
however, which resemble them a good deal, being laid
separately, and having also similar lateral floats, are cer-
tainly much more resistant, and may apparently be unin-
jured by being left high and dry for a month.

Carroll, Agramonte, and Lazear {Philadelphia Med.
Journ., October 27, 1900, p. 291), state that in experiment-
ing on conveyance of yellow fever by Mosquitoes (Stego-
myia, fasciata, Fabr.), they obtained a supply of this
species from Dr. Finlay. Thirty days previously, vide
Finlay, ova had been deposited by a female just at the
edge of the water in a small basin, whose contents had
been allowed to shghtly evaporate, so that these ova, at the
time of their visit, were entirely above contact with the
water. Notwithstanding this long interval, they were
promptly converted into the larval stage, after a short period,
by simply raising the level of the water in the basin.

The separately-laid eggs of Panoplites have a very pecu-
liar shape, like a long, narrow pear, one end being prolonged
into a sort of neck exactly like that of a Florence flask.

LIFE HISTORY AND SEASONAL PREVALENCE 127

According to Meinert, the ova of Gorethra are laid in
flat, round, jelly-like masses, each containing from 100 to
150 eggs, but Miall describes them as arranged in spiral
lines, so as to form a gelatinous sheet. We have no in-
formation as to the eggs of the other genera, and as a
matter of fact, our knowledge on this point is very scanty,
and further observations are much required.

The insects pair towards evening. The males assemble
in large numbers, flying hither and thither without travel-
ling far, and the females appear among them in smaller
numbers, so that the moment one appears she is clasped
by a male, allowing themselves to float in the air, or flying
together. The coupHng only lasts a few moments, and
when it is completed the pair separate, and the fertilised
female proceeds to deposit her eggs.

As already remarked, the time taken by the ova to hatch
out depends upon the species and on the circumstances of
its environment, but, in most, may be said to be about three
or four days. Even when newly escaped, they can be
easily distinguished by the naked eye, if favourably lighted,
as black dots with a transparent, wriggling tail. Being
mainly air-breathing animals, though it is possible that the
anal papillae may aid in respiration, by acting as gills, they
cannot remain under water for any length of time, and all
species pass the greater part of their time at the surface,
lying so that the respiratory opening is open to the air.
The larvaB of Culex, especially those species which possess a
long respiratory syphon, pose themselves in a nearly ver-
tical position, as if hung from the surface of the water, as
is well shown in the following illustration, which was
photographed from living larvae and pupae by my friend,
Mr. Koyle, of Kosa, N.W.P.

The larvae of Stegomyia, whose breathing tubes are
very short, lie much more obliquely, their position approach-
ing that of Anopheles, but this cannot be looked upon as a
generic peculiarity, as those of C. pulchriventer , which have
an equally short syphon, take up the same position, which
is presumably simply necessitated by the shortness of the
tube. The species figured by Christie in his useful little

128

GNATS OR MOSQUITOES — CHAPTER YII

book, "Mosquitoes and Malaria," is probably a Stegomyia
(pi. v., figs. 9, 10).

The larvae oi Anopheles on the other hand, lie absolutely
horizontally, with the respiratory orifices, which are almost

Fig. SO.^Larvse and pupto of C. fatigans, Wied., photographed
from life, about twice natural size.

flush with the rest of the dorsum, just emerged, and the
head barely submerged.

Lying in this position, they look just like scraps of straw-
blackened by long immersion, and as they often lie motion-

PLATE V.

LIFE HISTOEY AND SEASONAL PREVALENCE 1'29

less for a quarter of an hour or more, would be taken to be
such by any one not specially looking for them. As already
noted, they exhibit a great preference for the sides of the
tank or pool in which they may be, and may also be found
along floating sticks, leaves, or other such objects, lying at
right angles to, and with their tails touching it, as if moored
by the stern. When lying thus at the surface, they have
a quaint custom of screwing their heads completely round,
so that its ventral surface looks right upwards, and, though
otherwise motionless, if closely watched it will be seen that
the head is being slowly rotated and swept to and fro, while
by the rapid vibrations of the bristles of the " whorl organs,"
a current of water with its suspended nutritive particles is
kept flowing to it.

It is perhaps rather a misnomer to speak thus of the
moustache-like bunches of bristles that project from a pair
of plates hinged on to the fore corners of the clypeus, as the
bristles all work together, and do not bend, one by one in a
wave, hke the cilia of the Rotatoria, from a fancied resem-
blance to which they have got the name. Although capable
of being moved separately, the two brushes when in action
are brought together, as if grasping at something, and then
open out again, and repeating this movement with great
rapidity ; according to Nuttall {loc. cit.), three times a
second.

He describes, too, how they employ the peculiar bristles
on their mandibles to comb off, into the mouth, the nutri-
tive particles that have collected on the brushes. The
mouth-organs of Culices are used in exactly the same way,
only instead of sweeping the surface, for floating particles,
they depend on such as are suspended in the water beneath.
"When disturbed the larvae dart backwards into the depths
of their puddle, and hide themselves among the loose par-
ticles at the bottom, and then, after a few minutes, float
slowly back to the surface when the coast is clear. A good
deal has been written about these movements, and those
of the pupae, but as they are obviously purposive, they
naturally vary, and hardly merit detailed description.

The nymphs, as may be seen in Fig. 31, float in
9

130 GNATS OR MOSQUITOES — CHAPTER VII

much the same position as those of Culex, but perhaps
the breathing horns are held, sloped somewhat more fore-
wards, ^des larvae are said, by Dr. Lutz, to He in the
water in a sloping position, intermediate between that of
Anopheles and Culex, in other words, much as do those of
Stegomyia. According to Meinert, Corethra larvae are
much less restricted as to the character of the collections
of water in which their larvae are reared being often found
in deep still water containing but little vegetation, as well
as in very foul pools. Their position in the water is even
more absolutely horizontal than that of Anopheles, whose
position is slightly oblique, and they retain this horizontal
posture even during their excursions below the surface.
They are very voracious, devouring not only Daphnice and

Fig. 31. — Larva and Nymph of Anopheles, to show habitual resting posi-
tion. The head of the larva is shown looking up, as when feeding.

other small crustaceans, but also Dixa larvae and even young
fish and molluscous embryos ; nor do they even shrink from
cannibalism, smaller specimens of their own species being
devoured with gusto if they come within the reach of their
jaws. It is only the very young larvae that are so abso-
lutely transparent, as when older they develop a number of
patches of pigment, and the contents of the intestinal canal
obscure the complete transparency of the body. Still, it is
doubtless owing to this transparency that they are the only
gna.ts whose larvae live, as a rule, among fishes. All their
ecdyses are performed with singular rapidity, even the final
emergence of the imago taking only a few minutes to
accomplish. Their pupae assume a nearly vertical position
in the water, but otherwise do not differ greatly from
those of other senera.

LIFE HISTORY AXD SEASONAL PREVALENCE 131

Meinert states that the larvae of Mochlonyx affect the
waters of fields and woods, and that they are especially often
found in such places as ditches, where the water is clear,
but without any particular current. The larvae do not sur-
vive the winter, but he considers it probable that the eggs
do, as he has found newly-hatched larvae so early in the year
as to make it highly improbable that any of the perfect
insects could have emerged from their hybernation, even
assuming that the adults do hybernate. In a warm room,
in captivity, all the changes from egg to imago were gone
through in three weeks. Unlike Culex there appears to be
only a single generation of larvae developed in the year.
The position taken up by the larvae is very like that of
Corethra, but they are more in the habit of remaining
under water, They are voraciously carnivorous, and appear
to be even more arrant cannibals than the larvae of that
genus. The pupa also behaves much like that of Corethra ,
but it is not quite so rapid in its changes of skin.

The larvae of the Culicida swarm in spring and summer
in the stagnant water of tanks and other domestic collections
of water, where they may be found in abundance in Europe
from the time the ice is melted. Usually they keep at
the surface of the water or a little below it, in an inverted
position, the head being lowest, breathing by means of the
tube placed at the extremity of the abdomen. They are
extremely lively and easily disturbed by any movement of
the water, but soon resume their old position. Their body
is elongated and has no legs. At first greenish, they soon
become greyish and transparent. The larva undergoes
several moults, three taking place in the first two or
three weeks. In order to get rid of the old skin it places
itself horizontally on the surface of the water with the back
upwards. As a rule the change takes place through a rent
on the thorax, and extends after to the abdominal segments.
Like the larva the nymph is capable of swimming but
can take no nourishment. When in repose it is contracted
into a lenticular form, its abdomen being bent under the
thorax and kept closely applied to it. It lies vertically in
the water but in a different sense from the larva, as the

132 GNATS OR MOSQUITOES — CHAPTER VII

humped-np thorax forms the highest part, and it is from
the dorsal surface of that part of the body that the
respiratory syphons spring, being in the form of two horns
or asses' ears. Their upper extremity is cut obliquely, and
when the pupa is at rest is always kept above water.
The eyes are distinct, and beneath the thorax is found
a large mass consisting of the antennae, mouth parts and
legs. The abdomen is elongated, segmented, and ter-
minated by two oval plates. If the nymph desires to go
beneath the surface of the water it straightens itself and
gives a few strokes of the tail, but is soon carried back to
the surface, as soon as this has ceased to act, by its own
buoyancy. After passing five or ten days in this state, the
insect is ready for its last metamorphosis, but this is a most
critical period of its life, as if the nymph case upsets during
the process of the imago's freeing itself, the insect perishes
by drowning, as it is now entirely unable to survive con-
tact with the element in which, up to now,, it has passed
its life. The moult that frees the imago from the nymph
integuments takes place in the same manner as the pre-
ceding ones ; a rent appearing in the upper surface of the
thorax through which the gnat protrudes first the head
and thorax as much as possible above the aperture. The
posterior extremity of the body now contracts a little, and
extending itself immediately after, is gradually drawn out
in a perpendicular plane. Meanwhile, the old skin of the
nymph serves as a sort of boat of which its own body
serves as the mast, only a very small portion of the hinder
extremity touching it. Next, having drawn from their
sheaths the four anterior legs and then the hinder ones,
it carries them forward. Soon after it bends towards the
water and arranges its limbs, and thus assured of safety, it
unfolds its wings and flies off. When first it escapes it
is whitish with the thorax greenish, but it very quickly
after assumes the proper colours of the adult insect.

Both larvae and pupse show a remarkable degree of resis-
ting power to physical, and even chemical agencies, especi-
ally those of Culices, though those of Anopheles appear to
bear cold better. In Celli's experiments on them, it was found

LIFE HISTORY AND SEASONAL PREVALENCE 133

that freezing for a couple of days, killed all the larvae, while no
piipge survived more than twenty-four hours. Desiccation,
on the other hand, was better borne by the nymphs, which
in a few days, were transformed into very active Mosquitoes,
in spite of being placed in dry river sand, so that the drying
up of a pool does not stop the development of such insects
as have reached this stage, while the larvse were all dead
in two days if dried at 20° C, and both stages were killed
by two minutes' exposure to a temperature of 40°. Experi-
menting in India, I found that larvae were usually dead and
decomposed before the mud of the pool in which they had
lived had dried up by ordinary evaporation. Nearly all
species can live only in fresh water, but Bancroft found a
species, which has been named marinus, by Mr. Theobald,
living in sea water, and Ficalbi collected the larvae of C.
nemorosus (salinus) in a salt marsh.

Celli found that Anopheles larvae died within thirteen
hours in sea water, and could not live in mixtures stronger
than two parts of fresh to one of sea water, which agrees
with the results of Ficalbi's recent observations on the salt
marshes of Cervia, where he found that though they
abounded in neighbouring sweet water pools, they were
never present in the actual salt pools, and were only excep-
tionally found in even slightly brackish water. Stephens
and Christophers found that mangrove swamps never con-
tained Anopheles larvae, so that the idea that these salt-
water shallows are a cause of malaria is erroneous, though
one can well understand that a saturated atmosphere, reek-
ing with the emanations of vegetable putrefraction, may be
extremely unhealthy for other reasons. Moreover, although
they speak of them as being found in fresh water marshes,
it is evident from what they write that, strictly speaking,
they are no more to be found in such situations in Africa
than they are in India, as they proceed to explain that they
found them, not in the main marsh, but in the pools around
it. In running water, unless the current be very slow, they
cannot exist, and hence they are never found in rivers,
except in extremely sluggish streams, such as the Cam,
where Theobald mentions having once found them.

134 GNATS OK MOSQUITOES— CHAPTER VII

The length of time required to develop from egg to
imago varies greatly ; about a month in Italy, according to
Celli, in the case of Anopheles, but usually much more
rapidly in India.

In Mr. Aitken's paper, already quoted, he says : " I was
anxious to ascertain next the length of the larva life, but
found that it varied indefinitely with the conditions. Given
warmth and plenty of food a larva will come to maturity in
eight days, or perhaps less, but I have had one for more
than a fortnight, and then it died before becoming a pupa.
The time spent in the pupa state in all my specimens was
more than twenty-four, but less than forty-eight hours. So
I think we may put down the time which it takes to produce
a Mosquito at something between ten days and a fortnight,
from the laying of the egg ; and water, treated with kerosine
oil once a fortnight, should be perfectly safe. It is very
difficult to ascertain the normal life of the adult Mosquito,
because we cannot keep them in natural conditions. Some
of mine lived for ten days, but, as I have said, they would
not lay their eggs, and that alone could not but affect the
length of life. . . . To my astonishment I found y4?iOj9/<e/e.s
larvse swarming in a dirty drain filled with rotting straw,
which gave the water the colour of beer. In May, Colonel
Weir sent me two or three bottles of water from other
places in Bandora, with Anopheles larvae in them. The
breaking of the monsoon, of course, upsets the haunts of all
Mosquitoes for a time, but on the '2nd of July I went out to
Chinchpoogly and explored the quarries at the foot of the
hill, where Dr. Christy told me he had found larvae soine
time before. I found larvae in many of the pools and one
specimen in a little puddle, not more than four inches deep,
which must, I am sure, have been as dry as bone ten days
before, for we had a long period of hot, sunny weather, as
you will remember, in the middle of June."

This agrees with such scanty observations as I was
personally able to make in India, but, owing to the peri-
patetic nature of my duties, breeding-out observations were
rarely practicable.

In England, and also in the somewhat similar tempera-

LIFE HISTORY AND SEASONAL PREVALENCE 135

tare of Indian hill stations, Gulicidoi larvae develop even
more slowly than in Italy.

In Northern Europe all larvge and pupse that may not be
ripe for their remaining metamorphoses, perish during the
winter, but in Italy, according to Celli, the larvse of Ano-
pheles " hybernate " (CM., p. 78 and 166) and some, at any
rate, of the adult insects that appear early in the spring are
furnished by larvae that have passed the winter in that stage.

I am nearly sure that the same is the case in Northern
India, though in a tank in my garden at Shahjahanpur,
where I had found them present in swarms on my arrival,
and which I had always kept supplied with water, so as to
keep them under observation, they suddenly disappeared
about January 11th, without any apparent reason, as the
severest part of the "cold weather" was already past.
Unfortunately I was unable to find any other pool, in which
to continue observations, as a long spell of drought during
the last three months of 1900, had completely dried all
such pools as had remained under natural conditions. I
have on one or two occasions noticed the larvae, though
abundant a few days before in all stages, disappear from a
pool in the same unaccountable manner, and strongly
suspect that this must have been due to the outbreak among
them of some epidemic disease.

Although they were thus capable of maintaining them-
selves as larvae, no imagines were being produced, and for
six whole weeks not a single pupa could be found, though
carefully sought for ; nor, so far as I could judge, did the
individual larvae grow to any extent, as all sizes were present
from the first, and the large ones did not appear to become
more numerous.

The lowest temperature ever observed in this tank was
56' F. in the early morning, but the margin of temperature
inimical to pupiation must be a very narrow one as, after a
fall of rain accompanied by cloudy nights, during which the
night temperature could not fall to the usual minimum, the
higher temperature sufficed to bring about the appearance
of a few pupae.

Now the rain had not perceptibly altered the depth of

136 GNATS OR MOSQUITOES — CHAPTER VII

the tank, which was indeed filled periodically from the well
— and so fluctuated more from this cause than it could do
from any shower, — so that the occurrence has evidently no
connection with the rain pei- se. Brought into the house,
these pupae gave issue to their imagines, but I cannot say if
this is also the case with those in the open — at any rate,
none appeared in the bungalow.

Celli's observation may, however, be considered to prac-
tically settle the question, as it is little hkely that the larvae
are unable to survive the mild cold weather of Upper India,
if they can do so in the comparatively severe winter of Italy,
and hence it may be accepted that in subtropical and the
warmer parts of temperate climates, the survival of larvae
is one method whereby the permanence of the species is
secured, not only in Anopheles, but also in other gnats, as
the same thing was undoubtedly the case with both C.
fatigans, Wied., and G. impellens, Walk., although, in their
case, the imagines were more or less in evidence through the
entire period, and were, I believe, even breeding in a leisur-
able fashion whenever the weather turned a trifle warmer.

During the short Danish summer, Meinert finds that
Anopheles can find time for the rearing of but a couple of
broods, and it is natural that in a limited period such as
this, there should be more or less definite seasons of appear-
ance of larvae, and that the flying insects should appear in
periodical swarms ; but in Italy the process goes on more or
less irregularly throughout the much longer summer, and
the same is the case in Northern India.

In Italy, Celli says that the hybernating larvae pupate
and turn into Mosquitoes in April, and the first genera-
tions of new larvee are found in the first half of May.
With us, in Oudh, as we have seen, the new Mosquitoes put
in an earlier appearance, though they rarely can find many
opportunities to deposit their eggs till the advent of the
rains, as even their own winter quarters must usually be
dried up during the fierce heat of the months that intervene.
Just as the process commences earlier with us, so it con-
tinues into the later months of the year, so that active
Anopheles Mosquitoes are to be met with till quite the end

LIFE HISTORY AND SEASONAL PREVALENCE 137

of October, though neither they nor the two persistent
species of Culex, above noted, exhibit any tendency to
bite, so that there is no longer any need for the use of
Mosquito nets, whether on the score of comfort or safety,
for in India, just as is the case in England, we find that the
question as to whether or not Mosquitoes will attack human
beings is entirely one of air temperature. Celli says that
the adult Anopheles will bite as soon as it is out of the pupa
case, and after they have digested their meal, bite every two
days or so, and ovulate several times. As far as I can judge
by the rather desultory experiences of correspondents and
myself, I fancy this conveys rather too high an idea of the
gastronomic and reproductive capabilities of individual in-
sects : that they generally take some four days to renew
their appetite after a full feed, and that the laying of a
couple of batches of eggs is about the limit of the reproduc-
tive capabilities of most of them ; but doubtless they vary in
this, and their intolerance of captivity makes such matters
difficult of determination. However this may be, there can
be no doubt of the enormous capabilities of multiplication
possessed by the entire family, and this strongly tends
to neutralise any check that can be imposed upon it by
natural enemies.

Like the larvse, the adult insects may be roughly classi-
fied as domestic, field, and jungle species ; and they gener-
ally agree in these respects with the corresponding larvae,
though Celli emphatically states than An. maculipennis is
domestic, while An. bifurcatus is a field insect ; whereas
Ficalbi regards the habits of their larvae as the reverse.
It is impossible, however, to make distinctions of a rigid
character in such matters, and the looser our conceptions of
their signification, the better, though they are very useful
for general purposes if employed with such reservations.
Although some few species are active throughout the day,
e.g., An. bifurcatus, in Italy, taken as a family, gnats are
essentially nocturnal, and especially crepuscular insects.
Speaking generally, Mosquitoes start the day's work just
after sundown, and are very busy among the plants and
animal victims that furnish their food, till it falls dark,

138 GNATS OR MOSQUITOES — CHAPTER VII

by which time, though many continue in evidence through-
out the night, most of them have satisfied their needs, so
that there is a sort of lull in their activity till dawn, when
those that have failed to secure a breakfast reappear, in
vicious earnest to secure a supper. Once, however, is the
tropical sun above the horizon, and they may be seen
taking immediate measures to escape from the power of his
rays.

The window-door of my last little Indian work-room was
closed by means of Mosquito netting stretched over the
frame, so that it was possible to use the microscope with it
open, without being tormented with flies ; and I was often
much diverted by watching the persistent efforts, not only of
Anopheles, but of all the other domestic species about, to get
out to feed in the evening, and to return to shelter in the
morning. Shelter, as we have seen, they must have during
daylight, and speaking generally, an observer who confined
his observations to those hours might report that no such
insects existed in the tropics, so sluggish are they during the
heat of the day and so cunningly do they hide themselves in
dark corners. During the rains, however, the Stegomyiie,
which are common at that season, may be seen flying about
such places as orchards, and will occasionally even venture
upon short flights across patches of sunshine. Within the
bungalow, too, they will often bite viciously, even in the
middle of the day ; but as they are rather field than domestic
in their habits, they give less trouble in this way than would
otherwise be the case.

When settled on some surface, the position of these
insects is very characteristic. As a rule, they support them-
selves entirely on the four anterior legs, the hind pair not
being brought into requisition, but kept held up well above
them, so as to serve as extra feelers. During the active
hours of their existence they are kept slowly waving about
in the air, as if feeling for something ; but when quiescent,
though the hind legs are still rarely placed to ground, they
are kept quite still, so much so that I found little difficulty
in photographing the living insects, placed in the glass box
described in the first chapter, with even an entire minute's

PLATE VI.

.A>

^

^m^.

Photograiihs of living Anoi>helcs and Ciilc.v.

Face ]).

LIFE HISTORY AND SEASONAL PREVALENCE 139

exposure. The Cullces sit humped up, with the abdomen
drooping, or at most horizontal, and the proboscis held out
in front of them, so that even when they are looking at the
surface on which they are standing, it is at least kept off the
ground. The palpi are kept arched backwards and upwards,
and usually slightly separated from the proboscis, and the
antennae usually sloped well behind its line. In Anopheles,
on the other hand, there is a strong tendency to keep all the
appendages of the head in the same plane, and in both sexes,
the four basal joints of the palpi are kept closely held against
the proboscis. In the female the whole organ is so held,
but the clubbed terminal joints of those of the males are
kept spread out at an angle of about 45 degrees. On this
account the trunk of this genus of gnat looks almost as
thick as the rest of the body. In addition to this, instead of
bunching themselves up as the Culices do, they keep the
whole body and cephalic appendages nearly in one straight
line, obliquely, or it may be, almost vertically to the surface
on which they are resting, with the abdomen held up, and
the proboscis pointing at the surface on which they rest, as
if they contemplated boring themselves into it. This attitude
is so characteristic that it affords a ready means of recog-
nising the genus, when found settled. In the case of A71.
sinensis, An. Bossii, An. Jamesii, An. argyrotarsis, An.
albipes, and also, to judge even from Sambon's own figures,
in those of An. superpictus and An. maculipennis, there can
be no doubt that the habitual resting attitude is such as has
been described ; but I do not for a moment suppose that it
can be taken as an absolute rule for all species, either of
Anopheles or Culex ; and just before I left India I met with
a new species of Anopheles which entirely deviated from the
common rule in this respect, as every one of some dozen
living specimens of both sexes was found resting in a position
exactly such as is regarded as characteristic of Culex. The
resemblance is further aided by the fact that in the fresh
state the abdomen is conspicuously banded in a manner
most exceptional in the genus, so that when I arrived in the
rest-house in which I found them, late one night, I was
completely taken in by the counterfeit, mistaking them for

140 GNATS OR MOSQUITOES — CHAPTER VII

C. imj^ellens, to which they present a close resemblance.
On this account I have named the species A71. culicifacies
sp. n. On the other hand, I suspect that C. mimeticiis, Noe
simulates Anopheles, not only in costume, but in its habitual
position when at rest, though I have seen too few living
specimens to be sure. It must be remembered that these
remarks refer to habitual attitudes only, and that there is
nothing absolute in the application of any natural history
rule of the kind, so that there is nothing noteworthy, as
some writers appear to have imagined, in finding an indi-
vidual Culex with its tail cocked in the air, or an Anopheles
with the abdomen drooping ; for being living animals, they
may be occasionally found in any possible position, however
characteristic certain particular poses may be of their usual
ways.

When the season favourable for active existence and
breeding is over, the majority, at any rate of species, proceed
to find some suitable shelter, in which they may lay up,
protected from cold and injury, till the return of spring, or,
at the least, the surviving impregnated females do so. In
Northern Europe and similar climates this appears to be the
sole provision for the maintenance of the species through the
winter, and the males rarely or never hybernate ; but in the
south, as we have seen, this means is supplemented by the
hybernating of the larvae as well ; while in really tropical
regions there is no interval of inactivity, at any rate from
cold.

There can be no doubt, however, that not only cold, but
other climatic conditions unfavourable to the free multipli-
cation of a species, such as intense heat with drought, cause
Mosquitoes to retire into a similar condition of inactivity, as
the truth of the suggestion advanced in the first edition of
this book to that effect has been amply verified by my
observations during the past year in India. There can,
in fact, be no other way of accounting for the absolute
disappearance from public ken, for long periods, of certain
species, such as the Stegomyice, which are active only for
certain limited seasons. The larvae are certainly not con-
cerned in the matter, as none of the kind are to be found,

LIFE HISTORi' AND SEASONAL PREVALENCE 141

and water of a character suitable to their needs is nowhere
available, so that it is certain, that though nowhere in
evidence, they must undoubtedly exist somewhere in hiding ;
and though I cannot say I have actually traced them to
their lairs, the unlimited accommodation for secreting them-
selves afforded by the structural imperfections of an Indian
house make this in no way astonishing, and the same is the
case in the cold season with Anopheles, which appear to
have vanished from the earth, so well are their hiding places
chosen.

During the cold weather, of the twenty or thirty members
of the family that form the gnat family of the North-west
Provinces, the imagines of but two, with their larvae, and
the latter stage of a couple or so more, are alone to be found ;
and one of the former, I believe, also disappears into seclusion
when the fierce heat of the dry season commences, leaving
C. fatigans as the only adult Mosquito in evidence. Now,
as the insects are in no sense migratory, it is an obvious
conclusion that a sufficient number of adults, either
impregnated females only, or representatives of both sexes,
to maintain the breed, must be lying by somewhere. In
countries such as England, where buildings are carefully
finished and their ceilings fairly accessible, it is generally
possible to obtain living gnats at any period of the year,
provided one knows the kind of situations wherein to search
for them ; but in India, the ample space between the ceiling
cloth and the thatched roof, and the innumerable gaping
fissures in the rough mud plaster, that is all that covers the
carelessly laid bricks leave so many crannies and corners,
that it is by no means surprising that one may fail to un-
earth the insects from their hiding places without pretty
well pulling down the house, an operation by no means
unattended with risk, as one is apt to disturb other skulkers
of a larger and more immediately dangerous character.

This habit is spoken of as hybernation, and in Northern
Europe, where the males apparently all die before winter has
fairly set in, only impregnated females survive the winter, and
it is through their agency alone that the continuity of the
species is maintained. In more moderate climates, where

142 GNATS OR MOSQUITOES— CHAPTER VII

water rarely freezes during the winter, this expedient may
he assisted by the survival of larvae also, as we have seen
is the case with certainly some species of Anopheles, and
in that genus also, at least in India, it is not the females
alone, but the surviving members of both sexes that seclude
themselves in this way, for towards the end of the season
of activity, males are to be met with in houses, in far
larger numbers than their spouses, and are obviously sinking
into the same lethargic state. Mosquitoes, when hyber-
nating, creep into any hiding place that will afford warmth
and darkness, and prefer to utilise for the purpose narrow
crevices, into which one would hardly expect so delicate a
winged insect to creep.

When in this state they are extremely sluggish, and
may often be found stationary in the same place for long
periods, without having, to all appearance, moved for days
together; and provided one can find them, it is naturally
not difficult to catch them ; but they nevertheless retain
sufficient strength and alertness to escape from careless or
clumsy attempts to capture them, though, with such excep-
tions, they neither feed nor perform any of the other active
functions of life. All Mosquitoes are greatly affected by cold,
and refuse to move or feed on a cold morning, even on their
more habitual vegetable food, while the most troublesome
species never attack animals, except in warm weather, so
that although a couple of species are fairly common in
houses throughout the cold weather in Oudh, Mosquito
curtains can be safely laid aside.

It is only during the times of the year that they are
actively breeding that Mosquitoes attack animals, and the
habit is probably indulged in to supply the large amount of
nourishment required to supply material for the relatively
enormous bulk of eggs laid by them. Many species are, it
seems likely, unable to mature their eggs without having
first obtained a feed of blood. In Southern India, and
doubtless in similar climates elsewhere, it is never cold
enough to drive Mosquitoes into hybernation, and in places
within the equatorial belt of uniform climatic conditions,
the heat is rarely so extreme, or drought so prolonged, as to

LIFE HISTORY AND SEASONAL PREVALENCE 143

force them into retirement on the opposite account ; so that
in such places, all correspondents tell me that they are
present, in greater or less numbers, throughout the year ;
though even with them, the degree of prevalence is a good
deal influenced by such climatic variations as occur. Thus
there exists for many species, what may be called a southern
limit of hybernation, to the north of which it is compelled
to retire into seclusion during the winter ; whereas, further
south, they are able to remain in activity to a greater or less
extent throughout the year. In the case of our local species
in India of Anopheles, this line lies somewhere about the
26th parallel of latitude, for while they were certainly still
hybernating in Agra, in the middle of February, I found
them numerous and fairly active some hundred miles further
south, at Hoshangabad, in the Central Provinces. It should,
however, be needless to remark that it is impossible to lay
down any rigid limits in a case of this sort, and that it must
needs vary from year to year with temporary variations of
the character of the seasons.

Our knowledge of the seasonal prevalence of individual
species is very limited, and in any case, can only be stated
for areas of small extent, but it may be broadly stated that
in each locality it may be fairly predicated from an inspec-
tion of the annual weather reports, as while some are
tolerant of the greatest heat, others are confined to seasons
of abundant rainfall. Were we even much better informed
than we are, it is therefore obvious that no general account
could be afforded within any moderate limits of space, so I
confine myself, as an example of what may be observed in
some one region, to the following sketch of a year's experience
on this point in the North-west Provinces of India, and
Oudh. Presuming a visitor to arrive in January, he would
find about houses and in gardens, but two species, C.
fatigans, Wied, and C. impellens, Walker, and even these,
neither troublesome, nor numerous.

In the garden tanks he would find the larvae of these
species and also those of Anopheles Bossii, Mihi, and
sinensis, Wied, but, owing to almost all natural collections
of stagnant water of moderate size being already dried up.

144 GNATS OR MOSQUITOES — CHAPTER VII

be will find it difficult to find them away from the neigh-
bourhood of habitations.

These larvae may almost be said to be in a sort of resting
condition, or in other words, they are not growing, and, as
a rule, no pupae will be found, though the occurrence of a
day or two of warmer weather, will lead to their appearance,
especially in the case of the Culices. Towards the middle
of February, these two hardy species commence to renew
breeding operations in a leisurable way, but as far as I can
make out, it is only C. fatigans that dares to brave the fierce
heat of the dry season, C. impellens retiring into obscurity
in March, until the advent of the rain makes climatic con-
ditions more tolerable, alike for Mosquitoes and men. At
the end of March, C. fatigans was absolutely the only
species I could find, but it made up for the want of its
kindred by its enormous numbers, as the small, constantly
replenished tanks in the gardens, which afforded the only
possible nurseries for Mosquitoes of any sort, were well-nigh
solid with their larvae and pupae, and they so swarmed in
the bungalows, as to make the evening and night intoler-
able. In the open, or anywhere away from houses, it is
needless to remark that no Mosquitoes of any kind were
to be found, and they continued to be the only species
present in at all noticeable numbers, until the break of the
rains, in the end of June, or beginning of July.

In the middle of April, however, the Anopheles larvae
that have tided through the winter, pupate, and give birth
to a spring brood of adults, but as the climatic conditions
are unfavourable at this time to its breeding operations,
on account, I fancy, of the available collections of water
being too hot for the taste of the larvae ; which will swarm
in the very same situations a couple of months later, and
hence no fresh broods appear, and the imagines that made
a short appearance on the scene hide themselves for the
rest of the dry season : as none of my correspondents
make any mention of having met with them in our
part of the country at this time of the year. With the
advent of the rains, however, a renewed period of activity
commences. Anopheles larvae make their appearance in

LIFE HISTORY AND SEASONAL PREVALENCE 145

every available pool, and soon strange Culiciform larvae,
with short breathing tubes will be found in the pools of
fairly clean rain water that form in every depression of the
ground. Before long G. fa tig a ns, though still taking a con-
siderable share in the business of rendering human existence
less endurable, is thrown into the shade by numbers of
small Mosquitoes " brindled " with intensely contrasted
black and white, while the unobtrusive Anopheles, though
really present in large numbers are less importunate in
their attentions, and will not so readily be found by the
unpractised searcher. On this account, and because they
are more purely nocturnal than the Ciilices, those who do
not know where and when to look for them may be led to
report them as scarce or absent, though in reality they
may be present in large numbers. I suspect too that the
elevation of an upper storey affords a more secure haven
from their attacks than is the case with Culex.

Mr. Aitken, in the paper already quoted, remarks : —
"As regards the mature Anopheles, the most remark-
a,ble fact in my notes is that I have not seen one
during these last twelve months, excepting those that I
reared. This fact will give the best idea of what an insidious
■enemy we have to deal with. I lived most of the year at
the Bombay Club, within a stone's throw of the Frere
Fountain, in which Anopheles was being produced by the
thousand, but I never saw one, though I was often tormented
by Culex, the larvae of which were comparatively scarce.
Anopheles is a small, sHm Mosquito, of a pale, ashy-grey
colour, difficult to see at any time, and it appears to fly only
by night, so it is rarely seen. Add to this that its bite
appears to be almost painless, and you will see that one
may have malaria injected into him night after night while
he is sleeping without curtains, under the belief that there
are no Mosquitoes. I say that its bite is almost painless
on the authority of my own experiments only. I kept mine
in a bottle with thin muslin tied over the mouth, and if I
laid my arm on the muslin the females attacked me at once
and did not leave off till they were bloated with blood. I
sometimes felt a slight prick at the moment when they
10

146 GNATS OR MOSQUITOES — CHAPTER VII

punctured the skin, but there was Httle or no irritation
afterwards and no swelhng. Of course, others might have
been affected differently. The males never attempted to
suck my blood, but they fed freely, as did the females also,
on a slice of apple, fig, mango, or any juicy fruit. In default
of fruit I gave them jam, or even sugar and water. After
a meal of blood they seemed to feel heavy and indisposed
for active exercise, but were quite ready again in twenty-
four hours. It has been stated that Anopheles is mute.
This is certainly a mistake. It has a very shrill pipe."

The brindled Mosquitoes above alluded to are Stegomijice,
but, in addition to these, Tmiiorhynchus ager, Mihi,
Mucidus scatapliagoides, Mihi, and a number of Culices,
such as C. concolor, E. Desv., which are entirely absent to
all appearance at other times of the year, may be found in
greater or less numbers. In October, I took at Shahjahan-
pur, a solitary specimen of C. mimeticus, Noe, but I suspect
that this is rarely found during the rains in the plains, as
it was not included in any of the collections sent me for
identification from this part of India. I also took at this
period, the single specimen from which Corethra Asiatica,
sp. n. is described, but this species is so small and incon-
spicuous that it is easily overlooked, and it is probable that
the example I captured was a late specimen of a species
that may be common enough during the rainy season. With
the drying up of the rains, these species, peculiar to that
season of the year, cease to be seen, and retire into a
seclusion that cannot fall far short of seven or eight
months; but the Anopheletes linger longer, and may be
found, though in rapidly diminishing numbers, till late in
November. Towards the end of the period An. sinensis,
previously the less common, is far more often met with than
An. Bossii, but the very last specimen I was able to find
during the year was one of the latter species (on November
22nd). In a note dated two days previously, I find that
a male sinensis was so sluggish m the early morning
(temperature 56' F.) that he refused to fly, though he
revived somewhat during the day, when taken into the
sunny verandah to pose for his photograph. After this.

LIFE HISTORY AND SEASONAL PREVALENCE 147

no others than the two species mentioned as still present
in January could be found, and it is again not until April
that any were observed by my friend, Mr. Royle, who then
found them again in evidence in Shahjahanpur.

Such is roughly the seasonal prevalence of the family in
the Province to which my own observations have been
almost confined, but those who are not familiar with the
wide range of chmatic conditions found within the peninsula
must not be misled into supposing that the account can be
accepted for the entire country, or that it is even typical
of any large proportion of its area. In illustration of this,
I quote a further extract from Mr. Aitken's paper and a
few of the replies published in the Indian Medical Gazette,
in response to a circular of queries issued by Major
Buchanan, I.M.S., the editor of that journal. Mr. Aitken
says : " In the first place, you will note that I have found
Anopheles larvae in every month of the year, except
February, when I did not look for them. But I had
plenty in captivity during that month, which I had brought
home in January. So it appears that, in a place with a
moist climate, in which there is always some water to
be found, Mosquitoes can survive without hybernation and
may be found at all seasons. But as with other insects
generally in this Presidency, the time when they are most
abundant is the close of the rains."

Captain Giffard, I.M.S., notes that on the Coromandel
coast the malaria-bearing Mosquitoes were extraordinarily
prevalent, even at the end of the cold season, and existed
in thousands in every pool, well, and casuarina pool
examined.

Captain Cornwall, I.M.S., states that in Madras, Mosqui-
toes are never entirely absent, but they are most common,
when a sufficiency of water is lying about, i.e., in January,
February and March, after the rains. They decrease in
the hot weather and increase again in the showery months
of July, August and September. In the last three months
of the year, when the heavy rains have swamped the
breeding places, the Mosquitoes, both adult and larval, are
most difficult to find.

148 GNATS OR MOSQUITOES— CHAPTER VII

Captain C. J. Fearnside, I.M.S., writes that in Rajali-
mundri, Mosquitoes seem to exist all the year round, and
that he finds that Anopheles breeds anywhere, in a beaker
of water as well as in a puddle. He has also seen Anopheles
feed greedily in the daytime. With regard to the evolu-
tion of the malarial parasite, Captain Fearnside notes that
in some species of Anopheles crescents and spring tertian
parasites will not develop at all, the crescents may be " old
and impotent " (Grassi), but this will not explain the non-
development of the spring tertians. He also noted that
he had frequently found crescents in cases known to have
suffered from only mild attacks of fever.

Major J, Smyth, I. M.S., writes from Bangalore that
Mosquitoes are present throughout the year, but in
diminished numbers in January and December. Last year
Anopheles were present in certain localities in large
numbers, especially in July. The following note is of
special interest : last year at Bangalore some new plots
of land were opened out for the extension of the town, and
one of the new extensions became so malarious that it had
to be abandoned, all the children suffered from ague, and
most of them developed enlarged spleens. In this part of
the extensions Major Smyth found Anopheles larvae very
prevalent in some low-lying pools ; in two other extensions
no malaria prevailed and no Anopheles larvae could be found.

This observation at Bangalore is very interesting in
connection with the much discussed question of malarial
outbreaks among men employed in engineering and building
operations.

From Berhampur, Bengal, Major J. H. T. Walsh,
I. M.S., writes that Anopheles is present in small numbers
all the year, but only a few in the dry hot months. They
appear in large numbers during breaks in the rains, and
after the heavy floods of last September none were seen for
several days. Though Berhampur is a very " malarious "
district, very few cases of true "ague" were seen in the
jail and asylum. Anopheles larvae seem to breed every-
where he says, in tanks, or even in a bathroom.

Major C. R. M. Green, I.M.S., F.KC.S., writes from

LIFE HISTORY AND SEASONAL PEEVALENCE 149

Mozufferpore that Mosquitoes are present all the year, but
most common in September, October and November. The
adult Anopheles is most easily found in October and
November.

Captain Maddox writes from Chapra that August,
September and October are the worst months for Mos-
quitoes. He has found Ajiopheles larvae in small pools and
ditches near habitations.

Writing from Benguela, in Portuguese West Africa,
Dr. A. Yale Massey says that the hot, wet season lasts
from October to April, and that Mosquitoes appear in
November. There is comparatively little fever before
December, and cases occur as late as June, but January
and February are the worst months.

Speaking generally then, it may be safely asserted that,
for any given locality, the seasonal prevalence of Mosquitoes
may be accurately predicated from an inspection of its tables
of rainfall and temperature ; and that, provided Anopheletes
be included in its fauna, malaria will make its appearance,
in all warm climates, within a few weeks of the breaking of
the rains.

It only remains to add a few words on the natural
enemies of Mosquitoes in their various stages. Unfor-
tunately, as Celli remarks, these are not numerous, and are
rarely sufficiently numerous to perceptibly diminish the
numbers of so prolific a family of insects. In the adult
stage, birds, reptiles, frogs and certain insects devour them
whenever they have the chance, and are not tempted by
the superior attractions of larger game. The bright little
gecko lizard, which is so commonly found in our Indian
bungalows, should be always treated as an honoured guest
for his services in this direction, especially as he does not
strike work and go to sleep just when he is most wanted, as
most birds do. Each of these quaint, half-tame little beasts
may be looked upon as at least equal, in Mosquito-destroy-
ing efficiency to a fly-paper of the largest size, and their
company should be encouraged accordingly. In the larval
stage there can be no doubt of the efficiency of certain
species of fish, and the apparent contradictions that appear

150 GNATS OR MOSQUITOES — CHAPTER VII

in the communications one reads from various parts are
simply due to the fact that the tastes of different species of
fish as to their food differ Hke those of other orders of
animals. That fish and Mosquitoes are constantly found
in company in the Madras rice swamps simply shows that
the species found there do not happen to be larvivorous,
and in no way discredits the accuracy of other observers,
who find that gnat larvae are unable to exist in water that
contains fish of probably entirely different sorts. Per-
.sonuUy I have never found fish in the same pool with larvae

Fig. 32. —Caudal extremity of Anopheles Rosii infested by a parasitic
stalked infusorian.

though, like Mr. Aitken, I find that they and tadpoles, and
I may add frogs, seem to be able to live together as excel-
lent friends. Mr. Aitken says : " From my experience I
should say that, of all larvicides, the most effectual, in the
case of Anopheles, is little fishes. I have never found larvae
and fishes in the same pool. Once I put a large number
of larvae into two glass vessels and introduced a few gold
fish into each. Next morning there was not one larva
in either. They have many insect enemies; too, especially

LIFE HISTORY AND SEASONAL PREVALENCE 151

the larvae of dragonflies, but one fish will do more than a
hundred of these. Tadpoles do not eat them."

On the other hand, there is a general concensus of
evidence that the larvae of dragonflies are most efficient in
this respect, and I believe that their universal presence in
all collections of v^^ater of any size in upper India is the
explanation of the fact that we so rarely find gnat larvae
in such situations. The only parasite I have ever met with
infesting the larvae is the small stalked infusorian repre-
sented in the accompanying illustration. I have repeatedly
found every larva in a pool simply covered by these para-
sites, which lie crowded together in enormous numbers,
attaching themselves especially to the softer parts of the
integments, such as the angles between the anal tubercles,
and the soft membranes between the segments.

Larvae affected in this way have a peculiar, slimy appear-
ance, and seldom appear healthy, though it is difficult to
see how these ecto-parasites can be harmful, unless it may
be that being, at the ver}^ least, greedy mess-mates, they
may appropriate to themselves an undue share of the food
that would otherwise fall to the share of their hosts. Never-
theless, I strongly suspect that they may be the cause of
the inexplicable disappearance of larvae, already alluded
to, from situations where they were just before present
in abundance.

Since the above went to press, I have received from Dr.
J. Cropper, of Chepstow, a slide containing some immature
Acari which he found upon An. maculipennis, in Palestine.

They are of cordate outline, nearly as wide as they are
long, and are provided with a formidable suctorial mouth.
From their comparatively large size, they must be formid-
able parasites to so small an insect, but only three specimens
were observed to be infested in this way. They much
resemble the hexapod larvae of certain bird-ticks.

152

CHAPTER VIII.

On the Conditions Influencing the Prevalence of Mosquitoes,
and on the Prophylaxis of Malaria.

In the previous issue of this handbook, but Httle space
was devoted to this subject ; but though no more than a
year has elapsed, so much has been observed and M^ritten
on the subject, that it is impossible to give any adequate
idea of the present state of our knowledge in a few
paragraphs, incidental to the life history of the Culicidce,
so that it has become necessary to devote a special chapter
to the subject. To-day it may be fairly asserted that the
question of the connection between malaria and Mosquitoes
has passed beyond the tentative stage, and is indeed no
longer a " question," but an ascertained fact.

Not only has the casual connection between the Mos-
quito-carried parasite and malaria been proved be3'ond
question of reasonable cavil, but experiments on a large
scale in the practical application of our present knowledge
to the prevention of malarial disease have shown that,
given sufficient intelligent co-operation of the affected
population, and sufficient earnestness on the part of the
governing authorities, it is quite possible to so avail our-
selves of this newly gained knowledge as to greatly diminish
if not to actually "stamp out " the disease. Without this
co-operation, however, we are helpless, for science can only
help those who will consent to help themselves. The
primary discovery of Laveran has gradually been so followed
up by Marchiafava, Golgi, Plehn, Celli, Grassi, and many
other distinguished observers in Europe, as to place
Laveran's "epoch making" discovery on the soundest basis ;
establishing the fact that we have to deal with not one only,
but at least three distinct species of blood parasite and

CONDITIONS INFLUENCING PREVALENCE 153

elaborating and systematising our knowledge of their life
history within the blood-vessels of the human subject.

Following on these we have the undoubtedly pioneering
observations of Major Ross, I.M.S., on the life history
of the stage of the parasite which is passed within the
mosquito, which are all the more remarkable when the
conditions under which he worked are understood. Condi-
tions under w^hich those accustomed only to the luxurious
ease of a European laboratory with dozens of assistants
well-nigh as able as themselves to lighten the work, would
probably find so untenable that it may be doubted if they
would succeed in observing at all. But even under the
most favourable conditions, initial observations of the sort
must needs be always more or less incomplete and that
some inaccuracies should be discoverable in Ross's work is
merely equivalent to stating that the work was of an initial
and pioneering character. Ross's observations have been
now confirmed and elaborated by the admirable work of
Grassi, but there still remain some points to be elucidated
before our knowledge of the life history of the parasite
within the mosquito can be considered as complete. Finally,
it has been clearly demonstrated in Italy that man can be
inoculated with malaria by the bite of infected Mosquitoes
of the Genus Anopheles and quite recently Major Andrew
Buchanan, I.M.S., has recorded in the April issue of the
Indian Medical Gazette, a number of carefully conducted
experiments which establish the same fact for India.

The Italian observers, having the advantage of working
in a country, highly malarious, and yet provided with fully
equipped laboratories, and all the resources of an advanced
civilisation have been able to conduct their experiments
with an exactness of precaution quite unattainable in the
semi-civilised haunts of tropical malaria, and to the un-
prejudiced critic, leave no really fair ground for objection ;
but they still leave it open for the superficial objector to
suggest that, being inhabitants of a notoriously malarious
country, the subjects of experiment may have been infected
in some other way than through the agency of the experi-
mental Mosquito bite. The standpoint of these last

154 GNATS OE, MOSQUITOES — CHAPTER VIII

objectors is, however, entirely cut away by the truly crucial
experiments conducted in the London School of Tropical
Medicine under the direction of Dr. Manson which conclu-
sively demonstrate that malaria can be transmitted to man
through the agency of Mosquitoes. A number of Anojjheles
were allowed to bite a patient suffering from tertian ague
in Italy. They were then transported to England and
made to bite two healthy young English students. Both
these gentlemen developed tertian malarial fever, and the
characteristic parasites of the disease were found in their
blood.

It is difficult to find in this experiment any possible
source of fallacy. It is absolutely conclusive of the fact
that this is at the very least one of the methods of the
transmission and propagation of the disease ; and a very
little consideration will show any one conversant with the
data of parasitism that it is also necessarily the only one,
saving only by the intravenous injection of the blood of a
patient suffering from malaria into the vessels of a healthy
subject ; a method hardly likely to occur in nature.

The reason for our assurance of this is that the malaria)
parasite requires two successive hosts — a liuman being and
a Mosquito — to attain sexual maturity and propagation.
Ill the blood of the fever patient it multipHes non-sexually ;
in the tissues of the Mosquito it does so sexually. Now
there are a large number of parasites which have an
exactly parallel history, the most familiar being that of the
tape-worm, which lives and multiplies asexually in herhivora
and other eaten animals, and passes its sexually mature life
in the carnivora, and other animal-eating animals. Just as
it is possible to introduce asexually multiplying malarial
protozoa mechanically into the veins of a healthy man, so
would it, doubtless, be practicable, in these days of abdo-
minal surgery, to lay open the intestine and introduce into
it a living tape-worm, which would, doubtless, continue to
thrive in its new host. But in the ordinary plan of nature,
the eggs discharged from the bowel of the ea^w^-animal are
discharged in situations when they are hkely to be swallowed
by the eaten animal, and in the latter produce the asexually

CONDITIONS INFLUENCING PREVALENCE 155

multiplying bladder worm. This, when swallowed with its
eaten host, developes, in the flesh -eating animal, once more
into the sexually multiplying tape-w^orm.

Now, although we are acquainted with a large number
of parasitic life histories of this character, we know of no
instance in which a parasite with such a history is capable
of maintaining the continuity of the species in any other
manner, and it will be indeed astonishing if the malarial
parasite should prove an exception to what has been
hitherto found to be an unvarying law of parasitism; though
of course, however intrinsically improbable, it is within the
range of biological possibility, that besides finding its way
to its intermediate host along with the venom of the Mos-
quito, the parasite may also be capable of assuming the
form of a resting spore, or some kindred reproductive
mechanism, and being in this way conveyed to water, food,
&c., through the agency of the living or dead Mosquito.
There is not, however, a single recorded fact in the natural
history of the disease that suggests the probability of such
an occurrence, and practically speaking, everyone possessing
any special knowledge of helminthology will be convinced
that either the idea that the Mosquito is the alternative host
of the malarial parasite is a huge mistake ; or it is, under
natural circumstances, the one and only method of infection.
There is in reality no tenable middle position.

Most of the apparent exceptions depend on the fact that
like most other two-host life-history parasites, the host
carrying the asexual phase of the malarial parasite may do
so for years without any perceptible inconvenience. A
bladder worm may have to lie imbedded in the tissues of
an ox for years before the animal is turned into beef and
devoured by a man.

Then its opportunity has come and it developes into a
tape-worm each sexually mature, proglottis of which is a
complete, hermaphrodite, sexually mature animal.

So with the malarial parasite. An infected person may
have no visible symptoms, but lurking in his tissues are the
parasites ready to start again on their course of asexual
multiplication should any accident bring the resisting power
of the host sufficiently low.

156 GNATS OR MOSQUITOES — CHAPTER VIII

Hence, persons who have had no recent opportunity of
being bitten by Mosquitoes often do develop a typical ague,
but the fact remains that they must have been bitten at
some time, and as a matter of fact the interval is a concern
of but little moment to the parasite. The patient in fact,
though apparently' well, has latent malaria ; in other words,
he harbours but a harmless number of quiescent parasites,
and the exception is only apparent. The fact of the possi-
bility of the transmission of malaria in this way having
thus been now conclusively demonstrated, we may take it
as certain that every malarial patient has at some time
been bitten by an infected Mosquito. Further, it appears
probable that only Mosquitoes of the genus Anopheles are
capable of acting as the host of the sexual stage of the
parasite, but this is not certain.

Now the malarial parasite is responsible for by far the
greatest proportion of all sickness and death in the tropics.
Cholera and plague are the insignificant enemies that
perhaps kill a few thousands a year —in an impressive way
it is true ; but the quiet, insidious malaria sweeps off its
millions, and so habituated have we, native and European
alike, become to the danger, that we have come to look
upon the inconvenience of one or more " touches of fever "
during the year, as a necessary evil, inseparable from the
conditions of tropical residence, and no more to be escaped
than the occasional " cold " of more temperate climates.
Unfortunately there can, I fear, be no doubt that this
fatahstic frame of mind will, for a long time to come, con-
stitute one of the greatest obstacles to sanitary improve-
ment ; for such preconceptions are hard to eradicate, and
hence it comes that, while large sums are freely expended
in fighting in the dark against the unfamiliar terror of
plague, in the case of malaria, where we have already a
large basis of solid facts to work upon, our total expendi-
ture on the prevention of malaria appears to be comprised
in the vote of the absurd sum of Ks. 30 i^er mensem to
provide the salary of a man to destroy Mosquito-larvse with
kerosine, by the City fathers of Calcutta. I doubt if India
will ever be a pleasant residence for the white man for

CONDITIONS INFLUENCING PREVALENCE 157

the greater part of the year, but nevertheless it would
not compare very unfavourably with the temperate zone
from a health point of view, could we but do away with
malaria. While proposing, in the main, to treat the inter-
dependence of malaria and Mosquitoes as an established
fact, it may be well to devote a few words to certain
objections that have lately been, in various forms, advanced.
The first is that malaria may exist in places or at seasons
when Mosquitoes are rare or absent. Now in the first
place, putting aside certain little visited islands, gnats or
Mosquitoes are to be found practically everywhere, from
Greenland to the Equator, and it may be taken as certain
that abundance of these insects or their larvae would be
found by any one accustomed to the search in every one
of the cited cases of their absence. Moreover, it will be
generally found that such objections are raised by persons
who, however intelligent and highly educated they may be,
have had no practice in observations of the kind required
for the record of facts bearing on natural history, and they
are apt to forget that, in this, as in any other special
business, long training, much patience, and a certain apti-
tude, are required for the work. A writer, for example, in
a well-known Indian lay journal, a propos this question,
gravely propounded the astounding statement that Mos-
quitoes were extinct in India in the rains, and specially
troublesome in the cold weather, whereas, without for a
moment questioning the good faith of the writer, it is
needless to say that exactly the reverse is the case. In all
probability he had never made a single definite note on the
subject, or troubled himself as to whether few or many
Mosquitoes were about, since the time when they made
themselves painfully obvious to the newly landed journalist
during the time he was acquiring that indifference to their
bites which all of us sooner or later develop ; and under
such circumstances memory is naturally treacherous.

Another argument that has been used is that, whereas
the presence of malaria is dependent on that of man, and
should therefore be worst where population is thickest, the
reverse is the case, as towns enjoy a practical immunity,

158 GNATS OR MOSQUITOES — CHAPTER VIII

while some of the most deadly spots in the world have but
a scanty population. The reason why the conditions of
urban life are unfavourable to the spread of malaria will be
dealt with further on, and it is sufticient to point out that
the reason why such tracts of country as the notorious
Indian Terai are so deserted is that they are too malarious
for human occupation. In other words, the local conditions
are so favourable to the multiplication of the species of
mosquito concerned in the transmission of the disease that
the presence of a very small number of infected persons
suffices to infect an enormous number of Mosquitoes, and
to render well nigh certain the infection of any visitor who
exposes himself to the same conditions. Moreover, every
place of the sort that I have heard of has always been a
tract of close jungle or swamp, in which it is impossible
for the traveller to stray from certain beaten tracks, or to
make his camp elsewhere than at certain definite halting
places, where, however small may be its number, there is
always a permanent population, or at the least, passers
through are sufficiently numerous to maintain the infection.
That in such deserted tracts these foci of intense malaria are
purely local I do not entertain a doubt, and that a healthy
man who landed from a balloon a mile or two away from
them would, though equally pestered by Mosquitoes, take no
other harm ; but in such country man can push his way
but slowly, and there is always time for those who attempt
to open it up to carry the infection with them.

Moreover, in comparing the relative salubrity of neigh-
bouring places, it must be remembered that a certain tem-
perature is essential to the development of the parasite in
the body of the Mosquito, and that on this account a
difference of a few thousand feet above the level of the sea
is quite sufficient to account for a place being quite healthy,
though but a few miles of horizontal distance from foci of
intense virulence. It is this factor that is overlooked by Mr.
Guy Marshall, in his paper on " Mosquitoes and Malaria,"
in the Entomologist, August, 1900, p. 218, who points out
that Salisbury, in Mashonaland, though comparatively
thickly populated, is much less malarious than the sparsely

CONDITIONS INFLUENCING PREVALENCE 159

populated Umfuli district. Being a skilled entomologist,
Mr. Marshall found several species of Anopheles in both
localities, and asks why the more populated district should
not be the more malarious. He forgets that Salisbury lies
nearly 5,000 feet above the sea, more than 2,000 feet above
the malarious regions, and that, though elevation per se is
no absolute bar to malaria, the temperature of places so
elevated, in that latitude south, is always too low for the
development of really serious malaria. Curiously enough,
we find that almost simultaneously another writer from
Mashonaland, Dr. Ch. Todd, demonstrating (Journ. Trop.
Med. 1900, p. 92.) that there, as elsewhere, the curve of
malaria prevalence follows that of rainfall and therefore of
Mosquitoes ; the most rainy months being January and
February, and the most feverish, March and April.

Before proceeding to the consideration of the prophy-
laxis of malaria, it will be necessary to examine in detail
how far the prevalence of Mosquitoes and therefore of that
disease are influenced by climate, cultivation, and the other
incidents of everyday human environment. Working as I
have in India, it is natural that most of the examples cited
should be based on observations made in that country ; but
communications, personal and published, from observers in
other tropical regions, most of which cannot, however, be
quoted in any moderate space, convince me that our experi-
ence in that country may be taken as fairly typical of the
conditions present in other hot climates.

It will be seen too that, with modifications arising from
the differences of oriental surroundings, I come to practi-
cally the same conclusions as Professor Celli, in his admir-
able work on " Malaria in Italy," and if I quote less than
might be expected from that work, it is partly because most
of the notes on which the present chapter is based were
written before I had an opportunity of reading it, but
mainly because I thmk that everyone interested in the
subject should make a point of reading his book in extenso.

In certain cases, however, it is hardly practicable to
separate the consideration of the conditions favouring
malaria from the practical outcome of the facts noted, and.

lOO GNATS OR MOSQUITOES — CHAPTER VIII

where thus more convenient, points of prophylaxis will be
dealt with in their own connection.

The necessarily frequent allusions to climatic conditions
in India will be more easily understood by occasional
reference to the table on pp. 16'2 — 163, the materials of
which were kindly supplied me by Mr. J. Eliot, the
Meterological Eeporter to the Indian Government.

The range of cHmate within the peninsula is very wide,
almost every variety of tropical and subtropical condition
being represented, from the intense drought of the western
Punjab and Rajput desert, with their wide range of annual
temperature, to the uniform moist heat of Cochin and the
Burman littoral ; and it is obviously impossible to give any
complete account within the limits of a page or so of tabu-
lation. The references to the relative salubrity or other-
wise of the various places must, moreover, not be taken to
apply specially to the town mentioned in the table, but
rather to the region of which its climate is representative ;
and in speaking of the local malaria as " mild," " virulent,"
and so forth, I do not refer to the species of parasite (for
little else than the sestivo-autumnal fever is to be met with
in India), but to the malignancy or otherwise of the type of
the disease, which varies greatly in different places, and
indeed, from year to year, in the same place.

The places included in the list, being selected as fair
average examples of the climates or the regions in which
they are situate, by no means illustrate the extreme range
of variation within the limits of Indian jurisdiction ; and
hence do not include such situations as Cherra Punji
(said to be the wettest place in the world) where the rain
gauge has literally to be graduated to feet ; or Sibi, where,
as the tale has it, the smi has such power that the
European residents must needs assemble in the club, to
sit beneath the billiard table, still wearing their pith hats.

InjiiLence of Climate. — There can be no doubt that
climate is the most important of all the factors that
together contribute to render a given place malarious or
otherwise. The genus Anopheles has a world-wide distribu-
tion, extending much further north than malaria, which is

CONDITIONS INFLUENCING PREVALENCE 161

unknown in really cold countries, while, given the presence
of the malarial blood parasite, no hot climate is free from the
disease. If this be absent, there can of course be no malaria,
as the disease is no necessary concomitant of a tropical
climate, as is shown by the case of the Island of Mauri-
tius, which, previously healthy, suddenly became intensely
malarious. In this case there can be little doubt that
Mosquitoes of the suitable species were already present, for
the endemic developed with a rapidity quite inconsistent
with the idea of the establishment of the necessary species
from a few chance emigrants. All that was required was
the importation of the infected man ; and of this, before the
introduction of steam navigation, there was but a precarious
chance.

It has usually been suggested that the disease was in this
instance imported from India, but the type of the disease
seen there is very different. Never in India have I met
with cases exhibiting the absolutely classical malarial
paroxysm such as I have witnessed in soldiers sent to
Natal for change of air from Mauritius.

In other cases the immunity of places where every con-
dition of climate is favourable is due to the absence of
Mosquitoes.

In a letter to the Lancet, dated January 18th, 1901, by
Mr. H. D. O'Neill, an interesting observation by Robert
Louis Stevenson is referred to on the subject of Mosquitoes
and their association with filaria and malaria : "In Atuona
(Marquesas Islands), a village planted in a shore-side marsh,
the houses standing everywhere intermingled with the pools
of a taro-garden, we find every condition of tropical danger
and discomfort, and yet there are not even Mosquitoes, nor
even the hateful day-fly of Wuka-Niva, and fever and its
concomitant, the island fe'efe'e, are unknown."

It is a long-established fact that the northern limit of
malaria corresponds roughly with the summer maximum
isotherm of 76° F., or, according to Hirsch, to a mean
summer temperature of 15° — 16° C. (60° F.), which is much
the same thing. Eecent Italian researches show that the
development of the hgemosporidia within the Mosquito
11

Table Showing the Monthly Rainfall and

January

Februarj-

March

April

May

June

July

f'

r.

r.

S

f

K

«

Station

3

3

3

3

3

1
■5

Si

S

!s

S2

II

^5

Ir

1

II

ainfa

VIean
perat

F

F

' F

« £ 1

E

K ■ • E

X E

■"

"

■"

"

1 "

I

Simla ..

2-35

41 "s"

2-68

4i'5°

2-24

507"

1-90

59'7''

3-64

64-5°

679 68'o°

I7'55

65-0"

2

Peshawar (N.)

I '77

5i'7

0-Q8

S3-6

1-70

64-2

1-84

73 "7

075

83-6

0-35 : 91-0

1-79

g:?

Lahore (Mid.)

I -06

54 "4

i"io

57-1

o'73

69-2

0-46

807

I '03

87-4

1-84 1 927

667

]Multan(S.) ..

o-4a

56-3

0-38

59 "5

o-3a

72-1

0-07

27

0-42

9o'3

0-59 94-5

2-94

931

,

Meerut

1-27

57'4

0-79

61 -o

0-77

72-4

0-24

83-2

0-69

88-5

2'44 ! 91 'I

9'54

86-3

AuLhabad : '.

0-53

61 -o

0'2I

6S-0

0-31

77-2

o'i4

88"2

o-6o

93 7

2-54 \ 94'8

11-50

o"85

6o-8

0"28

6S-3

0-32

778

o-ii

88-1

o'39

22-4

5-69 1 92-6

i2'33

8S-4

Benares

0-79

61 -2

o'37

65-9

0-28

77-8

0.08

87-9

072

qi-0

5-13 91-6

1074

8s-S

Jhansi . .

<

0-59

63-3

033

67-3

0-35

79 "3

0-13

89-9

0-49

94'9

4'89 93'S

12-60

84-5

4

Patna . .

0-65

6i-3

o'53

5s-3

0-38

77 '4

0"26

87-0

1-97

88-6

7-34 1 88-4

1 1 75

8s.i

Hazaribagh ..

z

J

5

0-56

617

0-82

b5-8

075

76-3

0-41

85-2

226

86-3

7-63 ; 84-2

14-16

79-0

Calcutta

o-6o

66-2

i-s8

70-7

i'57

8o-o

i'74

8s-S

7-62

8s-2

1074 85-0

12-46

83-2

Dhubri

H
da
3

/o"40

63-S

o"53

66 -o

I -93

7S-6

4-83

794

13'97

79'4

24-53 i 81-0

16-17

83-1

Slbsagar

I "47

59'9

i-9b

62-9

5 '07

697

9-37

74-6

12-63

78-9

13-69 83-2

17-10

84-5

6

Jaipur . .

0-69

6i-i

0-19

63-9

o"39

75'4

0-09

84-9

0-45

90-9

2-49 1 91-4

9'37

84-4

7

Kurrachi

0-72

66-8

0-31

69-4

0-23

76-S

o'33

82-2

o-oo

3 '47

86-1

S

Deesa . .

0-17

67-1

70-2

0-05

79'9

85-9

0-25

91-8

2-i6 91-1

10-99

84-4

Khandwa . .

0-31

67-6

o-o6

7X-7

o'i3

81-3

0-17

89-3

0-45

93-1

6-05

877

8-82

81-0

Jubulpur

0-76

62 -s

0-47

66 '8

0-51

77-2

o'lS

86-2

071

91-6

9-IO

87 4

2o-So

80-1

Nagpur

o'55

69.2

0'27
O-QI
0-04

74"2

o-6i
0-03

ii3-i
79 '6

0-34
o-oi

90-8

o-8o
0-94

94'9

8-74
I9"37

£,79

i4"73

80-9

lO

Bombay

o'i3

75-1

75-5

82 7

85-2

83-3

ly7~

27-17

80-7

Hyderabad . .

C-09

71-0

76-8

075

83-6

0-67

S87

i'i5

90-4

4-85

6-90

78-6

Poena . .

^

o-o6

70-0

0-04

74-2

0-05

807

o"54

8VS

1-65

85-3

4'73

6-87

76-3

Beluaum

-

o-o6

7o"3

74 'o

o'35

78 -q

172

2-62

8o-s

6'50

74 3

15 37

71-2

Bellary

Z

0-13

76-0

0-04

79 "5

86-1

0-58

90-4

170

89-8

I -85

85-1

I '93

Bangalore

'

o-ig

07-9

72-0

o'54

77 "3

I-I5

8i-2

4-02

8o-i

3'45

75-6

4"59

73-7

Trichinopoly..

<
r

0-26

77-0

0-90

80 -o

o'55

»5-i

1-53

89-2

3-04

897

1-62

803

I "5°

87-1

12

Cochin (West
Coast)

g

0-59

8o-o

0-62

8i-2

2-44

83-7

4'37

847

i3'30

83-2

28-41

79'5

21-51

78-6

Madras (East

o-8q

76-0

0-28

77 '2

o"39

8o-6

0-62

8s-i

2-12

89"3

2-II

89-3

3-87

87-0

Coast)

13

Rangoon

0-17
o*o8

76-3
69-7

o"34
0-07

78-9

0-28

83-6

fS3

87-0

9-42

84-9

i7"Si

8.-3

21-68

80-3

14

Mandalay . .

74-8

0-2I

82-4

1-37

89-4

5-56

89-0

6-21

86-5

3"i7

...

Mean Temperature of Thirty-One Indian Stations.

August

Septem

ber

October

Novem

ber

December

[

No.

_

—

—

—

—

1

S 2

c

is
^1

-1

II

S2
E

•5

-'

"

■"

I

17-98

63-5°

6-56

62-4°

1-22

56-8°

0-54

49-7°

0-74

45-8°

Hill station, with practically tem-
perate climate. Not malarious.

2-70

88-4

0-64

82-8

o-ii

72 -Q

0-57

60-6

0-34

53-0

Punjab stations.— Intensely hot in

5-83

87-4

2-49

84-9

0-26

76-6

Q-IO

63-7

0-3S

S6-1

summer, quite cold in winter ; rainfall

1-58 ]

gi-i

0-42

88-5

0-00

79"9

o-io

67-8

58-6

scanty. Malaria rife from August to
November ; sometimes of a very viru-
lent type.

3

10-59

84-5

5-74

83-2

0-42

76-4

0-08

65-3

0-32

58-6

North-West Provinces.— Hot and

7-67

84-.

4-91

84-3

0-47

80-4

0-05

6Q-5

0-19

62-1

dry from April to mid-June; then to

84-1

6 -05

83-8

1-83 1

78-8

0-17

68-2

61-2

September, moderate rain ; cool with

11-83

84-3

6-59

84-2

2-30

79-3

0-36

68-7

0-24

61-4

bright sun, November to March.

12-50

82-6

6 -So

83-1

0-70

8o-5

0-12

70-4

0-13

64-3

Malarious from August to November,
but seldom of a severe type.

n-30

84-4

7-40

84-7

3'25

80-s

0-17

70-7

0-13

62-6

Upper Bengal.— Intermediate in cli-

13-11

78-3

8-76

78-3

3-41

75-0

0-29

67-2

60-9

mate and salubrity between N.W.P.
and Lower Bengal.

s

12-95 1

82-6

9-33

82-6

4-39

8o-s

0-66

72-9

0-24

66-1

Lower Bengal and Assam.— Moist,

13-76

82-3

13-35

81-4

3-50

79-0

0-26

71-8

o-io

6s-3

except for a few weeks in March and

16-19 1 83-8

12-22

82-6

4-84

78-0

0-98

6q-i

0.57

61-1

April ; heavy and prolonged rains, but

seldom with intense heat. Malaria

1

prolonged, and often of a severe type.

6

10-07

85-0

4-40

82-7

0-30

78-9

0-24

68-8

o-o8

62-8

Rajputana.- Closely resembles the
Southern Punjab.

7

1-55

83-8

0-54

83-6

..

0-09

75-0

0-16

69-0

Seaport of Sind.— Waterless and
desert ; but climate modified by prox-
imity to sea. Exceptionally little
malaria previously to the introduction
of a regular water supply.

7-6o 1

81-8

4-83

83-1

0-3S

81-5

0-16

75-6

0-06

68-6

Gujarat.— Scanty rainfall, July,
August ; heat of prolonged drought
modified by proximity to sea. Malaria
moderate, more or less throughout
the year, with two maxima : in Febru-
ary and October respectively.

9 ' 7' 14

79-8

7-56

80-2

1-73

78-0

0-31

70-4

0-56

65-3

Central India.— Prolonged dry sea-

79-3

8'77

79-9

2-07

75-6

0-50

66-6

0-38

60-6

son, intensely hot in May and June;

1 10-25

81-0

10-13

81 -2

2-95

78-9

0-64

66-8

ramfall moderate. Malaria autumnal.

prolonged far into cold weather, but

seldom particularly virulent.

TO

11-43

80-3

11-81

80-2

2-47

81-8

0-66

797

0-09

76-8

Considerable rainfall, almost con-

fined to three months. Not very

malarious.

II 8-17

78-4

5-99

78-4

3-08

77-3

1-76

72-5

0-27

69-1

Southern plateau.— Scanty rain-

3-22

75-7

5-21

76-3

4-80

77-5

1-31

72-4

0-26

68-6

fall, but no great annual variation of

8-74

71-2

4-64

71-9

6-39

73-7

2-II

71-6

0-13

6q-6

temperature ; intense heat of central

2-58

82-1

4-09

81-7

4-29

80-1

2-13

76-0

0-14

73-0

region modified by sea breezeas ghauts

5-8o

73-0

4-72

73-5

7-15

72-9

3-59

70-3

0-55

68-1

are approached. Malaria worst in

4-67

86-2

3-21

85-4

7 '49

5-37

79-1

2-55

76-7

August, prolonged far into cool season.

but rarely of virulent type.

12 I3'3i

78-7

9-38

79-2

14-01

80-1

6-77

8o-6

i-8i

8o-3

Southern litoral.— Climate uniform
and moist. No marked malaria-free

1 4-56

85-5

4-69

85-2

11-00

82-1

13-21

78-7

5-28

76-7

season, but disease seldom specially
severe.

13 18-19

80-3

16-04

80-7

6-74

81 -4

2-98

8o-i

0-09

77-5

Lower Burmah.— Resembles Indian
Southern litoral in climate, but disease

1

often of severe type.

.4 3-88

1
i

85-3

6-54

84-8

5-o8

83-1

1-28

76-9

0-28

70-5

Climate resembles that of Southern
Indian plateaiL Malaria from June to
December, worst in August ; disease
often of virulent type.

164 GNATS OR MOSQUITOES — CHAPTER VIII

cannot take place at a lower temperature than 20'' C.
(68° F.), or at a higher than 30' C. (86° ¥.), and in the
existence of this upper limit we find an explanation of the
fact that the hot dry weather in northern India, where for
months together the temperature rarely falls as low as
this, is, in spite of the unbearable heat, by far the healthiest
season of the year, and that during it, primary cases of
malaria are practically unknown.

As in these parts of the Peninsula the winter temperature,
for considerable periods falls considerably below the lower
limit, extreme climates such as we have there, are blessed
with two consecutive periods of immunity, malaria being
confined to the season of the " rains " and early autumn.
Strictly speaking, however. Northern India is sub-tropical.
In the Tropics, and especially in the Equatorial region the
lower limit is practically never reached and the existence of
the higher is of far greater practical importance, as during
the drier portions of the year it is, except in certain specially
favoured localities, always surpassed for considerable periods.
As, however, in the truly Equatorial regions there are two
dry and two rainy reasons in each year, the periods of
immunity are too short to bring about any very distinct
diminution in the amount of sickness due to malaria, as their
occurrence is masked by recurrent attacks which always
continue to vitiate the statistics of any period of immunity
for a longer period than the entire duration of these short
intervals.

While then in temperate climates the duration of the
winter is the most important climatic factor in securing a
prolonged period of immunity, in tropical and sub-tropical
regions it is the hot dry weather that exercises a sanitary
influence.

Second only to temperature is the amount and distribu-
tion of rainfall. It is needless, however, to go into any
lengthened considerations of its effects, as these naturally
follow from the premise that malaria is dependent on the
Mosquito for transmission from man to man; beyond
pointing out that besides being the main natural puddle-
producing agency it has the additional effect of keeping

CONDITIONS INFLUENCING- PREVALENCE

165

down the temperature, and so shortening or abrogating the
period of immunity due to heat, and that the more evenly
it be distributed the more serious will be its effects on
public health.

So greatly is malaria favoured by a copious rainfall that
in the majority of places the monthly incidence of new infec-
tions of malaria may he roughly said to be directly propor-
tional to the rainfall of the preceding month. The only
apparent exceptions to this rule occur in places where a
heavy rainfall occurs during months too hot or too cold for
the development of the parasite ; but of course in such
places the rule must be taken to apply only to those months
during w^hich the air temperature permits of the maturation
of the parasite within the Mosquito. Without encumbering
these pages with needless statistics, the two following exam-
ples may be taken as fairly typical, remembering always
that the figures given probably include a much larger number
of recurrent than of primary attacks.

Table Showing the Relation op Rainfall to Admissions to Hospital
FOR Malarial Fever (including Recurrent Cases) in Two Indian
Prisons.

1

1

<

1

B
3

■a

•-5

1
<

1

i

>

1

1

Admissions to

hospital for ■

31

27

28

12

15

14

31 38

50

50

50

37

\

malaria

1 Nagpiir Central

Rainfall during ]

the preceding Y

U

7

14

12

y

1

month

Admissions to )

I

hospital for ■

11-7

s-i

11-6

10

14-2

11-2

8-8

13-9

23-1

23-6

14-9

15 -y

malaria

Shabjahanpiir

Rainfall during ]

L District Jail.

the preceding ■
month

1-4

'2-6

2-5

0-9

0-V

1-4

6-4

12-6

14-1

6-3

0-9

0-4

The figures referring to the Nagpur Central Jail are
derived from a paper by Major A. Buchanan, I. M.S., pub-
lished in the Indian Medical Gazette, and both series
represent the averages of ten years. In the case of the
Shahjahanpur Jail the average strength of prisoners during
the period was 378"7.

Conditions of atmospheric pressure can, per se, have no
influence on the incidence of malaria.

166 GNATS OK MOSQUITOES — CHAPTER VIII

Hill stations are, of course, as a rale free from malaria,
"but the reason of this is not that because the barometer
stands some inches below the instrument in the plains, but
because they are cool and well drained. Given other favour-
able conditions, and malaria will develop at any elevation,
and as a matter of fact, the writer has observed and treated
undoubted malaria in Wakham in the upper Oxus Valley, at
an elevation of 9,000 feet above the sea, among the Tajik
tribesmen who rarely or never visit places at a lower level.

The Influence of Water. — In a certain sense, the
malariousness or otherwise of any locality is intimately
connected with its water supply ; but the question is one
o£ its relative abundance and distribution, and not of its
quahty. As has already been pointed out, it is, for biologi-
cal reasons, in the last degree improbable that there can
exist any alternative route of malarial infection than that
through the Mosquito, and it was not proposed to enter here
into the question on the possibility of drinking water form-
ing a vehicle of infection ; but old fallacies die hard, and as
an attempt has recently been made by Captain Leonard
Eogers, I.M.S., to resuscitate this one, it may be well to
devote some space to the consideration of the arguments
brought forward by him. Captain Rogers attempts to
show that those parts of Calcutta and neighbourhood which
receive a filtered water supply are less malarious than
neighbouring suburbs which draw their supply from the
river, from tanks, or from wells. He estimates the relative
malariousness of the compared localities by what he terms
their spleen ratio, a factor which he determines by counting
the number of enlarged spleens in a given number of persons
whom he regards as fairly representative of the general
population.

Now such a method of estimation is obviously open to a
variety of sources of fallacy, and is, at least, misleading.
As pointed out by Drs. Christophers and Stephens, and
confirmed by Koch, the true index of the malariousness of
.a place, is the average length of time required for the in-
fection of new comers ; and the most convenient class of
immigrant, because always ready to hand, are the young

CONDITIONS INFLUENCING PREVALENCE 167

children. Further, as will be pointed out further on, the
mere fact that Calcutta is a great city, in which municipa-
lisation on European lines has been carried much further
than in the "up-country" towns mentioned below, is alone
sufficient to account for the more crowded portions of the
towns which have been provided with a filtered water supply
being less malarious than the suburbs, even assuming that
Captain Eogers' researches may be taken as conclusive that
such is really the case. x\s a matter of fact, experiments on
a very large scale have been of late years conducted in
certain Indian prisons, where the entire drinking water
supply of the prisoners was carefully and systematically
boiled in the hope of diminishing malarial disease, but
without producing the least effect in the hoped for direction.

But apart from this, our experience in Northern India
strongly suggests that, so far from diminishing malaria, the
introduction of a piped and filtered water supply, has com-
monly quite the opposite effect. It is unfortunate that
before venturing on generalisation from what appears to be
such scanty data. Captain Eogers was not at the pains to
avail himself of the tabulated information on this point
readily available to him in the annual reports of the Sanitary
Commissioners of the Northern Provinces, where the effect
of the introduction of filtered water supplies has been a
matter of anxious observation for a considerable series of
years, for it is impossible to imagine that any diminution in
malarial fevers and splenic enlargement could fail to show
itself in a diminution of the general death-rate.

For the benefit of those to whom the above reports may
not be readily accessible, the figures bearing on the ques-
tion are extracted on next page : —

The addition of this table to the Sanitary Reports was
probably initiated with the view of illustrating the benefits
conferred by modern sanitation ; but if this be the case
the compilers must have been most disagreeably surprised,
for with hardly an exception a rise, and not a fall, of
mortality has followed. There are, of course, absolutely no
statistics extant of any value whatever as to the absolute
number of deaths referable to malaria in any Indian town,

168

GNATS OR MOSQUITOES — CHAPTER VIII

but it will not, I presume, be denied that the disease is
everywhere in India accountable for a considerable propor-
tion of the total mortality, or that any perceptible diminu-
tion in the prevalence of malaria could fail to make itself
evident in the total death-rate.

Town

Average annual

death-rate

since introduction

of a filtered

water-supply

Average annual
death-rate for the
five years' period

preceding
its introduction

Remarks

Cawnpore . .

Allahabad ..

Lucknow . .

Benares

Meerut

Agra

47-83
28-70
43-79
48-81
35-06
35-46

41-15

25-77

44-68 1

39-99

32-13

32-23

Only the data refer-
ring to towns provided
with a filtered supply
are included here ;
. places supplied with
unfiltered water have
been excluded, as also
have cantonments and
/ hill stations.

It is inconceivable that the introduction of a pure water
supply should raise the death-rate from cholera and bowel
complaints, and as a matter of fact the figures of such of
the towns as I have examined show improvement in this
respect.

Nor can the enhancement of the death-rate be fairly
ascribed to improved registration, as during the entire
period this has been admittedly fairly accurate as to total
mortality however worthless it may be as to detailed
causes, nor is there any evidence to show that there has
been any general improvement in this respect. Moreover
it must be remembered that the dates of introduction of the
regular water-supply differ widely in different places, and
that the increased mortality has followed immediately in by
far the majority of cases. Further, the increase though
immediate is not progressive, as may be seen by running
through the series of tables published up to the present
date.

It is therefore undeniable that in the North West
provinces at least, municipal malaria has increased and not
diminished coincidentally with the introduction of filtered
water-supplies.

CONDITIONS INFLUENCING PREVALENCE 169

For those who believe in the agency of the Mosquito
in the propagation of malaria, the explanation of this un-
expected and undesired result of modern sanitary enterprise
is not difficult.

In by far the majority of cases no attempt at improved
surface drainage has accompanied the spread of the water-
pipe.

Financial tightness has necessitated that the essentials
of sanitary reform should be taken in hand one by one ;
and the effort to introduce a pure water-supply has so
exhausted the resources of each municipality in which it
has been carried out, that the proportion of cases in which
the engineers have been able to so place their hydrants as
to secure a ready flowing away of waste water has been
perforce a very small one, and the result has been that often,
each hydrant is the source of a string of puddles of con-
stantly renewed, fresh cool water, and not unfrequently so
placed as to be the greater part of the day in the shadow of
tall buildings. In pools so fed and situated Anopheles larvse
may be found at times of the year, when but for the
hydrants, they would be rare as the dodo ; for these larvae
do not appear to be able to develop in water as hot as that
of the ordinary stagnant pool or tank in the hot dry weather.
At any rate it is only in such exceptionally conditioned
water that Anopheles larvse can be found in the N.W.
provinces in March and April, for the ordinary garden tanks,
which in the rains will harbour large numbers, are then full
of Ctdex larvae only. In this way a piped water-supply
extends the period of possible infections over several months,
which ordinarily yield but few fresh cases.

As even when confined to its normal times and seasons
malaria is responsible for a larger share of the total mor-
tahty than any other disease, the above explanation appears
to me to adequately explain the apparent failure of pure
water-supplies to improve the general health.

It is obvious, however, that the increased sickliness and
therefore presumably malariousness of these places are an
indirect result only of the introduction of a water-supply ; as
to the direct connection of which with malaria there exists
no tittle of proof.

170 GNATS OR MOSQUITOES — CHAPTER VIII

The case too of the French transport " Argo," which has
so often been quoted as an example of malarial infection
through the agency of water, has been conclusively shown
by Celli (CM., p. 95) to have been probably an outbreak of
acate poisoning, but certainly not malaria. Putting aside
then the possibility of water acting as the direct vehicle of
infection, let us see how far the prevalence of malaria is
indirectly influenced by its agency.

The presence of water is absolutely essential to the
multiplication of Mosquitoes, and further, it is now fully
recognised that it is not large bodies of water, such as
marshes, lakes and rivers, that form their favourite nurseries,
but small stagnant collections such as puddles and pools of
no great size that are favoured by them.

Further, these puddles must be sufficiently permanent to
persist for at least ten days. It is therefore the surface
distribution of water that is primarily of importance ; but as
this is largely influenced by the disposition of the ground
water the study of the depth and movements of the latter is
of the greatest importance, and the prevalence of malaria is
therefore influenced by this in exactly the same way as if it
were a truly water-borne disease. Of the older generalisa-
tions on the subject, perhaps the only one that can be said
to have stood the test of recent knowledge is the often
demonstrated fact that localities where the ground water
lies close to the surface are generally malarious. The
reason of this is that in such places the surface is easily
saturated and badly drained ; because, wherever the ground
water lies high, it is usually also either sluggish or stagnant.

Now a marsh is usually a place where the ground water
comes 10 the surface, the level of the soil being below that
of the ground water of the higher land around it ; but the
marsh itself, per se, is rarely concerned in favouring the
multiplication of Mosquitoes, the larvae of which are only
very rarely to be found in such situations. It is the stretch
of country immediately surrounding the marsh, where the
ground water level is not belov/, but almost coincides with
that of the soil, that is the true focus of " marsh miasmata,"
for in such places the smallest depression must needs
remain a puddle, in spite of the most rapid evaporation.

CONDITIONS INFLUENCING PREVALENCE 171

It is obvious therefore that hydrauHc sanitation can do
much to diminish malaria, though in some cases, as in the
Eoman Campagna, the works required may, as pointed
out by CeUi (CM., p. 126), be of such magnitude as to be
beyond our present financial and engineering resources ;
but there are many cases where much good can be effected.
The level of the true subsoil water can as a rule be
modified only by works of a public character, as the neces-
sary works must usually be undertaken at a considerable
distance from the properties they are intended to benefit.
What is usually spoken of by agriculturalists as " subsoil
drainage " only indirectly affects the ground water level,
as the pipes or rubble drains, &c., employed are placed at
but a little depth from the surface and nearly always much
above the true subsoil water level. The only works that can
directly affect the ground w^ater level are the straightening
and regrading of the channels which form its natural
outlets.

Harbour works for example, which involve the removal
of tidal bars, may have an important influence of this kind,
and there are other cases in which locks s"hutting off the
flood tide from estuaries may have an equally good effect.

It must be remembered in this connection that as a
rule the surface of a river is the lowest point of a section
of the ground water level across its basin at any given
point of its course. Now, whatever may be the fall of a
river bed, the main obstacle to its efficiency as a ground
water effluent is friction, and the more tortuous and there-
fore longer its course, the flatter will be its gradient and
the greater the total friction, which may also be greatly
increased by the presence of " snags," boulders and similar
obstructions, or by water plants, as in the instance of the
" sudd " of the upper Nile. Fortunately, in many cases,
works of this sort are not only of sanitary, but also of
obviously commercially economic importance, and it is very
possible that the desire to make "trade follow the flag"
which actuates our attempts to cut a way through the
" sudd" may result in a really far more economically im-
portant improvement in the health of the Soudan, whereby

172 GNATS OE MOSQUITOES — CHAPTER YIII

spots like Fashoda may become valuable possessions,
instead of merely affording worthless incitement to
international strife and jealousy, and early graves for the
Englishmen or Frenchmen who secure the pestiferous bone
of contention.

Surface drainage on the other hand is more usually a
matter of detailed small works, the collective effect of
which, however, as in the case of our own Fen country,
may have a most beneficial effect on the public health.
Apart from mere gutters, the most effective means of drying
the surface of the soil is the well-known agricultural system
of " subsoil drainage." Somewhat similar to this, but less
effective as a measure of malarial prophylaxis, is the system
of drainage employed on Assamese tea gardens, in which
the cultivation is divided into plots by straight drainage
cuts some five or six feet deep, and these in their turn are
divided and subdivided by progressively shallower cuttings.
As a measure for drying the soil to suit the needs of the
tea bush they are doubtless all that is required, but owing
to the impossibihty of accurately grading simple cuttings
in the soil, they are a fruitful source of puddles. Still they
on the whole appear to be beneficial, as they have un-
doubtedly diminished the malariousness of certain estates.

Influence of the Air. — On this point little need be said
as its effects are always rather those of climate than com-
position. Its chemical composition, the relative proportion
present of carbonic acid or ozone, &c., have nothing to do
with the case ; but still the word malaria can hardly be
said to be a complete misnomer, for it is the air that carries
the Mosquito which is the actual vehicle of the malarial
germ. The limits of distance to which they can be con-
veyed are, however, as we have seen very limited, and for
practical purposes the quality of the air of a place may
be left out of consideration.

Influence of Soil. — The comparative freedom of certain
sites from malarial disease has been well known from
the earliest times, though owing to ignorance of the
actual underlying causes, all attempts at generalisation
were rendered futile by the constant cropping up of
perplexing exceptions.

CONDITIONS INFLUENCING PREVALENCE 173

Always assuming that there be no water-logging of the
soil, it is obvious that the more porous it be, the less the
probability of the formation of puddles of sufficient per-
manence to admit of the rearing of a brood of Mosquitoes.
Sand for example, holds water so badly that short of com-
plete water-logging the formation of puddles is an absolute
impossibility ; but the deposition of a very thin layer of
fine silt by flood water will suffice to render such soils
sufficiently retentive. In the same way, rocky soils are
usually healthy, because they are generally found in hilly
regions, and are therefore associated with good surface
drainage ; but if the peculiarities of the rock are such as to
lead to the formation of basin-like depressions during the
process of weathering, such a soil may be highly favourable
to the development of malaria, always provided that the
rock be of uniform and impervious texture, so that the
water cannot soak away. For this reason the malariousness
of Hong Kong was, in our older books on hygiene, ascribed
to some mysterious influence of a soil consisting of de-
composed granite.

In a certain sense, of course, it really is the decomposi-
tion of granite that is at the root of the mischief, but this
is not because granite, whether intact or decomposed, can
have anything to do with the effect ascribed to it, but
because the denudation of such rocks leaves the undecom-
posed portion hollowed out into basins as watertight as the
best porcelain, and moreover, where such rocks form the
beds of water courses, they are particularly liable to be worn
into " pot-holes" by the action of pebbles retained in some
chance depression.

According to Drs. Stephens and Christophers, (B.S.M.C,
July, 1900, p. 43), the persistence of Aiwpheles during
the drier portions of the year in Freetown, Sierra Leone,
is due to puddles retained in rock basins of this description,
and similar conditions have been noted in Southern India,
in addition to which I have myself met with Anopheles
pools of this description at Jhansi.

Speaking generally, it is the character of the surface soil
that is of the greatest importance, as however deep the

174 GNATS OR MOSQUITOES — CHAPTER VIII

subsoil water may lie, puddles will necessarily form, provided
only the surface layer be sufficiently retentive. Hence sites
having a clayey surface are necessarily favourable to the
development of malaria, and even if the actual surface be
not of this character, it is quite possible for a thin inter-
mediate stratum of such material to so hold up the surface
water as to admit of the surface being practically water-
logged, in spite of the subsoil being quite pervious and the
ground water deep. An impervious stratum of this sort
is to be found in many parts of the Gangetic alluvium some
four or five feet below the surface, as is shown by the fact
that in such districts any attempt to increase water storage
by the deepening of existing tanks merely results in their
drying up.

In the Sitapur district (Oudh) for example, several tracts
of country of this description are to be found, and though
the subsoil water lies some thirty or forty feet from the
surface they are notoriously malarious.

I am not aware that such an experiment has ever been
tried, but in the case of limited areas, such as inhabited
sites, it seems possible that their sanitary condition might
in such cases be improved by the construction of a number
of blind wells carried through the impervious layer so as to
open up the porous subsoil beneath.

It would be impossible within any moderate limits to
give any account of the various kinds of soil that are to be
met with in India, as it is needless to say that in so vast a
country almost every possible combination of soil and
contour formation is to be met with, and malaria is rife
more or less throughout ; but a few words with respect to
the commoner formations may not be out of place.

Between the foot of the Himalayas and the old island
India of past geological times stretches an immense level
plain reclaimed from the sea by the silt deposited by the
mountain streams that now fall into the Ganges and Indus.
After this old narrow sea had completely silted up, the rivers
continued to raise their beds, wandering from side to side
whenever the detritus deposited in their shoals accumulated
sufficiently to raise them above the level of the land hard b3\

CONDITIONS INFLUENCING PREVALENCE 175

This process has been going on ever since, and is still in
progress ; till to-day, this enormous area of alluvium forms
an apparently level plain, stretching from sea to sea, and in
most places some hundred miles wide. Though apparently
as level as the ocean the imperceptible vi^atershed between
the Ganges and Jumna flowing east and the westward
bound rivers of the Punjab is really some 700 to 800 feet
above the sea, and the depth of the alluvium is in some situa-
tions enormous. The whole of this area consists of sand
and silt of various degrees of fineness, modified at the
surface with a variable amount of decomposed organic
matter. Saving where waters, rich with lime in solution,
have matted together vegetable fibre and sand into " kun-
kar," nothing of the nature of a stone is to be found.

At both the eastern and western limits of this wide
alluvial area, the ground water is necessarily close to the
surface, as hundreds of square miles of country are but a
few feet above the level of the sea ; but speaking generally,
as one travels further inland the level of the subsoil water
gets deeper and deeper, and on the watersheds between the
rivers may at times be as much as a hundred feet from the
surface.

The eastern or Gangetic half of this area is for the most
part naturally fertile, the natural rainfall being, in normal
years, sufficient to water a sufficient crop to support a large
population. In the easternmost portion indeed the normal
rainfall may be said to be excessive, and malaria is neces-
sarily rife and long continued.

On the other hand, once the country drained b}^ the
Indus is reached the rainfall becomes scanty and precarious.
Whether, however, we start from the Gangetic or Indus
delta, the subsoil water becomes deeper and deeper as
we travel inland so that in Oudh and the North-west
Provinces, forty or fifty feet is no uncommon depth for a
well, and in parts of the Punjab, the water may not be
reached for twice that depth. On the western side the
rainfall progressively diminishes, so that the country, even
where the ground water is at no great depth, is a water-
less desert ; and cultivation, apart from irrigation, an
impossibility.

176 GNATS OR MOSQUITOES — CHAPTER VlII

Now, except when its natural permeability is destroyed
by a layer of fine silt, as is the case with the cement-like
surface of the Punjab "put," this alluvium is considered
merely as a soil, by no means favourable to the development
of malaria, as it is for the most part so pervious that it
holds water badly. Owing, however, to its unbroken levels
the natural surface drainage is everywhere bad, and its
artificial improvement difficult or impracticable.

In spite, however, of these disadvantages, it may almost
be said that the haunts of Anopheles larva are throughout
its entire area mainly the work of man, and are therefore to
a great extent removable ; always provided that sufficient
funds and intelligence be available.

Passing south, the transition from the alluvium to the
broken and rocky ground of central India is often well-nigh
as abrupt as that from sea to land.

In place of the monotonous plains of Northern India, the
surface, at very least undulates and is often mountainous ;
and the soil, which is seldom of very great depth, is derived
from the decomposition of rocks in the immediate neigh-
bourhood which belong mostly to primary or metamorphic
formations. In the northern part of this area much of the
cultivable land consists of what is known as the black
" cotton soil," which possesses certain characters which
have an important influence on the local seasonal incidence
of malaria. It absorbs water like a sponge or like so much
"black cotton," and once thoroughly saturated holds it
well enough to favour the formation of puddles, which are
fairly permanent as long as the air remains damp. With
the return, however, of the dry weather it dries rapidly,
splitting up into a network of deep fissures which render
the existence of puddles, whether of domestic or natural
origin, well nigh an impossibility.

The greater part of this region is, however, to the south
of the line of hybernation lox Anopheles, so that adult insects
may be taken at all times of the year ; but the presence of
larvse is practically limited to the season of the rains, or
from the middle of June to the end of September.

As, however, the closure of the fissures by the sweUing

CONDITIONS INFLUENCING PREVALENCE 177

of the spongy soil as it drinks in the rains still leaves the
surface irregular, this soil is peculiarly favourable to the
multiplication of the species as long as the wet season lasts,
so that the physical characteristics of this sort of soil
necessarily are such as to favour greatly the development
of malaria for this short and limited period ; and as a
matter of fact, the seasonal incidence of malaria corres-
ponds well with these facts, as taken altogether, such sites
are fairly healthy, though malaria is rife and wide-spread
while it lasts.

Owing to the instability of such a foundation the con-
struction of permanent works of all kinds is a matter of the
greatest difficulty, and hence the surface drainage of towns
is always costly, and even when most carefully designed
requires continual regrading.

Below the black soil there is commonly, especially in
Kathiawar, a stratum of limestone, locally known as
moram (miliolite) intervening between it and the subsoil
water, which, though not very dense, is yet sufficiently
effective in holding up the surface moisture.

It is doubtful if even subsoil drainage would be of any
great use in combating malaria in land of this description,
as the peculiarity of this soil is that it dries rather by
evaporation from above than by the draining away of its
moisture from below, and it is obvious that under such
circumstances works of this sort would be not only expen-
sive, but probably ineffective.

With a highly civilised population it is possible that
some good might be effected by the systematic destruction
of the adult insects during the dry season, as the climate
and general characters of the country are such as to render
the shelter of houses almost indispensable to the mainten-
ance of the species ; but under existing circumstances this
appears one of those cases in which the free distribution of
quinine and the popularisation of its use can alone effect
much benefit, and is therefore the more fortunate that the
same physical peculiarities that so favour the development
of malaria for three or four months of the year also limit
its duration.

12

178 GNATS OR MOSQUITOES— CHAPTER YIII

It is needless to remark that besides the above ahuost
every variety of soil is to be met with in so large a country
as the Indian peninsula, but the above are the only two
cases with which I have any personal familiarity which are
sufficiently extensive and peculiar to require any special
mention. In all probability what has been remarked with
regard to the Gangetic alluvium will apply also to other
tropical and subtropical alluvia, such as those of the Nile
and Mississippi, but I have not been able to obtain any
special information on such points, except as regards
India.

Influence of Vegetation. — As the three essentials to the
well-being of all species of Mosquitoes, vegetable food,
shelter from the sun during the heat of the day, and the
presence of puddles wherein to rear their young, are all
either dependent upon or greatly favoured by the presence
of vegetation, it is obvious that its amount and character
is a factor that must always be considered in estimating the
potentialities of malaria in any given place.

Open grassy plains have long been known to be
unfavourable to malaria. In such situations gnats can find
no sufficient shade, and such puddles as form are hidden
from them by the closely crowded stems. I am inclined to
believe that the hiding of the surface of pools by close
vegetation tends to prevent Mosquitoes from using them as
breeding places, and that this fact might often be made use
of as a cheap and effective means of sanitation.

There can be no doubt e.g., that the presence of certain
water plants in some way prevents the appearance of
Mosquito larvae in the water covered by them, and the
only reason I can suggest for this is that the plants act in
this way because they hide the water from the female gnat
searching for a suitable place in which to deposit her eggs.
At any rate it is difficult to otherwise account for the
curious fact that in the Benares public gardens, where' there
are some scores of the small irrigation tanks described
below, Culex and Anopheles larvae, alone or in company,
were present in every tank save those that were covered
with a peculiar floating water plant, looking much like a

CONDITIONS INFLUENCING PREVALENCE 179

young lettuce, which is spoken of by the natives as the
jalkumi.

In the tanks so planted, the water was alive with young
leeches and nematodes, but in none of them could be
discovered a single Mosquito larva, while the others
swarmed with them. Introduced into a tank already con-
taining Mosquito larvse, however, the plants appeared to
exercise no hostile influence whatever on their develop-
ment, and for this reason, I conclude that the plants act
mechanically in the same way as an artificial cover. The
jalkumi floats on the surface of the water and so forms
a most effective screen ; but another plant resembling
the Canadian duckwood, which grows completely sub-
merged appeared almost as effectual, though in this case
the surface of the water was certainly not hidden in the
ordinary sense of the word, though the green coloration
may have masked its presence to the defective and short
ranged vision of a Mosquito. Whether the presence of
such plants would be equally effective in places where no
alternative pools are accessible is of course open to doubt,
and the vagrant nature of my employment as Sanitary
Commissioner prevented my being able to follow the point
up, but the matter is certainly worthy of investigation, as
if confirmed, it would afford an extremely simple and
inexpensive means of diminishing the number of available
nurseries for larvae.

A variety of plants, such as the castor oil and eucalyptus
have enjoyed the reputation of being protective against either
Mosquitoes and malaria, and it is possible that the scent of
certain strong smelling species may be obnoxious to the
insects; but the observations of Celli (C. M., p. 143) show
that so far from being destructive to them such plants may
form an excellent refuge for Mosquitoes. Our experience
in India is similar, as may be judged by the following
extract from the Pioneer (April 4, 1901) : —

To the Editor.

" Sir, — Some short time back, there appeared, in the
correspondence columns of your paper, I think, a recom-

180 GNATS OR MOSQUITOES — CHAPTER VIII

mendation to use the castor oil plant to keep a bungalow
free from Mosquitoes. I being a sufferer had six plants
placed in pots in my rooms. I fancy we must breed a
different variety of Mosquitoes than your correspondent, for
the castor oil plants are thickly covered with the insects by
day, who, at night time, seem to be actually invigorated by
the apparently stimulating effect of their new quarters. — D."

The influence of trees in especial has hitherto been
greatly misunderstood. A screen of trees was supposed, in
some occult way, to be capable of filtering out malarial
germs from the air ; and it is just possible that the presence
of a convenient shelter of this sort might in certain cases
prevent Mosquitoes wandering further to dwellings which
might otherwise have been their nearest convenient refuge ;
but it would obviously be bad policy to multiply such
shelters. Speaking generally indeed there can be no doubt,
that trees greatly favour the multiplication of Mosquitoes.

This they do in four ways : they afford shelter during
the day ; their shade prevents the drying up of puddles ;
their flowers often afford the staple food of the insects ; and
lastly, they prevent the growth of grass.

In 1884, the Italian authorities instituted an enquiry as to
the influence of disforesting on pubhc health (C. M. p. 141),
and the result of their investigations was that they were
unable to find any proof that disforesting was injurious to
health, but that some facts indicated an opposite effect.

Popular and professional opinion in India as to the
influence of trees on malaria has oscillated, but has gener-
ally been in favour of open sites, though no one doubts that
trees greatly favour the prevalence of Mosquitoes.

The history of the large military station of Meean Meer
is both curious and instructive in this respect. After the
annexation of the Punjab, the large force quartered close to
the great native city of Lahore suffered so terribly from
malaria that it became absolutely necessary to remove thein
to some more healthy site. At that time the, as we can
now see, perfectly defensible view that open sites are least
malarious was that in vogue and the General and his

CONDITIONS INFLUENCING PREVALENCE 181

advisers therefore selected Meeaii Meer, a barren plain,
almost uncultivated owing to the thinness of the soil.

The extensive building operations necessary for the hous-
ing of the troops, however, soon honeycombed the level
surface with excavations of all sizes and the rising barracks
and bungalows soon afforded abundant shelter of the effective
sort dearest to the Indian Anopheles ; and hence, like all
new places, before long occupation has gradually ameliorated
the untidiness inseparable from new and growing places ;
the change proved no better than from " frying pan to fire."
Changing opinion led to the unhealthiness of the place being
ascribed to the absence of trees, and at great trouble and
expense, a separate shaft, filled in with soil, having to be
sunk through the limestone for each sapling, large numbers
of trees were planted, with no very perceptible result. The
cantonment still bears an evil reputation for malaria, but
half a century of occupation has wrought great improve-
ment in surface drainage and other kindred ameliorations,
and though still decidedly malarious, it is to day neither
markedly better nor worse than other stations in the
Province. In India, as elsewhere, especially during the
rains, many species of Mosquitoes certainly habitually
harbour in trees during the day, and there can be no doubt
that the proximity of trees, however pleasant to the eye,
is most undesirable if one wishes to keep a house free from
these pests. Celli (C. M., p. 141) states that Aji. bifurcatus, in
Italy, lives preferably in thickets, and that persons sleeping
in such places, even during the day, are frequently bitten ;
though the experience of Dr. Sambon and Low {B.M.J.
Dec. 8, 1900, p. 162) were contradictory in this respect.

The explanation of this is probably that the observa-
tions were made in different months, for in India at any
rate our local species of Anopheles do not apparently find that
the shelter of trees is sufficient during the fierce heat of the
day, and prefer the deeper shade of buildings ; and, as far
as my somewhat limited experience extends, are not to be
found amongst trees during the day, though it is likely
enough that they may do so in parts of India where the
summer heat is less fierce than in the North West Provinces.

182 GNATS OR MOSQUITOES — CHAPTER VIII

All considered, I fear that great as is the solace of a shady
garden to the eye jaded with the fierce glare of a tropical
sun, it is a most undesirable adjunct to a residence, alike
for health and comfort, for besides sheltering and in every
way fostering the multiplication of Mosquitoes, whether
simply irritating or noxious, the trees, which practically
never shield the house from the direct rays of the sun, cut
off the breeze and so interfere with that free ventilation,
which is a sine qua non of tolerable existence.

Finally, it must be remembered that trees are in some
way capable of modifying the climate of a locality by
increasing the rainfall. How or why they are capable of
doing so is by no means clear, but the connection is
generally admitted, and some recent observations in India
appear to show that such an effect may be produced within
the limits of a comparatively restricted area. While they,
however, generally favour malaria, Celli justly points out,
(C. M., p. 142) that trees should be respected on hilly
ground, as by retarding the rapidity of drainage they tend
to prevent the flooding of the plains below by heavy rain-
falls on the hills.

Assuming then it to be possible to chose the site of a
tropical residence, it should certainly be placed on open
ground, and coolness and rest for the eye should be aimed
at by surrounding it with a stretch of well- watered grass,
the watering being conducted on some plan that does not
involve the use of tanks and other collections of standing
water ; but such a plan is always difficult and expensive, and
is in many parts of India, impossible.

The presence of thickets and undergrowth has long been
recognised as favourable to malaria. Growths of this sort
sufficiently luxuriant to be unmanageable, in and about
inhabited sites, presuppose a moist climate, free from the
extreme heat of the drier parts of the tropics, and in which
the habitations alike of Europeans and natives are con-
structed with an eye rather to the freest possible ventilation
than to keeping out the heat ; so that as far as that is con-
cerned, there is little to choose between the shade of a tree
and the best built of bungalows.

CONDITIONS INFLUENCING PREVALENCE 183

In such climates, e.g., in Assam and Burmah, the miilti-
phcation of all species of Mosquitoes, Anopheles included, can
hardly be otherwise than greatly favoured by growth of this
sort, and as a matter of fact, the clearing of jungle from
village sites is one of the few measures that come within
the range of practical pohtics in village sanitation of the
provinces in question. Costing as it does nothing more than
a little official persuasion, it is pretty generally carried into
effect, and is undoubtedly most beneficial.

When the District Officer in Assam has induced his
villagers to clear away the jungle from their huts and to
give their wells the annual cleaning, he justly feels that he
has done all that is humanly possible for the sanitation of
the unpromising human material he has to deal with ; for
the Government resolution that converts the Assamese
peasant to sanitary decency will have to be framed with
a stringency far beyond the ingenuity of any legislator, past
or present.

The consideration of the role played by plants in the
propagation of malaria leads naturally to the effects of
cultivation. As far as India is concerned the question of
the influence of cultivation on malaria resolves itself mainly
into that of the effects of irrigation. At any rate it is the
only one in which remedial measures on a large scale
can be attempted by Government, as the methods of the
indigenous cultivator are too ingrained and detailed to be
capable of modification by legislative measures, to say
nothing of the fact that agricultural experts are agreed that
in the main his system is but little capable of improvement,
and can therefore be hardly profitably meddled with.

The total area under irrigation in India is about 29,000
square miles, rather more than the area of Greece, or just
over 2 per cent, of the entire country ; but as the irrigated
areas are also very densely populated, their influence on
public health is much greater than such a proportion would
suggest, to say nothing of the fact that such areas serve as
foci from which the disease is constantly spread abroad by
human agency.

It is needless for me to quote here any evidence as to the

184 GNATS OK MOSQUITOES — CHAPTER VIII

untoward influence of canal irrigation on the public health
through the concomitant increase of malarial disease. The
matter has been the subject of several special investigations,
and as a matter of fact has been fully admitted alike by the
profession and the Indian Government long before we had
any exact knowledge of the way in which irrigation is
responsible for such evil effects.

In some cases these effects have been so serious as to
raise the question of the advisability of abandoning the
system ; but these have occurred mainly in situations where
old native canals have been utilised, and in which the
ahgnment contravenes the principles that will be explained
below ; but it may be at once admitted that no suggestion
involving the prohibition, or even restriction of irrigation
will be considered worthy of serious consideration by practical
men.

Starvation is a worse disease to bear than malarial fever,
and over much of the irrigated area, the very existence of
the population is dependent upon irrigated crops. In much
of the Punjab and Eajputana, and the whole of Sind, the
natural rainfall is so small that, but for irrigation from
canals or wells, the whole country would be an uninhabit-
able desert ; and the entire prosperity of the country and
the prevention of famine depend entirely on the energetic
extension of canals.

Though admittedly, as we -have seen, no unmixed blessing,
irrigation is in such places a necessary postulate to the
existence of any population at all, whether fever stricken or
healthy, and the question resolves itself merely into how
the malarious influences of canals can best be prevented, or
at least minimised.

Irrigation is so little needed in Europe, that in order to
convey to those who have not seen it in practice, why and
how far it favours the spread of malarial disease, some few
words of explanation as to how this form of cultivation is
carried out are necessary. In the greater part of Europe,
the natural surface soil is rather too wet than too dry for
the farmer's purposes, and his efforts at improvement
naturally take the form of surface and subsoil drainage, and

CONDITIONS INFLUENCING PEEVALENCE 185

SO are distinctly anti-malarial. In subtropical regions on
the contrary, especially far inland, the problem the cultivator
has to solve is how to keep the surface soil damp enough
to keep his crops alive.

Subsoil drainage would double the labours of his well
bullocks, and to prevent the escape of the share of the scanty
showers that fall on his patch of ground he subdivides his
holding into small, carefully levelled patches, each but a few
yards square, surrounding each patch with a low ridge of
earth a few inches high.

Naturally m such regions cultivation favours malaria.

Irrigation may be divided into two kinds : (a) Where the
water is raised to the surface by water lifts of various kinds ;
and (b) where it is brought on to the land by gravitation, or
in other words, by canals.

The first is always a domestic operation ; the second can
only be successfully carried out by large and expensive
engineering works. Whichever plan be adopted, the
method of applying the water to the land is the same. The
cultivator having levelled and subdivided his land in the
manner described above, the water is made to flow in turn
into each little square until the ground within its bound-
aries is covered to a depth suitable to the particular crop.
In practice (except for rice) the amount given is usually
absorbed in a few hours, and the careful levelling of each
patch is distinctly unfavourable to the formation of puddles.

In domestic irrigation the water is raised from wells by
bullock or human labour, or it may be scooped up in adroitly
swung baskets from rivers or swamps, but whatever its
source, the water is too laboriously gained to be wasted,
either in leakage or puddles. The quantity raised is usually
too small to perceptibly affect the level of the ground water;
but so far as it goes the tendency must be to lower the
spring leveJs, and on the whole I believe that the influences
of domestic irrigation are rather unfavourable to malaria
than otherwise.

In the case of canal irrigation the results are quite
different, as the enormous quantity of water used usually
seriously raises the level of the ground water and ma}' even

186 GXATS OR MOSQUITOES — CHAPTER VIII

cause water-logging, while the freedom of the supply, and
especially its independence of the individual efforts of the
user, leads to carelessness in its use, and hence to leakages
and puddles ; but it must be distinctly understood that, for
all crops except rice, the puddles are the result of the abuse
and not of the use of irrigation, and that the reason why
these puddles are sufficiently permanent to be harmful is
that the water is not absorbed on account of the saturation
of the soil with canal water, nearly three-quarters of which,
as will be seen, represents waste of some sort. In record-
ing my beliefs as to the harmlessness of domestic irrigation,
it must be understood that I refer only to field irrigatioi].
Gardening as carried on in the parts of India with which I
am familiar is conducted in a way that makes each garden
a paradise for Mosquitoes.

In considering the influence of canals in favouring the
production of puddles, and therefore of malaria, it must be
remembered that but a comparatively small proportion of
the water entering the canal at its head works ever reaches
the fields. In a departmental note " On the Irrigating Duty
of Bari Doab Canal," Mr. K. G. Kennedy, of the Indian
Public Works Department, states that out of every 100
cubic feet of water entering the canal during the winter
months of 1881-82, twenty were lost in the canal proper, six
in the larger distributories, twenty-one in water-courses,
twenty-five by waste in various ways, so that only twenty-
eight was left to do all the useful work of irrigation. That
is to say, that considering the canal as a machine, its
efficiency was but 28 per cent.

Now in this canal alone, which is by no means the largest
of our Indian irrigation works, the average flow of water
is over thirteen millions of gallons per minute, and this
enormous amount of water is distributed, in work and
waste, over 1,200 square miles, or, in other words, is
equivalent to a rainfall of about 40 inches per annum. But
this gives no adequate idea of what may be the local effects
of the system. In the case of the canal in question, for
example, the distribution of the water is very different at
different seasons of the year.

CONDITIONS INFLUENCING PREVALENCE 187

During the dry weather some 26 inches are spread over a
very large area, the security of the spring crops in the
Punjab being almost entirely dependent on irrigation (either
from canals or wells). For the autumn crops, however, the
cultivator relies mainly on the natural rainfall, and it is
mainly in the case of certain valuable crops, such as rice
and sugar, that he requires to supplement it by irrigation.

Hence, for the cultivation of the autumn crops, which
commences in the dry weather and is continued through the*
rains, some 56 inches of water are given, although the area
so treated is much smaller than that irrigated for the spring
crops ; and this, it must be remembered, is in addition to a
natural rainfall of some 20 inches, which is mainly concen-
trated in the same season of the year. Except in the case
of rice, however, the water never lies long enough on the
ground to admit of the development of a generation of
Mosquitoes, as for most crops only a few waterings and
those at long intervals are given, as may be seen from the
table given below, taken from Mr. Kennedy's above-quoted
paper.

Minimum Amount of Water Required to

Irrigate Different Crops.

Name of Crop

Number of
Waterings

Total Depth of
Watering in inches

Grain and fallow land

Millets ..

Wheat, barley, poppy, cotton, Indian corn .

Sugar cane

Rice

1

2

4

11

5

3-4

6-4

10-6

25-3

960

Senji (a fodder plant)

10-5

Mr. H. Frost, the present executive engineer of the same
canal, tells me that except in the case of rice it is rare for
the water given to the land to remain visible on the surface
for more than a day, and my own casual observation of
irrigated land has given me the same impression. It will
be observed in the above table that the number of waterings
given to rice is not stated. In other words, during the
earlier and more critical stages of the crop, it is kept con-
tinuously under water. The water is fairly clean, though

188 GNATS OR MOSQUITOES — CHAPTER VIII

rich ill vegetable growths, and the stems protect the surface
from the wind, so that we have here absolutely ideal con-
ditions for the development of the larvae of Anopheletes.

Personally I have seen but little of rice cultivation, as
irrigated rice is not greatly cultivated in the parts of India
in which I have served, but Captains James and Cornwall,
of the Madras branch of our service tell me that there, the
rice fields absolutely swarm with Aiiopheles larvae, and this
in spite of the presence of numbers of small fish, though
the latter do not appear to be often seen in the irrigated
rice of the Punjab. Eice is therefore the only irrigated
crop in which the used water is instrumental in the propa-
gation of Mosquitoes.

Putting aside the quantity lost by evaporation, whicli
must form a considerable proportion of the used water, the
whole of the canal intake, whether used or wasted, must
ultimately find its way to the subsoil and so raise the level
of the subterranean waters.

This raising of the spring level, with its attendajit increase
of malaria, does not take place at once, but is gradual and
progressive for many years after the introduction of canals,
until the subsoil waters are raised to a level at which canal
supply and drainage are at an equilibrium, but this may mean
the absolute water-logging of the soil and the progress of
these changes has been made the subject of careful observa-
tion by the irrigation branch of the Indian Public Works
Department. As an example I may quote a note by Mr. T.
Higham, Chief Engineer for irrigation in the Punjab, on
the " Spring Levels in the Dari Doab Circle," dated March
19th, 189G.

Summary of Kesults, Upper Sutley Canals Division.

" {a) On the upper two-thirds of the Katora canal, the
spring level is rising at the rate of 0.81 feet per annum, the
average depth from the surface being 16.2 feet in 1890-91.
After a lapse of fourteen years, the spring level may be
expected to average but five feet.

" (6) In the lower third of the same canal the spring level
is rising at the rate of 6.32 feet per annum, but as it is still

CONDITIONS INFLUENCING PEEVALENCE 189

on the average, about 20 feet from the surface, water-logging
is not likely to occur until a lapse of forty-seven years and
probably not even then.

"(c) There has been on the whole, during the last ten
years a general rise of spring level on the other canals of
the division but water-logging need not be apprehended
even in the remote future."

The departmental literature on this subject, would alone
more than fill the present book, and indisputably establish
that everywhere the introduction of canal irrigation has
brought about a serious raising of the spring level, but the
above quotation shows sufficiently for our purpose the
amount and considerable variation in extent of the changes
within a given small area.

The European engineer endeavours as far as possible,
to take his canals along the lines of the water-sheds, but
the reverse plan was followed as a rule by our native pre-
decessors, and when the old canals made by them have
been utilised the results on the health of the population
have often been so serious as to necessitate heavy expendi-
ture in altering their alignment ; as the untoward effects, of
canal irrigation in intensifying malarial diseases, has been
thoroughly recognised by the Indian Government for many
years, and the prevention of water-logging has been the
subject of anxious consideration by their canal engineers.

As regards the direct production of puddles, I suspect
that most of the 25 per cent, of waste " in various ways" is
spent in their production, as a very large proportion of this
item results from leakages due to carelessness on the part of
the cultivators. The water is allowed to flow on to the land
from the ultimate distribution channels by simply breaching
the bank, and is stopped by repairing the opening by means
of a few handfuls of mud. Although when carefully done
the plan is much more efficient than one would expect,
leakages, which necessarily result in the production of
puddles are certainly very common, and owing to the
water-logged condition of the soil resulting from the high
spring level, such puddles are likely to last for a consider-
able time.

190 GNATS OR MOSQUITOES— CHAPTER VIII

Of course the canal authorities do their best to lesson
such wastage of the precious fluid, and a cultivator may be
charged double rates if detected in permitting any con-
siderable waste, but the areas administered are so large, that
numbers of such cases must pass undetected, and a leakage
promptly repaired, would probably not be punished, though
it may easily have lasted long enough to produce a consider-
able puddle.

The means proposed by the Canal Department to
prevent water- logging of the land they irrigate are : —

(a) To carry their canals along the line of water-shed.

(h) To so arrange the minor channels as to avoid their
being carried across the natural lines of drainage.

(c) To limit the supply both as to quantity and time to
the amount absolutely required for the success of the crops.

(d) The making of drainage cuts along the natural lines
of outfall.

Of these the last is probably by far the most efficient,
but the attempts that have been made in this direction have
hitherto been of a tentative character and on no very large
scale.

With the water brought so near the surface as it often
is, it is obvious that it would be quite possible to use it
over again by raising it by pumping ; but under the present
conditions of Indian agriculture the cost of such a system
would obviously be prohibitory.

It will be observed that none of the above propositions
can have any influence in preventing loss of water from the
main canal and its permanent branches.

This loss, as we have seen, amounts to over a quarter of
the entire intake, and as the amount of water that can
be taken from the supplying river is limited, cannot be met
by increasing the speed of the flow, so that the water thus
wasted has a large and definite money value ; and the stop-
page of this loss becomes an object in which considerable
capital might profitably be expended. It is needless to say
this aspect of the question can hardly have escaped the con-
sideration of the administration of so successful a " pro-
ductive " department as that of our Indian canals, and I

CONDITIONS INFLUENCING PREVALENCE 191

understand that the question is not so much as to the
desirabihty of taking measures to stop the waste, as of how
to do it. At first sight the remedy seems obvious, namely,
to "revet" the channels with a layer of masonry or con-
crete, but unless the work is of a very solid description it has
been found that such a lining soon cracks and becomes
pervious. Doubtless, however, the ingenuity of our engineers
will ere long find a method of meeting the difficulty, and
whoever answers the question will not only have solved an
important economic problem, but will have done much to
diminish the untoward effect of canals in water-logging the
soil, with its inevitable concomitant of intensification of
malaria.

An occasional way in which canals may directly produce
breeding places for Mosquitoes is their being left empty for a
period long enough to admit of the rearing of a generation of
imagines from the egg. Necessarily a temporarily disused
canal is really a chain of pools ; and I have met with an
instance where a small canal supplying a tract of thin loam
lying on boulder alluvium, furnished the only situation in
which I could discover any Anopheles larvae, in the form
of pools in the beds of canal channels not in use. Moreover
the village in question was notoriously malarious.

It is obvious that the flushing of each empty channel
every week or ten days would prevent their serving as
Mosquito nurseries.

The above remarks apply exclusively to the large perennial
canals, but in many places another system is in use, viz.,
irrigation by " inundation canals." There are channels
carried from the river bank to any portions of the country
which may happen to lie lower than the flood level of the
river, and are so graded that they are filled only during times
of flood. Their object is in fact, exactly the opposite of the
system of protecting low-lying lands by dykes that have
been so efficient in diminishing malaria in Holland and in
the English fen country, and as neither the time nor
quantity of the supply can be controlled, their effect on
health can hardly be otherwise than disastrous.

Another reason why inundation canals specially favour

192 GNATS OR MOSQUITOES— CHAPTER VIII

malaria is that they carry and distribute over the surface of
the land an enormous amount of fine silt, the deposition of
which renders the surface of the soil impervious to water,
and so favours the permanence of puddles. The proportion
of solid matter suspended in the water at the intake of some
inundation canals may reach 3^^ of its weight, but so high a
proportion indicates a high velocity in the supplying river,
and therefore a coarser deposit, most of which will fall in
the earlier reaches of the canal before the water is dis-
tributed on the land. In spite of this, however, the amount
of fine deposit may be very large, especially in rivers such as
the Indus, which flood mainly owing to the melting of the
snows supplying the glaciers of the far distant mountains in
which they take their source ; and this really dangerously
fine, clogging silt, is naturally the most difficult to get rid
of. Col. Tremenhern, R.E. (" Roorkee Professional Papers,"
first series, vol. iii., p. 25), states that " Those inundation
canals in Sind which draw their supply from branches
separated from the main river by islands covered with
brushwood and long grass, contain a comparatively small
amount of material in suspension. The brushwood and
grass impede the velocity of the water and clarify it," and
the selection of a swampy tract of this sort, at a distance
from dwellings and cultivation, as the site for the intake of a
canal of this sort, could hardly fail to be useful in diminish-
ing its malarious tendencies.

I do not think that the importance of this water-
proofing effect of fine silt deposited by floods is sufficiently
appreciated as a malaria favouring agency ; but it appears
to me to be the true explanation of the way in which a
previously healthy place may be rendered permanently
malarious by a flood of but short duration. Anyone who
has been concerned in the management of large municipal
filter beds knows well how rapid and efficient is this
" staunching " action of fine silt. In the Calcutta water-
works, for example, after about a fortnight, the filters at
certain seasons become practically watertight to a head of
24 inches or so of water.

All canals, of course, carry more or less silt, but the

CONDITIONS INFLUENCING PREVALENCE 193

amount brought down by perennial canals is comparatively
insignificant.

It is, I believe, this staunching action of silt that renders
places liable to flooding so notoriously malarious. The
actual floodings are too occasional and temporary to exercise
much direct influence ; but the waterproofing of the surface
enormously favours the production of fairly permanent
puddles. The obvious remedy lies in the digging into the
soil of some more coarsely grained material, and in many
cases this would be not only practicable, but actually bene-
ficial to cultivation, for soil of this kind is necessarily im-
pervious not only to water but to air, and is therefore
unfavourable to the growth of plants. Not unfrequently the
thickness of this silt-choked layer of soil is but small, and
is immediately underlaid by coarse, pervious sand, so that
there is no need to bring material from a distance, and
that all that is required is digging, or even ploughing, to
a suflicient depth to break up the thin impervious super-
ficial layer, and to mix with it the abundant more coarsely
grained substratum.

Whatever the system employed, however, the extent to
which cultivation, whether irrigated or otherwise, favours
malaria depends almost entirely on the habits of the
cultivators. Where the agriculturist is neat and systematic
in his operations, and especially when he is anxious to
utihse every available foot of land, cultivation is harmless
because the waste of space and of water implied by the
presence of puddles is carefully avoided, while on the other
hand untidy cultivation necessarily spells opportunity for the
entire race of Mosquitoes.

It is for this reason that garden cultivation is as a rule
harmless, unless there be some special feature in the plan
of operations that provides the necessary collections of
water in some other way, and such is unfortunately the case
with the gardens that commonly surround our bungalows
in Northern India.

For the greater part of the year artificial watering is an
absolute necessity, and the plan employed is usually not
irrigation in the usual sense of the word, but watering by
13

194

GNATS OE MOSQUITOES — CHAPTER VIII

hand in the same way as is done in England. The water
is usually raised from a well by the agency of bullocks, and
as the gardens are generally tolerably large, is generally, to
save labour, distributed by masonry channels to a number
of small cemented tanks, such as that illustrated below,
fmm which the water is dipped up by the gardeners.

Fig. 33. — Typical garden tauk (one uf a dozeu) iu the gaideu of
European residence in yhahjahaupur, N.W.P., India.

Now as far as the Anglo-Indian resident of upper India
is concerned these tanks are par excellence the most fruitful
of breeding places for Mosquitoes of all kinds. During
the hot dry weather, they teem with the larvae of
Ciilex fatigans, whose imagines render life in the attached
bungalow well nigh unbearable ; when the rains come this
species is associated with Stegomyia fasciata, AnopheleteSy
and other Mosquitoes appertaining to the season ; and all
through the cold weather they serve as the last vantage
ground of the few Mosquitoes that remain active at that
time of the year.

To those who have lived only in Europe the remedy

CONDITIONS INFLUENCING PREVALENCE 195

appears simple and obvious :- Construct a single covered
tank at the well head, and distribute the water by means
of pipes provided with taps. But in practice the expense,
and the difficulty of obtaining sufficiently skilled workmen
to construct such an installation and keep it in order are
prohibitory. What might, however, be done, would be to
insist on each tank being carefully cleaned out and left
dry for a few hours once in every week ; but it is difficult
to persuade even Europeans to do this, and though in
smaller stations, where the houses are scattered, a good
deal of personal immunity might be secured by attention
to this point, in the larger places one would gain but little
unless one's neighbours could be induced to do the same.
In such places, however, the weekly cleaning out of all
such tanks should be made compulsory by municipal
bye-law, for in view of our present knowledge it is no
more justifiable to foster the multiphcation of Mosquitoes
than to permit the maintenance of the germs of cholera
or other zymotic disease.

How far care and neatness in cultivation can go to
diminish malaria is well shown in the case of Egypt. The
cHmate is well suited to the spread of malaria, the ground
water nowhere far from the surface, and the annual inunda-
tion of the entire inhabited area would, one would think,
render the country a perfect hot-bed of the disease.

During the actual rise of the Nile, when the whole
country becomes a shallow sea, the conditions are no doubt
not really favourable to the multiplication of Mosquitoes,
but as the waters recede ; but for the fact that the restricted
area available for cultivation renders every square foot of
value, the conditions are naturally ideally favourable to the
spread of the disease. Nevertheless, Dr. Sandwith, of Cairo^
assures me that, although present, malarial disease is neither
common nor serious in the Delta, and a few walks among
the fields near Cairo, showed me that the entire cultivation
was really of the best garden type, and that land and water
were afike regarded by the fellaheen as far too precious to
be wasted as sites and material for puddles.

Celli regards meadow land as unfavourable to the develop-

19G GNATS OR MOSQUITOES— CHAPTER VIII

ment of malaria. In the European sense of the word
cultivation of this description is so exceptional in the East
that it can have but little influence, but in Europe level
pieces of land with a high ground water level are commonly
chosen for pastoral purposes, and the careful levelling of
the fields, together with the thick covering of grass, must
necessarily go far to prevent the formation of permanent
puddles ; and the disappearance of malaria from England
is probably largely due to the substitution of pastoral for
arable agriculture throughout the country.

This disappearance, it will be noticed, dates from the time
when the improvement of means of transport so cheapened
imported corn as to render the cultivation of cereals in
England unprofitable. At that time quinine was so expen-
sive, and its proper administration so little understood, that
it was administered in doses so small that it could have had
but little effect in checking the disease which, nevertheless,
had practically disappeared from the country before the great
fall in the price of the drug rendered its full administration
practicable.

I have been told by a druggist who was in business in
the fen country while it was still malarious, that the
popular remedy was then the prophylactic use of opium.

It was quite common for a farmer's wife to carry home
along with her other weekly market day purchases, a half
pint of laudanum ; but he was but rarely asked for quinine.
Now while there is a certain amount of evidence tending to
show that the habitual use of opium has some protective
influence against malaria, it is certainly incapable of
destroying the parasite, and it appears on the whole unlikely
that the use of drugs can have had any appreciable effect
in stamping out the disease, which is more probably due
entirely to alterations and improvements in agriculture.

The Effects of Human Occupation. — Given the necessary
climatic conditions, it is man himself who is the most
efficient of all agencies in bringing about the conditions that
favour malaria. The human being is essentially a digging
animal, and, whether it be the savage, who scoops up a few
handfuls of mud wherewith to smear the wattles of his hut

CONDITIONS INFLUENCING PREVALENCE 197

or the modern engineer with his steam navy, the result is-
the same less or more, viz., depressions of the surface, and
hence puddles, and it cannot be too often insisted upon
that it is the small, unnoticed puddle close to the dwelling,
and not the large, but distant marsh or pond, that, for
practical purposes, is the real nursery of malaria. The
large tank or tanks that are to be found in almost every
Indian village are generally far too dirty for the taste of
even the none too particular Indian species of Anopheles,
but with us, as with most semi-civilised communities, mud
forms not only the main material of domestic architecture,
but also of house decoration. At frequent intervals the
careful Indian housewife smears the interior of her home
and the little platform before it with a mixture of mud and
the dung of the sacred cow, and the result is certainly to
give an appearance of cleanliness that could hardly be other-
wise obtained for so small an expenditure. In Zululand,
and, I believe, generally among African tribes, the same
mixture is employed. But though the Hindu lady likes to
keep her house neat in this way, both she and the goodman
are tolerant of untidy surroundings to an extent that will be
scarcely believed in Europe, and the result is that the ground
round the house soon becomes honeycombed with small
excavations, which in the rains get filled with fairly clean
water. Nor is the European one whit less blameworthy.

Apart from the special malaria-brewing tanks already des-
cribed, with which most of our gardens are furnished, there
will be found close to every building, whether it be the
cottage of the Eurasian clerk or the imposing barrack or
law court, an excavation of corresponding dimensions, from
which has been taken the earth for its plinth ; and rarely
or never is any attempt made to fill up or drain the
hollow so formed.

Further, as the surroundings of European dwellings
and official buildings are usually kept fairly clean, the results
to the inmates, as far as malaria is concerned, are even
worse than those of the tanks that have originated in the
same way round a native village.

The results of public works are equally disastrous on a

198 GNATS OR MOSQUITOES — CHAPTER VIII

'H^P

^^

feJiai,. '

tEf:

npSVHHS^^klpI

:,i*.

Fig. 34. — View in an Indian cantonment, showing, in the foreground, a
pond resulting from the excavation made to furnish the spoil required for
the plinth of a bungalow, which lies just behind the trees.

Fig, 35. — " Borrow pits " beside an Indian railway.

CONDITIONS INFLUENCING PREVALENCE 199

larger scale. Every road and railway has, on either side of
it, a continuous chain of " borrow pits," which, in rainy
weather, form simply ideal nurseries for mosquitoes of all
sorts.

It is obvious that, in many cases, these pits might easily
be converted into excellent surface drains, but unfortunately,
partly to facilitate the measurement of the work done by
the excavating gangs, and partly to prevent the scouring
effect that might be exercised by a continuous channel, the
engineers carefully avoid doing so, and leave them as a chain
of pools, which remain continuously full of water for months
together. The results, especially where a road or line passes
close to habitations, are so serious that, at very least in
such situations, the making of such undrained hollows should
be prohibited by departmental regulation.

The results even of avowedly sanitary works are unfortu-
nately too often no better. The untoward effects of pure
municipal water supplies, on modern lines, have already
been adverted to, and very often those of attempts at
surface drainage are no more fortunate.

On the next page are four photographs of pools in
the course of the local surface-drainage system, all taken
within a few hundred yards of my bungalow, and it would
have been perfectly easy to fill a large scrap-book with
similar prints. In hill stations, such as Naini Tal, the small
masonry tanks, such as are shown in the two upper illustra-
tions, form almost the only nurseries for Mosquitoes to be
found in such places, as the precipitous lay of the ground is
very unfavourable to the formation of natural collections of
water.

Celli (C. M., p. 147), also emphasises these undesired
effects of otherwise invaluable public works, and is particu-
larly emphatic on those of railways. Railways, and neces-
sarily also ordinary roads, when embanked, may also often
increase the malariousness of a locality in another manner ;
for when, as is often the case, they chance to be carried
across the natural lines of drainage, they inevitably bank
up the drainage of the land lying above them ; and this, too,
in spite of an apparently liberal allowance of culverts.

Fig. 36.— Collections of water in the surface drains of an Indian cantonment. In the two
upper photographs the pools are the result of an engineering expedient to break the fall of the
water ; in the two lower they are accidental.

CONDITIONS INFLUENCING PREVALENCE 201

When travelling by rail during the rains in India, nothing
is more common than to see all the country on one side
flooded, while the other is comparatively dry.

On this account railvs^ays, like canals, should, as far as
possible, be made to follow the water-shed, and where this
is out of the question, the provision of drainage openings
should be much more liberal than is usually the case. It
must be admitted that the European cannot be held
entirely blameless in such matters. Putting aside the
relationship of puddles to malaria, the presence of such
irregular collections of water has long been recognised as
unhealthy, and to say the very least they are untidy, and
would not be tolerated in any advanced European country
on this last score, if on no other. In countries such as
India, our public works should be made standing object-
lessons of the superiority of European methods and system.

To remedy the results of past carelessness would undoubt-
edly be extremely costly ; but, in the majority of cases, they
might have been entirely avoided by the exercise of a little
care and foresight, and that with little or no enhancement
of the first cost of the work. The mere fact that the
excavation of a continuous ditch, in place of a chain of
borrow pits, may endanger a railway embankment by
"scouring," shows what efficient drainage cuts they might
be made, and further conclusively demonstrates how badly
they must be needed. Engineers, moreover, admit that for
one embankment that succumbs to " scouring," a score
collapse from the mere water-logging of their foundations
in time of flood, so that the neglect to favour efficient drain-
age in every possible way can hardly be defensible, even
from a strictly technical point of view ; and I cannot doubt
that the ingenuity of our engineers is fully equal to devis-
ing a remedy for the scouring of drainage cuts placed beside
embankments, if their provision be once admitted as a
necessity. In the case of buildings again, in nine cases out
of ten, the earth required for their plinths might be advan-
tageously excavated so as to make drainage cuts to the
nearest natural line of outfall and so become a source of
improvement, instead of damage, but in practice the matter

20*2 GNATS OR MOSQUITOES — CHAPTER VIII

is mostly left to the native contractor who takes up the job,
and all he has to consider is how he may place the required
number of cubic feet of soil in the place indicated, with
the least possible expenditure of money or thought.

The extent to which this burrowing for building material
has been carried, in and about many Indian towns, is
incredible, and the haphazard way in which it has been
permitted to be carried on has resulted in the absolute
wasting of large areas of valuable, culturable land, in a
country where it can ill be spared. As a general rule, these
"tanks" are far too dirty to favour malaria, for it is not
uncommon to find that some zealous amateur sanitarian
has deliberately carried the foul bazaar drainage into them ;
but there are exceptions to this, and there can be no doubt
that the taking of earth for building and domestic purposes
should be systematised by local bye-law in every Indian
municipal area ; for there are other tropical diseases than
malaria, and the foul emanations of these lakes of putridity
cannot fail to be harmful, even where they do not form
nurseries for the transporters of malaria. The main obstacle
to the removal of these tanks lies in the difficulty of
obtaining, within any reasonable distance, sufficient soil to
fill them in ; but it is not really necessary to fill them up
to the level of the ground around them ; for, provided that
all surface drainage into it be diverted, any given area is
quite capable of absorbing all the rain that actually falls
within its own limits, and provided it be carefully levelled,
no permanent puddles will result, however low-lying it may
be, at any rate in Upper India. Now, without exception,
these hollows are very irregular, not only in outline, but in
depth, and it is only their deepest parts that remain full for
any considerable length of time, so that the only practicable
way of dealing with them is not to attempt to fill them up,
but to level the area they include ; obtaining the earth
required to fill in the deepest parts by cutting away from
the sides so as to bring them to a regular outline, and slope
of bank.

A good deal of spoil too can usually be obtained by
digging cuttings to cut off the drainage of the neighbouring

CONDITIONS INFLUENCING PREVALENCE

203

land, and in a few cases much more can be got by cutting
a ditch from the new level of the floor of the old tank site
to the nearest nullaJi.

The following photograph shows the progress of an ex-
periment that I made in this direction. At the time of my
taking over charge of the local jail, the site of the newly
levelled ground in the foreground was occupied by a tank of
some size, which had originated in the usual way ; in this
case to get earth for the building of the jail. Being within

Fig. 37. — Levelling up the site of aa actual Anopheles tank.

jail limits, although in a waste corner, the contents were
quite clean enough for the local Mosquito larva, and the
place formed one of the principal breeding places supplying
the jail. Some of the deepest parts must have been eight
or ten feet below the level at which the prisoners are seen
dumping the spoil gained from the banks ; and, one way and
another, I got about a couple feet more of material, with
the result that a plot of valuable garden land was substi-

204 GNATS OR MOSQUITOES — CHAPTER VIII

tuted for an injurious collection of water. Moreover, as
far as I can judge, the rental of such a plot would have paid
well for the amount of labour expended. I was led to try
the experiment by noticing a large levelled depression,
which had once been an old-standing brick field, and which
an enterprising native had converted into profitable garden
ground, and finding that, though without any proper out-
fall, the owner was in no way troubled with flooding.

In many cases in towns, hollows of this sort are grouped
together in considerable numbers, with intervening strips
of high ground, and, in such cases, these would supply
ample material for levelling, and the results could scarcely
fail to be remunerative, as cultivable land in such situations
always commands a high rental.

The effects of the surroundings that go to make up the
conditions of town as contrasted with country life are all
in favour of the urban resident. Where indeed could a
generation of Anopheles find a secure nursery for their larval
youth in the heart of such cities as London or Paris ?

Nearly forty years ago Dr. Wood, of Philadelphia,
pointed out "the extraordinary and very important fact,"
that " miasmata are neutralised, decomposed, or in some
other way rendered innocuous by the air of large cities.
Though malarious diseases may rage round a city, and
even penetrate the outskirts, yet they are unable to pene-
trate into the interior ; and individuals who never leave the
thickly populated parts almost always escape " (" Chambers
Encylop.," vi., p. 438, 1868). The true reason for this is, of
course, the careful utilisation of every square foot of surface ;
the grading and paving of the streets, and the resulting
impossibility of the existence of puddles sufficiently per-
manent to rear a brood of larvae ; and it follows from this
that it is only in towns in which the resources of pvilisation
are highly developed that this favourable influence of urban
life is very noticeable.

An unpaved, ill-drained town may be as malarious as
any country district, and as a matter of fact, few of our
Indian cities are sufficiently advanced in these respects to
gain any marked advantage over the surrounding country.

CONDITIONS INFLUENCING PREVALENCE 205

Most of them have arisen from the growth and amalgama-
tion of a number of contiguous villages, and strips and
islands of the old rural area are commonly left between the
more closely packed houses of the old village sites. The
only " up country " city, indeed, with which I am personally
acquainted which at all approximates to the conditions of
our large European towns is Benares, but even there the
back streets and lanes are mostly unpaved, and I have
found puddles teeming with Anopheles larvae in the very
heart of the sacred city, though most of these were fed from
the municipal hydrants, the untoward effects of which have
already been noticed in connection with the effects of water
supply on malaria.

Fortunate indeed it is that such is the case, as, malaria
or no malaria, the value of thase breathing spaces is in-
calculable, for a continuous mass of houses such as form
our large European towns, inhabited by a population of
oriental habits, would be a perfect hot-bed for the breeding
of plague and other bacterial infections.

Outside the bazaars, or business quarters, continuous
lines of houses are rare. In residential neighbourhoods,
owing to the necessity for privacy imposed by their social
system, each house is a hollow square of which one or more
sides are usually simple walls of no great height. As not
only the human inmates, but commonly also cattle and
horses, have to be accommodated, the size of this inner
court even in modest households is often considerable, and
the residences of well-to-do citizens have often considerable
gardens. The latter form ideal breeding places for mos-
quitoes of all sorts, but the domestic puddles of the more
usual enclosures are generally far too foul for the larvae of
even the none too fastidious An. Rossii, though C. fatigans
breeds in abundance about them.

The anti-malarial influence, then, of town life is less
marked in India than in Europe, but it is nevertheless, I
am inclined to think, quite perceptible, for though no
reliable statistics are available, most officers who have been
engaged in our civil medical administration appear to have
a general impression that their towns are less feverish than
the districts surrounding them.

206 GNATS OR MOSQUITOES — CHAPTER VIII

Judging from descriptions, this effect should be more
marked in Chinese cities than in other types of oriental
civilisation ; but whatever it may amount to, the advan-
tage must be purchased at too high a price.

When, however, the Indian townsman is driven to
overcrowding by special local conditions, he packs with a
closeness that puts the Western slum-dweller to shame, and
an enormous population may be concentrated in an area too
small to be beyond the influence of the nearest breeding
places even in its most central parts.

There are, for example, parts of Bombay where the
density of population probably exceeds that of the worst
European slums, but the dimensions of this " congested
area " are not considerable.

As, however, the domestic interiors are usually too
unsavoury to furnish nurseries for the really dangerous
species, it is certain that the paving and draining of our
large towns will do much to diminish municipal malaria.

A good deal indeed has been done, and is in progress in
this direction, but as the constantly recurring phrase of
conventional Indian self-depreciation has it, " We are very
poor folk," and improvements that appear the simplest
necessities of urban sanitation to the European expert are
quite beyond the pockets of the community, however
enlightened the views of the Government may be.

The paving of streets and the introduction of systematic
surface drainage are of course large and costly undertakings,
which can only be carried out gradually, but it must not be
supposed that nothing can be done in the mean time.

Nine-tenths of the nurseries of Anopheles larvse that I
have met with within municipal limits are of such small
dimensions that they might be put an end to by means of
a few shovels-full of earth, and a few men trained to
systematically fill up the puddles as they formed might
undoubtedly do a good deal to diminish municipal malaria.

Such a measure is of course rather of the category to
which belongs chemical disinfection, than to that of radical
prevention, for I am perfectly aware that new puddles
would form as fast as the old ones were filled up, but these
could do no harm if filled up in their turn.

CONDITIONS INFLUENCING PREVALENCE 207

Let it be at once admitted that the extermination of
Mosquitoes may be impracticable ; but it does not follow
that we should fold our hands and make no effort to pre-
vent their fullest multiplication. Every Anopheles puddle
filled up means one focus the less of infectible material, and
the mere fact that without any specially directed sanitary
efforts the city of Rome is malaria free, though standing in
the midst of most deadly surroundings, shows how much
may be effected.

There are millions of unvaccinated persons in India, and
everyone of them is a possible focus of variolous infection,,
but no one can doubt that vaccination has effected an enor-
mous diminution of small-pox in the country, though the
impossibility of securing universal vaccination might have
served equally well as an argument for attempting nothing
in the matter. In the wide expanse of the country in
general such detailed measures are certainly impractic-
able, but in limited areas, such as those of municipalities,
much might undoubtedly be done by the intelligent applica-
tion of our present knowledge, albeit of a " hand-to-mouth "
character, and that at but a trifling cost. Much also might
be done by the enforcement by municipalities of bye-laws
prohibiting the indiscnminate honeycombing of the surface
for earth for building and plastering purposes. Such a
regulation need not give rise to any real inconvenience,
as tanks and other excavations of a size unhkely to serve
as nurseries for larvae are so numerous in most such
municipal areas that it would cause no hardship to insist
on the earth required for such purposes being taken from
their banks. The systematic filling up of small depressions
with any hard rubbish that may be available is another
measure that obviously suggests itself, and as the most
dangerous collections are generally quite small and shallow,
need not be beyond the pecuniary resources of even small
places.

In the North-west Provinces and Punjab I have rarely
met with any large collection of water or "tank" within
the limits of a native town which contained Anopheles
larvae. Most of them, indeed, are too foul for even the least

208 GNATS OR MOSQUITOES — CHAPTER VIII

fastidious of Culices, though, within cantonnients, where
their contents may be Httle else than rainwater, they occa-
sionally do so. In provinces, on the other hand, where the
rainfall is heavy, and especially in such tanks as lie in the line
of a natural drainage depression, and are hence scoured out
by heavy falls, such tanks may contain larvae ; but I doubt
if they are likely to do so for any considerable portion of the
year, though it is hazardous to attempt to generalise in such
a matter, as exact local knowledge is alone of any value.

It must be remembered in this connection that, in Lower
Bengal, tanks arc largely used as sources of drinking water,
and such tanks are more or less guarded from pollution, for
callous as the Indian may be in this matter, there are few
who would drink the water of a town tank in Northern
India ; and any tank containing water that would be
regarded as drinkable, even from a native point of view,
would certainly form a congenial habitat for Anopheles
larvae, and the improvement of municipal sanitation against
malaria will be in such places proportionally difficult.

Where not indispensable as sources of drinking-water,
such tanks might doubtless be dealt with by the use of
larvicides ; but in such places these tanks are both large
and numerous, so that the expense would be a large and
constant one, and we are further met with a difficulty that
must be fatal to the success of all temporary, and therefore
continuous measures in India — the difficulty of providing
adequate intelligent supervision.

Another reason why not only cities but also smaller long-
inhabited sites tend to become less malarious is that the
ground level is being slowly but continuously raised by the
accumulation of the ruins of older buildings on which, from
time to time, new buildings are raised. Often, in India,
village sites are of an unknown antiquity, and in such the
tortuous village streets wind their way always up hill to its
centre, where often there still stands the more imposing
home of the headman. The older the hamlet, the higher
the hill, and some have lasted so long that the site of the hut
from which it grew may now be 50 feet or more above the
unbroken level of the surrounding plain. Now it is obvious

CONDITIONS INFLUENCING PREVALENCE 209

that this self-raising action must have a most beneficial
effect by favouring surface drainage, and this affords us a
hint that might well be taken advantage of in planning new
settlements on ground reclaimed from the desert by canal
irrigation ; only in place of gaining the earth by honey-
combing the plain hard by with a network of foul tanks, the
spoil should be taken from carefully planned drainage cuts
carried along the natural lines of drainage. Surely, con-
sidering the vast outlay involved in the construction of a
great canal, it should be a good investment to spend con-
siderable though comparatively trifling sums to secure
health for the colonists who come to reap the plenty brought
by the fertilising water.

Such being the conditions that influence the prevalence
of Mosquitoes, it may be asked whether the seasonal preva-
lence of these insects really corresponds with the intensity
of malarial disease ? On this point the figures given by Colli
are sufficiently convincing, but for many reasons it is difficult
to quote statistics of corresponding value for India, though,
speaking generally, no doubt can be entertained as to the
fact of the coincidence, or rather consequence.

The connection is, however, of a kind that is much more
obvious to the working physician that to the statistician.
The former well knows that, in Northern India for instance,
his really troublesome malarial cases occur between the
middle of August and the end of November, and that those
which are admitted between January and August are mostly
recurrences, generally lasting only a few days, and of a
comparatively mild type.

These relapses, however, go to swell the number of
admissions in months during which, in many parts of India
Anopheletes are, practically speaking, as rare as the dodo ;
and thus it happens that in spite of the infinite amount of
labour that is wasted on the statistics that lumber the
record rooms of our offices in India, we are still quite
without any really reliable information as to the seasonal
prevalence of malaria.

Apart from the fact that hitherto no attempt has been
made to distmguish between primary and recurrent attacks,
14

210

GNATS OR MOSQUITOES — CHAPTER VIII

a distinction which, it must be admitted, it would be very
difficult to make in practice, the diagnosis between the
various forms of pyrexial disturbance, that in India and
elsewhere in the tropics are grouped together as "fever,"
has hitherto been made in a very loose way.

No statistics as to malaria can be considered to be exact
and definite in which the diagnosis is not based on the
ascertained presence of the parasite in the blood, and as yet
such statistics on any considerable scale are entirely want-
ing. No doubt in the majority of cases the symptomatic
diagnosis of malaria is perfectly correct, but we are only
now commencing to differentiate exactly between remittent,
malarial, typhoid, and Malta fevers, and on the other hand
there can be no doubt that numbers of cases of transient
pyrexia, due really to digestive disturbance giving rise to
the absorption of toxic materials from the bowel, and to
various other causes, are returned as malaria. Such cases
are very common in the dry, hot season, and in a large
proportion of them certainly, the most careful examination
fails to demonstrate parasites in the blood.

A certain number of course are really malarial, but are
recurrent attacks which, as far as my limited experience
extends, appear in this part of India to be usually charac-
terised by the presence of the small, round, unpigmented
forms unaccompanied by crescents in the peripheral blood.

In spite, however, of these sources of fallacy the figures
at our disposal are really sufficiently conclusive for all but
the most exacting.

Taking the returns of intermittent fever as the least
open to diagnostic errors, we find that the last available
report of the Sanitary Commissioner with the Government
of India shows that in the European army the monthly
admissions were as follows : —

a

-?;

S

„■

>.

i

>.

^

-w

^•

o a;

Total

4

&,

s

<

S

H^

'^

■<

IB

O

t5 1 O

Beugal . .

368

181

2n3

316

493

230

387

899

666

754

446

323

5,272

Punjab ..

344

ITS

206

357

5S5

414

530

437

526

517

223

150

4,467

Madras . .

245

146

150

162

152

175

217

316

262

352

430

224

2,831

Bombay ..

296

235

197

216

427

355

317

360

270

306

212

162

3,353

All India . .

1,253

740

756

1,051

1,657

1,180

1,451

2,012

1,724

1,929

1,311 U59

15,923

CONDITIONS INFLUENCING PREVALENCE

211

These admissions took place on a strength of 67,697, so
that roughly speaking about a quarter of the entire force
suffered from malarial fever during the year.

In the native army the seasonal prevalence of the disease
follows a closely similar course.

Bengal
Punjab
Madras . .
Bombay ..
Hyderabad
Contingent

,480 2,417

1,704
044
65t)

1,167
1,532

1,596

2,112

696

802

124

1,339

'658
543

94

8,247
11,847
5,571
5,882

94 1,077

2,419] 33,434

■ This on a strength of 128,529, and again it will be seen
that roughly a quarter of the strength suffered from
malaria.

Our European and native armies in India are not strictly
comparable, as the latter is a long service force, necessarily
composed of men, older and more seasoned to soldiering
than their European comrades ; but in spite of this, while
rather less than a quarter of the former suffered, the latter
did so in the proportion of rather more than a quarter.
Something of the difference is doubtless due to the more
sanitary tendencies of European personal habits ; but mak-
ing all such allowances, it must be admitted that if the
native has acquired any immunity he has done so to so small
an extent that it is a factor of too trivial importance to be
worthy of practical consideration.

It is further noteworthy that in specially malarious dis-
tricts, such as the Bengal and Orissa group of stations, where
the relatively greater preponderance of genuinely malarious
cases tends to minimise the fallacy of included cases of
diseases simulating malaria, the seasonal variations in the
intensity of maiaria are much better shown than in less
malarious places.

In all stations where the seasonal prevalence of Mos-
quitoes has been made the subject of careful investigation,
the period of greatest intensity of malaria has been found

212 GNATS OR MOSQUITOES— CHAPTER VIII

to coincide with that of the greatest prevalence of Anophe-
letes.

As a striking example of this I may instance that
writing to one of my numerous correspondents, Lieut.
Glen Liston, I.M.S., who is quartered at Ellichpur in the
Berars, to ask him if there was any reason to believe that
one of the methods of securing permanence of the species
might be the survival of retarded larvae, he replied that
there. Anopheles was breeding already at the date of writing
(January), and sent me numbers of specimens including
some in tubes, one of which actually reached me alive.
Now it is a curious fact that in the above-quoted table,
Ellichpur is one of the few stations in which the largest
number of malarial admissions takes place in January, the
monthly admissions being 31, 12, 10, 3, 7, 2, 8, 14, 9, 29, 21,
15; total im.

In Indian jails the figures are closely similar. Out of an
average strength of 110,016, there were 37,776 admissions
for intermittent and 734 for remittent fever, the largest
number of admissions being in the months of July, August,
September and October. The proportion of admissions to
strength, about one-third, is somewhat higher than among
the troops, but this is only to be expected when the inferior
physical condition of the class of inmate is remembered.
The interiors of most Indian jails are models of cleanliness
and good sanitation ; but the boundaries of the hygienic
oasis are abruptly limited by the jail walls, and immediately
outside, the excavations that have yielded the material for
their construction, brick pits, and ill-contrived drains, too
often furnish breeding places in abundance for the malaria-
carrying Mosquito.

Moreover, the strict prohibition against the extra-mural
employment of convict labour greatly ties the hands of the
Superintendent in his efforts to improve the sanitary con-
ditions of any spot beyond the four walls.

Having now considered, as far as space will allow, the
conditions that influence the prevalence »or otherwise of
Mosquitoes, it remains to be considered what can be done
to diminish the pest. The malarial parasite has, it must be

CONDITIONS INFLUENCING PEEVALENCE 213

remembered, two distinct stages of parasitic life, each with
its special host : Mosquitoes of the genus Anopheles during
its period of sexual life and multiplication ; the human
subject during its non-sexual stage : and it is obvious that
the parasite may be attacked in either, or both of these
stages.

The stage of the parasite that is passed within the
human subject may be dismissed with a few words. As
the end of all our efforts is the preservation of the host,
all that can be done is to poison the parasite with quinine,
or to isolate cases of malaria in such a manner that they
cannot infect a relay of Mosquitoes. As regards the first
method, a good deal is being done by the Indian authorities
to popularise the use of quinine by distributing it through
the agency of the Post office at cost price through the
length and breadth of the land. At every Post office in
India one can buy for a farthing five grains of excellent
quinine, and though the amount so distributed, when
stated as so much per head of the population, is insignifi-
cant, the amount sold is already much more considerable
than most of us expected would be the case among a people
so slow to avail themselves of new advantages as that of
India. AYithin the present year a new effort in the same
direction has been made by enfisting the agency of land-
owners to distribute the drug in the same semi-gratuitous
fashion to their tenants ; and almost the last official duty
it fell to the writer's lot to perform was to compile an
indent for something like a hundred-weight of quinine for
distribution in this way in a single district, a quantity which
would have made his earlier administrative chiefs aghast
with astonishment. Added to this, the invaluable services
of Sir William King, F.K.S., late of our service, in estab-
lishing the cultivation of the cinchona plant in India, have
revolutionised the price of the drug, and thereby conferred
an incalculable boon not only on India but on all malaria-
stricken humanity. The absolutely gratuitous distribution
of quinine has been suggested, but personally I doubt if
the cost to the consumer would be perceptibly less, for
State benefits of this sort are woefully liable to be lost in

214 GNATS OR MOSQUITOES — CHAPTER VIII

lubricating the smaller administrative channels through
which they must needs pass after flowing beyond the ken
of the large European distributory. In advanced countries,
such as Italy, it is quite possible that the isolation of
malarious patients might become practicable, especially as
the large towns afford places free from the harmful species
of Mosquitoes in which patients could be treated in the
ordinary hospitals, without any special precautions for
keeping out the insects ; but apart from the stupendous
number of cases to be dealt with, such measures are out of
the question in any malarious British dependency, and most
emphatically so in India, where any attempt at enforcing
isolation would infallibly result in political disturbance of
the most serious character. To anyone who has passed a
few months in the East such a statement will seem a
superfluous truism, but so astounding and scathing are the
comments on Indian administrative matters that are made
by confident critics, as conversant with our social and
climatic conditions as they are of those of the planet Mars,
that it may not be out of place to record it.

Turning now to the stage passed within the Mosquito,
it is obvious that as we have no means of distinguishing
infected from healthy insects, the solution of the problem
lies either in the destruction of mosquitoes, or by avoiding
being bitten by them. The consideration of the second
class of precaution comes under the heading of personal
prophylaxis, while the first must depend mainly on com-
munal action. Measures designed to diminish the multi-
plication of Mosquitoes are of two classes, viz., those
designed to check their increase by doing awa}' with the
conditions that favour their breeding, and secondly, those
that aim at the extermination of the race while leaving
them to multiply at their own sweet will. Of the two,
the first is, without question, the more efticient, but the
practical sanitarian will not disdain to avail himself of the
other class of expedient whenever it is practicable. It will
be years before much can, however, be accompfished in the
way of the really radical measures of the first class, and
for the present we must content ourselves mainly with
temporary expedients.

CONDITIONS INFLUENCING PREVALENCE 215

And here it may not be out of place to comment on the
attitude of those who have been heaping cheap ridicule on
those who have been working on this subject. The favourite
expedient of this class of humorist is to impute to a man
opinions he has never expressed and then to demonstrate
how silly he must be to hold such doctrines. Applied to
their own persons and profession they would be the first
to resent the assumption that the shoemaker is, ex officio,
incapable of using a last ; and yet they will have it that
the fact of a man making a life-long study of sanitary
problems renders him absolutely incapable of forming a
rational judgment on the subject. The stock gibe of these
good folks is to accuse Major Eonald Eoss, and others
associated with him, with proposing to "exterminate Mos-
quitoes." Now I can assure them that though he labours
under what they regard the incurable disability of being a
scientific investigator, and a most distinguished one at that,
he is, even outside his laboratory, no simpleton ; and no one
but a fool would propose so impossible an undertaking,
least of all a naturalist who has made a close study of the
life-history of these insects. Between this and asserting
that it is possible to diminish their numbers there is a very
wide gap, though even on this score he has been singularly
moderate, for the utmost that he has ever suggested as
practicable is to check their free multiplication in certain
special localities ; and of the practicability of this there
can be no possible doubt.

While, however, none of us see any immediate prospect of
being able to "stamp out malaria" or of wiping out the
entire family of gnats, we find in that no reason for sitting
with fatalistically folded hands, or of neglecting to utilise
every possible method of keeping down their numbers,
whether by doing away with every removable breeding
place, or by directly destroying the insects in any stage of
their existence. The fact is that there still remain numbers
of people who regard disease as an inevitable infliction,
which can only be dealt with by drugging, and to whom the
conviction is strange that every disease must necessarily
have a definite cause, which, sooner or later, will be dis-

216 GNATS OE MOSQUITOES— CHAPTER VIII

covered in all, and is in a number of instances already
known. The older frame of mind saved all trouble in
thinking, and therefore, quite unconsciously, the modern
standpoint is repugnant to them, and they resent being
told that if they plan their gardens so as to be perfect incu-
bators for Mosquitoes and refuse to adopt the most simple
precautions against being bitten, they have only themselves
to thank if they get fever. That though now and again the
insect may have been reared elsewhere, the chances are nine
to one that the gnat that inoculated them would never have
seen the light had not they themselves provided his parents
with every possible convenience for rearing their family.

As a matter of fact, every naturalist would scout the
idea that we can hope to perceptibly diminish the number of
Mosquitoes in tropical regions ; and can probably advance
better grounds for the faith that is in him than those who
picture him as proposing to root out the race ; but he does
not share in popular notions as to these insects "appearing"
from no ascertainable origin, or that they can migrate or be
carried long distances by the wind. He knows on the contrary
that for all practical purposes every locality breeds its own
Mosquitoes, and that whenever he sees one of these insects
he may be sure that the water it was reared in is not a quarter
of a mile off. Given that a place be destitute of collections of
water suitable for rearing the larvae and it will be necessarily
free from Mosquitoes, even if there be places that swarm
with them within a mile's walk. From this it follows that to
secure a local immunity from them is almost always within
the range of possibility, and may in a few cases be a quite
simple matter, and resolves itself essentially into a question
of expense. Given that we have free access to the entire
area, and liberty to deal with every breeding pool we discover
as we think best, there are very few collections of water that
cannot be drained, filled in, or in some other way rendered
unfit for the rearing of Mosquito larvse; but in many cases the
cost would be prohibitory, as for example where the insects
larvate in rice fields, and if we desire to prevent the multipli-
cation of Anopheles, we must make, up our minds to prohibit
the cultivation of the staple food of the population — too

CONDITIONS INFLUENCING PREVALENCE 217

high a price to pay, even for immunity from fever. On
the other hand there are many places where the number of
possible breeding places are small in number and limited in
extent, and where accordingly immunity can be secured
at very trifling trouble or expense. The instance of the
successful use of kerosme as a larvicide thirty years ago,
in America, quoted in the first edition, must have been an
extreme example of this. In this case, L. 0. Howard
("Insect Life," vi, p. 90) describes how a residence was freed
from Mosquitoes by killing the larvae in a single pool some
4,000 feet square. In by far the majority of places the
experiment would, at best, have proved but partially success-
ful, owing to the existence of alternative, but unsuspected
breeding places.

To mention another instance : There is, I have no hesita-
tion in stating, no good reason why the European residents
of most stations in Upper India should be much troubled by
Mosquitoes in the hot weather. The fierce, dry heat has
dried up every possible natural breeding place, and the
swarms of Mosquitoes that render life a burden are bred
exclusively in the garden tanks described above, and in
other easily discoverable domestic collections of water. If
each resident would devote a few minutes once a week to
seeing that his servants emptied these and carefully
cleaned them out, the nuisance would be reduced to a
minimum. Moreover, in many places the individual houses
are so scattered that a considerable degree of personal
immunity may be secured by attention to one's own bunga-
low alone.

As the quarter in which they reside is usually quite
isolated from the houses of natives other than their own
domestic servants, the area is entirely under their own
control, and all that is required is a little friendly co-
operation.

From what has been said it will, I think, be clear that
it is impossible to prescribe any generally applicable method
of dealing with malaria by radical measures intended to
diminish the prevalence of Mosquitoes, and success can only
be hoped for by a careful study of local conditions, guided

218 GNATS OR MOSQUITOES — CHAPTER VIII

by a clear understanding of the conditions that favour the
multiplication of these troublesome insects.

The first step is to discover the pools where they breed
and the seasons at which they do so. This done, we have
to consider why these pools exist, and if possible to remove
the cause. Unfortunately in many cases the breeding
places, although small, are so numerous that the task of
dealing with them in detail is almost hopeless, and would
involve heavy and continuous expense. This may be due to
the character of the soil or to waterlogging, and in such cases
only extensive drainage works, surface and deep, can be
expected to afford any permanent benefit. As a rule it is too
costly to attempt anything but surface drainage, and even
this is by no means an easy matter, for unless the drains be
most carefully levelled and paved, they generally during the
rains becomQ chains of small puddles, which form the
favourite nurseries for the noxious "dapple-wing" Mos-
quitoes. But the paving of drains on any adequate scale
is a most costly business, and except in closely populated
municipal areas out of the question on the score of expense.
In most cases, therefore, they are better restricted to a few
deep cuttings, and to straightening and clearing the natural
outfalls, for I am convinced that the multiplication of what
are called in India " kachcha " drains, i.e., shallow unpaved
gutters, does more harm than good, and that in canton-
ments and in other places where sickliness is so costly to
the tax-payer as to justify any reasonable expense on sanita-
tion, it would be better to systematically drain the area by
means of agricultural " subsoil " drains. As already pointed
out, it is a misnomer to speak of these by their usual name,
as they are placed immediately beneath the surface, and so
cannot directly affect the level of the subsoil water, and
they really drain only the surface ; but as its contour
remains unaltered they cannot lead to the production of
puddles in the same way as mere open cuttings in the soil.
Deep cuttings, on the other hand, though equally irregular,
do not for some reason so often harbour Anopheles larv«.
It may be that this is because the water is hidden from the
female Mosquitoes when seeking for a place wherein to lay

CONDITIONS INFLUENCING PREVALENCE 219

their eggs, but a more probable suggestion, for which I am
indebted to Dr. St. George Gray, of St. Lucia, is that the
larvae require plenty of light.

It is in level country such as the Indian alluvia that
conditions such as the above are generally found. On
midulating ground surface drainage may usually mainly be
left to care for itself, and in such a place as Freetown,
Sierra Leone, to judge from the reports of the two Malaria
Commissions, no extensive drainage measures can be
expected to be of any proportionate value. Here the
breeding pools are basins of solid rock which necessarily
cannot be drained, and as far as one can judge, the
measures that suggest themselves are the regrading of the
banks of the rocky watercourses, and the improvement of
the roads and other places where the rock lies bare, and
the formation of these peculiar puddles thereby becomes
possible. Possibly, where not exposed to traffic, many of
these pools might be rendered innocuous for a considerable
period by filling them with sand ; but this aside, the main
reliance must be placed on larvicides, and I hardly under-
stand why Drs. Stephens and Christophers should be dis-
appointed in the results of their employment (K.S.M.C,
p. 43) on the ground that the larvas reappeared as soon as
the use of the larvicides was discontinued. The effect of
such agents is at most a matter of days, and there is no
possible reason for the Mosquitoes not returning to pools so
treated the moment their effects have disappeared. The
main objection to the use of these agents is not that they
are wanting in efficiency, but that success can only be
obtained by continuous trouble and expense, but where the
breeding pools are in manageable numbers, and they cannot
be done away with except at prohibitory expense, it is as
reasonable to keep up a staff of puddle oilers as of scavengers,
for the result of the labours of the street sweepers is no
more permanent than that of the larvae destroyers, and it
is as fair to object on this score to the continuous employ-
ment of the one municipal servant as the other. I feel
perfectly sure that Eoss never expected to obtain any
advantage from the use of such agents without their

220 GNATS OR MOSQUITOES— CHAPTER VIII

employment being continuously and systematically kept
up, and for the purposes of an initial experiment such as
his, or in any case where immediate results are demanded,
no other plan is applicable. Sanitation against malaria on
radical lines will everywhere require years of continuous
and progressive effort, and as long as it remains uncom-
pleted we must necessarily depend upon hand to mouth
expedients, such as the use of larvicides. But whatever
plan be adopted those who hold the purse strings must
reconcile themselves to the fact that the prevention of
malaria is impossible without considerable outlay, and they
must further be prepared to see a great deal of it expended,
though not really wasted, in futile attempts, for we shall no
more immediately hit upon the best method of dealing with
this difiiculty than we have in the deodorisation of sewage
or any other sanitary problem. For the present, while
radical measures should always be kept in view and under-
taken as funds and opportunity permit, we must content
ourselves with temporary methods, and it is well to remember
that though we may not be able even locally to " exter-
minate " the carriers of disease, it is always worth while to
diminish their numbers. Every breeding pool filled in or
otherwise dealt with, means one possible focus for dissemi-
nation the less, and a great deal might be soon accomplished
by the systematic attention to the immediate neighbourhood
of barracks and other dwellings.

Turning then to measures of this class, it is clear that
the insects might be attacked either in the aquatic or the
aerial stage of their existence, but that they may be far more
easily got at m the former. During their adult life, the only
period during which much advantage is likely to be gained
by attempts to destroy them is that of hybernation, and it
has been shown that, in uniformly warm chmates, they
cannot really be said to hybernate at all. Where, however,
they do so, their destruction is of the first importance, and,
owing to the sluggish condition of the insects, not so
impracticable as might be imagined. The impregnated
hybernating females are, it must be remembered, the main
hope of the race for the generation of the coming year, and

CONDITIONS INFLUENCING PREVALENCE 221

their destruction means that of their posterity. Their
numbers, moreover, are comparatively small, and the situa-
tions in which they are likely to be found can be predicated
with tolerable certainty. I do not, however, propose that
one should actually search for them, for the insects under-
stand too well their business to make any search profitable ;
but it may be taken as certain that the nooks and corners
of every room and native hut harbour numbers of them, and
it is perfectly easy to destroy them by fumigation with
sulphur employed in the same way as in ordinary disin-
fecting operations. The room to be dealt with should of
course be closed up as closely as possible, but no elaborate
arrangements for closing all crannies are necessary, as they
succumb to a proportion of the vapour in the air that would
be perfectly harmless to bacteria. It is very little use to
fumigate with pure sulphur, as it burns so uncertainly that
it generally goes out before half the material is consumed,
and it is even more futile to burn the sulphur by throwing
it on a charcoal braaier as is often done, because the result
is mainly vaporised sulphur, which is quite useless. A very
short exposure to the fumes suffices to kill the stoutest
Mosquito that ever buzzed, but as the buildings to be
dealt with are mostly extremely pervious, it is of the first
importance that the sulphur should be burned very quickly,
so as to produce a volume of sulphurous anhydride large
enough to be lethal before the product of the portion first
burned has time to escape. What is wanted, in fact, is a
mild firework which will burn out in a very short time,
but yet will not flare up to a dangerous extent. About one
part each of nitre and charcoal to eight of sulphur answers
well, the mixture being made up into pastilles weighing 4 oz.
each, by means of a little gum-water, and dried in the sun.
Pastilles of this description were made up for me by Messrs.
Waldie, of Cawnpore, and I found that even in thatched
buildings not a single insect was left alive. The floor of
a bath-room, in which hardly any Mosquitoes could be found
by any ordinary search, was found after fumigating in this
way covered with dead Mosquitoes, a circumstance which
gives a good idea of the effectual way in which they hide

222 GNATS OR MOSQUITOES — CHAPTER VIII

themselves. One pastille should be allowed for every
thousand cubic feet of space. Merely for the sake of per-
sonal comfort, it is well worth while occasionally burning
these pastilles in the hot season in bath-rooms and other
favourite lurking places, as to do so costs but little either
in cash or trouble, and the smell of the fumes disappears
as soon as the place has been aired ; but it is absolutely
useless to do so if active breeding places are allowed to exist
close by. To be effectual, fumigating should be done towards
the close of the hybernating season, and should be performed
during the heat of the day, when even the least sluggish
are sure to be under shelter. I trust that no playful critic
will indulge us with an arithmetical dissertation on the
number of tons of sulphur required for the annual fumi-
gation of every native home in India, on the gratuitous
assumption that I propose to enforce some such measure
at the point of the bayonet, for rudimentary statistical
exercises of this sort are within the powers of the most
modest mathematician, and it is needless to say that every-
one who has served in the sanitary department in India
knows far better than any others that it is quite useless to
try to force or even persuade the native to adopt sanitary
precautions of any sort within the precincts of his own
house. The caution is hardly needless, as, if I remember
rightly, some wit went to the pains of calculating the annual
enhancement of taxation per household necessary to provide
all native babies with a mosquito net, on no better grounds
than because some one of us had been emphasising the
importance of Europeans protecting themselves by their use.
But though we cannot fumigate the Indian continent, there
is no reason why the plan should not be adopted in the case
of all barracks and other quarters provided by Government,
and in cantonments where such natives as are permitted to
reside within their limits, do so on the distinct understanding
that their status is more or less one of sufferance, and on
the understanding that they are willing to submit to the
sanitary whims of those for whose use the area has been
set aside, however unreasonable they may appear to people
in their particular stage of civilisation ; it would be not

CONDITIONS INFLUENCING PREVALENCE 223

onl}^ justifiable, but feasible to insist on such a measure
being carried out.

As each pastille costs no more than a Lee-Metford cart-
ridge, the cost would not be prohibitory, and the annual bill
for invaliding the men who have been educated to use the
latter is so heavy that it would pay well to adopt any
measure likely to diminish it ; for the lessening of the
annual invaliding roll by one or two names would more
than meet the cost. In most parts of India the month of
February would be the best time to apply such a measure.
With this exception we possess no means likely to be of
service in destroying the adult insect, and therefore must in
the main confine our efforts to attacking the more vulner-
able larvee. In attempting this the species found in a pool,
or even their presence or absence, is a matter of little
importance, for almost any collection of water is a possible
breeding place ; and the rule should be to leave as few as
possible available. All small pools are certainly best dealt
with by filling them up ; care being of course taken not to
obtain the necessary material by excavating a new hollow.

In most cases sand from the nearest river bed will be
the best thing to employ, as it refuses to hold water, and no
harm is likely to result from digging it. A large proportion
of Anopheles pools are, however, so small that a dozen or so
may be done away with with a single cart-load of sand, and
as the results would be fairly permanent this would be far
more economical than the use of larvicides. For larger
collections of water, pending the adoption of more radical
measures, the use of the latter is the only feasible plan, and
the only question is as to what is the best material to
employ. Unfortunately the larvae are singularly resistant
to ordinary poisons, and are capable of disporting themselves
for hours in such energetic fluids as the Liq. hyd. bichlor.
B.P., or in Fowler's solution.

Celli (C. M., p. 19G, et. seq.) gives the results of an
extensive series of experiments as to the quantity required
to kill the larvae of a large number of agents, which may be
consulted by those who desire more detailed information on
the subject ; but their practical outcome is that with the

224 GNATS OR MOSQUITOES — CHAPTER VIII

exception of paraffin, and an aniline product called " larvi-
cide," prepared by Weiler-ter-Meer, of Uerdingen, none are
sufficiently powerful, in proportion to their cost, to be of
any use on a practical scale. To these may be added tar,
but the reports I have received as to the efficacy of this
latter are so contradictory that there can be no doubt that
it is very uncertain in its action, and it is hardly cheaper
than paraffin, as to the efficiency of which, when properly
applied there can be no doubt. Paraffin, it must be remem-
bered, kills the larvoB, not by acting as a poison, but by
choking them, the thin film which it forms on the surface
preventing their getting access to the air ; and therefore to
use it with success, care must be taken that it is made to
spread over the entire surface. No very large quantity is
required, but it is better to err on the side of liberality ;
and according to my experience, the best and easiest way is
to sprinkle it over the surface by one or two rapid sweeps
from a fine-rosed watering pot. So employed, I have never
known it fail, the larvse being all found dead in the course
of a few hours.

" Larvicide " on the other hand acts as a poison, killing
all larvEe within twenty-four hours, in a strength of about
one in 7,000. The consignment I sent for had not come to
hand when I left India, so that I know nothing of its powers
from personal experiment ; but Celli appears to regard it as
the only agent of the poisonous class that is cheap enough
for use out of the laboratory. He appears to consider it as
the most promising agent we have as yet at our disposal,
and states that the cost of disinfection per cubic metre of
water varies from lire 0'0056 to 0'0012, which is equivalent
to saying that a shillingswoxth is sufficient for from 7,350
to 34,300 cubic feet of water. It is said to be not in the
least injurious to plants, and that the water containing it
may be drunk with impunity by men and cattle, but never-
theless to be lethal to most insects which are injurious to
crops, so that it is particularly suitable for the treatment of
such portions of rice swamps as are dangerously near
habitations.

It has the further advantage over paraffin in the fact

CONDITIONS INFLUENCING PREVALENCE 225

that, not being volatile, its action is much more permanent,
lasting for as long as two months, and there can be no
doubt that it should be at least given a thorough trial.
Assuming it to stand the test of practical work in the open,
the question whether this or kerosine will be the cheaper
agent to employ will depend on the depth of the water to
be dealt with, as the entire bulk of the water must be
rendered lethal ; whereas with kerosine depth need not be
considered, and superficies is all that need be taken count
of; so that where the water is shallow in proportion to its
surface, paraffin will certainly be the cheaper ; while
deep pools will be more economically dealt with by the
new agent, the more so as the chemical would probably
remain unaltered in the surface mud when it dried up, and
more or less of it would be still available when redissolved on
the reformation of the pool after fresh rain.

It may be freely admitted taht in many places larvae-
bearing pools are so omnipresent that it may be practically
impossible to deal with them in this, or any other detailed
fashion, but it is equally true that many limited areas might
be dealt with at but httle cost or trouble ; and it is to be
hoped that some of our Colonial Governments may see
their way to giving a fair trial to the methods above
indicated in some selected localities.

Such being the state of our knowledge as to the sani-
tation of malaria on a large scale, it remains to be seen
what can be done in the way of personal hygiene ; and
here I believe it may be confidently asserted that much
may be accomplished, though not of course absolutely
without trouble, expense or other effort on the part of
those who wish to protect themselves against this disease.
Putting aside the prophylactic use of drugs, which may
be better left to purely medical treatises, personal hygiene
resolves itself into two classes of precautions, viz., by as
far as possible preventing Mosquitoes from breeding in the
nnmediate neighbourhood of our houses, and secondly, as
private efforts of the first category can be only partially
successful, to endeavour to avoid being bitten.

Precautions of the first class have already been sufH-
15

226 GNATS OR MOSQUITOES — CHAPTER VIII

ciently noticed, as they are merely such as have been
suggested as corporate operations, but on a smaller scale ;
but as the tropical resident usually lives in houses sur-
rounded by a considerable area under his own control, a
good deal more might be affected in this way than is usually
the case in Europe. By a very little personal trouble and
superintendence there should be no difficulty whatever in
doing away with all breeding places within one's own com-
pound, and an occasional round, followed by a coolie armed
with a can of kerosine, would do much to keep them down
in our immediate neighbourhood, while occasional fumiga-
tion with sulphur, especially of one's servants' houses, during
the hybernating season would help to minimise the number
of immigrants from the quarters of less careful folks. But
our main reliance must be placed on precautions against
being bitten, the principal of these being to endeavour to
keep Mosquitoes out of our houses ; and the wonderful
success obtained by Professor A. Celli in the case of certam
railway employes on the line from Rome to Solmona shows
that this is by no means as difficult as one would have
expected.

No one can read his recent pamphlet " Sulla nuova
profilassi della malaria," in which he gives an account of
these results, without being convinced of this ; unless they
prefer to regard the account as an effort of the imagination,
a supposition which no one who has had the pleasure
of talking with that distinguished hygienist would for a
moment countenance. Selecting a notoriously malarious
portion of the line, he had about half the cottages in which
the railway men with their families live made roughly
Mosquito-proof by protecting all the windows with fixed
screens of wire gauze, and by providing all entrances with
double spring doors of the same material, as shown in the
accompanying illustration, reproduced by Professor Celli's
courteous permission, while the other moiety were left in
their original condition. The experiments have been now in
progress for nearly three years, and counting each year as
a separate observation, the results may be epitomised
roughly as follows : In 25 protected cottages with a popu-

CONDITIONS INFLUENCING PREVALENCE

227

lation of 173 persons, only 8 were attacked with fever;
whereas in 30 unprotected, having a population of 220, only
17 escaped the disease.

In several instances the compared cottages are the same
building, unprotected in one year and protected in the
following, and it must be admitted that a much smaller
degree of success would warrant any reasonable person in
giving the plan a thorough trial. All the same, I know well
that what is comparatively easy in a small Italian cottage
will be by no means so easy in our rambling Indian bunga-
lows, with several doors to every room, and none of the
joinery made to fit with any approach to accuracy. House

Fig. 37a. — Railway servant's cottage in the Roman Campagna, protected
against the entry of Mosquitoes by Professor Celli's method.

building is as excellent in Italy as it is bad in India, and the
careless finishing of most of our bungalows, especially of
the roofs, would make it extremely dii^Hcult to keep out so
importunate an intruder as the Indian Anopheles without
extensive structural alterations. Then, too, I can easily
picture how systematically Gunga Din and Nabbi Baksh
would " forget " and carefully prop open the spring doors to
save the trouble of shouldering them open ; but that might
be got over by a little drilhng. In the type of bungalow

228 GNATS OR MOSQUITOES — CHAPTER VIII

common in the Punjab, especially those which are what we
speak of as " pacca " built, there w^ould be no great diffi-
culty, but in the ordinary thatched houses of the North -
West and lower India, it would tax the best ingenuity, and
is practically impossible, unless we can substitute something-
better for the abominable "ceiling-cloth."

These vile dirt and vermin traps are, however, I am glad
to say, rapidly disappearing. For the Burmese bungalow,
it is difficult to suggest any plan of applying the method
short of reconstruction.

I found that the cost of protecting the last bungalow I
occupied would be, at present prices in India, about 300
rupees (i?20), but Celli tells me that suitable wire gauze
(about 12 strands to the inch) of American manufacture is
obtainable in Italy at much cheaper rates than we have to
pay in India, and probably the price would lessen if there
were a sufficient demand. Presuming the house to be adapt-
able, I believe the simplest plan will be to completely
enclose the verandahs with gauze, providing each with a
single set of doors. All the other openings would of course
require fixed gauze frames, and doors opening elsewhere
than on the verandahs would require to be separately fitted
up. Possibly the bath-room doors might be left unaltered,
their use being prohibited between dawn and 8 a.m., but it
would certainly be preferable to fit them up also. The
first cost would be of course considerable, but a considerable
annual expenditure for " chiks " would be done away with.

In this, however, as in the case of every essential of
health and comfort, the main obstacle as far as officials are
concerned, is that we are so constantly moved about from
one station to another, that any attempts at improvements
of a permanent character are ruinous, as the houses are
rented from native landlords, who often are with difficulty
persuaded to keep the roof over our heads weather-tight.
In the case of permanent residents, however, I believe in
the long run a distinct saving would result. " Chiks " are at
the best a lame defence against either flies or mosquitoes,
but if nothing better be obtainable must be made the best
of, and this means that we must absolutely reverse our

CONDITIONS INFLUENCING PREVALENCE 229

present customs as to the times of openino; and closing
them, as we now plan these so as to keep out, not mos-
quitoes, but flies, the habits of which are the very antithesis
of those of mosquitoes.

At present they are kept more or less carefully closed
during the day, and are rolled up at night. From an hour
or two after sunrise till sunset, it little matters as regards
Mosquitoes what is done with them, as the insects settle
themselves down for their day's rest as soon as the sun is
well up, and will not leave the shelter they have chosen
unless they are obliged ; but at night every aperture must
be carefully guarded and especially the period from 4 to 6
a.m. In the early evening dusk it is better to have every-
thing opened, as just then Mosquitoes, if they can, leave
the house to obtain vegetable food, but for all the rest of
the darkness and dusk the more thoroughly a house can be
kept Mosquito-proof the better.

It will be a long time, however, before Mosquito-proof
houses become at all common in India, and for the present
most of us will have to be content with Mosquito curtains.
A compromise between the two would, however, be well
within most people's resources, and far better than the
stuffy old nets, which, in spite of my firm conviction of
their necessity, are well nigh intolerable during " a break in
the rains." For this reason no plan will ever be generally
adopted in India which cannot be combined with the use
of the punkah, and with the old-fashioned forms of the
latter this is most difficult, though the handy little electric
fans can be worked in almost any position. Unfortunately,
power of this sort can be obtained in very few places, but
it might be possible to employ the punkah coolies in work-
ing a small dynamo instead of pulling the rope ; though the
difiiculty of getting any small defect remedied would still
remain in all but the largest stations. A settled resident
would find it simplest to make his sleeping apartment insect
proof, but the vagabond ofdcial requires some contrivance
that he can carry about with him with his other furniture
when " transferred." For this I would suggest a portable
miniature room formed of light wooden frames filled with

230 GNATS OR MOSQUITOES — CHAPTER VIII

wire gauze, and hinged together into panels of convenient
size, two of which should, however, be filled with thin
planking, so as to protect the rest when packed up for
transport. About 8 ft. square would be a convenient size,
as to plan, but. to carry a punkah, the panels would have to
be at least 10 ft. high, and even then the bed must be
made as low as possible. Some contrivance to prevent it
being shifted by the pull of the punkah would of course be
required, and probably the best plan would be to secure
the solid panels to the wall through which it is usually
pulled ; but details may best be left to individual ingenuity.
Whether the room be temporary or permanent, however, it
is absolutely essential that all furniture and hangings should
be absolutely excluded, excepting the beds, chair and small
table, so that I fear the plan will hardly meet the aesthetic
tastes of the " Memsahib."

"When " in camp," or on other temporary absences, we
must still be content with the old curtains, and there is a
right and a wrong way of hanging these, the latter of which
is almost universally adopted, because it looks neater to
spread them over the iron or wooden frame, and thereby
leave gaps at the corners, which cannot be securely tucked
in on account of the poles, than to suspend them inside the
latter, so that these cannot obstruct complete security. The
weak point of the curtains, however, is that unless the bed
be very large and one can contrive to keep in the middle
of it, one is sure to be bitten through them, owing to the
limbs coming in contact with the net.

Many people have an idea that the punkah is an efficient
protection against Mosquitoes, but this is entirely erroneous,
and I have watched one filling herself, quite undisturbed by
the towel pinned to the punkah-frill, which flicked my knee
within a couple of inches of the spot she had selected, at
every swing ; but those who blame us for sacrificing security
to the comfort of a long-roped punkah, or for not swathing
our limbs in thick woollen putties with the thermometer at
98"" in the shade, can know nothing of the climatic condi-
tions under which we must contrive to exist, and forget that
a restless night is a bad preparation for a tropical day, and

CONDITIONS INFLUENCING PREVALENCE 231

that most of us have other concerns to attend to, so that it
is unreasonable to expect us to devote our lives to the
prophylaxis of a single tropical disease ; though this is no
argument against the adoption of such reasonable precau-
tions as may be practicable. In the " hot weather" of the
Punjab and North- West, though Culices, owing to our own
carelessness, are usually in swarms, Anopheletes are so
scarce that there is no real need of special precautions ; but
as soon as the rains break curtains are indispensable to safety.

This brings me to the question of clothing, which
may be made to afford considerable protection without
being otherwise unsuitable to the climate. I do not think
our Indian species of Anopheles ever bite during the day,
and the same seems to be true for West Africa, so that
practically it is only during the evening that there is much
chance of being bitten while awake, and not even then
while moving about. The really dangerous time is during
and after dinner, and the favourite point of attack the ankles,
which are bitten through the thin socks that are usually
worn. Unfortunately of late years, the sensible and cleanly
custom of dressing for dinner in white drill jacket and pants
is yielding to an absurd desire to assimilate our customs
to those of Europe ; though how men of refined habits can
reconcile themselves to wearing a garment which is nightly
saturated with perspiration till it is worn out, passes my
comprehension, and those who will sacrifice so, much are
hardly likely to adopt trowser-straps, which would not be
correct "form " in the black habiliments whereby the Indian
Memsahib loves to keep green the ideals of the old country,
though they formed a customary adjunct of the supplanted
white costume, and would, applied to either, afford complete
protection.

It is dangerous for a mere man to venture on the
subject of ladies' dress, and in any case absolutely bootless ;
but as fashion ordains that their evening costumes must
sweep the ground, they can obviously protect themselves
without straps, if they happen to think of it, and care to
deny us the pleasure of glancing at the neat contour of a
pointed shoe with an impossible heel; but the " low neck,"

232 GNATS OB MOSQUITOES — CHAPTER VIII

which is murderous enough in Europe, is clearly no better
in India, as it leaves a large surface unprotected just at
the time when Mosquitoes are most persistent in their
attacks. As regards children, mothers should deny them-
selves the pleasure of exhibiting their chubby limbs, at any
rate in the evening ; but as little boys really look " sweet "
in absolutely correct "sailor suits," with nautically cut but
long pantaloons, there is no real difficulty in their case.
The poor little girls, however, are again in a different case,
as long stockings thick enough to be any good would
cause unbearable irritation of the skin, and probably afford
staphylococci the opportunity to start a most debilitating
crop of boils ; and I fear it is no use pointing out that the
"pantalettes" portrayed in Leach's drawings, in which
their grandmothers disported themseves, would solve the
difficulty ; unless indeed we can contrive to initiate a sort
of Chippendale reaction for infantile millinery in their
favour.

As a rule the Anglo-Indian has no alternative in the
selection of his bungalow and must accept that occupied by
his predecessor, or camp out. The better class of native
gentlemen are beginning to see the superiority of the
European plan of villa to such an extent that it is often
difficult to find accommodation in what were once purely
European quarters. So much is this the case that the
provision of quarters built by Government is seriously
mooted ; but I should be sorry to serve in India, even if
housed in them gratuitously instead of compelled to rent
them, as, owing to the high standard of work demanded by
the department, the largest proportion of the money allotted
for their construction will be required to secure technically
perfect masonry, and a cramped heat trap, with no proper
plinth or adequate through ventilation, but of beautifully
pointed red brick, will necessarily result. There is a rule,
perfectly reasonable, were it not rigid, that in such cases
the rent must bear a certain proportion to the outlay, and
that the former must be no more than a certain percentage
of the average income of the occupants ; and probably a
sufficiently healthy and commodious building, durable

CONDITIONS INFLUENCINC; PREVALENCE 233

enough for practical purposes, might be built for the sum,
but the executive engineer would have no choice than to
reject it if tendered him by a contractor.

There can be no doubt, however, that the provision of
really healthy quarters for officials, as commodious as the
large but ill-planned bungalows now obtainable, would be
really a most remunerative investment even if the direct
returns fell short of the requirements of the Accountant-
General. Every station has its " Fever Hall, " most
several, with generally no alternative accommodation ; and
in one year I have seen three consecutive officers invalided
from a fever trap of this sort, none of whom had the least
desire to take leave, while the sum disbursed to them as
ineffective pay would have easily rebuilt the pestilential
hovel. Where, however, any choice can be exercised, a
house with an upper story should be selected, as Mosquitoes
never fly high, and even this amount of elevation affords
considerable protection.

The writer once built, on the flat concrete roof of the
house he occupied, a large grass hut, which was used by
himself and family right through the rainy season as a
sleeping apartment. Although fitted with punkahs, they
were never used, as while our neighbours were sweltering
through the night under them in the steamy, but sun-baked
houses, we were able to obtain refreshing rest in spite of
our Mosquito nets. This was long before the etiology of
malaria had been solved, and the fact that we not only
escaped fever but were all singularly healthy was a matter
of constant wonder to them, as to sleep, in such a situation
in the wet weather was looked upon as a certain invitation
to fever. As a matter of fact, however, a hut of this sort,
or a verandah, if deep enough to shelter from the drift of a
shower, is not a bit damper than the inside of a house, and
the only reason why to sleep in the latter is so notoriously
dangerous is that they swarm with Mosquitoes more thickly
than either within the house or in the open.

Given adequate protection from the bites of infected
Mosquitoes, it is far more healthy to sleep in such situations,
and so gain refreshing sleep with comparative coolness and

234 GNATS OR MOSQUITOES — CHAPTER VIII

the freest ventilation. Makeshift erections of this sort cost
so httle in India that it is well worth making one, even if
one can use it during but a short stay ; and I believe that
a portable Mosquito room of the sort suggested, sheltered
in a hut of this sort built on the roof where this is terraced,
is the most practical plan of availing oneself of our new
knowledge that I can offer. On several occasions we dined in
our airy quarters, but though singularly free from Mosquitoes,
the hghts attracted such numbers of other members of the
insect tribes as to ill compensate for the gain in coolness.

Mosquitoes exhibit a repugnance to almost all strong-
scented bodies, and almost every country has its own specific
for warding off their attacks by their use. I have tried
several of these reputed culicifuges and find that the most
efficient are bodies such as kerosine and turpentine, which
are as objectionable to the human nose as to the gnats.
Celli (C. M., p. 207) gives the results of a long series of
experiments with bodies of this kind, which clearly show
that none of them are sufficiently powerful to be of any
practical use for killing Mosquitoes, and I regard their
value when employed about the person with the view of
keeping them off as so small that it would be waste of space
to further consider the question.

Relied on in place of a Mosquito net, they are worse than
useless, as while they enable the user to get to sleep, their
action is so evanescent that they leave him unprotected as
soon as he has fairly landed in the land of dreams. For
practical prophylaxis, the only use that can be made of them
is to sprinkle them near persons sitting at table, and pro-
bably it very little matters what scent is used for the
purpose, the ordinary cosmetics being as good as any.

A good deal of protection, on the other hand, is un-
doubtedly afforded by smearing the skin with greasy sub-
stances, such as vaseline or oil, and I fancy that most of
the culicifuge salves and ointments that are sold owe their
value not to the scents, but to the fatty matter they contain.
As a remedy for the smarting of gnat bites, I find nothing
better than hazeline, and a mixture of eau de Cologne and
water also affords considerable relief. As a rule, the direct

CONDITIONS INFLUENCING PREVALENCE 235

effect of the bites are too trifling to require treatment
of any kind, but in some people they swell so much and
cause irritation for so long a period that it might be worth
while to try the use of weak tincture of iodine, which has
recently been stated to abort the effects of the bites.

One more point requires notice, viz., the suggestion
thrown out by Drs. Stephens and Christophers that the
European tropical resident should seek immunity from
malaria by isolating himself from the native.

The proposal to isolate the healthy instead of the sick
is a novel one, and I must confess that it is difficult to
understand how it is to be carried out, especially as in all
such places the climate makes it impossible for the
European to undertake his own domestic work, even if he
possessed the leisure to do so, which is rarely the case, as
his time is generally fully occupied in some occupation of a
supervisory character. European and native alike reside
in these insalubrious localities for the transaction of the
everyday business of life, and it is useless to ask either to
subordinate every business and social concern to the avoid-
ance of one only of the numerous tropical risks that he
has decided to brave. The native doubtless believes he
could spare the company of the European, but the latter is
absolutely dependent on the native for every necessity of
comfortable existence, and would be in a miserable plight
if deprived of his servants for even a few hours in the day.
In India there is little to choose in degree of malarial
infection between the two races, so that logically most of
us would have to throw in our lot with our " Aryan
brethren," and few but the last-joined "griffs" would be
left to inhabit the sanitary oasis. Apart from this, in India
at any rate, the declared acceptance by Government of the
duty of bearing what Kipling calls "the White man's
burden " would put the adoption of any policy, sanitary or
otherwise, intended for the sole benefit of Europeans,
entirely out of court, and I cannot beheve that the
authorities of any of our other tropical dependencies would
care to avow a different policy.

It is impossible to enter into a really exhaustive treat-

236 GNATS OR MOSQUITOES — CHAPTER VIII

ment of the prophylaxis of malaria within the limits of a
chapter like the present, or to notice all that has been
written on the subject, even within the last twelve months,
but it is hoped that the above sketch will suffice to give a
general idea of the present position of the subject. In
bringing it to a close, the writer trusts he may be absolved
from the charge of holding extravagant views as to what
can or should be done, and to avoid all misconception, would
repeat that while he neither thinks it possible to exter-
minate Mosquitoes, or to do away with all malarial disease,
he is convinced that, even with our present knowledge, it
would be practicable to enormously diminish the number
of cases in any given limited area taken in hand ; and
further, though it is of course impossible to ensure safetj^,
that anyone who will avail himself of a few by no means
onerous precautions may greatly diminish his chances of
becoming infected.

Since the above went to press, we have received good
accounts of the practical work of the last expedition sent
out by the Liverpool School of Tropical Medicine to
West Africa ; and I have also received an interesting letter
from Dr. Yale Massey, from Portuguese West Africa, in
which he attributes a diminution of fever in his station to
the filling up of pits left by building operations, and to the
liberal use of quinine among the children.

237

CHAPTEE IX.

On the Distribution of the Culicidae.

The Culicidce are a truly cosmopolitan family, and may
be found everywhere from the tropics to well within the
polar circle. Their commonness indeed depends rather on
the state of civilisation of a region than upon its geographi-
cal position. In new countries, in regions where the
severity of the climate is such that they cannot support a
sufficient population to undertake the complete drainage of
the area, and amongst people whose civilisation, however
old, has not reached the stage of "tidiness" and order,
gnats and Mosquitoes will be found to be numerous and
troublesome. As examples of this may be mentioned the
cases of Lapland and the north-western portions of British
America, where, during the short summer, they constitute
a veritable pest ; so that, in the former country, the
nomadic inhabitants are obliged to frequently change their
grazing grounds to enable themselves and their herds to
escape from their insect tormentors ; while in Manitoba,
it is not uncommon for horses and cattle to be "stampeded"
from the unbearable pertinacity of the indigenous species.
In tropical countries again, the commonness of Mosquitoes
is due far more to the difBculties of securing efficient
surface drainage, and to the careless domestic habits of the
people, than to any special favourableness of the climate.
On the other hand, in countries such as England, Northern
France and Germany, where centuries of human inhabita-
tion have perfected drainage, and domestic neatness has
reached almost to the position of a religious duty, gnats
are so rare that, when they appear in any numbers, thej^
are commonly suspected of being a recent importation.

Even in Holland, where the nature of the country
appears to be entirely in their favour, they are by no

'238 GNATS OR MOSQUITOES— CHAPTER IX

means so common as in many apparently less favourable
localities ; for in Holland drainage is a primary necessity
of occupation, and the domestic neatness of the people is
proverbial.

With the exception of a few rarely visited islands,
there are few parts of the world whence the Gulicidm have
not been recorded, and indeed their constant association
with man makes it almost impossible for any country that
is much frequented to long escape their importation, as
apart from their being carried in the larval state in ships'
tanks, their habit of hybernation, and of harbouring in
draperies while in that condition, makes their introduction
a very easy matter.

Mr. R. M'Lachlan, in his notes on the insects of Captain
Fielden's Arctic Expedition, mentions a species of Cidex,
which, he says, may be C. caspius, Pallas, as identified by
Curtis in the insects of Ross's Voyage (p. 66). Schiodle
identifies the same species with C. nigripes, Zett. The
latter, according to Staegel, occurs also in Greenland, and
is the same as C. pipiens O. Fabricius, nee Linne {" Fauna
Greenland," p. 201).

The late Professor H. N. Moseley, during the "Chal-
lenger " Expedition, described a species of " wingless
Culex " from Kerguelen's Island (" Proc. Linn. Soc," XII.,
p. 578), but it is almost needless to say that the identifica-
tion was a wrong one, as a wingless insect would not be a
gnat.

In the beginning of the century, the " Nouveau Dic-
tionnaire d'Histoire Naturelle," Tome VIII. , Paris, 1817,
states that only some fifteen species, mostly European, of
the family were known, but since then the number has
steadily increased.

Schiner, " Reise der Novara," notes that 132 species of
the family had been described (up to 1868). Of these 80
are European, 61 American, 21 Asiatic, 10 African, and 9
Australian, with one of unknown origin.

In 1889, Skuse estimated the number of described species
at 160, including no less than 21 new Australian forms
included in his paper ("S. A. C," p. 1,717); and subsequent

ON THE DISTRIBUTION OF THE CULICID^ 239

additions and the unearthing of many descriptions which
had not been accessible to him, brought the number of
descriptions collated in the first edition of this handbook to
a total of 242 species, of which 18 belong to the genus
Megarhina, 30 to Anopheles, 3 to Psorophora, 3 to
Sabethes, 160 to Culex, 13 to Mdes, 12 to Corethra, and
3 to Mochlonyx.

Of these 72 were European, out of which 24 were
recorded from England ; 20 from continental, and 29 from
the islands of Asia ; 41 from North, and 36 from South
America ; and 29 Australian. No better illustration of the
small attention that had hitherto been devoted to the group
can be given than the fact that but one species had been
originally described from India, and that but four were
recorded as having been found within its limits, putting aside
the species and records appearing for the first time. It was
obvious on comparing the various original descriptions then
brought together for the first time, that on the one hand,
many of the descriptions were so inadequate that they might
easily correspond in the few particulars mentioned to a
whole series of perfectly distinct species, while on the other,
it was equally clear that many must be mere synonyms.
It is, however, most dangerous to dabble in questions of
synonymy unless one can compare the actual types, or at
the least has available a large collection of locally-taken
specimens, and in reality, no collection of the family worthy
of the name existed ; and I therefore, with one or two
exceptions, confined myself to a guarded acceptance of the
efforts of others in this direction, most of which I may
remark, would, it appears, have better been left unnoticed,
as they have generally turned out to be wide of the mark,
and to have merely added to the existing confusion.

While I strongly suspected that many of the names
enumerated in the systematic portion of the book were
nothing more than redescriptions of species already known,
it was equally obvious that the determination of the gnat
fauna of vast regions of the globe was practically untouched,
and that, as a necessary corollary, a large number of
unknown species must remain to be discovered and des-

240 GNATS OR MOSQUITOES — CHAPTER IX

cribed. During the year that has elapsed since the appear-
ance of the first edition, the attention that has been drawn
to the family by the establishment of their instrumentality
in the transmission of at least three of the most serious of
the maladies peculiar to tropical regions has drawn minute
attention to the group in all parts of the world, and the
timely activity of the authorities of the British Museum
and of our Royal Society, has resulted in the acquisition by
the Museum of perhaps the largest collection of insects of
any one family that has ever been brought together, and
this has been further increased by the loan of collections
sent for comparison by naturalists from all parts of the
world. The examination of this enormous mass of material
has been entrusted to Mr. F. V. Theobald, whose reputation
as one of our most thorough dipterologists cannot fail to
be enhanced by the encyclopaedic Monograph, now in the
press, which has resulted from his labours.

As the result of his examination of the family, Mr.
Theobald summarises the outcome of his labours as
follows : —

" Total previously described good species, 164.

" Species recognised and redescribed, 116.

" Other descriptions, probably invalid owing to their
shortness and the apparent absence of types, 25.

" Described as distinct, but found to be synonymous
with other species, 80.

" New species described, 186.

" Therefore the number identified and redescribed and
the new species described in these volumes number 252. If
to this we add the 48 species not yet identified, we get the
total known number of GiilicidcB, 300."

It must not, however, be supposed that this represents
more than the total of species examined at the time of the
earlier sheets of his monograph going to press, nor must the
reader expect to find it correspond exactly to the number
given in the present edition, as the simultaneous passage of
both our tasks through the press has been continually
retarded by the receipt of fresh material, which not unfre-
quently has necessitated the reconsideration of the position

ON TH?: DISTRIBUTION OF THE CULICID^ '241

of an entire group. I understand from Mr. Theobald,
that ah'eady some thirty or forty more new forms have
turned up, which will necessitate the early issue of a con-
siderable appendix to the two heavy volumes already in
the press ; so that in place of the 242 species, good, bad,
and indifferent, of the first edition, we have at least 800,
whose validity is hardly likely to be questioned. More-
over, as there remain large areas from which no collections
have been received, a total of 500 species as the actual total
is certainly an underestimate, and I should not be surprised
were it found to rival that of the butterflies.

The remarkably wide distribution attained by certain
species such as Steg. fasciata, Fabr, (C. tceniata, Meig.)
has been found to be further illustrated in a most striking
way by those of many other species, certain of which have
reached the British Museum from the most widely distant
parts of the globe. A notable instance of this is G. cantans,
Meig., a well-known European species, which is now known
to be common in Northern America, from the " States" to
Manitoba, though this would hardly prepare one to find it
in the Nehilgerri hills of Southern India, from which un-
expected habitat specimens were sent me by Dr. Price,
of Conoor. Ano